Abstract
Student Competition Finalists
NR2B-N-METHYL-D-ASPARTATE RECEPTORS CONTRIBUTE TO NETWORK ASYNCHRONY AND LOSS OF LONG-TERM POTENTIATION FOLLOWING MILD MECHANICAL INJURY IN VITRO
Scott Ventre, University of Pennsylvania
Pallab Singh, Ph.D., University of Pennsylvania
David F. Meaney, Ph.D., University of Pennsylvania
To understand how mechanical injury affects neural circuits, we 1) built a multi-scale stochastic integrate-and-fire model of network activity, 2) simulated an injured network, 3) predicted an important role for the NR2B-NMDAR in mediating network asynchrony, and 4) tested our predictions using in vitro stretch injury.
Computational model: Population of pyramidal-like neurons were assembled into a network following the topology found in vitro cultures. Dendritic spines were populated with AMPARs and NR2A- and NR2B-NMDA receptors. Synaptic transmission was simulated by vesicular release of glutamate onto the spine head, and the activation of these receptors was modeled using a stochastic diffusion-reaction scheme. The resulting current, summed at the soma, was used to update the cell membrane potential. Injury was simulated by larger glutamate release and partial loss of NMDAR Mg2+ block, consistent with in vitro and in vivo observations.
In vitro experiments: Primary cortical neurons were plated on silastic membrane. Mature cultures were bulk loaded with fluo4-AM, and the calcium activity of the network was imaged pre- and post-stretch. The relative contribution of NR2A- and NR2B-receptors in altering network activity, functional connectivity and LTP potential of the post-injured network were assessed using subtype specific antagonists.
Since calcium is a crucial secondary messenger in neurons and serves as a proxy for underlying electrical activity, we first profiled the source of calcium and the temporal pattern of calcium dynamics of a spontaneously oscillating network. We found that, in mature cultures, the main sources of calcium during spontaneous activity are NR2A-NMDARs and VGCCs. In contrast, acutely following dynamic stretch injury, NR2B containing NMDARs become a significant source of calcium, likely due to partial loss of its native Mg2+ block.
To investigate the effects of enhanced NR2B activity on network dynamics, we built a multi-scale model of network activity (spines -> dendrites -> soma -> network) and simulated activity patterns of a population of neurons and matched to those of an in vitro culture (DIV 18–20). Next, we simulated the activity of an injured network and found that relief of NR2B Mg2+ block at the spine-level results in asynchrony at the network-level. Our simulations also predicted that the asynchronous activity in the simulated-injured region can propagate into adjacent regions in a distance dependent manner, influenced by the topology and connectivity of the network.
We conducted in vitro stretch injury experiments to validate our in silico predictions and found that following injury, 1) there is greater variability in the amplitude of calcium transients for a given neuron, 2) there is loss in synchrony and decline in functional connectivity in the injured region, 3) synchrony is rescued by antagonizing NR2B-containing NMDARs, and 4) disruptions of network activity in the mechanical penumbra are negligible beyond 700-μm.
Functionally, synaptic calcium asynchrony limits LTP potential in the injured region. However, NR2B antagonism during LTP induction allows a fraction of injured neurons to undergo chemical LTP as measured by persistent increase in the amplitude of Ca2+ transients.
Despite its enormous incidence, mild TBI is not well understood. One aspect that needs more definition is how mechanical energy disrupts neural circuitry, in the absence of overt cell death. In this study, we merged information from different perspectives (in silico and in vitro) and across different length scales (spines to networks) to uncover an important role of the NR2B-NMDAR in mediating network asynchrony. Traditionally, NR2B antagonism has been used to blunt pro-apoptotic pathways, but we show that in mild injuries, NR2B receptors may be an important target for reverting the injured network to its pre-injury state and rescue plasticity in an injured network.
We thank the contribution of Patricia Slobodnick (cell culture), and YungChia Chen and David Gullotti for helpful discussions. This work was funded by National Institues of Health and the Department of Defense (NIH RO1 NS35712, HD41699, and PO1 NS056202, W911NF-10-1-0526).
synchrony, NR2B-NMDA receptor, network activity
ADMINISTRATION OF A VEGF-A GENE THERAPY RESULTS IN ANGIOGENESIS, NEUROPROTECTION, FUNCTIONAL RECOVERY AND DECREASED NEUROPATHIC PAIN WHEN ADMINISTERED IN A DELAYED FASHION FOLLOWING SPINAL CORD INJURY
Yang Liu, MD, University Health Network, Dr. Fehlings' Lab
Kaye Spratt, PhD, Sangamo Biosciences Inc.
Gary Lee, PhD, Sangamo Biosciences Inc.
Dale Ando, MD, Sangamo Biosciences Inc.
Richard Surosky, PhD, Sangamo Biosciences Inc.
Martin Giedlin, PhD, Sangamo Biosciences Inc.
Michael G. Fehlings, MD, PhD, FRCSC, FACS, Toronto Western Research Institute
Traumatic spinal cord injury (SCI) results in disruption to surrounding tissues and the spinal microvasculature. In the current study, we utilized vascular endothelial growth factor (VEGF) – a pleiotropic factor known for vascular development and repair, and more recently for it's neuroprotective properties – as a potential therapy for SCI.
Briefly, female Wistar rats – under cyclosporin immunosuppression – were divided into four experimental groups: 1) Sham – laminectomy only, 2) Injured control – injury, no injections, 3) AdV-ZFP-VEGF, and 4) AdV-eGFP. AdV-ZFP-VEGF is an adenovirus that generates a bio-engineered zinc-finger transcription factor that promotes endogenous VEGF-A expression. AdV-eGFP was used as a control and reporter virus. Groups 2–4 received a 35g clip-compression injury, and Groups 3 and 4 were administered a delayed (24 hour post-injury) dosage (5×108 PFU) of AdV-ZFP-VEGF (treatment) or AdV-eGFP (control) by intraspinal injection, respectively. For molecular and vascular analysis (qRT-PCR, Western Blot, Immunohistochemistry), tissues were extracted at 3, 5 or 10 days post-SCI. For behavioural experiments (BBB, Catwalk, Neuropathic Pain), animals were studied for 8 consecutive weeks, and tissue was collected at 8 weeks post-SCI.
qRT-PCR analysis of tissue at 5 days post-injury showed a significant increase in VEGF-A mRNA in animals treated with AdV-ZFP-VEGF (p<0.01). Similarly, Western Blot analysis at 10 days post-SCI showed a significant increase in VEGF-A protein expression in animals receiving AdV-ZFP-VEGF (p<0.02). Furthermore, analysis of NF200 protein, TUNEL, and RECA-1 indicated that AdV-ZFP-VEGF treatment results in significant axonal preservation (p<0.05), reduction in cell death (p<0.01), and increase in the number of spinal cord blood vessels (p<0.01), respectively. In addition to the molecular results, functional locomotor analysis – using Catwalk – showed that AdV-ZFP-VEGF treatment dramatically improves hind-limb weight support (p<0.005), when compared to injured control animals. Finally, AdV-ZFP-VEGF administration provided a significant reduction in neuropathic pain, both at, and below the level of injury (p<0.01).
Although vascular repair mechanisms are initiated following SCI, endogenous efforts are insufficient to fully repair the damage. Therefore, therapies designed to reduce vascular damage and promote vascular regeneration may be beneficial. Here, we have shown that administration of AdV-ZFP-VEGF results in increased axon sparing, cell survival and number of vessels following SCI. These findings are further supported by our neurobehavioural data, which suggest AdV-ZFP-VEGF administration improves hind-limb weight support, as well as significantly attenuates neuropathic pain. In contrast to other VEGF therapies, AdV-ZFP-VEGF induces the endogenous production of VEGF-A, which results in the synthesis of multiple VEGF isoforms – this is a critical factor for proper vascular development and repair. Overall, the results of this study indicate that AdV-ZFP-VEGF administration can be delivered in a clinically relevant time-window following SCI (24 hours) and provide significant molecular and neurobehavioural benefits.
We wish to thank Ramak Khosravi, Michelle Legasto, Behzad Azad, and Spyros Karadimas. Also, thank you to Sangamo Biosciences Inc. for their financial support.
SCI, VEGF, gene therapy, neuroprotection
PERSISTENT INFLAMMATION IN THE CORPUS CALLOSUM FOR MANY YEARS FOLLOWING A SINGLE TBI IN HUMANS
Janice Stewart, B.S, Greater Glasgow and Clyde NHS
Douglas Smith, MD, The University of Pennsylvania
William Stewart, MBChB, PhD, Greater Glasgow and Clyde NHS/University of Glasgow
A single traumatic brain injury (TBI) is associated with hallmark Alzheimer pathologies in a proportion of patients surviving greater than a year after injury. However, little is known about the processes driving this pathology, in particular the potential role of persistent inflammation post TBI.
The inflammatory markers CR3/43 and CD68 were examined in sections of the corpus callosum derived from the TBI archive of the Department of Neuropathology, Glasgow. Cases were subdivided into survival cohorts as acute (10 hrs-2 wks; n=21), subacute (2 wks-1 yr; n=14) or long-term (1–47 years; n=39), and compared to age-matched uninjured controls (n=47). Indices of axonal injury and corpus callosum thickness were also assessed.
In this series, over one third of TBI cases with moderate/long-term survival displayed extensive, CR3/43- or CD68-positive, densely packed, reactive amoeboid microglia. No such pathology was identified in controls or in acute TBI cases. These amoeboid cells emerged months after injury, were present up to 16 years post-trauma and were associated with ongoing axonal degeneration (APP-immunoreactive axonal profiles) and, with survival >1yr post-TBI, a marked reduction in corpus callosum thickness (p<0.0001).
These data demonstrate enduring inflammatory changes in white matter associated with ongoing axonal degeneration even decades after TBI. These findings may provide support for persistent inflammation as a contributor to chronic progressive neurodegeneration following TBI.
Funding Source: NIH Grants R01-NS-038104, P01-NS-056202 (DHS).
TBI, inflammation, neurodegeneration, axonal-pathology, Alzheimer's
EFFECTS OF PENTOBARBITAL ON CEREBRAL MICRODIALYSIS IN SEVERE TRAUMATIC BRAIN INJURY
David L. McArthur, Ph.D, M.P.H., Department of Neurosurgery, UCLA
Paul Vespa, MD, FAAN, FACN, University of California, Los Angeles
Pentobarbital suppresses brain metabolism and metabolic demand, resulting in reduced cerebral blood flow and volume, which have beneficial effects on ICP and perfusion. The objective was to determine the effectiveness of pentobarbital on suppressing metabolism by analyzing microdialysis levels of glucose and lactate/pyruvate ratio (LPR) before and during pentobarbital infusion.
We retrospectively analyzed data from 12 patients with severe traumatic brain injury (TBI) in the first 7 days post-injury who were treated with pentobarbital for refractory ICP. Pentobarbital was rapidly titrated to induce burst-suppression coma. The microdialysis catheter was inserted in normal appearing frontal lobe white matter. Hourly samples of glucose, lactate, and pyruvate were taken and the LPR was calculated. We analyzed the cerebral microdialysis values of glucose and LPR for the 24 hours before pentobarbital infusion and the following 24 hours while on a continuous pentobarbital infusion. We performed a linear mixed model for the longitudinal data which compared each individual patient's values before infusion to during infusion to determine the time, treatment, and time×treatment effects. We hypothesized that pentobarbital induced burst-suppression would result in markedly higher glucose levels and a lower LPR.
The mean glucose value the 24 hours before pentobarbital infusion was 0.95±0.78 mmol/L (interquartile range 0.27 – 1.41) and the mean glucose during pentobarbital infusion was 1.29±0.93 mmol/L (interquartile range 0.49 – 1.99). The mixed model showed a significant effect on glucose level (time effect p<0.001, treatment effect p=0.010, time×treatment interaction p<0.001). The odds ratio for the time×treatment interaction was 1.05, meaning that for every hour of pentobarbital infusion we saw a 5% increase in the glucose concentration, independent of the normal variation we see over time. The mean LPR the 24 hours before pentobarbital infusion was 34.49±18.16 (interquartile range 23.30 – 40.44) and the mean LPR during pentobarbital infusion was 31.91±12.82 (interquartile range 21.91 – 37.29). The mixed model showed no difference on LPR (time effect p=0.299, treatment effect p=0.775, time×treatment effect p=0.551).
Pentobarbital is commonly used to treat refractory ICP following TBI by reducing cerebral metabolic demand and blood flow. In 12 patients we saw that pentobarbital infusion effectively increased the microdialysis glucose within 24 hours, but did not have a significant effect on lowering the LPR. These results suggest that more research needs to be done on the comparative efficacy of pentobarbital and other sedatives on suppressing cerebral metabolism.
The authors would like to acknowledge the staff and faculty of the UCLA Brain Injury Research Center (supported by NS-058489) for their contributions and work.
pentobarbital; lactate/pyruvate ratio; microdialysis; TBI
TRANSPLANTATION OF GENETICALLY-MODIFIED NEURAL PROGENITOR CELLS RELEASING A CHIMERIC NEUROTROPHIN LEADS TO INCREASED CELL SURVIVAL AND TARGETED MIGRATION AFTER TRAUMATIC BRAIN INJURY IN THE RAT
Pantelis Tsoulfas, MD, University of Miami Miller School of Medicine
Ofelia Furones-Alonso, BS, University of Miami Miller School of Medicine
Helen M. Bramlett, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton. Dietrich, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Cellular transplantation is a viable option for alleviating some of the hallmark deficits arising after traumatic brain injury (TBI). Cortical transplantation of multipotent progenitors has the potential to improve functional outcome in animal models of TBI. Furthermore, mature neurotrophins contribute to neuronal survival, differentiation, cognitive function, and neuroprotection after trauma.
Through the exchange of seven amino acids on a neurotrophin-3 (NT-3) backbone, we generated a chimeric neurotrophin with multiple neurotrophic specificities (MNTS1). This protein binds all tropomyosin-receptor-kinase (Trk) receptors and seems to support neuronal survival more so than any single neurotrophin individually. Using a lentivirus, we transduced neural progenitor cells (NPCs) isolated from embryonic rat cortical tissue with the MNTS1 construct and GFP. Adult rats received a moderate fluid percussion injury (FPI) or sham surgery. One week later, TBI and sham groups were transplanted with either 1) 400,000 NPCs releasing MNTS1 (NPCs+MNTS1/GFP), 2) 400,000 NPCs expressing GFP alone (NPCs+GFP), or 3) saline (Veh). Transplants were made 3.5mm below the surface of the dura into the pericontusional lesion. This transplant location was deemed conducive to white matter tract migration. Four weeks after transplantation, animals were tested for hippocampal-dependent memory capacity in the Morris water maze and then sacrificed for immunohistochemical analysis.
Five weeks post transplantation (six weeks post surgery), we found that only TBI animals transplanted with MNTS1-infected NPCs (TBI/MNTS1 group) had significantly greater exogenous cell survival and targeted migration compared to TBI animals transplanted with GFP-alone infected NPCs (TBI/GFP group). NPCs+MNTS1 were observed from 3.6 to 4.5mm posterior to bregma, 5.0 to 6.5mm lateral to midline, and approximately 4.5mm ventral from the needle tract, with deeper processes extending into the ipsilateral hippocampus and throughout the cortical contusion. The majority of these cells appeared to migrate via the external capsule to the contusion site induced by moderate FPI. In contrast, NPCs+GFP group had virtually no migration outside the needle tract and very little exogenous cell survival. Consistent with previous findings, only TBI animals transplanted with NPCs+MNTS1 had robust exogenous cell migration. Sham/MNTS1 animals had negligible migration with cells remaining in and around the needle tract only. Furthermore, we found that GFP-positive NPCs+MNTS1 colocalized mostly with NeuN (a mature neuron marker). Some NPCs also colocalized with nestin (an immature cell marker) and GFAP (an astrocyte/glia marker). These cells were negative for Aldh1 (an astrocytic marker), Iba1 (a microglia and macrophage marker), and Olig2 (an oligodendrocyte marker). Due to minimal exogenous cell survival, NPCs+GFP cells were not available for double-label immunofluorescent staining. Spatial memory testing via the Morris water maze (MWM) revealed that all animal groups receiving NPCs – irrespective of transduction profile – performed significantly better on the hidden platform task compared to control TBI/Veh animals. Escape latency differences between sham groups or TBI animals receiving NPCs and TBI/Vehicle controls were significant (p-value=.032). MWM assesses hippocampal-dependent spatial memory, which is strongly correlated with brain derived neurotrophic factor (BDNF) levels and TrkB activation.
We have found that genetically-modified NPCs transduced with a chimeric multineurotrophin had increased survival and targeted migration after transplantation into our animal model of TBI. Sham animals receiving MNTS1-transduced cells did not exhibit similar targeted migration patterns; however, exogenous cell survival was evident. NPCs exhibit tropism towards areas of injury, which may explain why only TBI animals had significant NPC migration. We have previously shown that cultured NPCs do not respond to MNTS1 release in vitro; perhaps in vivo NPCs differentiated into additional cell types thus gaining the ability to respond to MNTS1. Another possibility is that multineurotrophin release modulated the surrounding microenvironment permitting increased cell survival. The majority of NPCs transduced with MNTS1 colocalized with NeuN-positive cells, indicating that these cells differentiate primarily into mature neurons. Interestingly, all animals receiving NPCs performed on par with sham uninjured animals on the MWM spatial memory task, while the TBI/Vehicle group had significant learning deficits. Cellular transplantation coupled with multineurotrophin exposure represents a novel combinatorial approach to attenuate cognitive impairment after TBI. This multidimensional intervention may be more effective in improving outcome than one-dimensional therapies alone.
In addition to the contributing authors listed above, I would like to acknowledge Yunfang Wang and Dr. Juliana Sanchez. Supported by NS030291.
TBI, progenitor cells, transplantation, neurotrophins
ASSESSMENT OF INTRACEREBROVENTRICULAR INFUSION OF INSULIN-LIKE GROWTH FACTOR-1 AFTER CONTROLLED CORTICAL IMPACT IN MICE
Jennifer Brelsfoard, M.S., University of Kentucky
Dr. Kathryn Saatman, Ph.D., Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Traumatic brain injury produces neuronal dysfunction and loss, which can culminate in motor and cognitive impairment. Insulin-like growth factor-1 is a neurotrophic factor capable of mediating neuroprotective and neuroreparative mechanisms. We hypothesized that elevating brain levels of hIGF-1 attenuates behavioral dysfunction and cell death and promotes neurogenesis after brain injury.
C57BL/6 mice were subjected to 0.9mm depth controlled cortical impact (CCI) or sham injury and treated with human IGF-1 (hIGF-1) or vehicle via intracerebroventricular (ICV) infusion using the Alzet brain infusion kit attached to a 3d or 7d osmotic minipump. Treatment was initiated either 15 min or 6–8hrs post-injury by implantation of a primed or non-primed minipump. Motor and cognitive performance was evaluated by a modified neurological severity score (NSS) and a novel object recognition task (NOR), respectively. Brain levels of hIGF-1 were quantified using a hIGF-1 specific ELISA and downstream signaling activation of Akt was quantified by western blot. Hippocampal neurodegeneration was evaluated by fluorojade-C staining. Immunohistochemistry for the immature neuron marker doublecortin (DCx) was used to evaluate survival of immature hippocampal neurons.
Mice treated with 10μg/d hIGF-1 initiated 15 minutes post-injury had measurable brain hIGF-1 levels and significantly increased activation of Akt at 3d post-injury (n=4 hIGF-1 and n=3 vehicle), consistent with biological activity of IGF-1. In order to evaluate the efficacy of prolonged IGF-1 treatment over 7 d, mice received CCI or sham injury and hIGF-1 or vehicle via ICV infusion initiated 15 min post-injury (n=19 CCI/treatment, n=10 sham/treatment). Compared to vehicle-treated mice, IGF-1-treated mice exhibited significantly improved motor function (NSS) over the 7d period (p<0.05), as well as a trend toward improved cognitive function assessed through NOR (p=0.06). There was no overt difference in neocortical or hippocampal neurodegeneration between treatment groups, but quantitative analysis needs to be completed. However, DCx immunoreactivity in the subgranular zone was markedly increased in IGF-1 treated mice, consistent with increased hippocampal neurogenesis. A subset of infused mice showed an enhancement of trauma-induced brain swelling which appeared to be exacerbated by hIGF-1. Delaying the onset of ICV infusion for 6–8hrs post-injury (n=5 hIGF-1 and n=3 vehicle) did not notably decrease brain swelling, but IGF-1 mediated increases in DCx immunoreactivity were again observed, similar to acute onset IGF-1 infusion.
The data demonstrate that hIGF-1 centrally infused over 7 d after CCI brain injury improves neurobehavioral performance when compared to vehicle. IGF-1 infusion delayed as long as 6–8hr after injury resulted in increased DCx staining which may reflect neuroprotection of immature neurons and/or enhanced neurogenesis. However, ICV infusion exacerbated trauma-induced brain swelling, likely due to increased cerebral edema. Future experiments will quantify numbers of both immature neurons and degenerating cells within the hippocampus. Additional studies will explore reduced infusate volumes, as well as IGF-1 actions that may influence cerebral edema. Moreover, these studies highlight the need to pursue alternative routes of administration for IGF-1 after brain injury.
Supported by: KSCHIRT 7–20 and NIH R01 NS072302, P30 NS051220 and T32 DA022738.
TBI, neuroprotection, neurogenesis, growth factor, cognition
OSTEOPONTIN AND MMP UPREGULATION IN THE OLFACTORY BULB DURING SYNAPTIC REORGANIZATION INDUCED BY TRAUMATIC BRAIN INJURY
Thomas M. Reeves, PhD, Virginia Commonwealth University
Linda L. Phillips, PhD, Virginia Commonwealth University
Osteopontin (OPN) is a pleiotropic cytokine which can modulate CNS plasticity. Following traumatic brain injury (TBI), it supports synaptogenesis as a matrix metalloproteinase (MMP) substrate and microglial activator. A similar OPN role within injured olfactory bulbs is suggested by studies showing change in cytokine and MMP expression after deafferentation lesion.
Using the unilateral entorhinal cortex lesion (UEC) model, we demonstrated OPN transcript and protein increase during hippocampal synaptogenesis. OPN was significantly increased over controls 1 and 2d post-lesion, periods when MMP expression was increased. Notably, OPN protein and mRNA were localized to reactive microglia. Given these findings, we investigated OPN and MMP response in the olfactory bulb, a second CNS region exhibiting trauma-induced plasticity. Rats were subjected to moderate midline fluid percussion TBI or sham-injury and allowed to survive for 1 or 3d, after which olfactory bulb extracts were assayed by Western blot for OPN expression and gelatin zymography for MMP-2,9 activity. To confirm bulb injury, we profiled the extent of αII spectrin proteolysis within the same 1d extracts. Additional cohorts of animals were perfused with mixed aldehyde fixative at 1d post-injury, and the ultrastructure of the injured olfactory bulb compared to sham-injured controls.
As for UEC hippocampal deafferentation, we observed significant elevation (p<0.05) in the 66 kD full length form of OPN at 1 and 3d post-injury. One day after TBI, OPN protein was two-fold higher than control, and the cytokine showed a 28% increase at the 3d time point. We also observed that a 32 kD peptide fragment of OPN was increased at 1d (13% over control). Parallel gelatin zymography revealed that MMP-2 lytic activity did not change at either post-injury interval, however, MMP-9 gelatin proteolysis was significantly elevated by 78% over controls at both 1 and 3d after TBI. One day following TBI, we also observed a significant 38% elevation in the 150 kD proteolytic fragment of αII spectrin, but no change in the associated 145 and 120 kD fragments. This result not only confirms axonal injury, but indicates that acute calpain lysis may be more prominent than that produced by caspase within injured olfactory bulb. Electron microscopic analysis of the injured olfactory bulb revealed axonal swelling and membrane disruption of the afferent olfactory nerve fibers, confirming that TBI generated olfactory bulb deafferentation within our model.
These results support a significant role for OPN and MMP-9 in the olfactory bulb following diffuse TBI. OPN increase occurs over acute and sub-acute post-injury intervals, consistent with a role in the early degenerative phase of reactive synaptogenesis. Gelatinase activity in the injured bulb appears selective for MMP-9, and is correlated with acute elevation of an OPN peptide fragment. Importantly, both αII spectrin lysis and axonal disruption confirm TBI-induced deafferentation in the olfactory bulb. Collectively, the similarities between OPN/MMP-9 response in the olfactory bulb and hippocampus support the hypothesis that these molecules function in concert to facilitate reactive synaptogenesis after brain injury. Further, the timing of this response indicates this pair may act to foster debris clearance in the extracellular matrix and promote axonal sprouting. Ongoing studies with OPN KO mice will help confirm the importance of OPN/MMP-9 interaction during successful synaptic recovery.
Thanks to Lesley K. Harris, Raiford T. Black, and Nancy N. Lee. Support: NIH-NS44372, NS57758.
TBI, MMP, OSTEOPONTIN, SYNAPTIC PLASTICITY
A COMBINATION OF 17-BETA-ESTRADIOL AND MEMANTINE AFTER REPETITIVE, MILD TRAUMATIC BRAIN INJURY REDUCES INJURY SYNERGY
Elise Gill, MS, Columbia University
Barclay Morrison III., PhD, Columbia University
Clinical studies have shown that a first concussion greatly increases chances for subsequent concussions and brain injury. For athletes, who are at greater risk for multiple concussions, elucidating the necessary rest periods and therapies following mild traumatic brain injury (mTBI) is required for preventing and treating repetitive injury synergy (RIS).
The biomechanics of blunt head impact were simulated with a well-characterized in vitro model of TBI and organotypic hippocampal slice cultures (OHSC). Cultures received a mild stretch-injury (11.78±0.44% equibiaxial strain, 20 s−1 strain rate) or a sham-injury on days 0 and 1. Experimental groups received either a single stretch-injury, 2 stretch-injuries 24 hours apart, or no stretch-injury. All samples not scheduled to receive a stretch-injury at a given time point were sham-injured. For sham-injury, samples were clamped on the injury device; however the device was not fired. Additionally, OHSC receiving the double injury received either 1.5 μM memantine and 1.5 nM 17-beta-estradiol (E2) in combination or vehicle (PBS) 1 hour following both injuries. While on the injury device, cultures were kept at 37°C. Prior to and following injury, tissue health was evaluated by quantifying propidium iodide (PI) staining as the percent area above an intensity threshold. To induce maximal cell death for normalization purposes, at the end of experimentation and 4 days after the first injury, culture medium was removed and OHSC were incubated in 10 mM glutamate solution for 3 hours. At 24 hours following glutamate treatment, slices were analyzed for resultant cell death.
The stretch-injury model used here was able to recapitulate RIS in OHSC, which has been demonstrated previously in experimental and clinical studies of repetitive mTBI. OHSC exposed to 2 injuries 24 hours apart exhibited a significant and synergistic increase in cell death in all regions of the hippocampus 4 days following the first injury (n=17; DG 15±4%, CA3 20±7%, CA1 18±4%; *p<0.05) as compared to cultures receiving only 1 stimulus (n=13; DG 3±1%, CA3 2±1%, CA1 3±2%; *p<0.05) or the sham-injury (n=19; DG 0.7±0.5%, CA3 0.6±0.3%, CA1 0.8±0.4%; *p<0.05).
We postulate that activated biochemical cascades from multiple injuries in succession are additive, becoming supra-theshold for initiation of neurodegeneration. We hypothesized that drugs, which target these biochemical cascades, can intervene to prevent their super-position. Memantine, an uncompetitive antagonist of N-methyl-D-aspartate receptors, and E2, an estrogen and antioxidant, have proven to mitigate the effects of excitotoxicity and oxidative stress, respectively, following TBI. This combination therapy delivered 1 hour following both injuries significantly reduced cell death 4 days following the first injury (n=24; DG 3±0.7%, CA3 1±0.3%, CA1 2±0.6%; *p<0.05) as compared to vehicle treatment (n=43; DG 11±3%, CA3 8±2%, CA1 14±3%).
Our results suggest that the synergy between injury cascades activated by two mTBIs is reduced by a therapy targetting excitotoxic and oxidative stress pathways. Furthermore, the combination therapy was effective in preventing RIS when administered one hour after the first and second injuries. These results have implications for concussed atheletes who return to play 24 hours following injury and suggest that targeted treatment may reduce the potential for RIS during a period of heightened vulnerability to significant neurodegeneration. Our in vitro model of repetitive mTBI can be used to elucidate neurodegenerative pathways, novel therapeutic strategies to block those pathways, and important rest periods following injury to prevent synergy of biochemical cascades after multiple injuries.
concussion, hippocampus, in vitro, neuron
A METABOLOMICS APPROACH TO INVESTIGATING THE LINK BETWEEN ALZHEIMER'S DISEASE AND HUMAN TRAUMATIC BRAIN INJURY IN THE ACUTE PHASE
Kym F. Faull, Ph.D., University of California, Los Angeles/Dept. Psychiatry and Biobehavioral Sciences
John M. Ringman, M.D., M.S., UCLA, Department of Neurology
David L. McArthur, Ph.D, M.P.H., Department of Neurosurgery, UCLA
Neil Martin, M.D., Department of Neurosurgery UCLA
David A. Hovda, Ph.D., University of California, Los Angeles
Thomas C. Glenn, Ph.D., UCLA, Brain Injury Research Center, Department of Neurosurgery
This study investigates the similarity between the CSF global metabolic profile after TBI in the acute phase (post-injury hour 24–240) and the profile of patients presymptomatic and symptomatic for familial Alzheimer's disease (FAD) to test the idea that severe-TBI can predispose a patient to developing Alzherimer's disease.
CSF samples were collected from twenty severe-TBI patients (1 early and 1 late time point, mean 93.2 hours apart) (mean pr-GCS<7, mean age=46.45+/−18.77, 16:4 m:f) 24–240 hours after injury, eleven patients presymptomatic for FAD (1 sample/patient) (mean age=31.13+/−10.58, 3:8 m:f), ten patients symptomatic for FAD (1 sample/patient) (mean age=41.4+/−13.3, mean CDR score=0.75+/−0.48, 4:6 m:f), and eight control patients asymptomatic for FAD (1 sample/patient) (mean age=31.33+/−7.45, 2:6 m:f) from external ventricular drain (TBI) or lumbar puncture (all others). The trimethylsilyl oxime derivatives of the samples were subjected to electron ionization gas chromatography/mass spectrometry (Waters). Orthogonal partial least squares-descriminant analysis (OPLS-DA) was accomplished using the MarkerLynx XS (Waters) data analysis suite. Metabolite quantitation was done using the QuanLynx sub-routine of the MassLynx (Waters) suite, and other statistics were calculated using R. All p-values were calculated by Mann-Whitney U-test.
After application of the OPLS-DA model, a principal component analysis (PCA) of the comparison between the global CSF metabolic profiles of severe-TBI patients and those of the patients asymptomatic for FAD (normal controls) revealed a complete and robust separation between groups. Inspection of the associated S-plot identified the compounds that most significantly contributed to this group separation (correlation≥0.4, covariance≥0.05) and showed that the concentration of each of the following compounds is significantly lower in the TBI group: lactic acid, threonic acid, scyllo-inositol, citric acid, glyceric acid, stearic acid, and proline (p-value<0.0001 for all comparisons). An OPLS-DA supervised PCA of the comparison between the global CSF metabolic profiles of severe-TBI patients and those of the patients presymptomatic for FAD (carriers of a fully-penetrant PSEN1 mutation with CDR=0) also revealed a complete separation of the groups. Inspection of the associated S-plot identified that the concentration of each of the following compounds is significantly lower in the TBI cohort: lactic acid, threonic acid, scyllo-inositol, citric acid, glyceric acid, stearic acid, and proline (p-value<0.0001 for all comparisons). Finally, an OPLS-DA supervised PCA of the comparison between the global CSF metabolic profiles of the severe-TBI cohort and those of the patients symptomatic for FAD (carriers of a fully-penetrant PSEN1 mutation with CDR>0) once again revealed a complete separation between the groups. Inspection of the associated S-plot identified that the concentration of each of the following compounds is significantly lower in the TBI cohort: lactic acid, threonic acid, scyllo-inositol, citric acid, glyceric acid, stearic acid, and proline (p-value<0.0001 for all comparisons). Similar comparisons were made between the severe-TBI patients grouped by post-injury hour (24–240) and the presymptomatic and symptomatic groups, which resulted in similar findings.
This study reveals that in the acute phase after severe-TBI, regardless of the time point after injury, there is a significant difference between the global CSF metabolic profiles of severe-TBI patients and those of control patients, patients presymptomatic for FAD, and patients symptomatic for FAD. These data do not disprove that severe-TBI predisposes one to developing Alzheimer's disease; however, they do show that in the acute phase of injury the biochemical changes occurring in the CSF of FAD and TBI patients are significantly different. The samples collected from control and FAD patients were through lumbar puncture, while those collected from the TBI patients were through external ventricular drain.
This work was made possible by support from NS058489 and the UC Neurotrauma Initiative, as well as the UCLA Neurosurgical Intensive Care Unit staff.
Biomarker, Metabolomics, Alzheimer's Disease, Dementia
THE NEUROCHEMICAL PROFILE OF TRAUMATIC BRAIN INJURY ASSESSED WITH 1H-MRS AT 9.4T
Hung-Wen Yeh, PhD, University of Kansas Medical Center
In-Young Choi, PhD, University of Kansas Medical Center
Phil Lee, PhD, University of Kansas Medical Center
Nancy Berman, PhD, University of Kansas Medical Center
Russell Swerdlow, MD, University of Kansas Medical Center
Sorin Craciunas, MD, Carol Davila University of Medicine
William Brooks, PhD, University of Kansas Medical Center
The neurochemical profile detected by proton magnetic resonance spectroscopy (1H-MRS) provides a snapshot of the in vivo metabolic status of cells in the brain. Mounting evidence indicates that specific neurochemicals resolved with 1H-MRS may serve as biomarkers of brain injury mechanisms.
We used high field 1H-MRS to assess a neurochemical profile of 20 individual compounds in vivo after experimental TBI in Fischer rats. Controlled cortical impact parameters were: impact tip=5mm, velocity=3.5m/s, depth=2mm, contact time=300ms. MR spectra were acquired in the contused cortex and in the normal-appearing pericontusional hippocampus over a time course from 1 hour to 2 weeks post injury. Behavior was assessed with the rotorod and beam walk over the same period. MRS was collected with a water-suppressed STEAM sequence (Varian 9.4T spectrometer, TE=2ms, TR=4000ms). Neurochemicals fitted with LCModel with Cramér-Rao lower bounds (CRLB) ≤30% were accepted. For sample points with neurochemical concentrations below the detection sensitivity of our instrument (typically data with CRLB >30%), values were imputed by a uniform distribution between zero and the minimum reliably detected value. TBI effects were analyzed with a linear mixed-effects model.
TBI was confirmed by the presence of visible contusions on T2-weighted MRI and by persistent functional deficits on the rotorod and beam walk tests. We observed significant changes in 19 out of 20 neurochemicals in the contused cortex and 9 out of 20 neurochemicals in the pericontusional hippocampus after TBI. Neurochemical changes were typically similar in the two brain regions sampled, although of greater magnitude in the cortex, i.e., closer to the impact.
The neurochemical profile provides in vivo insight into several pathological mechanisms associated with TBI. For example, decreased levels of NAA and aspartate, and increased alanine, serine, and lactate together suggest serious disruptions in cellular bioenergetics. Decreased GABA and glutamate along with increased glutamine indicate disrupted neurotransmission and excitotoxicity. The dramatic depletion of ascorbate and glutathione levels in the injured cortex point to oxidative stress. Finally, increased NAAG in the pericontusional zone but not in the contusion core suggests a possible neuroprotective response.
Imputation allowed us to evaluate the temporal evolution of 20 neurochemicals after TBI, even though some neurochemicals fell below detection limits at some time points. For example, glutathione was reliably detected before injury, but fell sharply after injury to undetectable levels. Conversely, phosphocholine was difficult to detect in the uninjured brain but increased sharply after TBI.
Previous 1H-MRS studies of TBI have been limited to relatively few compounds, primarily NAA, glutamate, inositol, lactate, choline, and creatine. The increased resolution and sensitivity provided by 9.4T, ultra-short echo times, and efficient water suppression, provided a far more detailed neurochemical profile. Our results confirm and extend previous reports that1H-MRS detects neurochemical changes after TBI in brain tissue that appears normal with conventional MRI. Moreover, our results suggest that the neurochemical profile after TBI provides an in vivo measure of local injury severity.
These findings highlight the potential of 1H-MRS for monitoring the pathophysiological mechanisms of brain injury and for pre-clinical evaluation of novel neuroprotective strategies. As more sophisticated spectroscopic acquisitions become available on clinical MRI scanners, a multi-component neurochemical profile could be useful for diagnosis and as an outcome measure in clinical trials.
Supported in part by the KIDDRC (P30 HD002528) and pilot grants from the University of Kansas Medical Center and the KU ADC (P30 AG035982).
magnetic resonance spectroscopy secondary injury bioenergetics oxidative stress neurotransmission
SYSTEMATIC CONNECTOMIC ANALYSIS OF WHITE MATTER ATROPHY ASSOCIATED WITH SEVERE TRAUMATIC BRAIN INJURY
Micah Chambers, PhD, University of California, Los Angeles
Bo Wang, MS, University of Utah
Marcel Prastawa, PhD, University of Utah
Paul Vespa, MD, FAAN, FACN, University of California, Los Angeles
Dr. David A. Hovda, PhD, University of California, Los Angeles
Jeffry Alger, PhD, University of California, Los Angeles
Guido Gerig, PhD, University of Utah
Arthur Toga, PhD, University of California, Los Angeles
Ron Kikinis, MD, Harvard Medical School
John Van Horn, MEng., PhD, University of California, Los Angeles
Despite recent progress in the development of robust image analysis tools, it remains particularly difficult to quantify TBI-related changes in the connectome. Here we introduce a patient-tailored framework for connectomic mapping of brain circuitry, and hypothesize that it provides the ability to characterize atrophy due to TBI.
10 TBI patients recruited with IRB approval provided informed consent. Structural imaging was acquired acutely and chronically (3 and 180 days post-injury, respectively) using T1, T2, FLAIR, DTI, and SWI. Pathology metrics and segmentations of hemorrhage and edema were obtained. The freely available NA-MIC Kit and 3D Slicer platforms were used to quantify pathology and changes due to therapy and/or recovery. Image processing included segmentation of hemorrhage and edema using Atlas Based Classification (ABC). Longitudinal changes are assessed by registration and joint segmentation of baseline and follow-up data. Cortical parcellation yielded brain morphometrics and volumetrics, and DTI-based WM fiber tracking was conducted to map each patient's connectome. Tracts with statistically significant atrophy were identified, and volumetric atrophy measures were correlated against computed bifrontal (BFI), bicaudate (BCI), ventricular (VI), Evan's (EI) and Huckman's (HI) indices. Detailed descriptive metrics were computed and measures of network segregation/integration were calculated.
GM/WM atrophy (volume changes of 2.7±2.1% and 2.2±1.7% (μ±σ), respectively) was found in all subjects. The change in atrophy indices (μ±σ) was: 3.3±2.4% (BFI), 8.4±3.5% (BCI), 7.0±2.7% (VI), 6.2±3.2% (EI), 5.3±2.7% (HI). Changes in the integration/segregation of individual network nodes were systematically quantified for each patient. Importantly, metrics can be extracted for uni- and multivariate modeling to provide additional insights regarding the relationship between TBI-related neuroanatomical changes and clinical outcome variables predicting degree of change and/or recovery.
Our framework for atrophy quantification and cortical circuit exploration is more comprehensive compared to other methodologies, as it offers the ability to systematically describe changes in WM connectivity and circuit properties. Our methods may help clinicians and health care providers to design improved methods for patient monitoring and rehabilitation. With knowledge of general location, extent, and degree of change, such metrics can be associated with clinical measures and subsequently used to suggest viable treatment options for individual subjects against patterns that are typical TBI populations.
We acknowledge NIH grants 2U54EB005149 (to Ron Kikinis and John Van Horn), P01NS058489 (to Paul Vespa) and P01NS058489 (to David Hovda).
TBI, neuroimaging, connectome, MRI, DTI
PLECKSTRIN HOMOLOGY DOMAIN AND LEUCINE RICH REPEAT PROTEIN PHOSPHATASE 1 PLAYS A NEUROPROTECTIVE ROLE IN BRAIN INJURY
Jonathan Verrier, PhD, University of Pittsburgh
Ms. Keri Janesko-Feldman, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Dr. Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Pleckstrin homology domain and leucine rich repeat protein phosphatase (PHLPP1) is a potent inhibitor of extracellular regulated kinase (ERK). Mounting evidence supports a deleterious role for ERK activation following TBI. We recently established a PHLPP1 KO mouse colony. The effect of genotype on injury induced neuronal death was measured.
In vivo: Hippocampus, cortex, and cerebellum were harvested from 12 week old male WT and KO mice. Tissue was homogenized and prepared for protein analysis. Brain samples were run on SDS-PAGE gels, and transferred to PVDF membranes. Blots were probed with antibodies against PHLPP1, phospho-ERK, and ERK total. A second cohort of mice received controlled cortical impact (CCI; 6m/s velocity and 1.2mm depth). 24 hours after injury mice were euthanized, perfusion fixed, brains sectioned, and stained with hematoxylin and eosin. Dead cells in the hippocampal CA1 and CA3 subregions were counted by an investigator blinded to genotype and injury. Data were analyzed using ANOVA (n=5/group). In vitro: Primary cortical neurons harvested from E16 WT/KO mouse embryos were trypsinized, triturated, and seeded on 96-well assay plates. At day in vitro 12, cortical neurons were subjected to 100nM Staurosporine (STS) injury for 24 hours and viability measured.
Biochemical analysis of mouse tissue: Consistent with our previously published work in cultured primary rat neurons, loss of PHLPP1 increased basal ERK phosphorylation in the hippocampus of KO mice. No change in ERK total was observed between WT and KO mice. Genotype–dependent susceptibility to brain injury: No significant differences in CA1/CA3 H&E staining were noted between WT and KO sham (control) mice. Hippocampal injury was worse in PHLPP1 KO mice 24 hours after CCI. In the CA1 subregion, cell counts were significantly lower in WT-CCI compared to WT-sham. Moreover, significantly fewer cells were observed in CA1 from KO-CCI compared to WT-CCI. In the CA3 subregion, differences in cell counts comparing WT-sham to WT-CCI did not reach statistical significance. However, cell counts in KO-CCI were significantly below KO-sham. Genotype–dependent susceptibility of cultured neurons to injury: Primary cortical neurons (>95% pure) were harvested from WT and KO mice. Neurons were plated in 96-well format at high density (1.5×105well). Neurons were exposed for 24 hours to DMSO only (control), 10nM STS, 50nM STS, 100nM STS, 250nM STS, or 500nM STS. Neuronal viability was markedly reduced in DMSO control treated KO neurons compared to WT. Moreover, at low concentrations, STS induced significantly more cell death in KO neurons compared to WT.
The role of PHLPP1 in TBI has not been previously investigated. Here we show that PHLPP1 KO mice
Funding was provided by the American Heart Association and Laerdal Foundation for Acute Medicine. We thank Elizabeth Brough and Mu Xu for their technical support.
TBI, PHLPP1, CCI, Hippocampus, ERK
MICROGLIA/MACROPHAGES PHENOTYPIC HETEROGENEITY AND IMBALANCE IN M1/M2 ACTIVATION STATES IN THE AGED BRAIN FOLLOWING TRAUMATIC BRAIN INJURY
Bogdan A. Stoica, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Boris Sabirzhanov, PhD, Center for Shock, Trauma and Anesthesiology Research (STAR), University of Maryland, School of Medicine
Mark P. Burns, PhD, Georgetown University Medical Center
Alan I. Faden, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
David J. Loane, PhD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Considerable evidence indicates that outcomes after traumatic brain injury (TBI) are worse in the elderly. The aged and injured brain have increased microglial/macrophage (MØ) activation. MØ have multiple phenotypes, including a classically activated (M1) state that may contribute to neurotoxicity and an alternatively activated (M2) state that may promote repair.
In this study we hypothesized that an imbalance in M1/M2 MØ activation states and redox state disequilibrium cause worse outcomes in the aged brain followed TBI. To test this hypothesis, we anesthetized young (3 month old) or aged (24 month old) C57Bl/6 mice and subjected them to moderate-level controlled cortical impact TBI (6m/sec, 2mm depth), and collected tissue after 24 hours or 7 days for gene expression and histological analysis respectively. We performed unbiased stereological assessment of MØ activation phenotypes and post-traumatic neuronal cell loss in the cortex, hippocampus and thalamus, as well as histological assessment of Ym1 (M2 MØ activation) and oxidative stress markers.
Here, we report that aging modulated the expression profile of classical (M1) and alternative (M2) MØ activation markers after TBI; M1 (eg. TNFα, IL-1β & iNOS) and M2a (eg. Ym1 & Arg) genes were significantly increased whereas M2c (eg. IL-4Ra, SOCS3 & TGFβ) genes were significantly reduced in the aged TBI mice when compared to young TBI mice. This was associated with significant changes in MØ activation phenotypes, with highly activated MØ (hypertrophic and bushy phenotype) predominant in the cortex, hippocampus and thalamus of aged TBI samples, and less activated cells (ramified and hypertrophic phenotype) found in young TBI samples. Notably, our analysis of the M2 activation marker (Ym1), putatively involved in the resolution of inflammation and tissue repair after injury, revealed a heterogeneous phenotypic response in activated MØ. Ym1-positive cells displaying an amoeboid cellular morphology were present in perilesional cortical regions, whereas Ym1-positive cells displaying ramified/hypertrophic cellular morphology were detected at more distant sites (white matter and hippocampus) from the lesion area, with the latter being increased in young TBI samples when compared to aged TBI samples. Furthermore, the amoeboid Ym1-positive cells did not express the phagocytic marker, CD68, indicating altered functional capabilities of these MØ cells. We also evaluated redox signaling in these samples and found that aged TBI mice had increased expression of the NADPH oxidase sub-units (gp91phox) in MØ and increased lipid peroxidation (4-HNE) when compared to young TBI mice. These changes were accompanied by reduce anti-oxidant response capabilities (GPX1 and SOD1). Finally, the aged TBI mice had significantly increased lesion size and exaggerated neuronal loss in the hippocampus and thalamus when compared to young TBI mice.
These results suggest that in the aged injured brain there is an imbalance in M1/M2 MØ activation states as well as increased NADPH-oxidase dependent reactive oxygen species generation in highly reactive MØ, which may contribute to increased neurodegeneration and poorer outcomes in the aged brain following TBI.
This work is supported by a pilot award from the National Capital Area Rehabilitation Research Network (R24HD050845) (D.J.L.).
microglia/macrophage, alternative activation, Ym1.
FREQUENCY-DEPENDENT EFFECTS OF MEDIAL SEPTAL STIMULATION FOR COGNITIVE IMPROVEMENT FOLLOWING TRAUMATIC BRAIN INJURY
Ali Izadi, B.S., UC Davis
Dr. Gene Gurkoff, Ph.D., UC Davis
Professor Robert Berman, Ph.D., UC Davis
Arne Ekstrom, Ph.D., UC Davis
Professor Philip Schwartzkroin, Ph.D., UC Davis
Professor Paul Muizelaar, M.D., Ph.D., UC Davis
Kiarash Shahlaie, M.D., Ph.D., UC Davis Medical Center
Professor Bruce Lyeth, Ph.D., UC Davis
Elimination of hippocampal theta impairs rodent cognitive function. Following lateral fluid percussion (LFP) injury, TBI-injured animals exhibit reductions in theta power that correlate with cognitive dysfunction. Since the medial septum maintains and generates hippocampal theta, we hypothesize that medial septal theta stimulation following LFP will enhance behavioral outcomes.
105 male Sprague-Dawley rats (300–350g) underwent sham (n=28) or LFP injury (n=77, 2.12–2.15 atm), and were immediately implanted with a medial septal stimulating electrode and hippocampal recording electrode. On post-injury day (PID) 7, rats were tested with or without continuous medial septal stimulation in an object exploration task. Sham rats were stimulated at 7.7Hz with 80μA (n=5) or without stimulation (n=8), while injured rats received either no stimulation (n=11) or 7.7Hz stimulation at 20μA (n=5), 80μA (n=6) or 200μA (n=5). An additional group was tested with stimulation at 80μA in the gamma range (100Hz). Time spent exploring objects was analyzed. On PID 5–7, a second group of rats was tested twice daily on the Barnes Maze task (TBI control, n=25; sham control, n=15; TBI theta stimulation, n=8; TBI gamma stimulation, n=9; sham stimulation, n=8). Latency to finding the escape box, search strategy and number of errors were analyzed.
In the object exploration task, TBI-injured animals without stimulation spent less time exploring objects than sham-injured animals (14.2±2.6 and 59.8±7.7 seconds, respectively; p<0.001). TBI-injured animals stimulated at 7.7Hz at 80μA and 200μA spent longer times exploring (75.1±4.7 and 63.6±5.1 seconds, respectively, p<0.05) than TBI-injured animals without stimulation (14.2±2.6 seconds), while TBI animals stimulated at 20μA (28.3±4.8 seconds) were not different from unstimulated TBI animals. Similarly, TBI-injured animals stimulated at 100 Hz (18.5±4.4 seconds, p>0.05) were no different from TBI-injured animals without stimulation. Mean exploration time did not differ between sham-injured animals and TBI-injured animals stimulated with either 80μA or 200μA; however, only sham-injured animals demonstrated habituation. In the Barnes Maze task, sham-injured rats without stimulation found the escape box (latency) quicker than TBI-injured rats without stimulation (51.1±8.6 and 102.4±13.9 seconds, respectively, p<0.05). TBI-injured rats stimulated at 80μA and 7.7Hz (45.0±5.8 seconds, p<0.05) had a shorter latency than TBI-injured rats without stimulation. TBI-injured rats stimulated in the gamma frequency (100Hz) demonstrated a trend towards improvement (80.5±15.0 seconds, p>0.05), but had a longer latency than TBI-injured rats stimulated in the theta frequency range (p<0.05). Sham rats with or without stimulation did not perform differently (40.2±6.7 and 51.1±8.6 seconds). Analysis of search strategy demonstrated worse search strategies (percentage of random search strategies relative to spatial or peripheral search strategies) in TBI-injured animals relative to sham animals (70.7% and 40.3%, respectively, p<0.01). There was an improvement in strategy in TBI-injured animals with theta stimulation (34.8%, p<0.01) relative to unstimulated TBI animals, but not with gamma stimulation (66.7%). There were no significant differences in errors among the four groups.
Neuromodulation through deep brain stimulation has been recently studied as a clinical strategy to improve consciousness following TBI; however, it has not been explored as a therapy for persistent cognitive dysfunction. Following TBI, medial septal stimulation in the theta frequency band results in current-dependent increases in exploration of novel objects, but gamma frequency stimulation does not improve exploration. Stimulation of sham-injured rats does not alter exploration. Spatial working memory was impaired following LFP injury. Stimulation with theta, but not gamma, improved spatial learning in the Barnes Maze task. Interestingly, there is a double dissociation with theta and gamma frequency stimulation of sham animals relative to unstimulated shams demonstrating no significant alterations in memory. These findings suggest a potential role of frequency- and current-dependent neuromodulation for cognitive deficits following traumatic brain injury.
Medial septal stimulation, theta, memory
REAL-TIME MONITORING OF CHANGES IN EXTRACELLULAR SODIUM AND POTASSIUM CONCENTRATIONS AND THE EFFECT OF SELECTIVE VASOPRESSIN-1A (V1AR) RECEPTOR ANTAGONIST FOLLOWING FOCAL TRAUMATIC BRAIN INJURY IN RATS
Professor Clive M. Baumgarten, PhD, Virginia Commonwealth University, Department of Physiology and Biophysics
Dr. Christina R. Marmarou, PhD, Virginia Commonwealth University, Department of Neurosurgery
V1aR inhibitors diminish expression of both V1aR and aquaporin-4 and reduce astrocyte swelling and brain edema after focal traumatic brain injury. Effects of V1aR on osmotically active cations that regulate evolving brain edema are unknown. We determined whether V1aR inhibition after injury affects recovery of extracellular Na+ and K+.
Ion-selective Na+ and K+ electrodes were implanted stereotactically 1.5-mm deep in right parietal cortex of adult Sprague-Dawley rats (350–400 g). Extracellular Na+ and K+ concentrations ([Na+]e; [K+]e) were recorded and MABP, ICP and blood gases were monitored. Rats (n=5/group) were assigned randomly to either CCI-vehicle or CCI-SR49059-treated groups; SR49059 (2.76 mg/kg) was administered starting at 5 min post-injury for 5 h. Baseline data were recorded for 45 min prior to injury. TBI was produced by lateral controlled-cortical-impact (CCI; 5 mm probe, 6 m/s velocity, 1.5 mm injury depth), and data were continuously acquired for 5 h. [Na+]e and [K+]e changes in CCI-vehicle-treated rats were compared to CCI-SR49059. Electrodes were calibrated prior to experiments in mixed Na+ and K+ solutions at physiologic ionic strength. Preliminary experiments demonstrated that calibrations were not significantly influenced by electrode use. Data are expressed as mean±SEM.
Before application of SR49059, there were no significant differences in [Na+]e or [K+]e between CCI-vehicle and CCI-SR49059 groups immediately prior to injury (0 min; baseline). In the CCI-vehicle group, [Na+]e was 141.6±5.1 mM and [K+]e was 5.2±0.6 mM, whereas in the CCI-SR49059 group, [Na+]e was 135.3±3.9 mM and [K+]e was 5.7±0.7 mM. CCI at 5 min caused a significant fall in [Na+]e and increase in [K+]e from baseline in both groups (P<0.01); [Na+]e was 80.1±15 mM and [K+]e was 20.9±3.8 mM in CCI-vehicle, and [Na+]e was 87.9±7.9 mM and [K+]e was 13.4±3.4 mM in CCI-SR49059. Before drug administration but after injury (5 min), [Na+]e and [K+]e in both the CCI-vehicle and CCI-SR49059 groups remained indistinguishable. Importantly, treatment with the selective V1aR inhibitor accelerated the time course of [Na+]e recovery to baseline following injury. [Na+]e in CCI-SR49059 animals was significantly reduced post-injury (5–35 min) and increased to a value that was not significantly different from baseline (128.8±5.7 mM) at 45 min, whereas equivalent recovery in CCI-vehicle animals (125.2±9.1 mM) required 1 hr post-injury. In contrast, the time for recovery of [K+]e to baseline was 35 min for both the CCI-vehicle and CCI-SR49059 groups.
Large extracellular ionic changes rapidly occur after focal brain injury induced by a controlled cortical impact to the parietal cortex of rats. Extracellular Na+ decreases by ∼40% and extracellular K+ increases ∼300% within 5 min of injury, and both return slowly to normal values. Selectively inhibiting the V1aR with SR49059 allowed faster recovery of extracellular sodium concentration, whereas recovery of potassium was unaffected. Cellular edema may result from unbalanced influx and efflux of cations. By balancing the rate of osmotically active sodium and potassium recovery, SR49059 may thereby reduce water influx into cells after brain injury. These findings suggest SR49059 and V1aR inhibitors as potential therapeutic targets in treating cellular edema.
This research was supported by grant 5R01NS019235 from the National Institutes of Health, Bethesda, MD.
sodium, potassium, extracellular, v1aR, edema
ETHOSUXIMIDE DOSE-DEPENDENTLY ATTENUATED NONCONVULSIVE SEIZURES INDUCED BY PENETRATING BALLISTIC-LIKE BRAIN INJURY IN RATS
Xi-Chun M. Lu, Ph.D., Walter Reed Army Institute of Research
Weihong Yang, Walter Reed Army Institute of Research
Ying Cao, B.S., Walter Reed Army Institute of Research
Deborah A. Shear, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Traumatic brain injury can trigger non-convulsive seizures (NCS) that are detrimental to patients' prognosis. Identification of effective anti-epileptic drugs against NCS remains crucial to improve neurological outcome. We evaluated the dose-response effects of ethosuximide (ETX) against NCS in a rat model of penetrating ballistic-like brain injury.
Unilateral frontal PBBI was produced in the right hemisphere of isoflurane anesthetized rats (10% injury severity level). Spontaneously occurring NCS were detected by continuous electroencephalograph (cEEG) monitoring and video-recording for 72 h post-injury. ETX (25, 125, 250, and 375 mg/kg, i.v., n≥15/group) or vehicle was administered via a bolus injection 30 min post-injury followed by a maintenance dose (1/2 bolus dose) given twice daily thereafter. Animals were sacrificed 72 h post-injury and brain tissue was processed for histopathological assessment. The NCS were identified from offline review of the cEEG recordings and confirmed by investigators blind to treatment groups. ETX efficacy was evaluated on NCS parameters of incidence, frequency, episode duration, total seizure duration, and latency.
ETX attenuated NCS seizures in a dose-dependent manner. In vehicle-treated animals, 65% experienced NCS (averaging 9 NCS/rat) whereas ETX treatment at high doses (250 or 375 mg/kg) significantly reduced NCS incidence to 27% and 13% (p<0.05), and seizure frequency to 2.1 and 1.8 NCS/rat (p<0.05), respectively. By comparison, effects of two low ETX doses (25 and 125 mg/kg) were relatively moderate (47% and 53% incidence, and 5.0 and 3.0 NCS/rat, respectively) and failed to achieve statistical significance. Except for the lowest doses tested, ETX treatment significantly shortened NCS total duration by 50–80% compared to the vehicle (261 sec), and delayed NCS onset from 29 h (vehicle group) to 57–68h (ETX groups) post-injury. Overall, the two highest doses exerted equivalent therapeutic efficacy, yet severe sedative effects were observed at only the highest dose tested.All ETX doses did not significantly alter lesion volumes compared to vehicle-treated animals.
Ethosuximide is an FDA-approved anti-epileptic drug commonly used to treat absence (petit mal) seizures. It has previously been shown to prevent post-ischemic NCS in a rat model of middle cerebral artery occlusion at the comparable doses tested in this study. For the first time we demonstrate here that ETX is effective against spontaneously occurring NCS following penetrating ballistic-like brain injury. These data taking together indicate that ETX may be considered as a drug option for treating post-traumatic/ischemic NCS and improve patient outcome. It is widely accepted that the effects of ETX on absence seizure is mediated by its modulation of calcium channels functions. In this regard, the mechanisms of ETX on post-traumatic/ischemic NCS remain to be elucidated.
This research was funded by the Army Combat Casualty Care Research Program and DHP grant and performed while the author held an NRC research fellowship.
TBI, EEG, nonconvulsive seizure, Ethosuximide
CULTURED ENDOTHELIAL PROGENITOR CELLS PROTECT ISCHEMIC AXONS FROM DEGENERATING AFTER BRAIN INJURY
Eugene Park, PhD, St Michael's Hospital, Li Ka Shing Knowledge Institute, University of Toronto
Elaine Liu, MD, St Michael's Hospital, Li Ka Shing Knowledge Institute, University of Toronto
Andrew Baker, MD, St Michael's Hospital, Li Ka Shing Knowledge Institute, University of Toronto
White matter preservation is a strong predictor of mortality after brain injury. Disrupted microvasculature is accompanied by recruitment of endothelial progenitor cells to the injury. Blood vessels and axons are co-dependent during development and we propose also following injury recovery. We hypothesize that EPCs decrease axon degeneration after trauma.
Adult male Sprague-Dawley rats underwent a 2.5 atm midline fluid percussion injury (mFPI), and 24hrs later were venously injected with 5 million endothelial progenitor cells (EPCs). EPCs were derived from mononuclear cells isolated from rat bone marrow and cultured for eight days in Endothelial Growth Media (EGM-2). Rats were then survived for three days, and brains removed for immunolabelling. EPCs were labeled with a lipophillic dye, PKH26, for in vivo cell tracking.
We sought to delineate the effects of EPCs on mechanical versus ischemic injury after trauma using two sublethal in vitro injury models; oxygen-glucose deprivation and stretch-injury. Cortical neurons were harvested from E17 rat embryos and grown for 5 days before being injured. Neurons were co-cultured with EPCs or treated with EPC conditioned media.
Isolated mononuclear cells expressed CD133 (a marker for early EPCs), demonstrated metabolism of acetylated-LDL and ability to form microtubules in a Matrigel assay, consistent with characteristics of EPCs. In vivo analysis demonstrated increased beta-APP accumulation in the corpus callosum of injured untreated rats, indicating axon damage in the white matter tract. Rats treated with EPCs after injury had decreased beta-APP accumulation (number of swellings: FPI+vehicle 1309.83±281.14; FPI+EPC 520.79±91.26; sham 35.5±11.51 p<0.01). PKH26 labeled EPCs were detected in the injured cortex up to 5 days after injury but not at 30 days suggesting therapeutic effects may be mediated through realease of cytokines and growth factors. We are currently assessing functional recovery following EPC treatment using evoked compound action potential recordings from the corpus callosum of experimental rats.
Cultured cortical neurons subject to OGD treatment and co-cultured with EPCs had less axon degeneration than control neurons, indicating that EPCs are important in mediating axon survival in vitro. Neurons were cultured with EPC-conditioned media to determine if EPCs are required for axon survival or if trophic factors released from EPCs were sufficient to support axon maintenance. There was rescue of axon degeneration in neurons cultured with EPC-conditioned media after OGD but not after stretch injury, indicating that conditioned media was sufficient to rescue hypoxia-induced axon degeneration, but not mechanical injury. In cortical neurons subjected to OGD, there was increased cleaved-caspase 6 and cleaved-caspase 3 expression. Treatment with EPC conditioned media decreased the expression of these activated caspases indicating a role in inhibiting the signaling pathways leading to programmed axonal degeneration.
EPCs rescue axon degeneration after TBI both in vivo and in vitro. EPCs significantly decrease the amount of beta-APP accumulation in the corpus callosum in injured rats. This rescue appears to be mediated by signaling induced by the EPCs and not the EPCs themselves. This is evidenced by our in vitro results, where EPC-conditioned media was sufficient to protect cortical axons from degeneration following hypoxia induced injury. EPC-conditioned media was not sufficient however to rescue axon degeneration induced by mechanical stretch. This could indicate that separate pathways are involved in mechanically-induced injury versus hypoxia-induced injury and that these pathways respond differently to vascular signaling. Furthermore, results indicate that treatment with EPC conditioned media decreases activation of caspase 3 and caspase 6. These results suggest that EPC-derived signaling is an important mediator of white matter sparing after injury.
St Michael's Hospital, Keenan Research Centre, Li Ka Shing Knowledge Institute, University of Toronto
Axon degeneration endothelial progenitor cells
CURCUMIN REDUCES MICROVASCULAR INFLAMMATION FOLLOWING TRAUMATIC BRAIN INJURY - A LIVE ANIMAL STUDY
Joshua Marks, MD, Jefferson University School of Medicine
Mohammad Murcy, MD, University of Pennsylvania
Paymon Sanati, University of Pennsylvania
Rachel Eisenstadt, University of Pennsylvania
Wanfeng Gong, MD, University of Pennsylvania
Daniel Holena, MD, University of Pennsylvania
Carrie Sims, MD, MS, University of Pennsylvania
Douglas Smith, MD, The University of Pennsylvania
Jianning Zhang, MD, PhD, Department of Neurosurgery, Tianjin Medical University General Hospital
Jose Pascual, MD, PhD, FACS, FRCPS(C), University of Pennsylvania School of Medicine
Neuroinflammation of the cerebral microcirculation following traumatic brain injury (TBI) is a key element in the progression to secondary brain injury. Curcumin is known to reduce host-mediated inflammation following ischemia. We hypothesized curcumin would reduce in vivo endothelial cell (EC) interactions with leukocytes (LEU) 24 hours after TBI.
Male CD-1 mice (25–30g) underwent TBI by controlled cortical impact (CCI, speed 6m/s, depth 1mm) and then were randomly assigned to 300mg/kg curcumin treatment (CCM) or placebo (PCB; dimethylsulfoxide [DMSO]) immediately after TBI. Intravital microscopy of the pia was conducted 24 hours post TBI adjacent to the brain injury through a cranial window placed over the frontal-parietal cortex. Rhodamine 6G was injected intravenously to label LEU. Microvascular cerebral inflammation was assessed by determining the extent of live EC/LEU interactions and blood brain barrier (BBB) permeability was derived by the live microvascular leakage of fluorescent albumin (FITC). The quantification of EC/LEU interactions and albumin leakage was analyzed off-line using NIS software (Nikon). Mean LEU rolling, adhesion and BBB leakage in each group was compared with the Mann-u-Whitney statistic and significance was considered at p<0.05.
Baseline characteristics (craniotomy size, surgical time, blood pressure) were similar in curcumin and PCB groups. Curcumin significantly attenuated EC/LEU interactions, which was demonstrated by decreased LEU rolling (8.00±1.92 vs. 14.20±2.79, CCM vs. PCB, p=0.041) and adherence (1.54±0.52 vs. 3.17±0.55, CCM vs. PCB, p=0.015) along the EC lining the vessel. Concurrently, in vivo FITC albumin extravasation from the same live vessels demonstrated reduced BBB disruption after curcumin treatment (0.5673±0.0112 vs. 0.5947±0.0005, CCM vs. PCB, p=0.028).
Curcumin reduces microvascular inflammation following brain injury and concomitantly reduces microvascular permeability in brain. Curcumin thus presents an attractive potential neuroprotection that requires further investigation in TBI.
University of Pennsylvania
Traumatic brain injury, Inflammation, Curcumin
DIFFUSE BRAIN INJURY ALTERS SYNAPTOGENESIS OVER A TIME COURSE THAT CORRESPONDS TO LATE-ONSET BEHAVIORAL MORBIDITY
Travis P. Spaulding, University of Kentucky; Spinal Cord and Brain Injury Research Center
Lindsey E. Smith, University of Kentucky; Spinal Cord and Brain Injury Research Center
Dr. Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Despite preventative measures (helmets/seatbelts), diffuse traumatic brain injuries (dTBI) occur at a staggering rate and frequently result in late-onset post-traumatic neurological impairment, including sensory sensitivity. In rodents, dTBI leads to a late-onset, gain-of-function sensory sensitivity to whisker stimulation hypothesized to result from maladaptive circuit reorganization in the thalamocortical circuit.
For circuit reorganization to manifest in neurological impairment, synaptogenesis is necessary. Here, we hypothesize that midline fluid percussion injury (mFPI) induces a distinct temporal profile of synaptogenic gene expression and morphological change in the thalamic relay of the whisker circuit which may be a pivotal point to solidify late-onset neurological impairment. In these experiments, adult male, Sprague-Dawley rats were subjected to a single moderate severity (1.9 atm; 6–10 min righting reflex time) mFPI. Gene and protein expression were assessed using real-time PCR and Western blots to quantify synaptogenic and thrombospondin-related molecular changes in the somatosensory thalamic relay of the whisker-barrel circuit at multiple time points after mFPI (from 1–56 days post-injury). Golgi and Olmos aminocupric silver staining were used to examine time-relevant morphological changes and identify the time course of pathological events (respectively) in the same thalamic relay at 1, 7 and 28 days post-injury.
Thrombospondin 1 and 2 (TSP1 and 2) and their cognate receptor, α2δ-1, have been reported to play a role in circuit reorganization. After diffuse brain injury, gene expression for TSP1 increased significantly by 30-fold at days 5–7 post-injury compared to sham. At 1–7 days post-injury, a significant injury-induced reduction in the gene expression of TSP2, α2δ-1, synaptophysin and GAP43 occurred, which rebounded to sham levels by 28 days. Synaptophysin protein expression followed a similar dynamic. Golgi staining showed branched and uniformly distributed neuronal processes in sham animals. Through 1 and 7 days post-injury, processes become more blunted and less branched. At 28 days post-injury, bushy outgrowths of branches and spines become evident on primary processes. Silver staining shows marked neuropathology at 1 day (10% of VPM stained) that increased through 7 days (17%) with only partial resolution by 28 days post-injury (14%), compared to the complete absence of staining shams.
The initial decrease in synaptogenic gene and protein expression, which rebounded by 28 days post-injury, indicates a loss of synaptic content sub-acutely after injury followed by synaptic rebuilding. These data are supported by morphological observations of bushy outgrowths in Golgi stained tissue and unresolved neuropathological processes at 28 days post-injury. The time course of all these outcome measures corresponds to the manifestation of sensory sensitivity in the whisker-barrel circuit at 28 days post-injury. This is also the first indication that thrombospondins and their cognate receptor may be modulated after experimental diffuse brain injury, as has previously been reported to reorganize circuitry after experimental stroke. These data will serve to identify a rehabilitative and therapeutic window for diffuse brain injury treatment by focusing on the role of synaptogenesis in circuit reorganization.
Financial support from NIH R03 NS077098-01A1, NIH R01 NS-065052 and NIH P30 NS051220-01. This research was conducted at the University of Kentucky.
Synaptogenesis, thrombospondin, circuit reorganization
POST-TRAUMATIC STRESS DISORDER PHENOTYPE IDENTIFIED BY IMPAIRED EXTINCTION OF FEAR LEARNING AFTER CONTROLLED CORTICAL IMPACT IN MICE
Christopher Lee, BA, Department of Pediatrics, Massachusetts General Hospital
Jimmy Zhang, BA, Department of Pediatrics, Massachusetts General Hospital
Emad N. Eskandar, MD, Massachusetts General Hospital
Michael J. Whalen, MD, Harvard Medical School
Post-traumatic stress disorders (PTSD) is modeled by impaired extinction of fear learning. Traumatic brain injury may increase PTSD risk, however this link has not been established in animal models. Using a mouse controlled cortical impact (CCI) model, we tested the hypothesis that CCI induces the PTSD phenotype.
CCI was induced to the parietal cortex in isoflurane-anesthetized male C57bl/6 mice (2–4 mos old; 0.6 mm, 6 m/s, 100 msec). Control mice were naïve and sham operated (craniotomy only). Mice underwent fear-conditioning using three mild footshocks (0.6 mA, 2 s duration each) paired with a tone (30 s duration, 75 dB, 1500 Hz). Freezing (defined as lack of movement except respiratory effort) and open field testing was assessed using video tracking software. Mice were conditioned on day 1 with tone/shock pairings in context “A”. On days 2 and 3, mice were placed in chamber context “B” and subjected to extinction trials done with tone alone presentations (15 presentations per day). Fear data were reported as total time freezing (30s maximum). Fear learning and extinction data were analyzed by RM ANOVA with group and trial as dependent variables. Open field test data were analyzed by t-test.
All mice survived sham injury and CCI. Based on fear learning and extinction curves in naïve mice (n=8/group), 0.6 mA was chosen as the test stimulus to avoid ceiling and floor effects. Naïve and sham-injured mice did not differ in tests of locomotion or in fear learning/extinction. Naïve (n=12), sham injured (n=10) and CCI injured mice (n=12) had no differences in rate of fear acquisition (day 1 trials) or consolidation of fear memory assessed by comparable levels of freezing on the first extinction tone on day 2. However, CCI injured mice had deficits in rates of acquisition of extinction learning as evidenced by higher levels of freezing in extinction trials on days 2 and 3 (p<0.005 RM ANOVA). No differences between naïve/sham injured and CCI injured mice were observed in locomotor activity (total distance traveled, average speed, total line crossings) in open field testing, suggesting that the observed fear extinction deficits in CCI injured mice were not explained by effects of CCI on locomotor behavior.
Mice injured by CCI had normal acquisition of fear learning but impaired extinction, strongly suggesting the PTSD phenotype in our mouse CCI model. To our knowledge, this is the first demonstration of an association between TBI and the PTSD phenotype in an experimental TBI model with fear extinction. Additional experiments on anxiety are needed to support this idea. The data have implications for war fighters as well as civilians who suffer traumatic brain injury, and provide a platform with which to examine mechanisms of induced PTSD as well as putative therapies to reduce or eliminate the PTSD phenotype after TBI. Future experiments will address the contribution of CCI on other types of memory, such as trace-fear conditioning. Future studies should also examine whether other TBI models such as concussion and blast injury also induce the PTSD phenotype as defined by extinction of fear learning.
This work was supported by NIH RO1NS047447 and DOD/CIMIT.
Post traumatic stress disorder, mice
Open Communication Presentations
SENSITIVITY OF SUSCEPTIBILITY WEIGHTING IMAGING FOR THE DETECTION OF TRAUMATIC BRAIN LESIONS IN CHILDREN
Richard Beare, PhD, Murdoch Childrens Research Institute
Michael Ditchfield, MD, Southern Health Victoria
Lee Coleman, MD, Royal Children's Hospital Melbourne
Franz Babl, MD, Royal Children's Hospital Melbourne
Michael Kean, Murdoch Childrens Research Institute
Louise Crossley, Murdoch Childrens Research Institute
Cathy Catroppa, PhD, Murdoch Childrens Research Institute
Keith Yeates, PhD, Ohio State University
Vicki Anderson, PhD, Murdoch Childrens Research Institute
Novel neuroimaging techniques may have advantages over conventional MRI and CT in terms of reduced radiation exposure and greater sensitivity to detect brain injury. Susceptibility-weighted imaging (SWI) shows promise because of its heightened ability to detect hemorrhagic lesions; however, its use in pediatric TBI and clinical potential remain uncertain.
The ability of three imaging techniques (acute CT, conventional magnetic resonance imaging - cMRI, SWI) to reveal traumatic brain lesions was compared in 76 children (mean age 10.24, SD=2.50 years, range 5.75 – 14.67 years) with varying levels (mild, mild complicated, moderate, severe) of traumatic brain injury (TBI). A non-contrast CT scan was obtained according to standard head trauma hospital guidelines during the Emergency Department visit. MR images, including SWI, were acquired as soon as possible after consent to the study on a 3T Siemens Trio scanner using a 32 channel matrix head coil. All scans were coded for presence or absence, extent, and type of traumatic brain lesions by two experts blind to the clinical details at presentation and to the CT scan results. Number and volume of SWI lesions were obtained using manual segmentation and correlated with standard clinical outcomes.
Glasgow Coma Score was 13–15 in 54 patients (71%), 9–12 in 13 patients (17%) and<8 in 9 patients (12%). CTs were completed in the ED; cMRI and SWI were completed at a mean of 36.11 (SD=15.75) days post-injury. Detection of any lesions occurred in CT scan in 68%, cMRI in 54%, SWI in 86% of cases, and SWI detected additional lesions 30% of the time compared to CT and cMRI. Of note, there were 6 cases (18.9%) where only SWI detected lesions and all were in patients with mild TBI as defined by GCS 13 (1 case), 14 (3 cases), 15 (2 cases). In total, SWI was the most sensitive modality in 47.2% of cases, and more sensitive than CT, but at least as sensitive as cMRI in 21 cases (58.3%). An important observation of the cases where SWI was the most sensitive is that there were 11 cases (52.4%) where CT was normal and 7 cases where cMRI was normal (33.3%). Both lesion volume (min 23.68mm3, max 24636.89mm3) and lesion number (min 1, max 77) showed substantial variability. Segmentation procedures for SWI lesions were reliable, with an intra-rater ICC score of 0.987 (95% CI=0.911 to 0.999). A significant main effect of group was found for number of SWI lesions (p=.001) and for volume of SWI lesions (p=.009). Significant correlations were found between volume and number of SWI lesions and a several clinical variables, including GCS, surgical intervention, length of hospital stay, and length of intubation. Presence and volume of SWI lesions did not correlate significantly with age at injury, general/neurological symptoms, or duration of loss of consciousness.
The clinical and scientific utility of advanced MRI sequences for the characterization of intracranial brain lesions sustained via TBI is gaining increasing support. Exploring alternative neuroimaging techniques is especially relevant in pediatric populations, where serious concerns exist regarding increased morbidity following CT use. The results of this study suggest that in pediatric TBI, SWI is more sensitive in detecting traumatic lesions than CT or conventional MRI. They also support prior findings indicating a relationship between greater number and volume of SWI lesions and poor clinical outcomes, and extend these findings to a more generalizable sample of pediatric TBI including the full range of injury severity. These findings may be important for the ongoing management of pediatric TBIs and their prognosis.
We gratefully acknowledge the financial support of the Victoria Neurotrauma Initiative (#CO6E1) and the Canadian Institutes of Health Research.
TBI, pediatric, MRI, SWI, outcome
COMBAT BLAST AND NONBLAST MILD TRAUMATIC BRAIN INJURY: THE COMPARISON OF ACUTE SYMPTOMS AND DIAGNOSED POST CONCUSSION SYNDROME
Andrew J. MacGregor, MPH, PhD, Naval Health Research Center
Judy L. Dye, MS RN ANP, SAIC
Michael R. Galarneau, MS, Naval Health Research Center
Mild traumatic brain injuries (mTBI) in current military conflicts are primarily caused by blasts, yet mTBIs are also due to nonblast mechanisms. While the mechanisms of injury of blast and nonblast mTBI are different, it is unclear whether acute symptoms or outcomes of mTBI are unique to these mechanisms.
The Department of Medical Modeling, Simulation, and Mission Support at Naval Health Research Center maintains the Expeditionary Medical Encounter Database (EMED). The EMED contains near-point-of-injury information on wounded service members through medical records completed by military providers at forward-deployed treatment facilities. Unique components of this medical information include specific injury types and locations, detailed descriptions of injury mechanism, vital signs, physical exam findings, signs and symptoms, and medical treatments and procedures. Using the EMED, a sample of 491 service personnel with a clinical diagnosis code consistent with an mTBI, defined as injury with loss of consciousness, was selected and divided into two groups by the documented mechanism of injury; blast-related mTBI and nonblast mTBI. A review of EMED provider notes allowed for the abstraction of acute symptoms immediately postinjury. In addition, outpatient medical databases were used to identify diagnoses within 1 year postinjury.
Out of the 491 service personnel, 24.0% were injured by a nonblast mechanism and 76.0% were injured by a combat-related blast. Improvised explosive devices accounted for a majority of blast-related mTBI, while motor vehicle accidents were the most frequent cause of nonblast mTBI. Personnel with blast-related mTBI were more likely to present with acute headache (53% vs. 38%, p<0.01), nausea and vomiting (17.2% vs. 9.8%, p<0.05), tinnitus (32.2% vs. 1.7%, p<0.001), and other auditory symptoms (17% vs. 0%, p<0.001), while those with nonblast mTBI were more likely to have amnesia (20% vs. 9%, p<0.01). It was determined that personnel with blast-related mTBI relative to nonblast mTBI were significantly more likely to receive a postconcussion syndrome diagnosis in the 1 year postinjury (Adjusted Odds Ratio: 3.66, 95% Confidence Interval=1.74, 8.28, p<0.001).
To our knowledge, this is the first study to compare acute mTBI symptoms between blast- and nonblast-related mTBI using a distinct nonblast population. In comparing acute symptoms between blast- and nonblast-related mTBI, systemic symptoms such as headache, nausea and vomiting, and hearing deficits were more common in blast mTBIs, while memory problems were more common in nonblast mTBIs. When examining longer-term outcomes, a diagnosis of postconcussion syndrome was over three times more likely in blast-related mTBI compared to nonblast mTBI. These results are in contrast to previous studies comparing blast- and nonblast-related mTBI, with stronger evidence supporting a difference in outcomes between the two mechanisms of mTBI in the present study. This study's primary strength is the use of the EMED as a unique data source for point-of-injury information in deployed service personnel, which includes mTBI diagnosis by a medical provider and documentation of acute mTBI symptoms.
Study was supported by US Navy Bureau of Medicine and Surgery under Wounded, Ill and Injured/Psychological Health/Traumatic Brain Injury Program, Work Unit No. 60808.
mTBI, blast, nonblast, military
PRE-EMPTIVE DECOMPRESSIVE CRANIECTOMY: A COMPARATIVE EFFECTIVENESS STUDY BETWEEN A U.K. AND U.S. CENTER
Professor Ross Bullock, MD, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Chris Zacko, MD, MS, Penn State Hershey
Steve Vidgeon, DM, King's College Hospital
Thomas Ridder, MD, Virginia Commonwealth University
Rick Stanger, MD, Virginia Commonwealth University
Martin Fabricius, DMSc, Glostrup Hospital
Bruce Mathern, MD, Virginia Commonwealth University
Christos Tolias, MBBS, PhD, King's College Hospital
Clemens Pahl, DM, King's College Hospital
Anthony Strong, DM, King's College Hospital
The uses of decompressive craniectomy as either a primary pre-emptive procedure or as a secondary procedure indicated by refractory elevated intracranial pressure (ICP) are controversial in treatment of traumatic brain injury (TBI). A majority of candidates for surgery fall between these extremes and comparative effectiveness studies may elucidate best practices for such real-world scenarios.
We compared outcomes and surgical methods to treat TBI between two neurosurgical centers in the U.S. and U.K. Only patients with a clinical need for surgery were enrolled, and those with fixed, dilated pupils were excluded. Research protocols were approved by ethical boards. To measure injury severity, we used the prognostic score based on seven variables collected prospectively at hospital admission, as defined by the IMPACT study. Electrocorticographic recordings were collected during intensive care using a subdural electrode strip placed during surgery, as described previously, and ICP was measured when clinically indicated. Operative notes were reviewed to determine the primary indications for surgery and the surgical procedures performed. Post-surgical CT scans were evaluated by a neurosurgeon blinded to other data for quantification of craniotomy/-ectomy type and size. Bone flap areas were approximated using the formula for an ellipse. Six-month neurologic outcome was assessed by the Glasgow Outcome Scale (GOS).
Patients with surgical treatment of TBI were enrolled in the Co-Operative Study on Brain Injury Depolarizations at King's College Hospital (KCH, London, UK; n=27) and Virginia Commonwealth University (VCU, Richmond, VA; n=24) from 2004–2010. Baseline measures of severity were similar for VCU and KCH, including prognostic scores based on admission characteristics (p=0.22), maximal pre-operative ICP values (48 vs 44 mmHg, respectively, p=0.67), and indications for surgery (initial CT as primary factor in 58% and 56% of cases, respectively). However, at VCU patients were operated earlier (83% vs 52% within 24 h, p=0.02), craniectomies with removal of bone flap were performed more frequently (75% vs 44%, p=0.03) and craniotomy/-ectomy areas were 56% larger (mean: 82 vs 53 cm2, p<0.001). Post-operatively, VCU patients had lower maximal ICP values (22.5 vs 31.4 mmHg, p<0.01) and trended toward lower incidence of cortical spreading depolarizations (42% vs 63%, p=0.13) and more favorable outcomes (GOS 3–5; 67% vs 46%, p=0.14).
In order to determine whether differences are attributable to the use of prophylactic decompressive craniectomy at VCU, as opposed to smaller craniotomies for lesion evacuation only at KCH, we also compared only those patients requiring lesion evacuation within 24 h after injury. For 16 VCU and 14 KCH patients, prognostic scores, pre-operative ICP values, and timings of surgeries were similar, and the majority of both groups had subdural hematomas evacuated (VCU: 14/16 and KCH: 12/14). However, differences in surgical procedures and post-operative courses were even more accentuated in this subgroup. At VCU, craniectomies were performed more frequently (75% vs 36%, p=0.03) and craniotomy/-ectomy areas were 78% larger (mean: 85 vs 48 cm2, p<0.001). Following surgery, VCU patients had lower maximal ICP values (20.8 vs 30.2 mmHg, p=0.02) and lower incidences of cortical spreading depolarizations (31% vs 86%, p<0.01). Accordingly, VCU patients had significantly more favorable outcomes (69% vs 29%, p=0.03).
Our results confirm signficant variability in surgical management of TBI, with surgeries performed earlier, craniotomies larger, and bone flaps removed more frequently at a U.S. compared to a European center. These differences were even more accentuated in patients undergoing emergency surgery for lesion evacuation, and were associated with less post-operative intracranial pathology, evidenced by raised ICP and spreading depolarizations, and better outcomes at the U.S. center. Results support the use of early prophylactic decompressive craniectomy as a primary surgical intervention in patients with mass lesions. Furthermore, they illustrate the use of spreading depolarizations as a biomarker of intracranial pathology and indicator of worse outcome.3,4 A limitation of the study is that undocumented differences in clinical care may have contributed to the different outcomes observed. Nevertheless, such observational and comparative effectiveness research has advantages over controlled clinical trials in identifying best real-world management practices and may assist guideline development.
This work was funded in part by the U.S. Army CDMRP PH/TBI Research Program, Contract No. W81XWH-08-2-0016. AJS acknowledges support from HeadFirst.
decompressive craniectomy; spreading depression; surgery
ASSESSMENT OF SLEEP DISTURBANCES IN ADOLESCENTS WITH MILD TRAUMATIC BRAIN INJURY
Tonya Palermo, PhD, University of Washington
Monica Vavilala, MBBS, University of Washington
Frederick Rivara, MD, MPH, University of Washington
Few studies have examined the nature or persistence of sleep disturbances in pediatric traumatic brain injury (TBI). In light of the potential impact on functioning and rehabilitation, this study aimed to examine sleep patterns and behaviors in adolescents 3 to 12 months after mild TBI compared to a healthy cohort.
This is an ongoing study of two cohorts of adolescents, those with mild TBI and a healthy comparison sample. Adolescents with TBI are being recruited through review of the Harborview Medical Center Trauma Registry (Washington) and healthy adolescents are being recruited from the community. Adolescents are completing an assessment protocol that includes questionnaire measures of demographics and sleep behaviors, and objective sleep assessment through actigraphy monitoring for 10 days. Specific measures include the Adolescent Sleep Wake Scale to assess sleep quality, the Adolescent Sleep Hygiene Scale to assess sleep facilitation and sleep inhibiting behaviors, and the Pre-Sleep Arousal Scale to assess cognitive and somatic arousal symptoms prior to sleep onset. Adolescents undergo 10 days of actigraphy monitoring (Actiwatch 64) to assess sleep wake patterns. Actigraphy variables will be averaged over 10 days including sleep duration, wake time after sleep onset (WASO) and sleep efficiency.
Participants to date include 54 adolescents, 12 to 18 years old (67% males, M age=15.5, SD=2.1), 27 with mild TBI and 27 healthy adolescents. Assessments of adolescents with TBI were performed at a mean of 9 months after injury with a range of 5 to 12 months from injury date. 26 of the adolescents with TBI were reported to have a Glasgow Coma Scale (GCS) score of 15, one with a GCS score of 14. T tests revealed that adolescents with TBI reported worse sleep quality (M=3.46, SD 0.8) compared to healthy adolescents (M=4.39, SD 0.61, p=0.03). Adolescents with TBI also reported poorer sleep hygiene (M=4.6, SD 0.6) compared to healthy adolescents (M=4.8, SD 0.5, p=0.05). Additionally, adolescents with TBI reported experiencing higher levels of pre-sleep arousal (M=31.4, SD=10.5) compared to healthy adolescents (M=25.9, SD 6.8, p=0.03). Actigraphy measures were also significantly different across groups. The average sleep duration per night was 355 minutes (SD=63) for teens with TBI and 415 minutes (SD=33) for healthy teens (p<0.001). Adolescents with TBI displayed significantly higher mean WASO of 113 minutes (SD=50) compared to 63 minutes (SD=18) in the healthy group (p<0.001). Adolescents with TBI also displayed lower sleep efficiency of 74.5% (SD=10.5) compared to 84.6 % (M=4.8) in the healthy group (p<0.001).
Our preliminary findings suggest that sleep disturbances may be present at a higher rate after mild TBI compared to healthy adolescents. We are assessing sleep using a comprehensive battery of subjective and objective measures which demonstrate that during the first year after injury, sleep is affected in multiple ways. On subjective measures, poor sleep quality and sleep hygiene were reported. Moreover, actigraphy data suggest shorter sleep duration, frequent night wakings, and compromised sleep efficiency in youth with mild TBI. This suggests that the clinical management of youth with mild TBI should include a comprehensive assessment for sleep disturbances. Further analyses will include examining a predictive model of the role of clinical factors (e.g., presence of pain) and behavioral factors (e.g., depressive symptoms) on sleep patterns and behaviors. This will assist in identifying risk factors to target treatment to minimize the development and persistence of sleep disturbances in youth with mild TBI.
5T32GM086270-03 (PI: D. A. Schwinn) P30 NR011400 (PI: M. Heitkemper & M.V. Vitiello) RO1HD053431 (PI: T.M. Palermo)
Traumatic brain injury, sleep disturbances, adolescents
UNFAVORABLE OUTCOME IN ELDERLY PATIENTS FOLLOWING DECOMPRESSIVE CRANIECTOMY: A RETROSPECTIVE RISK-ADJUSTED ANALYSIS
Hester F. Lingsma, PhD, Department of Public Health, Erasmus MC, Rotterdam, Netherlands
Christine Martin, MSN, RN, San Francisco General Hospital, Brain and Spinal Injury Center Dept. of Neurological Surgery, University of California, San Francisco, USA
Atsuhiro Nakagawa, MD, PhD, San Francisco General Hospital, Brain and Spinal Injury Center Dept. of Neurological Surgery, University of California, San Francisco, USA
Adam R. Ferguson, PhD, San Francisco General Hospital, Brain and Spinal Injury Center Dept. of Neurological Surgery, University of California, San Francisco, USA
Geoffrey T. Manley, MD, PhD, San Francisco General Hospital, Brain and Spinal Injury Center Dept. of Neurological Surgery, University of California, San Francisco, USA
Despite the widespread use of decompressive hemicraniectomy (DHC) for the control of elevated intracranial pressure, the effect on outcome following traumatic brain injury (TBI) remains unknown. The therapeutic benefit of DHC in the elderly is particularly uncertain since they are usually excluded from clinical trials because age is a significant risk factor for poor outcome.
To evaluate the effect of DHC and standard craniotomy on outcome in elderly TBI patients using the IMPACT prediction model (http://www.tbi-impact.org) to adjust for known risk factors, including age. A cohort of elderly patients (defined as age>60 years old), with a Glasgow Coma Score (GCS) of 3 to 12, who underwent a DHC or craniotomy at a Level I Trauma Center were retrospectively identified. The IMPACT prediction model was used to adjust for covariates such as age, pupillary reactivity, motor score component of GCS, and CT findings. The predicted outcome (probability of mortality or unfavorable outcome at 6 months) for each patient was computed for each patient using the web-based IMPACT calculator. Predicted outcomes were compared to observed outcomes separately for DHC and craniotomy groups by dividing them into 5 equal groups based on their predicted probability.
We included 78 patients, 59 received (76%) received DHC and 19 (24%) received craniotomy. Both mortality and unfavorable outcome was highest in DHC group (81% vs 58%, p=.086 and 95% vs 74%, p=.008). When compared to the predictions from the IMPACT models, observed outcomes were worse than expected in DHC group (intercept for mortality=− 3.182 to 80.26, intercept for unfavorable outcome=28.52 to 83.82).
Although the retrospective nature of our study and small sample size allows only cautious conclusions, comparison of observed outcomes with risk-adjusted predictions from the IMPACT prognostic model does not support the use of DHC in elderly patients based on GCS alone. A larger prospective observational trial is needed to determine what factors, if any, favor the use of DHC versus standard craniotomy in this population.
Frederick Stephens, MD, Mathieu Laroche, MD
TBI, decompressive hemicraniectomy, IMPACT, elderly
STIMULATION OF NEURONAL RESPIRATION BY NEUROPROTECTIVE ALTERNATIVE BIOFUELS
Melissa Laird, PhD, University of Maryland School of Medicine
Pascaline Clerc, PhD, University of Maryland School of Medicine
Brian Polster, PhD, University of Maryland School of Medicine
Impaired cerebral energy metabolism contributes to the pathophysiology of brain injury. One approach toward neuroprotection is administration of potential alternative energy substrates (“biofuels”) that may overcome inhibition of glucose-based energy metabolism. This study tested the hypothesis that pyruvate, lactate, acetyl-L-carnitine, and β-hydroxybutyratestimulate neuronal O2 consumption in the absence or presence of excitotoxic glutamate.
Primary cultures of rat cortical neurons were prepared from embryonic day 18 fetuses and used at either day in vitro 14 or 7 for cell respirometry measurements using the Seahorse Bioscience metabolic flux analyzer. One hour prior to these measurements, normal culture medium was replaced with artificial cerebrospinal fluid containing 5 mM glucose. Maximal neuronal O2 consumption rate (OCR) was induced by the addition of the proton ionophore FCCP, which uncouples respiration from ATP synthesis. Comparisons were made of the maximal OCR with glucose alone and with glucose plus either pyruvate, lactate, acetyl-L-carnitine, or β-hydroxybutyrate at 1 – 10 mM. Additional experiments measured OCR after 30 min exposure of neurons to 100 μM glutamate to determine if alternative biofuels can overcome respiratory inhibition caused by excitotoxic glutamate.
Significant stimulation of uncoupled respiration by DIV14 neurons was observed with pyruvate at 1 – 10 mM, lactate at 2.5 - 10 mM, and β-hydroxybutyrate at 1 – 10 mM. Maximum stimulation was 60% for pyruvate and β-hydroxybutyrate and 45% for lactate. Respiration was not stimulated by the presence of 1 – 10 mM acetyl-L-carnitine. Similar results were obtained with DIV7 neurons except that lactate was less effective. In addition, maximal OCR for DIV7 neurons with glucose alone was significantly less (30%) than that of DIV14 neurons. Exposure of DIV14 neurons to an excitotoxic concentration of glutamate resulted in a 25% inhibition of the maximal OCR in the presence of glucose alone. Glutamate exposure also inhibited maximal respiration in the presence ofthe alternative fuels; however, rates obtained with 10 mM pyruvate or β-hydroxybutyrate were equal to or greater than that obtained with glucose alone in the absence of glutamate exposure. Glutamate exposure did not impair respiration by DIV7 neurons.
1. Exogenous pyruvate, β-hydroxybutyrate, and lactate at concentrations between 1 – 10 mM significantly stimulate maximal respiration by primary cultures of rat cortical neurons over and above the rate of respiration observed in the presence of 5 mM glucose. 2. Respiration on glucose alone is 30% lower for DIV7 compared to DIV14 neurons. 3. Transient exposure to an excitotoxic level of glutamate inhibits maximal glucose-based respiration by DIV14 but not DIV7 neurons. 4. Glutamate also inhibits respiration by DIV14 neurons supplemented with alternative biofuels; however, the levels of O2 consumption expressed in the presence of pyruvate or β-hydroxybutyrate are as high or higher than that observed with glucose alone in the absence of exposure to glutamate. 5. These results support the hypothesis that one mechanism of neuroprotection by pyruvate or ketone bodies, e.g., β-hydroxybutyrate is stimulation of aerobic neuronal energy metabolism.
This work was supported by NIH grants P01 HD16596 to G.F., R01 NS064978 to B.M.P. and T32 NS07275 to M.D.L.
Excitotoxicity pyruvate lactate β-hydroxybutyrate acetyl-L-carnitine
N-ACETYLCYSTEINE AMIDE (NACA) TREATMENT IMPROVED MITOCHONDRIAL BIOENERGETICS AND HINDLIMB FUNCTIONAL RECOVERY FOLLOWING CONTUSION SPINAL CORD INJURY
Jignesh Pandya, PhD, Spinal Cord & Brain Injury Research Center and Department of Anatomy & Neurobiology, University of Kentucky
Khalid Eldahan, Spinal Cord & Brain Injury Research Center, University of Kentucky
Dr. Patrick Sullivan, PhD, Spinal Cord & Brain Injury Research Center and Department of Anatomy & Neurobiology, University of Kentucky
Dr. Alexander Rabchevsky, PhD, Spinal Cord & Brain Injury Research Center and Department of Physiology, University of Kentucky
Mitochondrial dysfunction and oxidative stress after spinal cord injury (SCI) are linked to secondary pathophysiological events that directly contribute to ensuing histopathology. In this study, we hypothesized that treatment with a glutathione precursor, N-acetylcysteine amide (NACA), following SCI would foster neuroprotection by preventing mitochondrial dysfunction and reducing oxidative stress.
Adult female Sprague-Dawley rat spinal cords were contused at the L1/L2 spinal level (250 kdyn) using the IH impactor®. Vehicle (saline) or NACA (75, 150, 300 or 600 mg/kg) was administered (i.p.) 15 min post-injury, followed by a booster dose after 6 hrs. At 24 hr post-injury, total, synaptic and non-synaptic mitochondrial populations were each isolated simultaneously from injured spinal cords, as well as from naïve animals, to assess mitochondrial respiration using the Seahorse Biosciences XF24 Flux Analyzer®, as well as activities of NADH dehydrogenase (complex I), cytochrome oxidase (complex IV) and pyruvate dehydrogenase (PDHC) using spectrophotometry. In subsequent experiments, long-term hindlimb functional recovery was assessed using the maximum effective NACA dosage administered 15 min post-injury, along with osmotic pumps inserted subcutaneously to deliver NACA for one week.
Compared to naïve animals, SCI resulted in significantly compromised mitochondrial respiration in all three populations of mitochondria; notably, state III (ADP phosphorylation) and state V-complex I (complex I-driven maximum electron transport) respiration rates. Conversely, for total mitochondria, NACA treatments at 75, 150 and 300 mg/kg improved mitochondrial respiration in a dose-dependent manner, with maximum restoration (p<0.05) at 300 mg/kg, compared to vehicle. Critically, the highest 600 mg/kg NACA dosage did not improve respiration compared to vehicle. Moreover, for both synaptic and non-synaptic populations, only the 300 mg/kg NACA dosage significantly maintained respiration near normal levels. Similar to the respiration data, SCI significantly (p<0.05) decreased activities of all three enzyme complexes. For total mitochondria, NACA administration significantly (p<0.05) improved activities of complexes I, IV and PDHC compared to vehicle treatment. Alternatively, for synaptic and non-synaptic mitochondria, only the 300 mg/kg NACA dosage showed significantly (p<0.05) increased activities of all three enzyme complexes to near normal levels. Consequently, we assessed behavioral recovery following SCI with and without prolonged treatment with 300 mg/kg NACA. As early as one week post-injury, NACA-treated rats showed improved hindlimb function compared to saline. Critically, after 7 weeks, they demonstrated consistent plantar stepping with frequent forelimb-hindlimb coordination (∼13 BBB), whereas vehicle-treated injured rats demonstrated primarily dorsal and only occasional plantar stepping without coordination (∼10 BBB). Ongoing analyses include histological assessments of tissue sparing and kinematic analyses from the behavioral studies, in addition to quantitative assessments of glutathione levels and oxidative markers in mitochondria isolated from the acute experimental groups.
Collectively, our results show that acute administration of NACA at optimal dosages after severe contusion SCI significantly maintains bioenergetics in all three populations of mitochondria, and that prolonged continuous NACA delivery significantly improves long-term hindlimb functional recovery.
Supported by NIH/NINDS R01NS069633 (AGR&PGS), P30 NS051220, Craig H Neilsen Foundation 190115 (AGR), and a donation from Michael and Helen Schaffer Foundation (AGR).
neuroprotection, mitochondrial dysfunction, locomotor recovery
TOLL-LIKE RECEPTOR 2 AND DECTIN-1: TWO RECEPTORS WITH DICHOTOMOUS ROLES IN MACROPHAGE-MEDIATED REPAIR AND PATHOLOGY IN CNS TRAUMA
Zhen Guan, The Ohio State University
Kyle Beckwith, The Ohio State University
Bryan Brautigam, The Ohio State University
Phillip Popovich, PhD, The Ohio State University/Department of Neuroscience/Center for Brain and Spinal Cord Repair/Institute for Behavioral Medicine Research
Pro-regenerative and pathological macrophages are hallmarks of CNS trauma. The signaling pathways underlying this diversity are unknown but zymosan activation, through TLR2 and dectin-1, simulates these dichotomous macrophage functions. Here, we provide insight into macrophage-mediated repair and neuropathology by examining these receptors in the context of SCI.
To determine the consequences of different modes of macrophage activation, intraspinal microinjection of different agonists: zymosan (100nl; 10mg/ml), PAM2CSK4 (TLR2 agonist; 1ul; 1mg/ml), heat killed Saccharomyces cerevisiae (HKSC; dectin-1 agonist; 100nl; 108 cells/ml), zymosan depleted of TLR2 agonist (100nl; 10mg/ml) or vehicle (saline), were delivered to female Sprague Dawley rats (200–225 gm, Harlan, Indianapolis, IN). Lesion volume and macrophage activation was quantified using standard techniques. Similar agonists were delivered after dorsal column crush injuries (see Horn et al., 2008 PMID: 18799667) to determine the effects of macrophage activation on axon retraction. Dectin-1 expression and immunohistochemical distribution was examined after mid-thoracic mouse contusion injury. To further elucidate the role of different modes of activation on axon growth and cell death, supernatant from bone marrow derived macrophages were prepared and placed on neurons in vitro and cells analyzed as described in Gensel et al., (2009 PMID: 19321792).
All intraspinally delivered agonists elicited robust macrophage responses. Dectin-1 agonists (zymosan, zymosan depleted, HKSC) produced dense areas of macrophages associated with demyelination, axon loss, and retraction bulbs. The TLR2 agonist (PAM2CSK4) elicited a robust macrophage response associated with little frank pathology or axon loss. Quantifying the ratio of pathology to macrophage volume revealed significantly less pathology associated with TLR2 vs. other modes of activation.
To confirm the role macrophages are playing in pathology we tested the direct effects of TLR or dectin-1 activated macrophages on neuron survival by stimulating adult DRG neurons in vitro with conditioned medium (CM) from stimulated macrophages. As reported previously, prolonged exposure of adult DRG neurons to zymosan-activated macrophage CM caused significant cell loss and axon degeneration. Similar effects were detected with CM from agonists activating dectin-1. Conversely, neurons exposed CM from TLR2 stimulated macrophages had no identifiable pathology.
We utilized a dorsal column crush injury model that causes macrophage-mediated axon retraction to test whether stimulating macrophages could affect axon growth in vivo. Two days after SCI, we intraspinally microinjected TLR2 or dectin-1 agonists into the injury site. Axon retraction was significantly reduced in spinal cords enriched with TLR2, but not dectin-1, activated macrophages.
To determine if dectin-1 could be playing a role in pathogenesis after SCI we examined temporal changes in dectin-1 expression. PCR and immunohistochemical analysis revealed a significant and maintained increase in dectin-1 by 14 after SCI. Dectin-1 labeling was examined in bone-marrow chimeras, for which we could differentiate between monocyte and microglia-derived macrophages, and found to be expressed by both monocyte and microglia-derived macrophages. Dectin-1 clusters were detected suggesting the receptor is being activated after SCI. We are conducting loss of function experiments to determine if dectin-1 activation contributes to SCI pathology.
Here, we demonstrate that TLR2 and dectin-1 receptor pathways are activated after CNS trauma and contribute to pathogenesis and wound repair. Specifically, activating TLR2 after spinal cord injury blunts the detrimental effects of endogenous macrophage activation on axon dieback. This is consistent with our previous observations after SCI; TLR2 KO mice have worse recovery and increased lesion size compared to wild-type controls (Kigerl et al., 2007 PMID: 17403033). Conversely, endogenous dectin-1 macrophage activation causes macrophage-mediated pathology in the CNS. Collectively, these data provide novel insight into macrophage-mediated repair mechanisms after CNS trauma that can be harnessed for therapeutic development.
Supported by The Craig H. Neilsen Foundation (JCG), American Academy of Neurology (KAB), NINDS NS037846 (PGP) and The Ray W. Poppleton Endowment (PGP).
microglia, regeneration, axon, macrophage, neurotrauma
IMAGING INFLAMMATION AND WHITE MATTER INTEGRITY IN MOUSE SCI
Joong Hee Kim, PH.D., Washington Univsersity in St. Louis
Ms. Xiaojie Wang, B.S., Washington University
The contusion injury induced inflammatory responses play a crucial role for secondary degeneration exacerbating axonal and neuronal damage. As development of anti-inflammatory and axonal preservation therapies, there is a need for non-invasive and quantitative assessment of both axonal integrity and inflammatory response guiding therapies and predicting neurological outcomes.
18 female C57BL/6 mice weighing 20 – 22 g received sham, mild, and severe dorsal-to-ventral contusion injury at T9, n=6 per group. Randomly selected 3 animals per study group were perfusion fixed using PFA at 3 days post injury (DPI) where the others underwent behavior assessment until 14 DPI and were perfusion fixed. Ex vivo diffusion magnetic resonance imaging was performed on the harvested spine analyzed using (a) conventional diffusion tensor imaging (DTI), and (b) diffusion basis spectrum imaging (DBSI, a newly developed novel imaging modality in our lab modeling diffusion MRI signal as a linear combination of isotropic and anisotropic tensor components). The spared white matter was segmented by axial diffusivity derived using DTI and DBSI. The DTI/DBSI determined spared white matter content was compared with phosphorylated neurofilament positive (SMI-31) immunohistochemical (IHC) staining.
The ex vivo DBSI at 3 and 14 DPI produced signal ratio maps for cellularity, extracellular water, and fiber, reflecting inflammatory cell, vasogenic edema, and spared white matter respectively. The DBSI derived parameter maps exhibited high spatial resolution as those of the conventional DTI allowing quantitative analysis of white matter pathologies. The ex vivo DBSI derived cell and extra cellular water ratio map at 3 DPI increased extensively near the central canal. The fiber ratio was relatively high at the peripheral region of ventrolateral white matter suggesting peripheral preservation, consistent with the injury pattern of dorsal-to-ventral contusion. The DBSI derived cell and extra cellular water ratio at 3 DPI correlated positively and linearly with injury severity and BMS at 14 DPI. Both DTI and DBSI estimated spared white matter, based on axial diffusivity, at 3 DPI linearly correlated with injury severity (negative) and BMS at 14 DPI (positive). However, there was a non-negligible discrepancy in the estimated extent of spared white matter extent between DTI and DBSI. The DTI and DBSI determined spared white matter at 14 DPI (endpoint) was compared to SMI-31 on a pixel-to-pixel basis. A non-negligible discrepancy between axial diffusivity determined (for both DTI and DBSI) spared white matter extent and that determined by SMI-31 was observed. DBSI estimated extent of spared white matter was closer to that estimated by SMI-31. The spared white matter extent estimated by combining DBSI derived fiber ratio with DBSI derived fiber axial diffusivity matched that determined by SMI-31.
In this study, the trauma induced inflammatory cellular response, vasogenic edema, and axonal integrity was examined using conventional DTI and the newly developed DBSI. DBSI derived parameters allowed quantitative analysis of inflammatory response that was not available by conventional DTI. The DBSI derived cell and extracellular water ratio showed potential to be used as biomarker of inflammation. Sub-acute to chronic DTI derived axial diffusivity failed to reflect axonal status in SCI. The DBSI derived fiber ratio or fiber axial diffusivity alone may not be sufficient to reflect axonal integrity in chronic SCI. However, the combination of fiber ratio and axial diffusivity accurately detected spared white matter as determined by SMI-31. In vivo DBSI has recently been performed on EAE mice reflecting pathologies identified by IHC. Human translation of DBSI has also been pursued. The in vivo application of DBSI in SCI will be feasible in the near future.
This study is supported by NIH R01 NS 047592.
Inflammation, Cellularity, Vasogenic edema, Biomarker, Axonal injury, Spinal cord injury
FASL MEDIATES THE DISRUPTION OF BLOOD-SPINAL CORD BARRIER AND INFLAMMATORY RESPONSE AFTER SPINAL CORD INJURY
Professor Michael G. Fehlings, MD, PhD, Toronto Western Research institute
The disruption of the blood spinal cord barrier (BSCB) following spinal cord injury (SCI) potentiates the influx of inflammatory mediators and recruitment of circulating immune cells into the spinal cord parenchyma. A better understanding of the mechanisms relating BSCB disruption and inflammation following SCI could lead to novel therapeutic strategies.
We examined extravascular expression of fibronectin as a measure of BSCB permeability, the expression of cytokines/chemokines using confocal immunohistchemistry and counted the number of Von Willebrand Factor (VWF) positive vessels in post-mortem human postmortem spinal cord tissue from acute and chronic SCI cases and ten control cases. Complementary studies were conducted using the in vivo Fejota clip compression model of SCI in FasL-deficient B6Smn.C3-Tnfsf6gld/J (gld) and wild-type (Wt) mice. We used this model to test FasL-mediated BSCB disruption and inflammation by immunohistochemistry, Western blotting, and ELISA with Mouse 32-plex cytokine/chemokine panel bead immunoassay.
We report novel evidence that shows that increased levels of fibronectin were present outside of blood vessels and increased IL-1, IL-10 and CX3CR1 expression occurred in the injury epicenter in cases of acute and subacute human SCI but not in chronic SCI or in control cases. We also found significantly reduced expression of fibronectin at 3 days post injury and increased expression of fibronectin at 7 and 14 days post injury in gld mice when compared to Wt mice after SCI. We showed significantly increased numbers of CD4 positive T cells and increased levels of G-CSF, IL-7, MIP and IL-6 expression. Levels of IL-2, M-CSF, MCP-1, IP-10, MIG, and IL-12(40) expression were reduced in B6Smn.C3-Tnfsf6gld/J when compared to Wt mice after SCI.
We report multiple lines of evidence that indicate that FasL activation plays a pivotal role in mediating BSCB disruption, the inflammatory response and neurodegeneration after SCI. These data provided a compelling rationale for therapeutically targeting Fas/FasL in human SCI.
This work was supported by the Krembil Chair in Neural Repair and Regeneration.
FasL, BSCB disruption, inflammation, SCI
SERUM MICRORNAS AS BIOMARKER FOR BLAST INDUCED TRAUMATIC BRAIN INJURY
Nagaraja Balakathiresan, PhD, USUHS
Raghavendar Chandran, MS, USUHS
Mikulas Chavko, PhD, Naval Medical Research Center
Richard McCarron, PhD, Naval Medical Research Center
Radha Maheshwari, PhD, Dep of Pathology, USUHS
Blast induced Traumatic brain Injury (bTBI) is termed as the signature injury of war. The diagnosis of bTBI is difficult due to limitations of the imaging techniques and lack of clinically approved biomarkers. In this study we have studied the potential of serum microRNAs as biomarker for diagnosis of TBI.
We have studied the effect of blast overpressure injury (BOP) on the miRNA modulation in the serum and cerebrospinal fluid (CSF) of rats. Two injury groups were used to analyze the short and long term effect of injury on miRNA expression. Short interval injury (SII) group were exposed to 3 serial blast exposures of 120kPa using a shockwave tube at an interval of 3hr whereas in long interval injury (LII) a 24hr time was allowed between successive exposures. Blood and cerebrospinal fluid were collected at various time points after the injury. Total RNA was isolated; reverse transcribed using stem loop primers and amplified. MiRNA modulation was analyzed using real time PCR using Taqman low density array cards according to manufacturer's protocol. Candidate miRNAs were validated using specific primers by individual real time PCR. Functional and network analysis of the modulated miRNAs was done using Ingenuity pathway analysis.
Serum samples were analyzed for miRNA modulation in response to BOP injury. In case of SII, 123 and 75 miRNAs were dysregulated in samples collected at 3hr and 24hr post-BOP injury respectively. In case of LII, 17 and 19 miRNAs were dysregulated in serum samples at 3hr and 24hr of last BOP injury respectively. Five miRNAs were selected for analysis which includes miR-let-7i, miR-122, miR-340-5p, miR-200b* and miR-872 since these miRNAs were modulated both injury groups. Among these members of let-7 miRNA family are highly enriched in brain specifically in hippocampus and frontal cortex. Hence miR-let-7i was selected for further validation. Specific PCR was performed to avoid any bias introduced by pre-amplification step in the previous screening. We found that miR-let-7i expression was upregulated in SII by more than 10 folds in both the sub groups. In case of LII, no significant change in the expression of miR-let-7i was observed. Further, CSF samples from the same group of animals were analyzed using specific real time miRNA assay for miR-let-7i. An over expression of miR-let-7i in CSF samples in SII group by more than 8 and 5 fold was observed (p<0.05) whereas in LII only 3hr injury group showed an increase of 2 folds and the group with injury 24hr apart and sample collected after 24hr last BOP exposure did not show any significant increase over the control samples. Pathway analysis using ingenuity pathway analysis indicates a potential role of microRNA let7i in regulating important proteins implicated in TBI such as UCH-L1, S100b and inflammatory cytokines. These data suggest that microRNA let7i may be an important regulatory molecule during TBI and hence can be used as a biomarker for diagnosis.
We have for the first time shown modulation of miRNAs in serum and CSF in response to BOP injury. Specifically, miR- let-7i appeared at elevated levels in the serum as early as 3hr post injury and was up regulated in SII. Moreover, we found the elevated levels of miR-let-7i in the CSF samples in SII whereas a modest upregulation was also observed in LII. Together, this indicate that miR-let-7i can be used a biomarker for an early detection of blast induced TBI. Validation of its role may lead to new therapeutic interventions for TBI involving miR-let-7i.
This work was supported by funding from Defense Medical Research and Development Program.
TBI, Blast, Biomarker, microRNA, Serum
MICROVASCULAR SHUNT FLOW AFTER TRAUMATIC BRAIN INJURY (TBI) WITH INTRACRANIAL HYPERTENSION IN RATS
Gloria Statom, B.S., Department of Neurosurgery, University of New Mexico School of Medicine
Edwin M. Nemoto, Ph.D., Department of Neurosurgery, University of New Mexico School of Medicine
We recently showed that intracranial hypertension in the uninjured rat brain induces a transition from capillary to microvascular shunt (MVS) flow resulting in hypoxia, edema and increased blood brain barrier (BBB) permeability. We now examine the role of intracranial hypertension-induced MVS flow in the traumatized brain.
Using in vivo 2-photon laser scanning microscopy through a cranial window over parietal cortex in rats, we measured microvascular red blood cell flow velocity (fluorescein dextran), tissue oxygenation (NADH autofluorescence) and BBB integrity (fluorescein dye leakage). Doppler cortical flow, rectal and cranial temperatures, intracranial pressure (ICP), arterial pressure and arterial blood gases were monitored. Brain edema and lesion volume were measured by proton MRI. Fluid-percussion injury (FPI) induced TBI was optimized to induce a stable increase in ICP without cortical hemorrhage under the optical window. After baseline recording, TBI was induced by a lateral fluid-percussion transient pulse (1.5 ATA, 100 ms) with a custom built gas-driven device on the left hemisphere through a 5 mm craniotomy. Physiological variables were continuously monitored for four hours after TBI.
Fluid percussion injury resulted in a sustained, significant (P<0.05), increase in ICP to 30.8±4.7 mmHg (mean±SEM, n=7) above the pre-injury level of 10.4±3.6 mmHg. Arterial pressure was unaltered over 4 hours. Posttraumatic increase in ICP resulted in a significant (P<0.01) redistribution of microvascular flow from low velocity capillary to high velocity microvascular shunt flow reflected by a decrease in the capillary/shunt flow ratio. The capillary/shunt flow ratio decreased from 1.83±0.15 before injury to 0.92±0.25, 0.92±0.21, 0.77±0.16, and 0.68±0.20 and 0.74±0.23 at 0, 1, 2, 3 and 4 hours, respectively. These changes were accompanied with marked NADH reduction to 59.32±9.2% from baseline (P<0.01) reflecting tissue hypoxia; and with fluorescein dye extravasation reflecting BBB degradation. Proton MRI showed tissue damage in the left hemisphere. Cortical water content increased after injury from 77.6±0.35 to 80.7±0.41% and in the hippocampus, from 75.4±0.38 to 77.6±0.36%, respectively.
Our studies show that post-traumatic intracranial hypertension induces a transition from capillary to MVS flow as previously observed with high ICP in the uninjured rat brain. Importantly, this transition to MVS flow was associated with tissue hypoxia and BBB opening; characteristic of non-nutritive MVS flow. Tissue edema and intracranial hypertension results in the transition from capillary to microvascular shunt flow by virtue of increased capillary resistance which shunts flow through the lower resistance microvascular shunts. The increase in cerebral venous pressure secondary to an increase in ICP, results in an increase in capillary resistance and decrease in capillary flow. As a result, precapillary arteriole dilation occurs increasing capillary pressure and decreasing the arteriole-venular pressure gradient with higher capillary and venous pressures. The transition from capillary to MVS flow may be used to better define the critical cerebral perfusion pressure in the management of high ICP.
This work was supported by the NIH CoBRE P30 Pilot Project (8P30GM103400-01) and Pilot Project Grant from the UNM SOM Dedicated Health Research Funds.
intracranial hypertension, microvascular shunts, edema
THE EFFECTS OF TBI ON HIPPOCAMPAL SNARE-COMPLEX ASSEMBLY AND ATTENUATION BY CHRONIC LITHIUM THERAPY
Hong Q. Yan, M.D., University of Pittsburgh/School of Medicine/Neurosurgery
Larry W. Jenkins, Ph.D., University of Pittsburgh/School of Medicine/Neurosurgery
Xiecheng Ma, M.D., University of Pittsburgh/School of Medicine/Neurosurgery
Youming Li, University of Pittsburgh/School of Medicine/Neurosurgery
Samuel S. Shin, Ph.D., University of Pittsburgh/School of Medicine/Center for Neuroscience, and Neurosurgery
C. Edward Dixon, Ph.D., University of Pittsburgh/School of Medicine/Neurosurgery and VA Pittsburgh Health System
The pathology of soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex proteins may play an important role in TBI-induced neurotransmission disturbances. We examined the effects of TBI on SNARE complexes and rescue by lithium which promotes SNARE-complex assembly by enhancing cysteine string protein alpha (CSPα) expression.
Male Sprague-Dawley rats were anesthetized and surgically prepared for controlled cortical impact (CCI) injury (4 m/sec, 2.6 mm deformation) or sham surgery and sacrificed at six hours, one day, one week, two weeks and four weeks post-injury (N=6). Ipsilateral hippocampal sections were assessed using Western blot analysis after tissue was either boiled or un-boiled to identify high molecular weight complexes and the effect of TBI on these complexes. In separate groups, injured rats were treated with lithium (1mmol/kg, i.p.) daily and sacrificed at 7 days post injury. In both studies monomer and higher molecular weight complexes of syntaxin, SNAP-25, and CSPα were measured by Western blots and semi-quantified using NIH ImageJ software.
At six hours, one day, one week, two weeks and four weeks post-injury, syntaxin complexe levels were respectively decreased 20.1%, 0.2%, 46%, 54.1%, and 13.2% compared to time-matched shams. SNAP-25 complex levels were increased 32.2% and 23.1% at 6 hours and 1 day and decreased 33.4%, 44.9%, and 15% at 1 week, 2 weeks, and 4 weeks after CCI compared to time-matched shams. After daily lithium treatment, SNAP-25 and syntaxin complex reductions were partially attenuated at one week after TBI. At 1 week after CCI, there was a significant decrease in hippocampal CSPα levels to 43.5±10% of sham controls. Lithium attenuated the loss of CSPα to 69.6±12% of sham controls.
This is the first study to demonstrate a change in high molecular weight SNARE protein complexes after experimental TBI. Reductions in syntaxin and SNAP-25 complexes were greatest at one and two weeks after TBI. A reduction in the ability of form SNARE complexes may contribute to impaired vesicular docking and subsequent neurotransmitter release deficits after TBI. CSPα promotes SNARE-complex assembly by chaperoning SNAP-25 during synaptic activity. It has been shown that lithium can enhance the expression of CSPα. The present study indicates that lithium can attenuate the loss of SNARE protein complexes within the hippocampus after TBI. This occurs with a concominent lithium-induced attenuation of CSPα after TBI. Further work is warranted to evaluate the role of SNARE-proteins in post-injury morbidity and to determine the therapeutic potential of lithium via both classic (i.e. GSK3β deactivation, BDNF, etc.) and novel (CSPα upregulation) mechanisms of action.
Support: NIH/5R01NS060672, NIH/1F30NS067731, VA/RR&D #B6761R
TBI; vesicles; lithium; hippocampus
CROSS MODEL COMPARISON OF BEHAVIOR, NEUROPATHOLOGY, AND SERUM BIOMARKERS AFTER CONTROLLED CORTICAL IMPACT, PARASAGITTAL FLUID PERCUSSION, AND PENETRATING BALLISTIC-LIKE BRAIN INJURY: RESULTS FROM OPERATION BRAIN TRAUMA THERAPY
Helen Bramlett, PhD, University of Miami Miller School of Medicine
C. Edward Dixon, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Deborah A Shear, PhD, Walter Reed Army Institute of Research
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Kara Schmid, PhD, Walter Reed Army Institute of Research
W. Dalton Dietrich, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Kevin Wang, PhD , Department of Psychiatry, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
Frank C Tortella, PhD, Walter Reed Army Institute of Research
Operation Brain Trauma Therapy (OBTT) is a unique multi-center pre-clinical drug-screening consortium and the first systematic multi-center cross-model comparison in the TBI field. Hypothesis: controlled cortical impact (CCI), fluid percussion (FPI), and penetrating ballistic-like brain injury (PBBI) yield differences in neuropathology and biomarker profiles at injury levels generating behavioral deficits.
In all three models, male, adult Sprague-Dawley rats (total n=140) were subjected to TBI using the established protocol at each center. Rats studied in the FPI model also had indwelling arterial catheters to monitor acute effects of drug administration on MAP and blood gases, and in each site, body temperature was controlled. In each model, established outcomes were evaluated including various motor and cognitive tasks over 21d. The motor tasks assessments were completed by d10 at all sites. Subsequently, Morris water maze (MWM) was used to assess cognitive function, including a hidden platform paradigm and probe trial across sites. At 21d, rats were sacrificed and in each model serial brain sections were assessed for lesion volume and either hemispheric or cortical tissue volume loss. Serum was obtained at 4h post-TBI to quantify initial levels of UCHL-1 and GFAP as assessments of neuronal and astrocyte damage resulting from primary injury, respectively.
Motor testing revealed deficits in 1) beam walking and beam balance tasks after CCI, 2) rotarod performance after PBBI, and 3) cylinder task performance after FPI (all P<0.05 vs. sham). Average latency to find the hidden platform (across all testing days) on the MWM paradigm was increased 19±12%, 121±13%, and 18±14% of sham in CCI, PBBI, and FPI, respectively (P<0.05 vs. sham in CCI and PBBI). Lesion volume was 11.18±1.16%, and 15.01±2.38% of contralateral hemisphere in CCI, and PBBI, respectively and 2.84±0.57% of the contralateral cortex in FPI. Lesion volume in both CCI and PBBI were greater than FPI, P<0.05. Surprisingly, serum UCHL-1 levels at 4h after injury were highest after FPI (P<0.004 vs. CCI and P<0.001 vs. PBBI) while serum GFAP levels at 4h were highest after CCI (P<0.0001 vs. FPI), intermediate in PBBI (P<0.0003 vs. FPI), but also modestly increased in FPI (P=0.04 vs. sham).
We conclude that CCI, FPI, and PBBI produced distinct functional deficits, neuropathology, and early biomarker profiles. Our findings support the concept that cellular injury mechanisms and therapeutic targets may differ across TBI models and/or injury severities suggesting that multi-model therapeutic screening could provide unique insight into therapy development for clinical translation.
We thank the United States Army (W81XWH-10-1-0623) for support.
traumatic brain injury, neuroprotection, consortium
THE ROLE OF THE BLOOD-CEREBROSPINAL FLUID BARRIER IN POST-TRAUMATIC INVASION OF INFLAMMATORY CELLS
Jessica R. Gandy, B.Sc., Brown University
Andrew Varone, B.Sc., Brown University
Nathalie Strazielle, Ph.D., INSERM
Brian J. Zink, M.D., Department of Emergency Medicine, Alpert Medical School of Brown University
Jean-François Ghersi-Egea, Ph.D., INSERM
Adam Chodobski, Ph.D., Department of Emergency Medicine, Alpert Medical School of Brown University
The blood-brain barrier is the predominant route for leukocyte invasion occurring after TBI; however, increasing evidence suggests that the blood-CSF barrier also plays an important role in post-traumatic neuroinflammation. This study defines a role of this latter barrier in recruitment of neutrophils and monocytes to the injured brain.
The major site for the blood-CSF barrier is the choroid plexus, a highly vascularized organ located in cerebral ventricles. Our analysis focused on the lateral ventricle choroid plexus. The controlled cortical impact model of TBI in rats was used. Post-traumatic changes in choroidal expression of neutrophil and monocyte chemoattractants were analyzed by real-time RT-PCR and Western blotting. CSF was sampled from the cisterna magna and the chemokine levels were measured by ELISA. Immunohistochemistry was used to characterize the distribution of chemokines in the choroidal tissue, and transmission electron microscopic (TEM) analysis was performed to track the movement of inflammatory cells across the blood-CSF barrier. An in vitro model of the blood-CSF barrier was used to characterize the kinetics and polarity of chemokine secretion by the choroidal epithelium.
RT-PCR analysis demonstrated a gradual and transient increase in synthesis of neutrophil (CXCL1–3) and monocyte (CCL2) chemoattractants predominantly in the ipsilateral choroid plexus peaking at 6 h after injury and then dropping rather abruptly at 1 d post-TBI. No changes in chemokine synthesis in the contralateral choroid plexus, except a transient increase in CXCL3 expression early after injury, were observed. This increase in choroidal chemokine synthesis was followed by a significant increase in their CSF levels (e.g., 50-fold elevation of CCL2 concentration at 6 h after TBI compared to sham-injured animals). Immunohistochemical analysis showed the intense Golgi-associated staining for CXC and CC chemokines in clusters of epithelial cells scattered through the ipsilateral choroid plexus. The chemokine-immunopositive staining was not associated with other types of cells normally present in the choroidal tissue, such as endothelial and epiplexus cells or stromal macrophages. The localization of chemokines to the Golgi complex was consistent with increased synthesis of these secreted proteins observed after TBI. To characterize the kinetics of chemokine synthesis by the choroid plexus epithelium and to determine the direction of chemokine release (across the apical or CSF-facing versus basolateral or blood-facing membrane), an in vitro polarized model of the blood-CSF barrier, which reproduces the in vivo properties of choroidal epithelium, was used. Both CXC and CC chemokines were released across the apical and basolateral membranes of the choroidal epithelium, a pattern of chemokine secretion that is a prerequisite for leukocyte migration across epithelial barriers. The TEM analysis of the ipsilateral choroid plexus showed that neutrophils and monocytes were able to reach the intercellular space between the epithelial cells and were apparently moving, occasionally in tandem, toward the apical domain of choroidal epithelium.The movement of these inflammatory cells between epithelial cells did not appear to affect the integrity of tight junctions.
This is the first detailed analysis of the blood-CSF barrier function related to trafficking of leukocytes. Our results provide evidence for the pathophysiological role of this barrier in neuroinflammation observed after TBI and further support the involvement of CSF space in post-traumatic invasion of inflammatory cells. Thus, post-traumatic dysfunction of both the blood-CSF barrier and the blood-brain barrier residing in pial microvessels may synergistically promote leukocyte accumulation in the CSF space from where these inflammatory cells can invade the traumatized brain parenchyma.
Supported by grants NS49479 (NIH) and HEALTH-F2-2009-241778 (EU), and by funds from the Department of Emergency Medicine at the Alpert Medical School of Brown University.
TBI, brain barriers, neuroinflamation
Poster Presentation Sessions
NEUROPROTECTION BY GENIPIN AGAINST FREE-RADICAL MEDIATED INJURY IN ORGANOTYPIC HIPPOCAMPAL SLICE CULTURES
Victoria A. Silva, Columbia University
Ijaz Ahmed, Ph.D., Rutgers University
David I. Shreiber, Ph.D., Rutgers University
Barclay Morrison III., Ph.D., Columbia University
Genipin, from traditional Chinese medicine, has been shown to be a multi-potent neuroprotective agent, but its mechanisms of protection are poorly understood. In the current study, genipin reduced cell loss in a model of oxidative damage using organotypic hippocampal slice cultures (OHSC).
Hippocampi of post natal day 9–10 rat pups were removed aseptically, cut into 400 um thick sections, and plated on Millipore membranes. Cultures were maintained for 11 days under standard conditions (37°C and 5% CO2) prior to injury with tert-butyl hydroperoxide (tBHP). Slices were incubated in 1 mM tBHP for 24 hours and simultaneously treated with 50 uM genipin or vehicle (0.1% DMSO). At 24h tBHP was removed, and slices were treated with fresh genipin or vehicle. Control cultures received fresh medium at the appropriate time points (0 and 24h). In a 1 hour delay treatment paradigm, treatment with 50 uM genipin or vehicle began 1h after tBHP injury onset. Slices were imaged with Sytox Green at 0, 24, and 48h to calculate cell death as the percent area staining above a threshold within a given anatomical region.
When administered at the same time as tBHP, genipin significantly reduced cell death (p<0.05) measured 48h after injury onset by 66.7% compared to vehicle-treated cultures. When genipin administration was delayed 1h after tBHP injury onset, it significantly reduced cell death by 48.7% compared to vehicle-treated cultures.
Our results indicate that genipin was protective against free-radical mediated damage in OHSCs. Treatment with genipin was effective even when delayed 1 hour after injury onset, producing a similar level of protection when treatment was initiated at injury-onset. These results suggest that genipin may hold promise as a therapy for traumatic brain injury since a clinically viable treatment must be effective when treatment is delayed. Previous studies using rat brain homogenates or dissociated cell-cultures suggest that genipin does not act as a direct antioxidant, but confers neuroprotection by inhibiting enzymatic production of reactive oxygen species. In future studies, the mechanism of genipin protection will be investigated further using our OHSC model to identify which enzymes are inhibited.
This research was supported by the New Jersey Commission on Brain Injury Research Award #10-3215-BIR-E-0.
Genipin, Neuroprotection, Organotypic, ROS, Delay
EFFECT OF PERFLUOROCARBONS IN PRESERVING MEMBRANE INTEGRITY OF IN VITRO STRETCH INJURED POSTNATAL RAT NEUROGLIAL CULTURES
Shyam Gajavelli, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shani Jones, Bachelor of Neuroscience, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Natalie Aguirre, High School, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Helen M. Bramlett, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
In addition to improved oxygen delivery, there is possibility that Perfluorocarbon (PFC) may directly improve membrane integrity following TBI/stretch injury. We propose using the well-characterized and established in vitro mild TBI model. This model replicates the combination of mechanical stretch insults and subsequent ionic and metabolic changes seen in vivo.
Postnatal day 4 rat brains were dissected and cultured on special stretchable dishes. Adherent mixed cultures of neurons and astrocytes were subjected to mild mechanical neuronal injury by direct millisecond stretching (30% stretch) of the flexible silastic membrane using an air pulse. Immediately following stretch, cultures were exposed to in vitro oxygenated PFC. The following experimental groups were be employed: Group 1, uninjured control, the cultures were not stretched nor exposed to PFC. Group 2, the cultures were subjected to stretch but not exposed to PFC (injured vehicle). Group 3, immediately following stretch, cultures were exposed to 10 (v/v) oxygenated PFC. For each group, we used four culture wells as replicates. Twenty four hours after the stretch injury, cell death was assessed with ethidium homodimer live-dead assays. Cultures were counterstained with nuclear stain. Fixed cultures were imaged and dead cells identified by colocalization of ethidium homodimer and nuclear counterstain.
The number of dead cells in uninjured control cultures was only 1.67%. In contrast the number of dead cells in group 2 was 43%. PFC treated cultures showed a slightly fewer dead cells ∼30%.
Perfluorocarbon (PFC) appears to protect about third of the injured cells by directly improving cell membrane integrity.
These studies were made possible by support from CDMRP award PTO74521 (W81XWH-08-1-0419).
PFC, Live-death assay, in vitro
PROTECTIVE EFFECT OF TRH AGAINST GLUTAMATE TOXICITY IN HIPPOCAMPAL NEURONS
Professor Deborah Watson, PhD, University of Pennsylvania
Thyrotropin Releasing Hormone (TRH) is a neuropeptide that regulates the TSH-thyroid axis. TRH has shown neuroprotective effect in different CNS injury models and many neurodegenerative disorders. The goal of this study is to identify the signaling pathway mediating the neuroprotective effect of TRH in hippocampal neurons.
Primary hippocampal neurons derived from embryonic day 17 Sprague Dawley rat were cultured for 11 days in vitro (DIV) and transduced with AAV2/1-TRH vector for one week. Neurons were then treated with 30μm glutamate for 24 hours followed by cytotoxicity assay. Lactic acid dehydrogenase (LDH) release from TRH transduced neurons were measured and compared with AAV2/1 vector control. For TRH peptide treatment, neurons were cultured for 3 weeks in vitro and then treated with different concentration of TRH for 24 hours followed by100μm glutamate treatment for 24 hours. Western blot was used to detect the expression level of TRH, tubulin and Akt. The activity of Akt was measured using antibody against phospho-AKT (Serine 473). Intracellular calcium concentration was measured with Fura-2 calcium imaging.
Hippocampal neurons transduced with AAV2/1-TRH vector showed increased expression of TRH protein, no change was found for tubulin protein after one week transduction. Compared with AAV2/1 vector control, TRH transduction significantly decreased glutamate induced toxicity, demonstrated by decreased LDH release, suggesting the neuroprotective effect of TRH in hippocampal neurons. TRH neuroprotection was further confirmed by TRH peptide treatment. We found TRH (50ng/ml and 100ng/ml) pretreatment for 24 hours significantly attenuated glutamate induced LDH release. TRH pretreatment for 30 min or administered 30 min after glutamate treatment had no effect on LDH release. In order to investigate the signaling pathway mediating the neuroprotective effect of TRH, inhibitors of pro-survival (PI3-K/Akt and MAPK/ERK1/2) pathway were tested. We found inhibitor of PI3-K (LY294002), but not MAPK/ERK1/2 (U0126) attenuated the neuroprotective effect of TRH against glutamate toxicity. In order to confirm the involvement of PI3-K/Akt pathway in TRH neuroprotection, we measured the activity of Akt by looking at the level of phospho-Akt, an active form of Akt. We found TRH (100ng/ml) upregulated the level of phospho-Akt (Serine 473) in a time-dependent manner, with no change found in total Akt protein. Further study showed that new protein synthesis is needed for TRH neuroprotection as protein synthesis inhibitor cycloheximide (10μg/ml and 100μg/ml) inhibited the protective effect of TRH against glutamate toxicity. In order to know whether TRH protects neuronal cell death by inhibiting glutamate induced calcium influx, we performed Fura-2 calcium imaging on hippocampal neurons. Treatment with 100μm glutamate and glycine elevated intracellular calcium level in hippocampal neurons. However, the presence of TRH (100ng/ml) had no effect on glutamate induced calcium influx. TRH itself doesn't affect intracellular level of calcium.
TRH protects hippocampal neurons from glutamate toxicity. This effect was achieved both by TRH transduction and TRH peptide given exogenously over 24 hour period of time. TRH neuroprotection is mediated by PI3-K/Akt pathway and needs new protein synthesis. Reduced calcium influx doesn't contribute to the neuroprotective effect of TRH.
This work was supported in part by a Brain Injury Research Training Grant NRSA fellowship on NIH-T32-NS043126 (To Y.D.), and NIH-RO1-NS040978.
TRH, glutamate toxicity, PI3-K/Akt pathway
DECREASED SYNAPTIC AMPA RECEPTOR LOCALIZATION AND ALTERED AMYLOID BETA (Aß) DYNAMICS FOLLOWING MECHANICAL INJURY IN CORTICAL NEURONS
Carrie F. Ferrario, Ph.D., University of Michigan
Jack Hunt, B.A., University of Michigan
Margaret Ragsdale, B.S., University of Michigan
Enming J. Su, Ph.D., University of Michigan
Daniel Lawrence, Ph.D., University of Michigan
Henry L. Paulson, M.D., Ph.D., University of Michigan
Leslie S. Satin, Ph.D., University of Michigan
TBI-induced deficits in cortical processing are mediated, in part, by dysfunctional synaptic transmission, yet the cellular mechanisms underlying synaptic alteration following injury are unresolved. In addition, TBI is associated with altered Aß dynamics that may contribute to and/or reflect changes in neuronal activity and synaptic transmission.
Using whole-cell patch-clamp electrophysiology, we examined excitatory postsynaptic currents (EPSCs) in cultured cortical neurons following in vitro stretch-injury and in the prefrontal cortex (PFC) of brain slices prepared from animals subjected to controlled cortical impact (CCI). The effect of in vitro mechanical-injury on AMPAR subunit expression and extracellular Aß42 levels was examined in cortical cultures, using immunocytochemistry and ELISA assays, respectively.
We previously demonstrated reduced excitatory synaptic transmission in cortical pyramidal neurons following mechanical injury in vitro. The reduction in spontaneous and miniature EPSC amplitude occurs immediately post-injury, persists for at least 2 days, and reflects a reduction in the synaptic activity of GluA2-containing AMPARs. We observed a similar injury-induced reduction in EPSC amplitude in PFC layer II/III pyramidal neurons 2 days after CCI. Using immunocytochemistry to investigate whether injury affects AMPAR expression, we observed alterations in the spatial distribution of specific AMPAR subunits in cultured cortical pyramidal neurons after stretch-injury. Injury induced an early internalization of AMPARs and resulted in reduced synaptic localization with increased extrasynaptic expression of GluA2 at 2d post-injury. Reduced efficacy of excitatory transmission following injury would be expected to contribute to decreased neuronal activity, and decreased neuronal activity correlates with a reduction in CSF levels of Aß protein, including Aß42, in both human TBI patients and in animal models of TBI (Brody et al., 2008; Schwetye et al., 2010). We also found decreased levels of extracellular Aß42 in cortical cultures 2 days following mechanical injury in vitro.
Our data suggest that aberrant AMPAR trafficking to synapses contributes to the reduction in excitatory synaptic transmission that is observed following traumatic injury. The injury-induced reduction in excitatory synaptic activity likely contributes to altered excitatory/inhibitory balance and decreased neuronal activity, which in turn may hinder Aß42 protein secretion. We are continuing to explore the relationship between AMPAR function, excitatory neurotransmission and Aß42 dynamics following injury and are investigating whether the reduction in extracellular Aß42 is accompanied by increased intraneuronal accumulation of Aß42, which in turn may lead to synapse degeneration.
Supported by NINDS NS49519 to LS and a research grant from the Michigan Alzheimer's Disease Center.
AMPAR, amyloid beta, synaptic transmission
VALIDITY OF CORTICAL SOMATOSENSORY EVOKED POTENTIALS RESPONSES IN PREDICTING OUTCOME AFTER CARDIAC ARREST DURING HYPOTHERMIA COMPARED TO NORMOTHERMIA
Alexa Richie, Mayo Clinic Hospital in Jacksonville, FL
Michelle Freeman, Mayo Clinic Hospital in Jacksonville, FL
Kevin Barrett, Mayo Clinic Hospital in Jacksonville, FL
William Freeman, Mayo Clinic Hospital in Jacksonville, FL
We hypothesize that hypothermic cortical N20 SSEP responses assessed between 24 and 72 hours in post cardiac arrest survivors are predictive of outcome similar to normothermic data.
After IRB approval, we retrospectively reviewed all patients who had successful return of spontaneous circulation (ROSC) after cardiopulmonary arrest (CPA) in or out of the hospital, of any entry cardiac rhythm (VFIB, VTACH arrest, PEA or asystole). Patients were included if the SSEP data was obtained 1 to 3 days post CPA in normothermia (NT, Temperature 36 – 37°C) or mild-to-moderate hypothermia (HT, 32 – 35°C). Patients were excluded if they endured profound hypothermia, lower than 32°C, at the time SSEP data was obtained. Statistical comparison was made between NT and HT SSEP groups against outcomes which were defined by the Cerebral Pittsburgh Performance (CPC) outcome scale, where good outcomes indicated CPC=1 or 2, whereas poor outcome defined as CPC=3–5. Negative and positive predictive values (NPV and PPV) of SSEP, as well as sensitivity and specificity of SSEP between the two outcome groups were compared.
Over 6 years (2006 to 2012), we identified 91 cardiac arrests patient treated with therapeutic hypothermia at our hospital during the study period. Of those patients, 27 had reliable SSEP data, 2 patients being excluded due to technical difficulties. 10 SSEP data were during HT, and 17 during NT. Sensitivity, specificity, PPV, NPV during HT were 100%, 56%, 20% and 100% respectively, whereas during NT were 100%, 53%, 22% and 100%.
In this series of SSEP studies obtained for neurological prognosis after cardiac arrest, sensitivity for good prognosis was 100% in HT but had poor specificity (56%) similar to NT SSEP obtained data (53%). Although the data are small, similar results appear feasible when SSEP was obtained in hypothermia-treated patients compared to normothermic cardiac arrest patients. Technical factors limit prognostic accuracy due to limb edema and other factors. Larger multimodal prognostic models with clinical, biomarkers and SSEP should be considered in hypothermic treated cardiac arrest patients.
Mayo Clinic Hospital at Jacksonville, Florida
hypothermia, cardiac arrest, prognosis, SSEP
PROGESTERONE ATTENUATES INFARCT AND BEHAVIORAL DEFICITS IN MIDDLE-AGED RATS AFTER TRANSIENT ISCHEMIC STROKE: A DOSE-DEPENDENT STUDY
Fahim Atif, Ph.D, Emory University
Iqbal Sayeed, Ph.D, Emory University
Bushra Wali, Ph.D, Emory University
Fang Hua, M.D, Ph.D, Emory University
Huiling Tang, Ph.D, Emory University
Donald Gerald Stein, Ph.D, Emory University
Previous studies demonstrate the neuroprotective effects of progesterone (PROG) in numerous animal injury models, but a systematic dose-response study in an ischemic stroke model is lacking. We investigated the effects of PROG post-treatment on the severity of stroke-induced infarct and functional deficits in middle-aged rats.
Transient focal cerebral ischemia was induced in male middle-aged (12 months) Sprague-Dawley rats by right middle cerebral artery occlusion for 2 h followed by reperfusion. Rats received intraperitoneal injections of either PROG (8, 16, and 32 mg/kg) or vehicle (β-cyclodextrin) 2 h post-occlusion followed by subcutaneous injections at 6 h and then every 24 h post-injury for 7 days. Behaviors were evaluated at repeated intervals for 22 days on motor, sensory and cognitive tests (Rota-Rod, Grip Strength, Sensory Neglect, Morris Water Maze (MWM), and Gait Analysis). Cresyl violet staining was used to evaluate infarct size.
For grip-strength, rotarod and sensory neglect tests, repeated measures ANOVA showed significant group (F(4,44)=13.80, p<0.001; F(4,43)=2.69, p<0.05: F(4,36)=2.56, p<0.05) differences. Compared to vehicle, PROG treatment showed significant improvement in functional outcomes at all doses, but the 8 mg/kg dose led to maximum improvement. Performance on the spatial memory task (MWM) revealed that all PROG-treated groups spent more time spent in the platform quadrant compared to vehicle-treated rats; however, the 8 mg/kg dose showed maximum (82.30%) improvement in long-term memory compared to vehicle. Significant gait impairments were observed in the vehicle group compared to shams. Rats given PROG at 8 mg/kg showed maximum gait improvement in the affected limb, assessed by stand duration (76%), print length (72%) and maximum intensity (77%) compared to vehicle. Post-hoc analysis revealed that P8 and P16 showed significant attenuation in infarct volume compared to the vehicle group (1.93±0.39 and 1.61±0.63 vs. 11.58±3.76 respectively). PROG at 32 mg/kg did not show any effect on the infarct volume (4.19±1.4).
Stroke is the third most costly disease, leaving 70% of its survivors with severe functional disability. Inclusion of behavioral end-points in stroke studies enhances their clinical relevance, because for stroke patients maintenance or restoration of brain functions after injury is the prime objective. Our findings show that PROG (8 mg/kg) improves functional deficits and reduces the extent of infarct after ischemic injury in clinically relevant middle-aged rats. These results suggest that PROG warrants further testing in a clinical trial for ischemic stroke.
This work was supported by NIH/NINDS grant UO1NSO62676 to DGS, who also receives royalty payments/support from BHR Pharma for progesterone treatment in brain injury.
Stroke, Progesterone, Pharmacology, Behavior
EFFECTS OF GGF2 ON MICROSCOPIC CHANGES IN THE SKELETAL MUSCLE AND SCIATIC NERVE FOLLOWING PERIPHERAL NERVE INJURY IN RATS
Anindita Ganguly, PhD, Acorda Therapeutics
Corissa McEwen, MS, Acorda Therapeutics
Maya Srinivas, PhD, Acorda Therapeutics
Andrea Vecchione, BS, Acorda Therapeutics
Patrick Sarmiere, PhD, Acorda Therapeutics
Tao Hu, PhD, Acorda Therapeutics
Jennifer Iaci, MS, Acorda Therapeutics
Craig Hackett, PhD, Acorda Therapeutics
Donald Button, PhD, Acorda Therapeutics
Anthony Caggiano, MD, PhD, Acorda Therapeutics
Tom J. Parry, PhD, Acorda Therapeutics
As neuregulin-1 is known to impact Schwann cell function in peripheral nerves and have myogenic and metabolic effects in skeletal muscle, the effects of the neuregulin glial growth factor 2 (GGF2) on microscopic changes in both sciatic nerves and target muscles following sciatic nerve crush injury were evaluated.
Male rats underwent sciatic nerve crush or received sham surgery on the right hind limb. Animals with nerve crush were treated with a single bolus dose of GGF2 (2.6 mg/kg, iv) or vehicle 24 h after injury and then twice weekly for the duration of the study (2, 4 or 6 w). Animals were anesthetized with ketamine/xylazine and euthanized by transcardial perfusion of phosphate buffered saline followed by 4% paraformaldehyde perfusion. The lateral gastrocnemius muscle and the distal sciatic nerve were collected, embedded and sectioned. Muscle sections were stained with H&E and Masson's trichrome for morphometric evaluation of muscle fibrosis and myofiber cross sectional area distribution. Further, the distal sciatic nerves were embedded in plastic and sectioned. Separate sections were stained with H&E, luxol fast and toluidine blue to evaluate the extent of demyelination following crush injury.
Previous studies in our laboratory showed gastrocnemius muscle atrophy following crush and transection injuries compared to sham animals. Results from the current study showed increased muscle fibrosis with sciatic nerve crush injury over time. GGF2 significantly reduced muscle fibrosis at 4 and 6 weeks after injury. Additionally, the myofiber size distribution was skewed toward smaller cross sectional areas in crush-injured nerves compared to sham controls. GGF2 treatment promoted a shift in the myofiber area toward increased size (i.e. more like sham controls) at 4 and 6 weeks after injury. Preliminary analysis of injured distal nerve stumps showed that crush injury produced axonal demyelination at 2, 4 and 6 weeks post-injury. In addition, axonal swelling, Wallerian degeneration, mild inflammation and fibrosis between the axons occurred following injury. Axonal damage was noted in greater than half of the nerve bundle in cross section. Preliminary data suggest that GGF2 improves remyelination at 4 and 6 weeks in the injured distal sciatic nerve compared to vehicle-treated controls.
Although further studies are needed to examine the direct effects of GGF2 on sciatic nerve myelination as well as skeletal muscle fibrosis and myofiber size (e.g. complete sciatic nerve transection), the present results suggest that GGF2 may benefit both nerve myelination and target muscles following peripheral nerve crush injury.
Comparative Biosciences, Inc. contributed to this work.
neuregulin, neuropathy, PNI, muscle, myelination
PARALLELS BETWEEN AXON STRETCH GROWTH AND REGENERATION FOLLOWING INJURY
Dr. Bryan J. Pfister, PhD, New Jersey Institute of Technology
Axon Stretch Growth (ASG) is the process in which applied biomechanical forces stimulate and sustain the growth of axons in coordination with body growth. Here, we investigate the morphological changes of the soma, as well as identify unique similarities in gene expression between ASG and regeneration following injury.
Utilizing custom motion-controlled bioreactors, we developed a method to perform high magnification imaging of neuronal somata undergoing ASG in vitro. First, embryonic rat Dorsal Root Ganglia (DRG) were dissociated and plated onto poly-d-lysine coated cover glass in 15μL drops. Next, Aclar manipulating substrates were positioned 200μm adjacent to the plated cells. Axons were allowed to extend and adhere to the manipulating substrates over 5 day while the system was held statically. Finally, ASG was initiated and sustained for 2 weeks prior to fixation and visualization of somata.
For mRNA collection, DRG explants were plated directly onto the Aclar substrates. Axons were stretch grown for 8 days similarly to dissociated cells and lysed using standard mRNA isolation protocols. Matching control explants were cultured without manipulation and isolated simultaneously with stretch grown samples. Gene expression profiling was performed using the Affymetrix GeneChip Rat Gene 1.0 ST DNA Microarray.
Briefly, conjugated wheat germ agglutinin was used to label plasma membrane of both perikarya and axolemma. Neurotrace™ (Invitrogen, Carlsbad, CA) was used to label Nissl bodies, while DAPI was used to counterstain nuclei. Confocal sections were taken using a 60x objective and magnified further utilizing field zoom. Sections were rendered as 2D images and cross-sectional areas of organelles were calculated by tracing their circumference using ImageJ.
Preliminary data analysis revealed enlargement of the cytoplasm in both stretch grown and stretch-axotomized neurons. Perikarya of control neurons grew from 148±60 μm2 to 356±137 μm2 over 19 days, while ASG samples reached 454±120 μm2 as they were stretch grown for 14 out of 19 days in culture. Similarly, stretch-axotomized neurons reached 472±229 μm2 as they were stretch grown for 9 out of 19 days in culture and axotomized on day 10.
Control nuclei grew from 73±28 μm2 to 104±25 μm2 over 19 days similarly to ASG nuclei which reached 103±15 μm2 following stretch growth. Conversely, stretch-axotomized nuclei grew to 120±39 μm2.
Total mRNA yield from explants undergoing ASG was up to 5-fold higher compared with controls. Preliminary gene expression profiling revealed significant upregulation of heat shock proteins HSPA1b and HSPA5, which correspond with gene expression following axotomy. In addition, several Regenerative Associated Genes (RAGs) commonly upregulated following axotomy were selectively regulated by stretch growth. The transcription factors LIF, C-Jun, and ATF-3 were upregulated, while we saw no change in Sox11, Gap43, Cap23, or SPRR1A.
We hypothesize that the perpetual state of growth induced by continuous axon stretching may be similar to the temporary regenerative growth cascade following axotomy. Our results suggest the cytoplasm of stretch grown and stretch-axotomized neurons swell similarly, and are larger than those of control neurons. These results suggest ASG may initiate the classic chromatolysis cascade, and lead to growth through similar regenerative signaling pathways.
Our preliminary gene expression profiling of stretch grown neurons support several components of the regenerative growth program that follows axotomy. Some differences, such as GAP-43 expression, were expected since ASG does not result in interruption of retrograde transport as does axotomy. Differences in retrograde signaling (negative and positive factors) that ensue following stretch or axotomy will undoubtedly lead to different gene expression, and influence the capacity of the neuron to regenerate or undergo apoptosis. Further study of ASG may help to isolate signals responsible for regenerative growth.
Work was supported by the New Jersey Commission on Brain Injury Research 07-3204-BIR-E-0 and NSF CAREER CBET-0747615.
axon stretch growth, regeneration
EARLY VERSUS LATER SURGICAL DECOMPRESSION OF SPINAL CORD AFTER TRAUMATIC CERVICAL SPINAL CORD INJURY: A COST-UTILITY ANALYSIS AND FEASIBILITY STUDY
Julio C. Furlan, MD, MBA, MSc, PhD, Toronto Western Research Institute
Michael G. Fehlings, MD, PhD, FRCSC, FACS, Toronto Western Research Institute
This study compares early (≤24 hours) versus later spinal cord decompression in order to determine which approach is more cost effective in the management of patients with acute cervical spinal cord injury (SCI) and to examine the potential barriers and ideal timelines for each step to early surgery.
A cost-utility analysis (CUA) was performed using data for the first year after cervical SCI. The perspective of a public health care insurer was adopted. Utilities are from the Surgical Trial in Acute Spinal Cord Injury Study. The reasons for delays in the management steps were classified into: (a) healthcare-related (“extrinsic”) and (b) patient-related (“intrinsic”) factors. In the benchmarking analysis, the patients were grouped into patients who underwent early surgery (Group-I) and individuals who underwent later surgery (Group-II).
When considering the late decompression as the baseline strategy, the incremental cost-effectiveness ratio is US$8,523,852 per quality-adjusted life year (QALY) for patients with complete SCI and US$275,390/QALY for patients with incomplete SCI. The probabilistic analysis indicated that there is no clearly dominant strategy.
Group-I (n=21) and Group-II (n=42) were comparable regarding prehospital time, time in a second general hospital and time in a spinal center emergency. However, Group-I had significantly shorter waiting time in a general hospital (p=0.0181), shorter waiting time for assessment by spine surgeon (p<0.0001) and shorter waiting time for the decision on surgical management (p<0.0001). While both patient groups showed comparable time periods related to intrinsic factors, patients who underwent early surgery had a significantly shorter time period associated with extrinsic factors when compared with individuals who underwent later surgery (p<0.0001).
In conclusion, the results of our economic analysis suggests that, although no strategy is clearly superior to the other, early decompression of spinal cord can be more cost effective than delayed surgery in approximately one quarter of the patients with complete SCI and one third of the individuals with incomplete SCI. Our benchmarking analysis suggests that health-related factors are key determinants of the timing from SCI to spinal cord decompression. Time in the general hospital and time of waiting for surgical decision were the most important causes of delays for surgical spinal cord decompression. Early surgery is feasible in the vast majority of the cases.
Supported by the Cervical Spine Research Society and Spinal Cord Injury Solutions Network Rapid Response Award from the Rick Hansen Foundation.
benchmarking analysis; cost-utility analysis
BISPEROXOVANADIUM DIFFERENTIALLY AFFECTS CELLULAR AKT AND ERK ACTIVITY AND PROMOTES OLIGODENDROCYTE AND MYELIN SPARING AFTER HEMI-CONTUSIVE CERVICAL SPINAL CORD INJURY
Nai-Kui Liu, M.D., Ph.D., Indiana University School of Medicine
Xiao-Ming Xu, M.D., Ph.D., Indiana University School of Medicine
Following spinal cord injury (SCI), bisperoxovanadium (bpV) promotes neuroprotection, functional recovery, and enhanced Akt/mTOR activity. However, bpV's effect on myelination and oligodendrocyte survival, and Akt/mTOR crosstalk with Erk 1/2 signaling is unknown. In this study, we hypothesized that bpV spares myelin and oligodendrocytes, and Erk 1/2 modulation may be involved.
To test these hypotheses, we administered bpV (400 μg/kg/day) or saline twice daily intraperitoneally (IP) to Sprague-Dawley rats (200–250g) for 1 or 7 days post-SCI (C5 hemi-contusion, NYU Impactor), and animals were sacrificed 1 day post-SCI for investigating bpV-mediated effects on Erk 1/2 activity, and 6 weeks following injury for assessing oligodendrocyte survival and myelination. Luxol fast blue was used to assess myelin sparing. Erk inhibition was accomplished through IP injections of the inhibitor, U0126. For Akt/mTOR and Erk 1/2 time course investigation, animals were sacrificed 1 day, 3 days, and 7 days post-C5 SCI. Antibodies against PTEN, phospho-PTEN, phospho-Erk 1/2, total Erk 1/2, phospho-Akt, phospho-ribosomal protein S6, phospho-4E binding protein-1, cleaved-caspase 3, and β-tubulin were used in protein analysis and/or histological and immunohistochemical staining for investigating bpV-mediated or time course-related effects post-injury.
As we previously demonstrated that bpV promotes Akt/mTOR signaling, neuroprotection and functional recovery post-SCI, we investigated whether Erk activity is affected by bpV treatment, and if this involvement is potentially detrimental or beneficial. Our results showed that total Akt activity decreased while Erk activation increased during the first week post-injury in both neurons and oligodendrocytes. bpV treatment rescued Akt activity and reduced Erk activity in these cell types, suggesting that bpV may operate through multi-factorial neuroprotective mechanisms. In support of these findings, bpV treatment also increased oligodendrocytes and myelinated white matter area, and decreased the expression of an apoptotic activity marker, cleaved-caspase 3. Finally, an inhibitor that blocks Erk 1/2 activity, U0126, also diminished injured spinal cord caspase 3 activity, as determined through Western blot analysis.
In summary, we have demonstrated that bpV's neuroprotective effects include myelin sparing and oligodendrocyte survival, and downregulation of Erk 1/2 activity may be part of its action. Whether Akt and Erk activity are directly related requires further investigation. This study widens our knowledge of bpV's neuroprotective effects and further unravels its mechanism of action, advancing bpV's potential as a clinical therapy for SCI.
This work was supported by the National Institutes of Health (NS36350, NS52290, and NS50243); Mari Hulman George Endowment Fund; and the State of Indiana.
SCI, bpV, PTEN, mTOR, neuroprotection
ANATOMICAL AND FUNCTIONAL CHARACTERISTICS OF VAGAL PREGANGLIONIC MOTORNEURONS FOLLOWING EXPERIMENTAL SPINAL CORD INJURY
Gregory Holmes, BA, MA, PhD, Penn State University College of Medicine
Spinal cord injury (SCI) impairs vagally-mediated gastric reflexes by reducing vagal afferent input to the nucleus tractus solitarius (NTS). However, gastric reflexes also require functional efferent output from the dorsal motor nucleus of the vagus (DMV). We hypothesized that vagal efferent signaling diminishes after SCI.
Experiment 1: Anaesthetized male Wistar rats were subjected to a midline contusion injury at spinal T3. Control rats received laminectomy only. Three days or 1 week later, the rats were re-anaesthetized, gastric motility and tone were recorded by a gastric strain gauge. Thyrotropin releasing hormone (TRH; 0, 3, 10, 30 or 100 pmoles/60nl) was microinjected in the left DMV adjacent to the area postrema.
Experiment 2: Six days prior to spinal surgery, rats (n=8) were anaesthetized, the stomach isolated, and the retrograde neuronal tracer Cholera toxin B (CTB) was injected into the anterior gastric corpus. Nine days following SCI or control surgery rats were transcardially perfused and brainstem sections were prepared for CTB and choline acetyltransferase (ChAT) immunohistochemistry.
Three days after spinal surgery, central microinjection of TRH increased gastric tone and motility in both SCI and control rats (p>0.05). At 1 week after spinal surgery, control rats demonstrated higher motility only at the 10 pmol dose (p<0.05).
Neurons comprising the rat DMV are known to respond more severely to axonal insult than other peripherally projecting neurons. CTB immunohistochemistry revealed no differences in the number of CTB immuno-positive neurons of SCI or control rats (n=6 per group; p>0.05) in the region of the DMV which corresponds to our TRH microinjection site.
ChAT immunohistochemistry revealed that the number of ChAT immuno-positive neurons was significantly higher in the caudal DMV of SCI rats compared to control (p<0.05). However, in the region of the DMV corresponding to our TRH microinjection site (adjacent to area postrema) no significant differences in the number of ChAT immuno-positive neurons was observed.
The neural control of gastric motility is dominated by the vagal parasympathetic reflex arc. Derangements in the vagal afferent limb contribute to gastric dysmotility following experimental SCI. However, integrity of the efferent limb had not been determined. TRH profoundly modulates gastric motility by a direct depolarization of DMV preganglionic motorneurons and offers a pharmacological tool for testing DMV efferent integrity. Our TRH data suggests that the vagal efferent limb remains capable of eliciting gastric motility following SCI.
Our immunohistochemical results revealed no differences in the number of gastric-projecting DMV neurons within the region of the brainstem corresponding to our TRH microinjection site. The functional implication of the increase in caudal DMV ChAT immuno-positive neurons of SCI rats remains to be determined. These data suggest that diminished vagal afferent input to the brainstem neurocircuitry is responsible for diminished gastric reflexes after SCI.
Support: NS 49177
Gastroparesis; TRH; NTS; DMV; Vago-vagal
OMEGA-3 FATTY ACIDS DIETARY SUPPLEMENTATION COUNTERACTS THE EFFECTS OF TBI ON THE SPINAL CORD
Cameron Feng, BS, Department of Integrative Biology and Physiology, UCLA
Zhe Ying, BS, Department of Integrative Biology and Physiology, UCLA
Rahul Agrawal, Ph.D, Department of Integrative Biology and Physiology, UCLA
Ethika Tyagi, Ph.D, Department of Integrative Biology and Physiology, UCLA
Traumatic brain injury (TBI) is often associated with locomotor deficits likely involving spinal cord (SC) centers.We have embarked to determine the influence of TBI on the SC and the potential of dietary interventions to counteract these effects.
We assessed the effects of a diet high in omega-3 fatty acid docosahexaenoic acid (DHA) on the lumbar enlargement region of the SC following moderate fluid-percussion injury (FPI). Pregnant mothers and their offspring (three months post-weaning) were exposed to diet containing DHA (1.2) or no DHA.
Animals on the DHA deficient diet showed reduced levels of omega-3 fatty acids (DPA-3 and DHA) while elevated levels of omega-6 fatty acids (DPA-6 and arachidonic acid) in their SC. FPI significantly reduced levels of synaptic plasticity related molecules (BDNF, pTrkB, and pCREB) and increased the marker for lipid peroxidation 4-HNE.The effects of the FPI were more pronounced in the DHA-deficient animals, particularly for pTrkB, pCREB, and 4-HNE.The membrane homeostasis related molecules (syntaxin-3 and iPLA-2) were only significantly decreased when FPI was performed in the DHA-deficient group.
These results indicate the pervasive effects of TBI in the spinal cord, and the capacity of dietary DHA to counteract these effects.
NIH RO1 NS056413, RO1 NS 050465
rat, diet, spinal cord, docosahexaenoic, membrane
THE GELATINASES, MMP-2 AND MMP-9, AS MODIFIERS OF ANGIOGENESIS AND VASCULAR STABILITY IN THE INJURED SPINAL CORD
Haoqian Zhang, PhD, University of California, San Francisco
Adanma Ekeledo, BA, University of California, San Francisco
Sang Mi Lee, PhD, University of California, San Francisco
Zena Werb, PhD, University of California, San Francisco
Kimberly Topp, PhD, PT, University of California, San Francisco
Prof. Linda J. Noble-Haeusslein, PhD, University of California, San Francisco
Endothelial cell death is evident in the spinal cord within the first day after spinal cord injury (SCI) followed by an angiogenic phase that persists upto at least 28 days. The gelatinases are known modulators of angiogenesis but have yet to be fully evaluated in the context of SCI.
All procedures were approved by the Institutional Animal Care and Use Committee at the University of California, San Francisco. Matrix metalloproteinase (MMP-2) wildtype (WT) and knockout (KO) female mice (n=56) were anesthetized and subjected to a moderate contusion injury to the lower thoracic spinal cord. Groups were randomized and analyses were conducted blinded to the experimental conditions. To identify dividing cells, bromodeoxyuridine (BrdU) was given 24 h prior to euthanasia at 3, 7, 14, and 21 days. Every fifth section was labeled for endothelial cells (PECAM), nuclei (Dapi) and BrdU. Three non-overlapping images were captured at the epicenter by confocal microscopy and analyzed using Metamorph software. MMP-2 and -9 activities were assessed by zymography. The dependency of endothelial proliferation, migration, tube formation and regression on the gelatinases was evaluated in vitro using brain-derived capillary endothelial cells (RBCEC4), exposed to pharmacologic inhibitors of these proteases.
To determine the dependency of angiogenesis on MMP-2, the number of proliferating endothelial cells was evaluated in spinal cord injured MMP-2 WT and KO mice. Proliferating endothelial cells increased at 7 days post-injury within both genotypes. Complimentary measures of vascularity (vascular density, area and length) likewise revealed an increase in these measures at 7 days in MMP-2 KO mice that were similar to that of WTs. However, subsequent comparisons between the numbers of endothelial cells at 7 and 21 days revealed a significant decline in the MMP-2 KO group at the later time point. This decline corresponded to a compensatory up regulation of MMP-9, demonstrated by zymography, in the MMP-2 KO group. We next evaluated endothelial proliferation, migration and tube formation in cultured RBCEC4 cells that express MMP-2 and can be induced to express MMP-9 in the presence of tumor necrosis factor a (TNFa). Using selective pharmacologic inhibitors of MMP-2 and -9, we found that endothelial proliferation showed a preference for MMP-2, while MMP-9 modulated endothelial tube formation. However, prolonged exposure to TNFa, a condition producing a sustained up regulation of MMP-9, led to vascular regression. Similar findings were demonstrated after prolonged exposure to exogenous MMP-9.
Taken together, these data demonstrate compensatory mechanisms for angiogenesis in the injured spinal cord that is deficient in MMP-2. Our data suggest that MMP-9 is a likely candidate for supporting angiogenesis, in part through its compensatory up regulation in the absence of MMP-2. However, we provide the first evidence that prolonged exposure to MMP-9, results in vascular regression, a process that while studied in the field of brain tumors, has yet to be fully described in spinal cord injury. We believe that excessive MMP-9 activity overrides other factors that normally contribute to vascular stability. Such findings emphasize the need to further explore gelatinases from the perspective of potential “switches”, triggered by changes in activity that may in turn coordinate undesirable outcomes.
This work was supported by NINDS R01 NS039278.
MMP-2, MMP-9, vascular regression, proliferation
CEREBELLAR REORGANIZATION FOLLOWING SPINAL CORD INJURY
Prof. Melanie L.. McEwen, Ph.D., University of Kentucky Spinal Cord and Brain Injury Research Center
Nishant Visavadiya, Ph.D., University of Kentucky Spinal Cord and Brain Injury Research Center
Spinal cord injury (SCI)-induced changes in cerebellar circuitry and the role these events play in recovery of sensorimotor function have received limited attention. In this study, we examined changes in cerebellar circuitry following SCI with a focus on excitatory granule cell and inferior olive inputs to cerebellar Purkinje cells.
Adult male and female Long-Evans rats received sham surgery or a moderate-to-severe T10 SCI using the Infinite Horizons impactor and euthanized at 42 days. Immunofluorescence staining was performed on sagittal cerebellar sections using SMI-31, synapsin-1, vGLUT2, and calbindin. Analysis of the staining patterns was conducted in cerebellar lobule III, which contains granule cells receiving direct spinocerebellar input from Clarke's nucleus located in T1-L2. Additional animals were euthanized at one and seven days post-injury and cerebellar samples analyzed by western blot using an antibody to Cbln1, which is essential for parallel fiber-Purkinje cell synaptic integrity in the mature cerebellum.
Immunofluorescence staining with SMI-31 revealed a prominent loss of parallel fibers in the molecular layer of lobule III in sections from injured animals compared to sham. This finding was accompanied by an increase in punctate synapsin-1 staining in the Purkinje cell body layer, which receives excitatory input from the contralateral inferior olive. Punctate vGLUT2 staining was observed in lobule III of both sham and injured animals. However, there was an expansion of vGLUT2 staining into the vacated parallel fiber zone. This pattern of staining was similar to that reported for Cbln1 gene deletion studies. Several studies have demonstrated that granule cell axon terminals (parallel fibers) release Cbln1. A loss of Cbln1 expression results in a decrease in parallel fiber synaptic contacts on Purkinje cell dendrites and expansion of excitatory inferior olive inputs (climbing fibers) into the vacated parallel fiber zone. Consistent with a role for Cbln1, western blot studies revealed a decrease in Cbln1 levels in the cerebellum as early as seven days following SCI.
We propose that these changes in cerebellar circuitry are initiated by the loss of excitatory signaling originating, in part, from Clarke's nucleus via the dorsal spinocerebellar tract to cerebellar granule cells. As a consequence, a decrease in granule cell excitation from dorsal spinocerebellar inputs reduces Cbln1 levels, leading to a loss of parallel fiber synaptic contacts on Purkinje cells and expansion of excitatory inferior olive contacts into the denervated parallel fiber zones. Future studies will investigate if the SCI-induced changes observed in the present study resulted in altered synaptic activity within the cerebellum or impacted cerebellar transmission of non-conscious proprioceptive information to higher motor centers.
Supported by an endowment from Cardinal Hill Rehabilitation Hospital and a grant from the Craig H. Neilsen Foundation to JES.
neuroplasticity, spinocerebellar tract, non-conscious proprioception
STABILITY, DISPOSITION, AND PENETRATING ABILITY OF THE CATALYTIC ANTIOXIDANTS MNTBAP AND EUK-134 AFTER EXPERIMENTAL SPINAL CORD INJURY: COMPARISON WITH METHYLPREDNISOLONE
Liqin Wu, Ph.D., University of Texas Medical Branch
Yichu Shan, Ph.D., University of Texas Medical Branch
Antioxidant therapy reduces oxidative damage in secondary SCI. This study characterized the in vivo stability, disposition and blood–spinal cord-barrier (BSB) penetration of the Mn-containing catalytic antioxidants Mn (III) tetrakis (4-benzoic acid) porphyrin (MnTBAP) and EUK-134 to evaluate their potential for treating SCI. Parallel measurements of methylprednisolone provided reference information.
Following laminectomy, the rat spinal cord was injured by dropping a 10-g rod 2.5 cm onto the cord. Immediately after injury, 6.4 mg/kg MnTBAP or 30 mg/kg methylprednisolone sodium succinate, a water-soluble derivative of methylprednisolone, was administered over a 10-min period into the femoral vein through a PE-10 tube. Methylprednisolone is the only drug used for treating SCI. Since EUK-134 at 15 mg/kg was not detectable several minutes after administration by the same method, this dose EUK-134 was given intraperitoneally. Blood was sampled from the tail artery through a PE-10 tube, and cerebrospinal fluid (CSF) was sampled from a hole made in the atlanto-occipital membrane through a PE-10 tube affixed to the membrane. Samples were collected at 0.5, 1, 2, 4, 6, 8, 10, and 24 h post-injury. The MnTBAP, EUK-134, and methylprednisolone were extracted from the collected samples and analyzed by HPLC with UV detection.
By measuring the time courses of concentration changes following administration of the agents, this study kinetically characterized the stability, disposition, and penetrating ability of MnTBAP, EUK-134, and methylprednisolone. MnTBAP was stable in both CSF and blood for the periods over which samples were collected (up to 10 h for CSF and 24 h for blood) post-administration. EUK-134 and methylprednisolone concentrations sharply declined to near basal values at 4 h and 2 h post-administration. Thus, the order of stability in CSF and blood was MnTBAP >> EUK-134>methylprednisolone. The maximum CSF/blood concentration ratios of EUK-134, methylprednisolone, and MnTBAP post-administration were 32±3.1%, 19.2±6.4%, and 4.42±0.73% in the injured rats, and 22±6.5%, 17.8±2.9%, and 1.0 0.5% in the sham control animals. Thus, the order of penetration of the BSB was EUK-134>methylprednisolone >> MnTBAP.
This study represents the first in vivo measurement of the stability, disposition, and BSB penetrating ability of MnTBAP and EUK-134 — critical information for designing further experiments to test their potential for treating SCI. Even though MnTBAP penetrated the BSB much less than did EUK-134 and methylprednisolone, because of MnTBAP's higher stability it produces a higher concentration in the CSF at a lower dose compared with EUK-134 and methylprednisolone at higher doses. This supports further exploration of the potential use of MnTBAP for therapy after SCI. Measuring the post-administration time courses of concentration changes of an agent by continuously sampling CSF and blood in one group of animals not only provides information regarding the agent's stability and BSB penetration after administration, but also obviates the need to sacrifice separate groups of animals at every post-administration time point for obtaining spinal cord tissue.
The authors thank the National Institutes of Health (National Institute of Neurological Disorders and Stroke, RO1 NS 44324 to Danxia Liu) for financial support.
MnTBAP, EUK-134, BSB, spinal injury
PATIENT SPECIFIC ELECTRICAL STIMULATION FOR GUIDING NEUROPLASTIC CHANGES
Bennett Groshong, Ph.D., Self-Empolyed
David Wiley, Ph.D., Stratovan
Transcranial electrical stimulation (TES) is an effective therapy for guiding neuroplastic changes after traumatic brain injury (TBI). TES may treat depression, tinnitus, and memory loss, but demonstrates significant variability in individual outcomes. This study examines the value of individualized conductivity modeling for reducing outcome variability.
After approval by the Institutional Review Board. Multiple 3Tesla MRIs were captured from the heads of 20 normal subjects according to the Aaken Protocol (T1, T2, proton density, diffusion weighted, and magnetic resonance angiography). The recordings were obtained with a 1mm by 1mm pixel by 1mm slice spacing. Diffusion weighted recordings were acquired at 2mm by 2mm pixel by 2mm slice spacing. These images were fused into a single 3D representation and a conversion equation was applied to obtain a resistivity index at each point within the individual's head. The resistivity models were then analyzed with Finite Element analysis to determine current density throughout the individual's head based on virtually placed electrodes. Current predictions were calibrated with 25 surface measurements from three subjects. Virtual electroes were placed according to the 10–20 system at four configurations: F8-P2, C3-C4, F8-F7, and Fpz-O1.
Electric current was seen in all regions of the brain. Seventy-five regions were segmented for each individual and examined for each of the four electrode configurations resulting in 300 mean values and 80 three-dimensional image simulations. There was as much as a six-fold difference in current density recevied between individuals in regions near the electrodes. The highest current levels and greatest variance between individuals were found with the F8-P2 configuration. With this electrode pair, the highest current levels within a structure were found in the par-sobitalis, par-stianularis, right precentral and right transverse temporal regions. All of these regions are areas close to the stimulating electrodes. In most nstances, the regions with the lowest current values also had the lowest variance and were found to be the most distant from the stimulating electrodes. This is particularly true when the stimulating electrodes were placed close together. As might be expected, the C3-C4 and F7-F8 electrode positions had the lowest current values overall and least variance in those regions that were distant from the electrodes. There were exceptions to this distance rule however. Regions at the interface of high- and low-resistivity tissues (e.g. bone and CSF) often showed areas of surprisingly high current densities and variance. These “hot spots” were difficult to predict and changed between electrodes configurations and individuals. The presence of cerebral spinal fluid in sulci or ventricles appears to draw in the current, but a gyrus had the opposite effect. Larger blood vessels and the high conductance of CSF are major factors in deep brain structures. This is particularly true at the interface of CSF and higher resistance tissues.
The highest current variability was found in cortical structures directly under and adjacent to the stimulating electrodes. This “near electrode” current variability was dependent on the local architecture of the cranium. Small anomalies, such as the type and thickness of the bone, significantly impacted current levels. Current level variability received by patients is likely a major factor in determining patient outcome. The large variations in the current received should be considered as a factor in determining patient dosing. Thus, adjusting electrode position and stimulation levels to account for individual variation in the current received will likely improve patient outcomes.
Using modeling as a pretreatment planning tool for optimal electrode placement could reduce individual variance, and guide neuroplastic changes in the brain. The application of TES to TBI patients is likely to improve individualized electrical stimulation modeling.
This work was done with the support of the Sutter Institute for Medical Research and Radiological Associates of Sacramento.
Neuroplastic, Recovery, Stimulation, Outcomes, Treatment
HYPOTHERMIA FOLLOWING PEDIATRIC SEVERE TRAUMATIC BRAIN INJURY (PEDIATRIC TRAUMATIC BRAIN INJURY CONSORTIUM: HYPOTHERMIA)
John Beca, MD, Starship Children's Hospital
Stephen R. Wisniewski, PhD, University of Pittsburgh
Sue R. Beers, PhD, Children's Hospital of Pittsburgh of UPMC
Deborah Hirtz, MD, National Institute of Neurological Disorders and Stroke
S. Danielle Brown, RN, MS, Barrow Neurological Institute at Phoenix Children's Hospital
Claire Sherring, RN, Starship Children's Hospital
Following positive preliminary results of a small Phase II randomized clinical trial (RCT) for moderate therapeutic hypothermia (TH) (32–33°C) following severe traumatic brain injury (TBI) in children, we aimed to assess outcome through a Phase III multicenter, multinational RCT utilizing TH for 48 hours with slow rewarming.
The Pediatric TBI Consortium (PTBIC): Hypothermia (“Cool Kids Trial”) was a randomized, multicenter clinical trial of pediatric (<18 years old) patients with severe TBI who were enrolled within 6 hours of injury at 15 sites in the USA, New Zealand, and Australia as compared to normothermia (NORM) (36–37°C). TH patients were cooled to rapidly to 32–33°C following randomization using initially “iced” saline (4°C) to 34–35°C and then surface cooled to target for 48 hours. They were then gradually rewarmed beginning at 48 hours following reaching target temperature at 0.5–1.0°C every 12-24 hours. The primary outcome measure was mortality at 3 months post injury; further global outcomes using the Glasgow Outcome Score-Extended (Pediatrics) (GOSE-Peds) were obtained at 3 months post injury. Investigators who assessed the outcome measures were masked to treatment allocation. Analysis was by intention to treat.
Enrollment occurred from November, 2007, to February, 2011, when the trial was terminated for futility by the independent data and safety monitoring board (DSMB). Follow-up for the primary outcome measure was from February, 2008, to May, 2011. Seventy-seven patients (39 in the TH group and 38 in the NORM group) were randomized at mean time of 5:08 hours:minutes (SD 0:55) after injury. The mean time to target temperature of 33°C for the TH group was 8:27(2:23) hours:minutes (SD 6:58). The primary outcome of mortality assessed at 3 months post TBI showed that in the TH group 6/39 (15.4%) died as compared with 2/38 (5.3%) in NORM (p=0.15). Of those randomized, no child in the NORM group but two children in TH who later died were found to have been ineligible (undocumented non-accidental trauma, severe hypotension in the field) and were included in the intent to treat analysis. Global outcome assessed by the GOSE-Peds was no different between TH and NORM groups for poor (upper and lower severe disability, vegetative state, or death) (45.9% and 44.5% respectively, p>0.99). During the 5-day ICP monitoring period, TH patients had lower average ICP (SE), 2.1 mm Hg (2.0) though was not statistically significant (p=0.23). Lastly, there were group differences with regard to 2nd Tier therapy treatment in that NORM patients underwent decompressive craniectomy for control of their ICP more frequently than TH patients (17 (44.7%) vs. 7 (17.9%), respectively).
TH was not shown to be efficacious as a primary neuroprotective strategy from mortality or early global outcome in children with severe TBI though was effective in lowering ICP as part of 2nd Tier therapy.
NINDS/ NIH [UO1 NS# 052478] (PDA) for funding this project and all the members of the PTBIC for their participation in this study.
Clinical Trial, Pediatric, Hypothermia, TBI
HUMORAL AND NEUROTRANSMITTER MECHANISMS IN SEVERE BRAIN INJURIES
Ms. Evgenia Alexandrova, M.D., Burdenko NSI
Alexander Anatolievich. Sychev, Ph.D, M.D., Burdenko NSI
Yuryi Vasilevich. Vorobiov, Ph.D, M.D., Burdenko NSI
Inna Trubina, Ph.D, Burdenko NSI
Vladimir Grigorievich Voronov, Burdenko NSI
Alexander Alexandrovich Potapov, Burdenko NSI
At present it had been proven the neuroprotective effects of progesterone and other neurosteroids in severe traumatic brain injuries (TBI) and ProTECT III (Emory-led multicenter studying to evaluate the progesterone effectiveness on acute TBI patients) is continued. The study goal was to compare hormonal (progesterone and iodothyronines) dynamics with clinical condition, especially neurological and consciousness recovery data.
Seventy-five male (12 died) and 26 female (4 died) Pts with GCS≤8 on admission were enrolled. Patient age was 16–72 y.o. (32,0 /-12.7 y.o.). Progesterone (Prog) and other steroids (estradiol, testosterone, cortisol), thyroid hormones (triiodothyronine - T3, thyroxine - T4 and their free forms) were examined by immunochemiluminescent analysis (Immulite 2000) in dynamics. Plasma catecholamines (Noradrenaline-NA, Dopamine-DA, Adrenaline-A) levels were evaluated in dynamics by HPLC with electrochemical detector (BASi, USA). Degree and localization of brain damage was evaluated by KT and 1,5-3T MRI. Clinical estimation included consciousness level, motor function, hemodynamics, presence of dysautonomia (vegetative paroxisms). Besides, hemodynamic (HD) patterns by transpulmonary thermodilution (PiCCO) method were evaluated.
We found the significant decreased as Prog and thyroid hormones levels (down to undetected values) in the most patients, regardless of gender (in 10 females and 26 males) and in Pts with more severe states. Correlation coefficient are less than 0,05 between Groups depending of severity states. However, Pts groups depending on the Prog decreasing degree significantly differed from each other in the frequency of vegetative paroxysms. There were low or low/normal Prog levels in 58,3% and 61,4% accordingly in patients with focal brain contusions. But there was predominantly normal Prog levels in Pts with diffuse brain injuries (in 42,8%). In these patient groups did not differ for the duration of coma, vegetable status and other stages unconsciousness and somatic state (unstable hemodynamics, intracranial cerebral pressure duration, brain edema). Comparison of thyroid hormones levels with respect to violations of Prog levels were found the significantly lower levels in the patient groups with the lowest progesterone levels.
Traditionally, progesterone and thyroid hormones are not attributed to the acute period of stress. This opinion led to conventional point of view on these biologically active molecules as a special interest of endocrinologists but not intensivists. Meanwhile, according to modern concepts, T3 is a neurotransmitter which coexists with classical neurotransmitters, namely noradrenaline. Moreover, thyronergic function T3 is involved in processes of regeneration after brain injury and apoptosis inhibition. T3 participates in astroglial and immune responses to injury, in calcium homeostasis, axonal transport during regeneration etc. Serum progesterone levels are known to decrease after stress, moreover progesterone biosynthesis in the brain and beneficial effects of exogenous progesterone after brain injury are proved. These provisions served as the stimuli to start to examine the relationship between T3 and Prog on the one hand and their role in the functioning of the brain on the other hand. Probably, future research should study the simultaneous application of progesterone and triiodothyronine in acute head injury, because as progesterone and triiodothyronine have neuroprotective properties.
Traumatic brain injury, progesterone, iodothyronines
MANAGEMENT STRATEGIES FOR REDUCED BRAIN OXYGEN IN SEVERE TRAUMATIC BRAIN INJURY: THE BOOST-2 TRIAL
Peter LeRoux, MD, University of Pennsylvania
Professor Ross Bullock, MD, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Randall Chesnut, MD, University of Washington
Lori Shutter, MD, University of Cincinnati
Christopher Madden, MD, University of Texas Southwestern Medical Center
Nancy Temkin, PhD, University of Washington
Dick Moberg, PhD, Moberg Research, Inc.
Carol Moore, MA, Center for Neuroscience and Regenerative Medicine/Uniformed Services University
Brain tissue oxygen (PbtO2) based care combined with management of intracranial pressure (ICP) may result in improved outcome after severe traumatic brain injury (TBI). The Brain Oxygen Optimization in Severe TBI Phase 2 (BOOST-2) trial is a randomized controlled trial of PbtO2 based therapy.
BOOST-2 was designed to demonstrate that PbtO2 based therapy is effective in reducing the duration and severity of brain tissue hypoxia. In planning for this study, BOOST-2 investigators collaboratively designed a management strategy for reduced PbtO2 that could be consistently applied across multiple clinical sites. Instrumentation designed to continuous measure ICP, PbtO2, CPP, and other physiologic parameters was also developed. The BOOST-2 investigators identified publications that described PbtO2 - based care and use of PbtO2 monitors. Described therapies and management practiced by the investigators were then refined through a series of iterative processes seeking greater than 90% consensual agreement from the investigative group.
BOOST-2 currently consists of 10 clinical sites throughout the United States. A PbtO2 of 20 mmHg was defined as the threshold to initiate treatment. Four treatment scenarios were defined: a) ICP (≤20 mmHg) and PbtO2 (≥20 mmHg); b) ICP>20 mmHg and PbtO2≥20 mmHg – management directed toward ICP; c) ICP≤20mmHg and PbtO2<20 mmHg – therapy directed at PbtO2; and d) ICP>20mmHg and PbtO2<20 mmHg – management directed at both ICP and PbtO2. ICP, PbtO2, MAP temperature, and HR are recorded continuously. A tiered approach is utilized, starting with simpler strategies first and proceeding to more invasive and riskier interventions. Interventions are targeted at patient specific physiology. In each tier therapy is started within 15 minutes of an abnormality in ICP or PbtO2, and several options may be used in any order or combination. If interventions in a given tier are ineffective, no more than 60 minutes are spent before moving on to the next tier. Continuous recordings identify spontaneous oscillations in ICP and pbtO2. There is no consistent relationship between ICP and PbtO2.
Abnormal PbtO2 may occur in patients with severe TBI, despite efforts to maintain CPP. A management strategy based on a tiered approach that is targeted to patient physiology is being tested in a Phase 2 randomized controlled trial, which aims to demonstrate physiologic efficacy and feasibility.
Supported by NIH R01 NS061860
PbtO2; ICP; physiologic monitoring
VALIDITY OF PEDIATRIC GLASGOW OUTCOME SCALE-EXTENDED 12 MONTHS POST TBI
Peter L. Stavinoha, PhD, ABPP, University of Texas Southwestern Medical Center
Ana Hernandez, MS, Children's Medical Center Dallas
Rong Huang, MS, Children's Medical Center Dallas
Katrina van de Bruinhorst, MA, Perot Family Center for the Care of Brain and Nerve Injuries-Children's Medical Center Dallas
Lakshmi Raman, MD, Children's Medical Center Dallas
Pediatric Glasgow Outcome Scale-Extended is derived from the GOS and GOS-E which are widely used to assess TBI outcomes in adults. The measure provides a cost effective method to quantify outcome in pediatric TBI. The current study examines validity of the GOS-E Peds relative to cognitive outcome.
This study analyzed prospective data from a study of TBI in children from a large metropolitan children's hospital. Inclusion criteria were: (1) moderate to severe non-penetrating TBI defined by initial GCS score between 3 and 12 (2) birth to 17 years of age at injury; (3) English or Spanish as the child's language of origin. Instruments administered 12-months post injury included GOS-E Peds and age appropriate measures of general intelligence (IQ) assessed by the Mullen Scales of Early Learning for young children and the age-appropriate Wechsler Scale for the remainder of the sample (i.e., WPPSI-III, WISC-IV). IQ and GOS-E Peds scores were examined to determine whether a significant relationship existed between these variables. Subjects were also grouped according to initial injury (moderate or severe) for comparison of GOS-E Peds and IQ to determine sensitivity to categorization of initial injury.
Subjects included 64 children, ranging from 13 to 173 months of age, with males (72%) outnumbering females (28%). Seventy-five percent of the sample sustained TBI secondary to motor vehicle accident as a vehicle occupant or as pedestrian. Mean initial GCS for the sample was 6.61 with a standard deviation of 2.83, with the majority of the sample classified as severely injured (72%). Neither IQ nor GOS-E Peds at 12 months correlated with initial GCS from the time of injury. However, IQ and GOS-E Peds were strongly inversely correlated (r=−.81, p<.0001) suggesting validity of the GOS-E Peds in terms of sensitivity to overall cognitive ability following pediatric TBI. While there was not a significant correlation between initial GCS and either IQ or GOS-E Peds at 12 months post injury, subjects were grouped by severity of initial injury (Moderate Injury n=18; Severe Injury n=46) in order to compare outcomes using a group comparison design. Results of t-tests indicated statistically significant differences between groups, with severely injured subjects performing considerably lower than those with moderate injuries in terms of IQ (Satterthwaite procedure for unequal variances t=−3.16, p<.0027) as well as on the GOS-E Peds (Satterthwaite procedure for unequal variances t=3.25, p<.0021).
This study provides evidence of validity of the GOS-E Peds 12 months following initial injury. The strong relationship between GOS-E Peds and objectively measured IQ indicates the GOS-E Peds is sensitive to differences in cognitive outcome 12 months post pediatric TBI. Further, on average, severely injured children performed significantly poorer on an objective measure of cognitive ability 12 months following TBI relative to those with moderate injuries. Similarly, the severely injured group also exhibited significantly poorer scores on GOS-E Peds at 12 months, again reinforcing that the GOS-E Peds is sensitive to severity of initial injury at the 12 month time point. Future research is necessary to further establish validity of the GOS-E Peds and may include evaluating the measure's sensitivity to other cognitive, adaptive behavioral, and behavioral domains as these have been shown to be at high risk for pathology following pediatric TBI.
Perot Family Center for the Care of Brain and Nerve Injuries at Children's Medical Center Dallas
GOS-E Peds, Children, TBI, IQ
THE JAPANESE TRANSITION OF SEVERE BRAIN INJURY WITH DRINKING: A REPORT FROM THE JAPAN NEUROTRAUMA DATA BANK
Takeki Ogawa, MD, PhD, Department of Emergency Medicine/ Jikei University School of Medicine
Alcohol is an important factor for the mechanism of head injury, because alcohol influences imbalance, the reduction and delay of judgement and reflexes. To investigate the efficacy of the increased the penalties for drinking and driving offences; we compared the data of drunken severe brain injured patients on JNTDB.
The Japan Neurotrauma Data Bank (JNTDB) Committee was founded within the Japan Society of Neurotraumatology.
The committee carried out joint clinical studies to evaluate traumatic brain injury in Japan as Project 1998 (P1998; 1,002 cases from 1998 to 2001 at 10 emergency medical centres) and Project 2004 (P2004; 1,101 cases from 2004 to 2006 at 19 emergency and neurosurgery medical centers).
In 2002, a revision to part of the Road Traffic Act drastically increased the penalties for drinking and driving offences in Japan.
These two projects were enrolled before and after the law changing and were affected on the decreased accident.
187 cases that were checked as a drunk patient on JNTDB P1998 and 172 cases on JNTDB P2004 were in this study.
The comparison of age, sex and mechanism of injury between the projects were examined.
According to the traffic accident statistics, these increased penalties for drinking and driving offences dramatically decreased the traffic accident with drinking by years, especially in 10–20’S. Our results show that traffic accident's patients with drinking significantly decreased from P1998 to 2004 (P<0.001). In traffic accident on P1998, there were two peaks in the 10–20's and the 50's. On P2004, traumatic accident patients with drinking significantly decreased in the 10's and 20's (p<0.001). Otherwise, the non-traffic accident patients with drinking significantly increased in the 70's. The leading reason of the non-traffic accident with drinking is tumbling and falling.
In 2002, a revision to part of the Road Traffic Act drastically increased the penalties for drinking and driving offences in Japan. According to the traffic accident statistics, these increased penalties for drinking and driving offences dramatically decreased the traffic accident with drinking by years, especially in 10–20’S. The tendency of reduction of the traffic accident patients affected to our results. So the Japanese increased penalties for drinking and driving offences significantly decrease the amount of traffic severe brain injured patients with drinking.
Meanwhile, for reduction of the non-traffic accident with drinking, we should make the prevention measure for imbalance of elderly patients.
The Japan Neurotrauma Data Bank Committee, The Japan Society of Neurotraumatology, Japan Council of Traffic Science, Tokyo, Japan
Alcohol, Severe brain injury
CLINICAL ANALYSIS OF TRAUMATIC MULTIPLE INTRACRANIAL LESIONS: REPORT OF 118 CASES
PENGCHENG LI, master, Department of Neurosurgery, West China Hospital of Sichuan University, Chengdu, P.R.China.
QIANG LI, M.D, Department of Neurosurgery, West China Hospital of Sichuan University, Chengdu, P.R.China.
JUNPENG MA, M.D, Department of Neurosurgery, West China Hospital of Sichuan University, Chengdu, P.R.China.
CONG WU, M.D, Department of Neurosurgery, West China Hospital of Sichuan University, Chengdu, P.R.China.
CHAO YOU, M.D, Department of Neurosurgery, West China Hospital of Sichuan University, Chengdu, P.R.China.
The article is aimed to study the clinical features of traumatic multiple intracranial lesions and surgical treatment.
The clinical data of 118 patients with traumatic multiple intracranial lesions were collected in West China Hospital of Sichuan University, and the patient's age, gender, cause of injury, diagnosis, treatment, and outcomes were analyzed retrospectively.
According to GCS on admission, 6 patients were with mild head injury, 10 moderate injury, and 102 severe injury. There were 12 patients with different types of injury in the same site, 69 one type of injury in different compartments, and 37 different types of injury in different compartments. One hundred and six patients had obliteration of basal cisterns, of which 34 patients had midline shift more than 5 mm at the same time. Eighty-nine patients underwent one site operation, 19 two sites operation, 10 non-operation. According to GOS at 6 months post injury, good recovery was achieved in 37 patients, moderate disability in 4, severe disability in 3, Vegetative State in 41, and death in 33. Obliteration of basal cisterns was a strong indication for surgery (OR=48.067, 95%CI, 9.482–243.669). One site operation or two sites operation had no significant relationship with outcomes (P>0.05).
Most of traumatic multiple intracranial lesions was severe injury. The surgical indication mainly depends on the status of basal cisterns, ICP, and patient's consiousness. After evacuation the major lesion, the other lesions could be treated conservatively. Most of the patients only need one site surgical treatment.
none
traumatic brain injury; surgery
IS AN EARLY ONSET AND CONTINUOUS CHAIN OF REHABILITATION BENEFICIAL FOR SEVERE TRAUMATIC BRAIN INJURY IN FIVE YEARS PERSPECTIVE?
Prof. Cecilie Roe, MD, PhD,
In our recently published study, early onset of rehabilitation in a continuous chain of treatment improves functional outcomes one year after severe traumatic brain injury (TBI). The aim of this study is to evaluate whether the functional improvement related to early rehabilitation persist at five-year follow-up.
This prospective cohort study included patients with acute TBI admitted to the Trauma Referral Center in South-East region of Norway (2005–2007). Inclusion criteria's were age 16–55 years, Glasgow Coma Scale (GCS) ≤ 8, need of neurointensive care (i.e. neuromonitoring to optimize conditions for neuronal survival), and survival one year post-injury. As the ethical justification for randomizing patients seemed untenable for researching acute care of TBI we used a quasi-experimental study design. In general, the capacity in the Intensive Care Unit (ICU) determined the assignment to the early rehabilitation (Group A) (i.e. rehabilitation integrated in ICU and a continuous chain of treatment) or not (Group B) (i.e. broken chain of treatment and delayed admission to inpatient rehabilitation). The functional outcome was assessed by Glasgow Outcome Scale Extended (GOSE). A binary logistic regression analysis was used to quantify the relationship between the type of rehabilitation and GOSE.
A total of 61 survivors with severe TBI were included in the study. Five patients dropped-out from the one- to five-year follow-up. Of the remaining patients, 29 patients were assigned to Group A and 27 patients were assigned to Group B. No statistically significant differences were found between Group A and B regarding demographic variables (gender, p = 0.79; age, p = 0.11) and cause of injury (p = 0.23). The groups were also comparable for injury severity variables such as GCS (p = 0.35), pupil condition (p = 0.95), hypoxemia (oxygen saturation ≤ 90%) (p = 0.63), hypotension (mean arterial pressure ≤ 60 mmHg) (p = 0.39), Abbreviated Head Injury Scale (AIS head) (p = 0.87), Injury Severity Scale (ISS) (p = 0.39), intracranial injuries on the worst head CT scan performed within 24 h post-injury (p = 0.76), the rate of ICP monitoring (p = 0.11) and intracranial surgery (p = 0.36). A favorable outcome (GOSE 6–8) was found in 75% of the patients from Group A vs. 44% in Group B (p = 0.028). In the unadjusted regression model, Group A had a significantly increased probability of a better functional outcome (unadjusted OR 3.93; 95% CI (1.26–12.28), p = 0.019). The trend of the increased probability of a better outcome in Group A was also upheld when the model was adjusted for age (OR 3.67; 95% CI (1.15–11.69), p = 0.028), as well as in the final regression model adjusted for age, severity of the intracranial injury and length of the stay in rehabilitation hospitals (OR 3.14; 95% CI (0.94–10.45), p = 0.06). The model predicted a favorable GOSE outcome in 79% of the cases.
Patients with severe TBI who received a delayed admission to inpatient brain injury rehabilitation centers had poorer functional outcomes five years post-injury compared to patients who did receive rehabilitation at early onset, as well as a continuous chain of rehabilitation. Our results support other studies that suggest that early rehabilitation is broadly beneficial for severe TBI. The cost-effectiveness of the early rehabilitation model is currently under investigation and results will be presented at the conference.
Thanks to Tone Jerstad for the CT assessments and Morten Hestnes for the extraction of trauma scores from the Hospital`s Trauma Register.
THE EXPERIENCE OF MINIMAL INVASIVE EVACUATION OF THE INTRACRANIAL HEMATOMAS IN TREATMENT THE PATIENTS WITH MULTIPLE TRAUMA
Vladimir Sorokovikov, M.D., Irkutsk Institute of Traumatology
Alexander Zhivotenko, Irkutsk City Hospital
The current work is dedicated to rendering neurosurgery aid to patients suffering a multiple brain injury accompanied by the hematoma brain compression. The research aims to study possibilities of applying the restricted invasion drainage method to the intracranial hematoma, with consequent local fibrinolysis.
It was comparison the results of neurosurgical treatment of 2 groups patients with multiple trauma and intracranial hematomas: group A – 14 patients, who were subjecting the minimal invasive evacuation of the intracranial hematomas by puncture, drain and local lysis of clots of blood in compliance with original method (V.Krylov, S.Burov et al., Sklifosovsky institute, Moscow, 2009); group B – 14 patients, who were subjecting traditional neurosurgical operations. For local fibrinolysis of clots of blood we used the second generation fibrinolitic - recombinant prourokinase. All the patients were random selecting with a glance of intracranial hematoma localization, extracranial injures localization and severity of injury evaluation by ISS and RTS. In addition we evaluate the severity of injury by index, that was introducing by us previously. Both the groups were comparison by two criteria's: lethality and average ball of Glasgow coma scale after two weeks for survive patients.
We found that lethality in group A was 42,9 %, and in group B – 71,4 % (Student's index t=1,53); average ball of Glasgow coma scale in two weeks after accidents for survive patients in group A was 12,5 (±1,6), and in group B – 10,5 (±2,1) (Student's index t=1,49), but we received best statistic results after minimal invasive evacuation just of the intracerebral hematomas.
Our experience showed that method of minimal invasive evacuation intracranial hematomas by local lysis of clots of blood is applicable for treatment the patients with multiple trauma, exactly for evacuation the intracerebral hematomas. We are in need of observation more patients with multiple trauma and intracranial hematomas for deeper statistic analysis.
B.Braun company
fibrinolysis, multiple trauma, intracranial hematoma
PERSISTENT DECREMENTS IN ANAM PERFORMANCE IN RECENTLY REDEPLOYED MARINES
Alia Creason, Ph.D., BUMED
Justin Campbell, Ph.D., LCDR, MSC, USN, SPAWARSYSCEN-PACIFIC
Aimee Alphonso, B.A., The Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc.
Lauren Stentz, BUMED TBI Programs
Mikas Wolde, BUMED TBI Programs
Sherry Wing, BUMED TBI Programs
Kevin Garcia, BUMED TBI Programs
Jack Tsao, M.D., D. Phil., CDR, MC, USN, BUMED
Redeployed Marine units with high rates of mild traumatic brain injury (mTBI) and/or blast exposure were administered post-deployment Automated Neurocognitive Assessment Metrics (ANAM) to determine the usefulness of routine post-deployment ANAM testing.
705 recently deployed Marines were administered the ANAM 4 TBI-MIL battery pre- and post-deployment (within 2 weeks to 2 months after return). The ANAM is an automated neurocognitive testing modality that includes six subtests: Simple Reaction Time, Coded Substitution, Procedural Reaction Time, Mathematical Processing, Matching to Sample, Coded Substitution Delayed (CDS), and Simple Reaction Time Repeated. The battery also includes a TBI questionnaire from which participants were divided into groups based on their responses. Mild TBI (mTBI) was defined by when individuals reported an injury event accompanied by an alteration of consciousness. This included endorsement of at least one of the following: feeling dazed and confused, experiencing loss of consciousness, or experiencing loss of memory for the injury. Throughput scores, which represent the correct number of responses per minute were analyzed using multiple repeated measures ANOVA.
168 Marines met the mTBI group definition. Results indicated significant main effects for Time (p<.000), Group (p<.000), and Time x Group Interaction (p<.000). Post-hoc analysis indicated that there were no significant differences between the TBI groups and non-mTBI group pre-deployment scores for all subtests (p>.05) except for CDS (p=0.042). Post-hoc analysis indicated the mTBI group had significantly worse post-deployment scores for all subtests (p<.000) than the non-mTBI group. Post-hoc analysis indicated that both the mTBI group and non-TBI group had significantly worse post-deployment scores than pre-deployment scores for all subtests (p<.000).
Results indicate that for some Marines with self-reported mTBI demonstrate declines in ANAM performance when compared to Marines without mTBI. This suggests that the ANAM might be reasonably sensitive to peristent decreases in ANAM performance. However, it should be noted that both the mTBI group and the non-mTBI group demonstrated declines in ANAM from pre- to post-deployment. These decrements indicate a possible “deployment effect” on cognitive functioning that must be expected and accounted for statistically if ANAM performance is to be used to track recovery over time.
BUMED M9, Wounded, III, and Injured Program
military, TBI, ANAM, concussion, neurocognitve
EFFECT OF PERFLUOROCARBON OXYGEN TRANSPORTERS, ON COAGULATION: IMPLICATIONS FOR TBI
Carlos Bidot Jr., BS, University of Miami
Shyam Gajavelli, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Mark Johansen, BS, University of Miami
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Wenche Ji, Ph.D., University of Miami
Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Cerebral ischemia is associated with high mortality in TBI. Monitoring of brain oxygen tension (PtiO2) shown reductions to <25 mm Hg02 in∼35% of severe TBI, and in this group outcome is significantly worse. Although PFCs are biologically inert, controversy about their possible effect on coagulation, may limit their use.
Compare the effect of two chemically and structurally different PFCs (designated PFC-A and PFC-B) on the thromboelastographic profile in blood from normal human subjects. Material and Methods: To evaluate the effect of these two PFCs on coagulation we used a Thromboelastography analyzer (TEG) which measures 5 parameters of clot formation and lysis: R (clotting time), K (time to reach 20 mm), Angle deg (A) (rate of clotting), MA (maximum amplitude) and LY30 (% lysis during 30 min after MA) - a close approximation of clotting in vivo. We used a physiological dose of 10% v/v of PFC/saline in whole blood samples, at 37°C. Samples were drawn from 12 healthy volunteers age between 23-43 y (4F/8M). We compared the effect of the PFCs and saline (as control) in each blood samples using T Test.
As shown in table 1, the parameters of hemostasis for PFC B were significantly more affected than controls, or PFC A (p<0.001) for all the parameters measured by TEG except for R-the initial clot formation time. At the doses used, and based on K and Angle the PFC B produce an slowed progression of the clot, a reduced clot strength (low MA) and increased clot lysis detected by the LY-30. These results suggest that PFC B have an anticoagulant effect.
Further studies, especially a dose response curve, and studies using blood from TBI patients, are needed to determine the optimal scenario for their use in TBI patients.
These studies were supported by CDMRP award PTO74521 (W81XWH-08-1-0419).
perfluorocarbon, coagulation, hypoxia, ischemia, TBI
A STUDY ON COMPLICATION OUTCOMES BETWEEN POLYETHERETHERKETONE AND TITANIUM CRANIOPLASTY
Mr. Nicolas KK King, MRCS (England), PhD, Neurosurgery, National Neuroscience Institute, Singapore
Prof. Beng It Ang, MBBS, FRCSGlasg, FRCSEd (SN), FAMS, Neurosurgery, National Neuroscience Institute, Singapore
Prof. IVAN NG, MBBS, FRCSE, FRCS (Surgical Neurology), FAMS, Neurosurgery, National Neuroscience Institute, Singapore
Polyetheretherketone (PEEK) is available as a prefabricated patient specific implant for cranioplasty (CP). At present, it remains unknown whether PEEK CP has lower complication and failure rates, when compared to titanium CP. In this study, we aim to investigate the complication outcomes between PEEK CP and titanium CP.
This is a retrospective study which involved patients who underwent PEEK CP and titanium CP at the National Neuroscience Institute, Singapore from January 2001 to February 2012. Data was collected by reviewing medical notes of the patients, which included parameters from previous craniectomy (indication for craniectomy, Glasgow Outcome Score, site and size, and any complications post-operatively) and parameters from CP (time interval to CP, type of implant used, type of complications, time to complications and duration of follow-up). Data was analysed using SPSS Statistics (IBM Corp, Armonk, New York, USA). The level of statistical significance was set at p<0.05.
134 patients (male/female ratio 0.64), with an average age of 41.7 (±16.0) years were included in this study, where 24 PEEK CP and 110 titanium CP were performed. The predominant indication for craniectomy was trauma and the mean time interval to cranioplasty was 11.2 months. The overall complication rate for all CP was 26.1%. Individual complication rates of 25.0% for PEEK CP and 26.4% for titanium CP were indentified. The predominant complication for both PEEK and titanium CP was exposed implant (4.2% and 13.6% respectively). Cranioplasty failure (defined as removal or revision of cranioplasty) rates of 12.5% for PEEK CP and 22.7% for titanium CP were noted. There were no significant statistical differences in complication and failure rates between the PEEK CP and titanium CP. Patients with previous cranial deep infection were at an increased risk for developing complications (OR 8.41, CI 0.034 to 0.52, p-value=0.004).
This study represents the largest to date in which PEEK CP and its complications have been evaluated, and compared with titanium CP. The overall complication rate of 26.1% is consistent with those reported in the literature and is not negligible. At present, complication outcomes between PEEK CP and titanium CP are comparable.
The staff at Neurosurgery Department, National Neuroscience Institute for their help with data collection.
cranioplasty, polyethertherketone, PEEK, titanium, complication
SERUM LEVELS OF SPECTRIN BREAKDOWN PRODUCT 150 (SBDP150) DISTINGUISH MILD TRAUMATIC BRAIN INJURY FROM TRAUMA AND UNINJURED CONTROLS AND PREDICT INTRACRANIAL INJURIES ON CT AND NEUROSURGICAL INTERVENTION
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Jason Demery, PhD, University of Florida
Salvatore Silvestri, MD, Orlando Regional Medical Center
Philip Giordano, MD, Orlando Regional Medical Center
Jay Falk, MD, Orlando Regional Medical Center
Kara Schmid, PhD, Walter Reed Army Institute of Research
Frank C Tortella, PhD, Walter Reed Army Institute of Research
Ronald Hayes, PhD, Banyan Biomarkers
Claudia Robertson, MD, Baylor College of Medicine
This study examined whether early serum levels of SBDP150 could distinguish: 1) mild TBI from 3 control groups; 2) those with and without traumatic intracranial lesions on CT (+CT vs −CT); and 2) those having a neurosurgical intervention (+NSG vs −NSG) in mild and moderate TBI (MMTBI).
This prospective cohort study enrolled adult patients presenting to 2 Level 1 Trauma Centers following mild and moderate TBI with blunt head trauma with loss of consciousness, amnesia, or disorientation and a GCS 9–15. Control groups included uninjured controls and trauma controls presenting to the ED with orthopedic injuries or an MVC without TBI. Mild TBI was defined as GCS 15 and moderate TBI as having a GCS <15. Blood samples were obtained in all patients within 4 hours of injury and measured by ELISA for SBDP150 (ng/ml). The main outcomes were: 1) the ability of SBDP150 to distinguish mild TBI from 3 control groups; 2) to distinguish +CT from −CT and; 3) to distinguish +NSG from −NSG. Data were expressed as means with 95%CI, and performance was tested by ROC curves (AUC and 95%CI).
There were 275 patients enrolled, 54 TBI patients (42 GCS 15, 12 GCS 9–14), 23 trauma controls (16 MVC controls and 7 orthopedic controls) and 198 uninjured controls. The mean age of TBI's was 39 years (range 19–70) with 63 males. Fourteen (14) had a CT and 9 had NSG. Mean serum SBDP150 levels were 0.764 (95CI 0.561–0.968) in normal controls, 1.035 (0.091–2.291) in orthopedic controls, 1.209 (0.236–2.181) in MVC controls, 2.764 (1.700–3.827) in mild TBI with GCS 15 and 5.227 (0.837–9.617) in TBI with GCS 9–14 (P<0.001). The AUC for distinguishing mild TBI from both controls was 0.83 (95CI 0.68–0.99). Mean SBDP150 levels in patients with −CT versus CT were 2.170 (1.340–3.000) and 6.797 (2.227–11.368) respectively (P<0.001) with AUC=0.78 (95CI 0.61–0.95). Mean SBDP150 levels in patients with −NSG versus NSG were 2.492 (1.391–3.593) and 6.867 (3.891–9.843) respectively (P<0.001) with AUC=0.88 (95CI 0.77–0.98).
SBDP150 is a breakdown product of the cytoskeletal protein alpha-II-spectrin found in neurons. To date this novel marker has been detected in severe TBI. In this study we found that SBDP150 was detectable in serum acutely within 4 hours after injury in MMTBI patients and was associated with measures of injury severity. Serum SBDP150 was able to distinguish mild TBI patients from control patients and was significantly higher in patients with intracranial lesions on CT and in those with neurosurgical intervention. Further study is required to validate these findings before clinical application.
This study was supported in part by DoD Award number DoD W81XWH-06-1-0517 and in part by NINDS Award Number R01NS057676.
Serum, biomarker, neuronal, TBI, neuroimaging
THE INFLUENCE OF IMAGING FINDINGS ON PATIENT SCREENING FOR THINC STUDY: AN INVESTIGATION OF mTBI IN A COMMUNITY HOSPITAL
Mary J. Rucker, B.S.N., Suburban Hospital Johns Hopkins Medicine
Jessica L. DeStefano, B.S., Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Martin R. Cota, B.A., Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Lawrence L. Latour, Ph.D., Stroke Branch, NINDS, NIH
Imaging-based research studies of mTBI sometimes include the diagnosis of concussion or mTBI as inclusion criteria. We have found that the clinical diagnosis of concussion by the ED may be imprecise and therefore sought to estimate the discordance between clinical diagnosis and evidence of injury on MRI.
Patients with acute head injury presenting to Suburban Hospital (Bethesda, MD) trauma/ED were enrolled in the THINC study. All patients with acute head injury within 48 hours were considered including patients that did not receive a diagnosis of mTBI, concussion, or meet the diagnostic criteria. CT was obtained in most for clinical purpose. Following enrollment, subjects underwent a standardized, targeted MRI protocol for detection of intracranial abnormalities related to trauma. Based on the findings, subjects were categorized based on the radiology report as CT+/− for 1) parenchymal injury or 2) SAH or extra-axial hemorrhage, and MRI +/−based on research evaluation for 3) parenchymal injury, and 4) SAH or extra-axial hemorrhage. The clinical diagnosis was then compared to the categorization of imaging. Values reported are median (IQR).
During the 14-month period from October 2010 to February 2012, 108 subjects were enrolled in the THINC study. The age was 45 (30–61), 37 (34%) were female. Time from injury to triage was 41(19–75) minutes and 88 (81.5%) were admitted to the hospital. 106 (98.2%) received a CT during ED admission. Of the 108 subjects, 32 (29.5%) received a diagnosis of concussion, 2 (6.3%) of which were CT+ for parenchymal injury, 1 (3.1%) CT+ for SAH, 8 (25%) MR+ for parenchymal injury. Of the 108 patients, 76 (70.4%) did not receive a diagnosis of concussion, 11 (14.5%) of which were CT+ for parenchymal injury, 14 (18.4%) CT+ for SAH, 29 (38.2%) MR+ for parenchymal injury. It is of note that patients could have had multiple image findings (parenchymal injury as well as hemorrhage) and that the results do not differentiate overlapping patient findings.
In recent years, TBI/concussion awareness has increased, which prompted a speculative review of concussion diagnosis in a local community hospital. It was observed by clinical staff that some head-injured patients were not given concussion diagnosis upon discharge from the ED. This observation was further supported by data collected for research purposes. Of the subjects in the study that did not have a diagnosis of concussion or TBI in the charts, ∼15% had injury to brain on clinical CT and ∼38% had injury to the brain on MRI. These results point toward an opportunity to implement a standardized pathway for screening for concussion and better communicating care instructions to the patient.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
TBI, concussion, MRI, CT, Imaging
CT SCAN INTERPRETATION IN TBI: IMPLICATIONS FOR RESIDENT TRAINING
Brittany Lewis, BS, UC Davis School of Medicine
Matthew Bobinski, MD, PhD, UC Davis Department of Neuroradiology
Anna Nidecker, MD, UC Davis Department of Neuroradiology
Nancy Rudisill, RN, MSN, CN IV, UC Davis Medical Center
Gene G. Gurkoff, PhD, UC Davis Department of Neurosurgery
Marike Zwienenberg-Lee, MD, UC Davis Medical Center
Paul Muizelaar, M.D., Ph.D., UC Davis
Kiarash Shahlaie, MD, PhD, UC Davis Medical Center
Resident interpretation of cranial CT scans has been reported in various subspecialties, but no such study exists in the neurosurgery literature. We compared neurosurgery resident trauma head CT interpretations with those of a faculty member, with emphasis on year of training and features comprising the Rotterdam and Marshall CT scores.
Data was prospectively collected from traumatic brain injury (TBI) patients at a level 1 trauma center between September 2010 and November 2011. Enrollment criteria included adult and pediatric patients with blunt or penetrating TBI and abnormal admission head CT scans, or persistent neurological deficit and a negative head CT. All patients were evaluated by the neurosurgery resident on call. Resident interpretations were entered into an electronic consult note that required responses to variables such as the presence of parenchymal hematoma, contusions, intraventricular hemorrhage (IVH), epidural hematoma (EDH), subdural hematoma (SDH), subarachnoid hemorrhage (SAH), midline shift, cisternal compression, and skull fractures. An attending neurosurgeon who was blinded to resident interpretations reviewed each CT scan and recorded his interpretations. Marshall and Rotterdam scores were assigned to each CT scan interpretation using data gathered from the consult notes. Members blinded to the final analysis entered the data into a secure centralized database.
615 patients were evaluated for TBI at our level 1 trauma center between September 2010 and November 2011. Twenty-nine patients were excluded from the study due to unavailability of their image, resulting in a final study population of 586 patients. This group had an average age of 37 years, was comprised of 414 men and 172 women, and 238 (40.6%) of them were transferred from an outside hospital. Junior residents saw the majority of consults: 40% by PGY1s, 32% by PGY2s, and 22% by PGY3s. There were significant differences between resident and faculty interpretations with regard to the presence of contusions, IPH, and SAH. When comparing junior residents (PGY1 and PGY2) to faculty, the residents recorded more IPH and SDH, whereas faculty recorded more contusions, IVH, and SAH. More senior residents (PGY3 and PGY4) did not differ significantly from faculty with regard to identifying the presence of IVH or SAH. The presence of EDH and midline shift was also not significantly different between residents and faculty. The effects of these differences on CT scoring systems were also investigated. Resident interpretations resulted in more patients with Rotterdam scores of 2 and less scores of 3 as compared to faculty. When stratified by year of training, this difference diminished with each successive year of training. Resident scores were nearly identical to that of faculty by PGY3. Marshall CT scores demonstrated an opposite trend, with resident interpretations yielding a higher rate of false positive interpretations when scans were actually normal. This became less significant with increasing year of training. Type 3 injuries became more similar with each training year.
Neurosurgery residents generally perform well with respect to interpreting key features of trauma head CTs; and they are often on par with that of faculty. Differences between resident and faculty interpretation highlight the importance of targeted education, and may influence accurate patient prognostication and/or enrollment in clinical trials that are based on admission CT criteria. However, these differences in interpretation do not seem to affect acute clinical decision making, as even our junior-most residents were adept at identifying large extra-axial lesions and midline shift, features that may prompt immediate surgical intervention and triaging. This large blinded investigational study confirms neurosurgery residents' abilities to critically and accurately interpret trauma head CT scans.
n/a
computed tomography, resident education, Traumatic Brain Injury
ADVANCED MAGNETIC RESONANCE IMAGING IN THE ACUTE AND SUBACUTE PHASE OF MILD TRAUMATIC BRAIN INJURY: CAN WE SEE THE DIFFERENCE?
Noemi Kovacs, M.D., Department of Neurosurgery, University of Pécs
Gábor Perlaki, MSc., University of Pécs Medical School
Gergely Orsi, MSc., Diagnostic Centre of Pécs
Mihály Aradi, M.D., Diagnostic Centre of Pécs
Hedvig Komáromy, M.D., Diagnostic Centre of Pécs
Péter Bukovics, Ph.D., Department of Neurosurgery, University of Pécs
Orsolya Farkas, M.D., Department of Radiology, University of Pécs
Prof. Tamás Dóczi, M.D., Ph.D., Department of Neurosurgery, University of Pécs
Prof. József Janszky, M.D., Ph.D., Department of Neurology, University of Pécs
Attila Schwarcz, M.D., Ph.D., Department of Neurosurgery, University of Pécs
Andras Buki, M.D., Ph.D., Pecs University
Advanced MRI methods were proven sensitive in detecting structural and functional alterations that follow mild traumatic brain injury (mTBI). This work investigates changes of the brain during the first month after mTBI by diffusion tensor imaging (DTI), volumetric analysis, susceptibility weighted imaging (SWI) and functional MRI.
Seven uncomplicated mTBI patients underwent the MRI protocol at two time points: within 72 hours (ranging from 12 h to 72 h) after injury and one month later (ranging from 28 to 43 days). An age and sex matched control group with similar acquisition time frame was also involved. Voxelwise analysis of DTI indices mean diffusivity and fractional anisotropy was carried out using Tract Based Spatial Statistics (TBSS), part of FSL (FMRIB's Software Library, www.fmrib.ox.ac.uk/fsl). T1 weighted images were fed into FreeSurfer (Athinoula A. Martinos Center for Biomedical Imaging, 2005) whole brain volumetric segmentation. SWI images were searched for micro haemorrhage by a board certified neuroradiologist. Covert word generation and Roland's Hometown walking tasks were performed during fMRI, first and higher level statistics were calculated in FSL. MRI data were compared across groups and time points as well.
T1-, T2- and susceptibility weighted imaging revealed no pathology. TBSS showed mean diffusivity to be significantly (corrected p<0.05) higher in the mTBI group in several white matter tracts (42721 voxels) compared to controls within 72 hours after injury and still one month later, but in a smaller extent (10101 voxels). Longitudinal analysis revealed significant decrease (i.e. normalization) of mean diffusivity over one month dominantly in the left hemisphere (6116 voxels). Significant (p<0.05) decrease in cortical volumes (mean 1,49%) and increase in ventricular volumes (mean 6,95%) appeared comparing 72h to one month brain volumes of the mTBI group. Reduced blood oxygen level dependent signal was detected by higher level analysis during Hometown walking task at 72h after injury compared to one month over areas as parahippocampal gyrus and temporal lobe which were involved in the task induced activation. During covert word generation task in the mTBI group, the 72h blood oxygen level dependent signal was found higher than one month later over the frontal medial cortex that was not involved in word generation, instead was included in the deactivation areas.
Our findings present dynamic structural and functional changes occurring in the acute to subacute phase of mTBI, in patients lacking micro haemorrhage detectable by SWI. DTI data suggest an asymmetric regenerative process following an initial traumatic disintegration of the white matter tracts. The reduction of overall brain tissue volume based on cortical volume decrease and ventricle expansion over one month may represent recovery from initial oedema. Cortical activation changes revealed by fMRI indicate altered recruitment of neural resources. These alterations of the injured brain revealed by MRI may lie behind cognitive disturbances and post-traumatic clinical symptoms. Our results underscore the importance of strict image acquisition time points when performing MRI studies on patients with mTBI.
This work was supported by Developing Competitiveness of Universities in the South Transdanubian Region (SROP-4.2.1.B-10/2/KONV-2010-0002).
mild traumatic brain injury, diffusion tensor imaging, susceptibility weighted imaging, volumetry, functional MRI
UCH-L1 AND MAP-2 BIOMARKERS IMPROVE 6-MONTH FUNCTIONAL OUTCOME PREDICTION IN SEVERE TRAUMATIC BRAIN INJURY
H. Julia Hannay, PhD, University of Houston
Linda Papa, MD, MSc, Orlando Regional Medical Center
Callie Tyner, MS, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
Claudia Robertson, MD, Baylor College of Medicine
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Steven Robicsek, MD, PhD, University of Florida
Identification of biomarkers predictive of mortality and functional outcome is important for clinical management of severe TBI. We examined whether acute levels of biomarkers UCH-L1 or MAP-2, in combination with patient demographic characteristics and acute injury variables, could predict a range of functional outcomes six months post-injury.
Data was obtained from adult severe closed head injury patients presenting to two Level I Trauma Centers. Patients had a GCS score<8 and required intracranial pressure (ICP) monitoring. Ventricular CSF was sampled within 24 hours of injury and analyzed for biomarkers using ELISA. Biomarker levels were measured in ng/ml. Two stepwise hierarchical linear regressions were constructed to model functional outcome prediction including either UCH-L1 or MAP-2. Both models evaluated predictive contributions of 3 demographic variables (age, pre-injury occupational status, educational attainment) and 4 acute injury variables (post-resuscitation GCS motor and pupillary response, Rotterdam score, ISS), in conjunction with either the UCH-L1 or MAP-2 biomarker level. Both models predicted 6 month outcome measured by the Disability Rating Scale (DRS), which quantifies functioning across 4 domains and yields a total score between 0 (no disability) and 29 (maximal disability), with 30 to indicate deceased.
There were a total of 131 patients enrolled in the broader multi-site prospective study. Patients had a mean age of 38 years and 78% were male. As expected given the injury severity enrollment criteria, 68% had CT Marshall Classification of Diffuse Injury I-IV and 32% had Mass lesions. At 6 months post-injury, 104 patients had available follow-up outcome data. Of these, 84 patients had CSF collected for biomarker analysis within 24 hours of injury. Thirty (33%) patients did not survive to 6 months and were assigned a DRS score of 30. The remaining 54 surviving patients had a mean DRS score of 6.7 at the 6 month post-injury time point, with scores ranging from 0 (no disability) to 28 (extreme vegetative state) and a group median value of 4 (moderate disability). Both of the predictive models were statistically significant (p<0.001) and accounted for 54% of the variance in DRS scores at 6 months post-injury. Surprisingly, pre-injury educational attainment, post-resuscitative GCS motor score, post-resuscitative pupillary response, and Injury Severity Score (ISS) did not contribute to either predictive model. The stepwise hierarchical regression which included UCH-L1 yielded a model comprised of the following variables in order of the outcome variance they explained: pre-injury occupational status (R2 change=0.255), Rotterdam Score (R2 change=0.166), age (R2 change=0.109), and UCH-L1 (R2 change=0.039). The regression model which included MAP-2 yielded a very similar model, with only slightly different levels of variance explained by the predictive variables: pre-injury occupational status (R2 change=0.247), Rotterdam score (R2 change=0.159), age (R2 change=0.110), and MAP-2 (R2 change=0.028).
The results of this study suggest that the inclusion of UCH-L1 or MAP-2 biomarkers drawn within 24 hours of severe TBI can help to predict broad functional outcome of patients 6 months post-injury. This is an important contribution to previous findings that indicate these biomarkers may aid prediction of mortality or more basic dichotomous (good/bad) outcomes after TBI. Using a model comprised of patient demographic characteristics, acute injury variables, and either UCH-L1 or MAP-2 successfully predicted over half (54%) of the functional outcome variance in the current TBI sample. Somewhat surprisingly, some of the acute injury variables traditionally included in outcome prediction models (i.e., GCS motor score, pupillary response, and ISS) did not contribute to the prognostic model. Conversely, patient's pre-injury occupational status accounted for the largest amount (25%) of variance in 6 month functional outcome, highlighting the importance of considering pre-injury functioning when making functional outcome predictions. Future studies are needed to explore the prognostic value of similar models in predicting functional outcome specifically among survivors. Additional studies should also explore the optimal combination of specific predictors and functional outcome measures to provide the most useful information regarding individual patient prognosis after severe TBI.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
UCH-L1, MAP-2, Biomarker, Outcome, TBI
PREDICTING MORTALITY IN SEVERE TBI USING THE IMPACT SCORE AND EARLY CSF MAP-2 LEVELS
Linda Papa, MD, Orlando Regional Medical Center
Andrea Gabrielli, MD, University of Florida
Shelley Heaton, PhD, University of Florida
H. Julia Hannay, PhD, University of Houston
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Ronald Hayes, PhD, Banyan Biomarkers
Claudia Robertson, MD, Baylor College of Medicine
An assessment of whether early levels of biomarkers measured in CSF within 24 hours of injury would improve upon the clinical prediction of mortality in patients with severe TBI.
A prospective observational study design that enrolled adults with severe TBI presenting to two Level I trauma centers (Gainesville, FL and Houston, TX). Patients were included following blunt head injury with a GCS score of 8 or less and required a ventriculostomy. Ventricular CSF was sampled from each patient within 24 hours of injury and analyzed for candidate biomarkers using ELISA. Biomarker levels were measured in ng/ml. The IMPACT score (www.tbi-impact.org) was calculated for each patient to determine risk of mortality at 6 months. ROC curves were constructed to determine if the biomarkers added prognostic value to the Core IMPACT score (age, pupillary reactivity and GCS motor score), Extended IMPACT score (Core + hypoxia, hypotension, CT findings), Lab IMPACT score (Extended + glucose and hemoglobin).
There were 131 patients enrolled in the study. Mean age of patients was 38 years and 78% were male; 68% had CT Marshall Classification of Diffuse Injury I-IV and 32% were with mass lesions. At 6 months 104 patients had follow-up data and, of these, 84 had CSF collected for biomarker analysis within 24 hours of injury. Thirty (36%) patients did not survive to 6 months. The area under the ROC curve (AUC) for the Core, Extended and Lab IMPACT Models for predicting mortality at 6 months were 0.77 (0.68–0.87) for Core, 0.77 (0.67–0.88) for Extended, 0.74 (0.63–0.84) for Lab. The AUC for MAP-2 drawn within 24 hours of injury was 0.65 (0.52–0.77). However, when the MAP-2 was added to the Core IMPACT model the AUC for predicting mortality increased to 0.80 (0.70–0.90).
These data suggest that early levels of this novel biomarker provided added prognostic information to clinical data about mortality risk at 6 months in patients with severe TBI.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
Biomarker, TBI, IMPACT, prediction
INFLAMMASOME PROTEINS IN CEREBROSPINAL FLUID OF BRAIN INJURED PATIENTS ARE BIOMARKERS OF FUNCTIONAL OUTCOME
Gordon Dale, B.S., University of Miami
Juan Pablo de Rivero Vaccari, Ph.D., Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Professor Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton Dietrich, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Robert W. Keane, Ph.D., Department of Physiology and Biophysics, University of Miami Miller School of Medicine
To evaluate TBI severity and prognosis, physicians rely on clinical variables. Here we seek objective, biochemical markers reflecting molecular injury mechanisms specific to the CNS as more accurate measurements of injury severity and outcome. One such secondary injury mechanism, the innate immune response, is regulated by the inflammasome.
We investigated whether inflammasome components are present in the cerebrospinal fluid (CSF) of TBI patients, and whether levels of inflammasome components correlate with outcome. We performed immunoblot analysis of CSF samples from TBI patients and non-trauma controls and assessed outcome five months post-injury by the Glasgow Outcome Scale (GOS).
Patients with severe or moderate cranial trauma exhibited significantly higher CSF levels of the inflammasome proteins apoptosis-associated speck-like protein (ASC), caspase-1, and NAcht leucine-rich-repeat protein-1 (NALP-1) compared to non-trauma controls. ASC, caspase-1, and NALP-1 were significantly higher in the CSF of patients with unfavorable outcomes, including death and severe disability.
NALP-1 inflammasome proteins are potential biomarkers to assess TBI severity, outcome, and the secondary injury mechanisms impeding recovery, serving as adjuncts to clinical predictors.
Miami Project to Cure Paralysis
biomarkers, inflammation
THE INFLUENCE OF ISOLATED HEAD INJURY STATUS ON INFLAMMATION CYTOKINE LEVELS POST TRAUMATIC BRAIN INJURY
Krutika Amin, BS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Christian Niyonkuru, MS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Rachel Berger, MD, MPH, Children's Hospital of Pittsburgh
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Samuel A. Tisherman, MD, University of Pittsburgh
Amy Wagner, MD, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Neuroinflammation is a major contributor of secondary injury occurring after traumatic brain injury (TBI), and inflammatory pathways can have dual actions that contribute to cell damage as well as support cell survival. Major extra-cerebral trauma is accompanied by a significant inflammatory response. However the effects of extra-cerebral trauma on neuroinflammation and the role of early inflammation on TBI outcome are not well studied.
We measured cerebrospinal fluid (CSF) and serum cytokine levels in 97 adults during the first six days after severe TBI. IL-1b, IL-5, IL-6, IL-8, IL-10, IL-12, TNF-a, ICAM-1, and VCAM-1 were measured by Luminex™ multiple bead assays. Isolated/Non-Isolated head injury status was determined using the anatomic injury scale (AIS) across multiple body regions. Isolated TBI was defined as head AIS≥3 in patients with AIS<3 for other anatomic regions. Non-Isolated TBI was defined as head AIS≥3, as well as an AIS score of ≥3 for at least one other anatomic region that was not the head, neck, or face. We compared cytokine levels (averaged over the first 6 days post-TBI) to clinical variables, demographics, and outcomes for subjects with isolated TBI versus those with non-isolated TBI. CSF and serum collected from healthy controls was used as a comparison for biomarker levels from both TBI groups. Acute care mortality as well as Glasgow Outcome Scale (GOS) and Disability Rating Scale scores (DRS) were assessed at 6 months after injury.
The study cohort consisted of 97 subjects, 76 males and 21 females. The mean Glasgow Coma Scale (GCS) score for the entire cohort was 5.79. 53 subjects were categorized as having a non-isolated head injury and 44 subjects had an isolated head injury. The GCS scores tended to be higher in the isolated head injury cohort (median GCS of 6.5), compared to the non-isolated cohort (median GCS of 6) (p=0.051). In total, 353 CSF samples from 70 subjects, and 303 serum samples from 81 subjects were analyzed. Compared to controls, all cytokine levels (except serum IL-12, CSF IL-5, and CSF VCAM-1) were much higher in subjects with TBI vs. healthy controls, (p<0.010 all comparisons). Despite GCS scores that were modestly better than the non-isolated TBI cohort, it was the isolated TBI cohort that had significantly higher CSF levels of IL-5, IL-6, IL-8, IL-10, ICAM-1 and VCAM-1. However, the isolated TBI group also had lower CSF levels for IL1-b, IL-12 and TNF-a. The non-isolated TBI cohort had higher serum cytokine levels for all cytokines, reaching significance in IL1-b, IL-5, IL-6, and IL-10, (p<0.017 all comparisons) except IL-12 and VCAM-1. Interestingly, there were no differences in outcomes for either the isolated or non-isolated TBI group with regard to acute mortality, global outcome, disability, length of stay hospital/ICU, number of complications, or incidence of sepsis.
Extracerebral injury appears to play a significant role in the evolution of neuro-inflammation by influencing the transport of inflammatory factors from the peripheral circulation to the CNS. We speculate that the differential inflammatory responses in brain and the peripheral circulation in adults with isolated and non-isolated TBI both contribute to impaired outcome.
DOD W81XWH0710701; CDC R49 CCR 323155-03; NIH 5P01 NS030318-16
TBI, Inflammation, Cytokines, Recovery
PRE-MORBID AND HOSPITAL COURSE FACTORS AFFECTING INTRACRANIAL HEMORRHAGE EXPANSION AND MORTALITY IN THE ELDERLY POST TRAUMATIC BRAIN INJURY
Gillian Harrison, BS, MS, University of Pittsburgh School of Medicine
Ramesh Grandhi, MD, University of Pittsburgh Medical Center Dept. of Neurological Surgery
Zoya Voronovich, BS, University of Pittsburgh School of Medicine
Ava Puccio, RN, PhD, University of Pittsburgh School of Medicine Dept. of Neurological Surgery
David O. Okonkwo, MD/PhD, University of Pittsburgh
As the U.S. population ages, the number of elderly people on anti-thrombotic medications increases. Previous studies of traumatic brain injury (TBI) outcomes in this demographic have yielded conflicting results. The purpose of this study is to examine the pre-hospital and in-hospital factors affecting the progression of intracranial hemorrhage.
We performed a retrospective analysis of elderly TBI patients (>65 years) with evidence of brain hemorrhage on computed tomography (CT) scan at our institution from 2006–2010. Patient demographics, injury severity (Head Abbreviated Injury Score (AIS), Marshall Score, and Glasgow Coma Scale (GCS) score), clinical course (intracranial progression, infection, ventilator days, medications), hospital and ICU length of stay (LOS), and mortality/discharge disposition were collected. Descriptive patient data were stratified by pre-morbid Oral Antithrombotic (OAT) use, and analyzed via non-parametric methods; Kruskal-Wallis for continuous variables and Fisher's Exact for categorical variables. Multiple regression analyses were performed to compare groups and identify predictors of mortality, complication, infection, neurosurgical intervention, and radiographic progression. Covariates accounted for in the regression model include: age, gender, ISS, Head AIS, Trauma Injury Severity Score (TRISS), admit GCS, pre-existing comorbidities, both Place and Mechanism of Injury, OAT reversal, radiographic intracranial pathology, and Marshall Score.
1552 patients were identified with 543 aspirin only, 97 clopidogrel only, 218 warfarin only, 193 clopidogrel/aspirin, and 501 patients on no antithrombotic. Significant differences existed in AIS (p=0.012), GCS score (p=0.013), and Marshall Score (p<0.001), at time of presentation. Blood products (Vitamin K, FFP, and/or Platelets) were administered to reverse coagulopathy in 77.3% of patients. After adjusting for covariates, including medication reversal, antithrombotic use was associated with increased mortality (p=0.03). Subgroup multiple regression analysis indicated that warfarin (OR 2.53, p=0.003), but not aspirin and/or clopidogrel (p=0.622), is associated with higher mortality after a TBI. Rates of neurosurgical interventions (Craniotomy/Craniectomy, External Ventricular Drain, and/or insertion of intraparenchymal pressure monitor) (p=0.677) did not differ between groups. Survivor subset analysis demonstrated that CT-identified hemorrhage progression was not associated with preinjury antithrombotic therapy, nor were rates of complication or infection development, hospital/ICU LOS, ventilator days, or discharge disposition. When stratifying for severe and moderate TBI only, use of antithrombotics did not affect patient outcomes.
Preinjury use of warfarin, but not antiplatelet medications, influences survival in elderly patients admitted with TBI. Outcomes including radiographic progression, neurosurgical interventions, and morbidity are not affected. The importance of pre-morbid antithrombotic therapy seems to lie in its impact on initial injury severity, which, in turn, is predictive of increased morbidity and mortality.
University of Pittsburgh Medical Center, University of Pittsburgh School of Medicine, Brain Trauma Research Center
Anti-Thrombotics, Radiographic Progression, Elderly, Mortality
PREVALENCE AND RISK FACTORS FOR INTRAOPERATIVE HYPOTENSION DURING CRANIOTOMY FOR TRAUMATIC BRAIN INJURY
Michelle Brown, BS, University of Washington
Sakura Noda, BA, University of Washington
Randall Chesnut, MD, University of Washington
Monica Vavilala, MD, University of Washington
Hypotension after traumatic brain injury (TBI) is associated with poor outcomes. However, data on intraoperative hypotension (IH) are scarce and the effect of anesthetic agents on IH is unknown. We examined the prevalence and risk factors for IH, including the effect of anesthetic agents during emergent craniotomy for isolated TBI.
Following Institutional Review Board approval, a retrospective cohort study of patients ≥18 years who underwent emergent craniotomy for isolated TBI at Harborview Medical Center (level-1 trauma center) between October 2007 and January 2010 was conducted. Patients undergoing repeat/non-emergent craniotomy, burr hole evacuation and patients with polytrauma were excluded. Demographic, clinical and radiographic characteristics, hemodynamic and anesthetic data were abstracted from medical and electronic anesthesia records. Hypotension was defined as systolic blood pressure (SBP) <90 mmHg. The electronic anesthesia records displayed the hemodynamic parameters once every five minutes and every displayed reading of SBP<90 mmHg was counted as an “episode” of IH. Univariate analyses were performed to compare the clinical characteristics of patients with and without IH and multiple logistic regression analysis was used to determine independent risk factors for IH.
One-hundered thirteen patients (48±19 years) were included in the final analysis. Fifty-three percent had severe TBI (Glasgow Coma Scale score≤8) and 4 had admission hypotension. One patient had diabetes mellitus, 5 had ischemic heart disease, 5 others had hypertension and 13 had some combination of the three. Propofol and etomidate were used for induction in 64 and 4 patients, respectively while inhalation of volatile anesthetic through indwelling tracheal tube was used in 42 patients. Only 3 patients received propofol for maintenance and the remaining received volatile anesthesia. Seventy-three patients (65) had at least one episode of IH. Thirty-four (30) patients had 1–4 episodes and 39 (35) patients had > 5 episodes of IH. The mean lowest intraoperative SBP was 68±11 mmHg (range 50–85 mmHg) for the IH group and 97±8 mmHg (range 90–120 mmHg) for non-IH group. The median number of episodes of IH was 5 (range 0–40). The most common vasopressor used was phenylephrine. The TBI severity was similar in patients with and without IH (p=0.62). There was no difference in the distribution of co-morbidities (p=0.54) or use of ACE inhibitors, beta-blockers, furosemide or any combination of these medications (p=0.17). There was no difference in the prevalence of IH based on induction agents (p=0.87), maintenance anesthetic agents (p=0.37) or opioid administration (p=0.45). Univariate risk factors for IH were subdural hematoma (SDH), multiple computed tomographic (CT) lesions, immediate preoperative SBP, and anesthesia duration (Table 3). Independent risk factors for IH were multiple CT lesions (AOR 19.1 [95 CI: 2.08–175.99]; p=0.009), SDH (AOR 17.9 [95 CI: 2.97–108.10]; p=0.002), maximum CT lesion thickness (AOR 1.1 [95 CI: 1.01–1.13]; p=0.016), and anesthesia duration (AOR 1.1 [95 CI: 1.01–1.30]; p=0.009).
Intraoperative hypotension was common in adult patients with isolated TBI undergoing emergent craniotomy despite the absence of preoperative hypotension and was not associated with the choice of anesthetic agents. The injury related factors and radiographic features appear to be more significant in contributing to IH in TBI rather than pharmacological factors. Anesthesiologists should anticipate the need to treat IH in patients with multiple CT lesions or SDH requiring longer duration of anesthesia.
None
TBI, anesthesia, hypotension, craniotomy
PERFORMANCE ON SIMILAR TASKS OF THE DEFENSE AUTOMATED NEUROBEHAVIORAL ASSESSMENT (DANA) AND AUTOMATED NEUROCOGNITIVE ASSESSMENT METRICS (ANAM) IS COMPARABLE
Eusebio Flores, PA-C, JD, 1/3
Alia Creason, PhD, BUMED
Jay Haran, PhD, LT, MSC, USN, Humans Systems Department, Naval Air Systems Command
The Defense Automated Neurobehavioral Assessment (DANA) is a series of cognitive tasks on a portable handheld computer newly developed for frontline battlefield use to detect cognitive changes after concussion. Performance on DANA cognitive tasks also tested by the Automated Neurocognitive Assessment Metrics (ANAM) was compared.
40 Marines took both ANAM and DANA sequentially. Both ANAM and DANA have tasks measuring simple reaction time, procedural reaction time, and code substitution. Results were analyzed using paired t-tests with Bonferroni correction. p<0.0167 was defined as being significant.
Analysis of group scores among the test takers indicate that the 3 tests of cognitive function which DANA and ANAM have in common were not significantly different. There also was no difference between individual performances on the tasks.
These results indicate that performance on the tests DANA and ANAM have in common are similar with DANA. If battlefield testing also demonstrates no significant difference between these cognitive tests, forward-deployed units may be able to use automated neurocognitive testing reliably to aid with return to duty determinations after concussion.
These are the private views of the authors and are not to be construed as official or reflecting the views of the DoD.
Neurocognitive, testing, automated, concussion
POTENTIALLY INAPPROPRIATE MEDICATION USE IN CRITICALLY ILL ELDERLY PATIENTS WITH TRAUMATIC BRAIN INJURY
Patricia Slattum, PharmD, PhD, Virginia Commonwealth University, Medical College of Virginia
Perry Taylor, PharmD, Virgnia Commonwealth University Health System
Spencer Harpe, PharmD, PhD, MPH, Virginia Commonwealth University, Medical College of Virginia
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Potentially inappropriate medications (PIMs) have been shown to impact outcomes in elderly patients. PIMs are medications that may increase cognitive burden in older adults such as opioids, antidepressants, antipsychotics, and benzodiazepines. This study evaluates the use of PIMs and associated clinical outcomes in elderly patients with traumatic brain injury (TBI).
This is a retrospective study of patients ≥65 years of age admitted to the neuroscience intensive care unit (NSICU) of Virginia Commonwealth University Health System (VCUHS) between March 1, 2011 and July 31, 2011. Electronic medical records were used to obtain the number of PIMs administered during NSICU admission, LOS and discharge disposition. The change in neurological status from pre- to post dose when given sedatives or opioids, as determined by the Glasgow Coma Scale (GCS) and Richmond Agitation and Sedation Scale (RASS), was also measured. Patient variables were summarized using median and interquartile range (IQR) for continuous variables or percentages for categorical variables. Fisher's Exact test was used to compare groups.
Of 112 critically ill older patients admitted to the NSICU during the 5 month study period, 33 patients (29.5) had a diagnosis of TBI. The median age was 81.2 years (IQR, 74.7 – 86.4 years), 42 were male, median admit GCS was 14 (IQR, 13 – 15) and mean (SD) NSICU LOS was 3(2) days. Type of TBI included subdural hematoma (n=18), subarachnoid hemorrhage (n=9), intraparenchymal hemorrhage/contusion (n=5), and intracerebral hemorrhage (n=1). Antithrombotic agents were reported upon NSICU admission in 24 of patients (anticoagulation therapy, n=3; antiplatelet therapy, n=5). Overall, 79 (n=27) of patients were given at least 1 PIM. During NSICU admission, 18 received no intermittent PIMs, 30 of patients received 1 intermittent PIM, 24 received 2 intermittent PIMs and 27 received ≥3 intermittent PIMs. Twelve percent of patients were given at least 1continuous intravenous infusion (cIV) PIM. A total of 278 intermittent PIM doses were given during NSICU admission and a decrease in GCS from pre- to post-dose was recorded 15 of the time. A decrease in GCS>2 points occurred after 2 of intermittent PIM doses. A change in RASS score was reported after 36 of intermittent PIM doses and median (IQR) recovery time to a goal RASS score of 0 or −1 was 28 hrs (20 – 47). The most common PIMs associated with a change in RASS were famotidine, propofol and fentanyl. Mean (SD) NSICU LOS was 4 (3.1) days in patients with ≤1 intermittent PIM and 2 (1.3) days in those receiving ≥2 intermittent PIMs. In patients receiving ≤1 intermittent PIM, 93 survived to discharge compared to 83 of patients who received ≥2 intermittent PIMs (p=0.4).
A majority of TBI patients receive at least 1 intermittent PIM while in the NSICU. GCS and RASS scores did change post PIM dose and median time to recovery was greater than 24 hours. Additional studies are needed to determine associations between increased PIM use and clinical outcomes in elderly TBI patients.
None
Elderly, TBI, potentially inappropriate medication
EARLY RESPONSE OF NEURAL VASCULATURE TO MILD TRAUMATIC BRAIN INJURY AT THE ACUTE STAGE
Hardik Doshi, M.S., Wayne State University School of Medicine
Valerie Mika, Ph.D. Candidate, Wayne State University School of Medicine
Jie Yang, Ph.D., Wayne State University School of Medicine
Grace Ma, M.D., Wayne State University School of Medicine
Prof. Robert Welch, M.D., Wayne State University School of Medicine
Prof. Mark Haacke, Ph.D., Wayne State University School of Medicine
Mild Traumatic Brain Injury (mTBI) accounts for over 1 million emergency visits each year. The diagnosis is still a challenge at the acute setting. The purpose of this study was to evaluate the neural vascular response to brain injury at the acute stage.
Sixteen mTBI patients and twenty age and gender matched healthy controls were recruited at the acute stage. The patients underwent MRI scan and a short neurocognitive test, Standard Assessment of Concussion (SAC). The MR imaging protocol includes susceptibility weighted imaging (SWI) and ASL and other sequences.
For ASL image processing, all relative cerebral blood flow (rCBF) images are co-registered onto T1 atlas with prior defined anatomical regions of interest. Then these ROIs were inversely transformed to each individual subject's rCBF images for the measurement of blood flow in each major region. For quantitative susceptibility mapping, a series of image processing has been performed as described in our preliminary work to extract the susceptibility signal of eight major veins. Therefore, the susceptibility signal is proportional to the blood deoxygenation.
Our data demonstrate that: a) patient group has significant lower susceptibility signal than control group in left thalamus striatal veins, which implies that patient group has lower deoxygenation in thalamus region than controls; b) Patient group has significant higher CBF levels than control group in left frontal lobe and left striatal region, including caudate, putamen and globus pallidus; c) patient group demonstrated significantly reduced SAC scores than published normative data, especially in delayed recall; and d) the imaging data is not correlated with SAC data.
This results imply that, after brain concussion, the neural vascular system tend to increase the blood flow to overcome the consequences or help neurons to recover itself at the acute stage. Therefore, it manifest either higher level of rCBF in artery or lower level deoxyhemoglobin (more oxyhemoglobin left over) in veins. This result is consistent with the literature in severe TBI that, at the acute stage, brain tends to be hyperactive. However, it is in contrast with the published data on decreased CBF in chronic mTBI patients. Taken together, it demonstrates an overshoot and then undershoot profile of cerebral vascular response to brain injury in brain concussion.
Our data demonstrated increased neural vasculature response to brain injury at the acute stage in mild TBI patients.
This project is sponsored by the Department of Defense grant.
traumatic Brain Injury,
MULTIPLE BLAST EXPOSURES EXACERBATE SYSTEMIC LEAKAGE OF NEURO-GLIAL BIOMARKERS IN A RAT MODEL OF BLAST INJURY
Dancia Scharf, MS, Banyan Biomarkers
Hector Gutierrez, Ph.D., Florida Institute of Technology
Daniel Kirk, Ph.D., Florida Institute of Technology
Artem Svetlov, Banyan Biomarkers, Inc.
Kenneth Curley, Ph.D., U.S. Army Medical Research and Materiel Command
Ronald Hayes, PhD, Banyan Biomarkers
Stanislav Svetlov, Ph.D., M.D., Banyan Biomarkers, Inc.
The current TBI paradigm suggests that the blood–brain barrier (BBB) is often compromised during brain trauma, which permits concomitant release of brain-specific proteins into circulation. Although exposure to repeated low level blasts is a common hazard for war zone personnel and civilian populations, the cumulative effects of multiple blasts on brain injury have not been investigated. As a result, identifying biochemical markers after repeated blast exposures is vital to the development of diagnostics and risk assessment for blast-induced mild TBI.
Our previous studies characterized the importance of positional interaction between the blast wave and rat head/body and roles of head hyperacceleration in a rat model of overpressure-induced TBI. Here, we compared the effects of a series of multiple rat exposures to a single overpressure event inducing brain injury by a controlled primary blast wave load. Assessment of previously characterized and potential novel TBI biomarkers was done by ELISA, antibody microarrays, and Western blot. Rats were exposed to a primary blast wave with 365 kPa overpressure and total duration 75 μsec at the target. Multiple blasts were performed as a series of 3 exposures, with a 45 min to 1 hr recovery period between each blast.
High speed imaging revealed a negligible degree of acceleration at rat positions “off-axis” toward shock tube, thus confirming primary blast kinematics. We examined blood accumulation of glial markers (including GFAP and CNPase) and neural markers (including UCH-L1 and NSE) at 1 day and 7 days post-blast. Remarkable elevation of all proteins studied was observed at 1 day after either single or multiple exposures to blast. Of note, multiple blasts significantly increased blood levels of GFAP, UCH-L1, and NSE vs. single blast exposure 1 day post-blast, but at 7 days the effects of multiple blasts were much lower compared to a single blast. On the other hand, serum CNPase after multiple blasts was significantly elevated compared to single blast both at 1 day and 7 days post exposure.
In summary, the time-course of serum biomarkers of neuro-glial injury was characterized. Biomarker levels rose significantly as a rapid response at day one post-blast, with the CNPase, NSE, and UCH-L1 levels after repeated blast exposures elevating further over single blast. The appearance of characteristic proteins in circulation may reflect deterioration of the BBB and can be used for assessment of injury accumulation. However, multiple blast exposures did not enhance biomarker increases detected at 7 days post exposure compared to a single blast. At this time point, their elevated levels depend on the cell-specific origin of biomarkers and the stage of injury, rather than reflect a cumulative blast load.
This work was supported by grants W81XWH-8-1-0376 and W81XWH-07-01-0701 from Department of Defense.
Blast; TBI; Neuro-glial; Biomarkers; Circulation
DIFFERENTIAL EFFECT OF NICOTINAMIDE ON SERUM DAMAGE MARKER PROFILES FOLLOWING CONTROLLED CORTICAL IMPACT, PARASAGITTAL FLUID PERCUSSION, AND PENETRATING BALLISTIC-LIKE BRAIN INJURY: RESULTS FROM OPERATION BRAIN TRAUMA THERAPY
Helen M. Bramlett, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Dr. C. Edward Dixon, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Deborah A. Shear, PhD, Walter Reed Army Institute of Research
Zhiqun Zhang, MD/PhD, Banyan Biomarkers, Inc.
Susie Zoltewicz, PhD, Banyan Biomarkers
Dr. Kara Schmid, PhD, Walter Reed Army Institute of Research
W. Dalton Dietrich, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Kevin K.W. Wang, PhD, University of Florida
Dr. Ronald Hayes, PhD, Banyan Biomarkers
Dr. Frank C. Tortella, PhD, Walter Reed Army Institute of Research
Dr. Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Preclinical models can help define TBI pathobiology and test new therapies. The multi-center, pre-clinical Operation Brain Trauma Therapy (OBTT) is evaluating drugs and biomarkers in three TBI models to prioritize for clinical translation. We evaluated the effect of nicotinamide across models on post-injury biomarker profiles and their neuropathological correlations.
Adult male Sprague-Dawley rats were subjected to controlled cortical impact (CCI), fluid percussion (FPI), or penetrating ballistic-like brain injury (PBBI) and treated with nicotinamide or vehicle at 15 min and 24 h post-injury. Sham rats in each model underwent all manipulations with the exception that they did not receivetrauma. In each model, animals were classified into 4 groups: (1): Sham, (2): TBI + Vehicle, (3): TBI + nicotinamide (50mg/kg), (4): TBI + nicotinamide (500mg/kg). At 21 days, rats were sacrificed and brain sections were assessed for lesion volume and either hemispheric or cortical tissue volume loss. The glial marker glial fibrillary acidic protein (GFAP) and the neuronal marker ubiquitin C-terminal hydrolase (UCH-L1) were measured by ELISA in serum at 4, 24 hours and 21 days after injury.
At 4h post-injury, serum concentrations of UCH-L1 and GFAP were significantly increased compared to shams across all 3 TBI models. Initial GFAP levels were higher in CCI and PBBI model than in FPI model, whereas initial UCH-L1 concentration was highest after FPI. No significant treatment effect was detected in initial biomarker concentrations. At 24h post-injury, nicotinamide dose-dependently reduced serum GFAP levels in CCI and PBBI (but not FPI) with the greatest effect detected with the high dosage (500mg/kg). These findings were correlated with histological measures of lesion size and brain tissue loss. GFAP concentrations at 24h post-injury in CCI and PBBI strongly correlated with lesion volume (R=0.84, P<0.0001 and R=0.53, P<0.0001, respectively) and hemispheric tissue loss (R=0.87, P<0.0001 and R=0.59, P<0.0001). In FPI UCH-L1 concentrations at 4h post-injury correlated with lesion volume and hemispheric tissue loss (R=0.63, P<0.0001 and R=0.50, P=0.002, respectively). However, nicotinamide had no impact on serum UCH-L1 levels.
Our results indicate that nicotinamide reduces GFAP levels following CCI and PBBI but not FPI, and that biomarker concentrations are predictive of lesion size and brain tissue loss. Differential effects of nicotinamide may be a reflection of the different animal models used and their associated type of injury. Our findings suggest the potential use of serum GFAP as a surrogate for volumetric analyses in pre-clinical therapeutic screening and the possibility that serum biomarker profiles may be useful in matching experimental models with patient subgroups in human TBI.
We thank the United States Army (W81XWH-10-1-0623) for support.
TBI, Biomarker, Neuroprotection, Rat, Model
FUNCTIONAL AND STRUCTURAL DYNAMICS OF MILD TRAUMATIC BRAIN INJURY
Jun Maruta, PhD, Brain Trauma Foundation
Miss Zarah Iqbal, BA, Brain Trauma Foundation
Miss Charvi Shetty, BS, University of California, San Francisco Department of Radiology and Biomedical Imaging
Julia P. Owen, PhD, University of California, San Francisco Department of Radiology and Biomedical Imaging
Jamshid Ghajar, MD, PhD, Brain Trauma Foundation
Pratik Mukherjee, MD, PhD, University of California, San Francisco Department of Radiology and Biomedical Imaging
The cognitive functions of people who have sustained a mild traumatic brain injury (mTBI) are often impaired. We studied these cognitive impairments at different points in time after injury and assessed the white matter integrity of tracts that may underlie them.
We measured cognitive functions in acute and chronic mTBI subjects, and analyzed diffusion tensor imaging (DTI) in the chronic mTBI group. The chronic mTBI group, consisting of 29 patients with persistent postconcussive syndrome (PPCS), was compared to a normal group of age-, education- and gender-matched controls. All chronic mTBI patients were studied at least three months from injury. All reported experiencing some symptoms, and all but one reported experiencing some cognitive symptoms. The acute mTBI group consisted of 11 patients, whose cognitive functions were studied longitudinally within one month of injury. In all groups, cognitive functions were assessed with a battery of tests, including the Attention Network Test (ANT), circular visual tracking and the Spatial Span subtest of the Wechsler Memory Scale-III. White matter integrity was assessed in terms of fractional anisotropy (FA) values in predefined regions of interest (ROIs) with Tract-Based Spatial Statistics.
Two measures from the ANT, conflict and mean reaction time, were significantly greater for chronic mTBI patients with PPCS. All other measures of cognitive ability in the chronic mTBI patients were indistinguishable from those in controls. This is in contrast to our previously reported PPCS group, which showed significant differences in cognitive testing, including circular visual tracking, compared to controls. The current measures from controls were similar to those in previous estimates.
The ROI-based analysis showed that white matter integrity in the right anterior limb of the internal capsule, as measured by FA values, was diminished relative to controls (p<0.01). This tract contains connections to regions of the prefrontal cortex, which have been shown to relate to many different cognitive abilities. FA values in many other tracts, including some that have been found to differ from controls in other studies, were indistinguishable from those of controls.
The acute mTBI patient group exhibited significant differences in circular visual tracking from the control group. Nearly all acute patients showed improvements within one month of injury, including some resolution of symptoms. The magnitudes of the improvements were related to the initial performance such that patients who initially performed poorly showed the largest improvements. The improvements were generally large enough to bring the performance measures for patients within the range of those for controls.
Our results suggest that mTBI patients recover much of their cognitive ability between the acute and chronic phases. Considering that chronic mTBI patients perform as well as controls despite lower FA values in a tract that is likely related to cognitive function, they may rely on other tracts within the same functional networks subserved by this tract. Acute mTBI patients consistently show abnormal visual tracking and impairment on other attention tests. Both of these improve within one month of injury. In those acute mTBI patients who become chronic mTBI patients, only limited deficits in attention are likely to remain.
This work was supported by Department of Defense grant W81XWH-08-1-0646 and by the James S. McDonnell Foundation grant #220020185.
concussion, neuropsychology, eye movements, neuroimaging
THE MACROPHAGE RESPONSE AFTER FOCAL TRAUMATIC BRAIN INJURY IN THE RAT
Lindsay Janes, BS, Center for Neuroscience and Regenerative Medicine
Eric Gold, BA, Center for Neuroscience and Regenerative Medicine
Matthew Budde, PhD, Medical College of Wisconsin
Tiziana Coppola, BS, NIH
Dana Dean, PhD, NIH
Jacob Lescher, BA, Center for Neuroscience and Regenerative Medicine, NIH
Joseph Frank, MD, MS, NIH
After spinal cord injury, inflammatory macrophages (M1) predominate over anti-inflammatory macrophages (M2) (Kigerl 2009). The temporal profile of M1/M2 phenotypes after traumatic brain injury is unknown. We subjected female rats to severe controlled cortical injury (CCI) and examined the post-injury M1/M2 time course in brain.
The motor cortex (2.5 mm left lateral, 1.0 mm anterior of Bregma) of anesthetized female Wistar rats (ages 8–10 weeks; n=49) underwent severe CCI with a 5 mm impactor tip driven by an electromagnetic piston (velocity 5 m/s, depth 2.5 mm, dwell time 100 msec). Rats were euthanized for a post-injury time course at 24 hours, at days 3 and 5, and at weeks 1, 2, 4, and 8 for brain perfusion/fixation for immunohistochemistry (IHC) and ex vivo MRI, flash freezing for RNA/protein, and cell harvesting for flow cytometry (n=3/time point per group). Perilesional brain tissue lysates were analyzed with Bio-Plex cytokine assays.
By IHC, the greatest macrophage response was seen between days 3 and 7, with a small peak of M2-associated staining at 3 days, followed by a shift to an M1 phenotype. By flow cytometry, the majority of macrophages and microglia at all time points were of M1 phenotype (mean±SD: 82±6%), with a smaller percentage of M2 phenotype (10±6%), which peaked at Day 5. Cytokine analysis showed an increase in TNF-alpha relative to control brains within the first week after injury, and increases in MIP-1a and GRO KC within the first 24 hours after injury.
The macrophage/microglial response to severe CCI is maximal between days 3 and 7. The predominant response to injury is of an M1 phenotype, although there is an early rise of an M2 phenotype.
These studies were supported by the Department of Defense (Center for Neuroscience and Regenerative Medicine) and the National Institutes of Health (Intramural Research Program).
TBI, CCI, macrophage, microglia, cytokine
BLAST-INDUCED DISRUPTION OF AN IN VITRO BLOOD-BRAIN BARRIER MODEL
Kiet V. Vo, Columbia University
Gwen B. Effgen, M.S., Columbia University
Edward Vogel III, Columbia University
Matthew B. Panzer, M.A.Sc., Duke University
Cameron R. Dale Bass, Ph.D., Duke University
David F. Meaney, Ph.D., University of Pennsylvania
Barclay Morrison III., Ph.D., Columbia University
Blast-induced traumatic brain injury (bTBI) is a serious wound sustained by warfighters and civilians; however, its consequences on the blood-brain barrier (BBB) are poorly understood. Given its fine structure – 300 nm thick with 20 m2 surface area – BBB damage could be a primary mechanism of bTBI.
bEnd.3 mouse brain microvascular endothelial cells were used to generate in vitro cultures representing the BBB. Cells were seeded on Transwell inserts and cultured for 8 days to form monolayers with tight junctions. Blast overpressures were generated using a compressed helium-driven shock tube with a 76 mm-diameter, 50 mm-length driver section, and 1240 mm-long driven section. BBB cultures were placed in a fluid-filled receiver designed to recapitulate the intracranial pressure history and mitigate wave reflections. Acute TEER changes were measured with an Endohm-12 electrode chamber and an EVOMX Epithelial Voltohmmeter (WPI). Hydraulic conductivity was measured using a custom device to quantify fluid flow across each monolayer. ZO-1 immunofluorescence (Invitrogen 61–7300) images were acquired to assess tight junction integrity. Sham controls were processed identically to injured cultures but not exposed to blast.
Exposure of bEnd.3 cultures to controlled blast acutely disrupted monolayer integrity. TEER acutely decreased in a dose-dependent manner as blast severity increased from: 376 to 571 kPa peak incident overpressure, 0.90 to 1.06 ms duration, and 96 to 186 kPa·ms impulse in-air. Following the 571 kPa peak overpressure blast, TEER significantly (p<0.05) decreased within 30 minutes of injury to 65±14 % (mean±SEM) compared to 105±5 % in controls. For the same injury, hydraulic conductivity significantly increased to 1.1×10−5±1.5×10−6cm/s/cmH2O compared to 3.0×10−6±8.7×10−7 cm/s/cmH2O in controls. ZO-1 immunofluorescence was disrupted after blast exposure, indicating compromised barrier integrity.
These data demonstrate that barrier integrity of an endothelial monolayer (BBB model) was disrupted by blast overpressure. A major function of the BBB is to restrict movement of water and ions between the brain and systemic compartments, making TEER and hydraulic conductivity functional indicators of monolayer integrity. The acute TEER dose response observed post-injury suggested that BBB disruption can occur in bTBI and that a tentative threshold for barrier opening exists between 469 kPa and 571 kPa peak overpressure. Significantly increased hydraulic conductivity of cultures injured at the highest overpressure level indicated compromised tight junctions, which was confirmed by disrupted morphology and reduced ZO-1 immunofluorescence. These results suggest that disrupted tight junction binding may influence BBB neuropathology associated with bTBI. Fundamental outcomes envisioned from this research are to determine an injury threshold for blast-induced BBB disruption, mechanisms of breakdown, and the time course for spontaneous recovery.
This work was supported by a Multidisciplinary University Research Initiative (W911MF-10-1-0526) from the Army Research Office.
blast injury, blood-brain barrier, bEnd.3
A LOCOMOTOR TASK TO STRATIFY SEVERITY OF EXPERIMENTAL TRAUMATIC BRAIN INJURY
H. Francis Farhadi, MD, PhD, The Ohio State University Medical Center
Lateral fluid percussion injury (FPI) is a standard experimental model of traumatic brain injury (TBI). Reflex righting time (RRT) is routinely used to assess the severity of injury immediately after FPI in mice. We analyzed locomotor function as measured by the activity box to further characterize the extent of injury.
One day prior to surgery, female C57BL/6 mice were placed in an activity monitor to acclimate them to the environment. Mice were then anesthetized and a right-sided craniectomy was performed midway between bregma and lambda. While under light isofluorane anesthesia, animals were subjected to a right lateral FPI or sham injury. RRT was recorded immediately after administration of the fluid pulse. One day after injury and weekly for 4 weeks, animals were placed in the activity box for 30 minutes. Horizontal and vertical beam breaks were recorded to give indices including distance moved, time spent moving, and time engaged in stereotypy behaviors. Animal activity was correlated with RRT at various time-points after injury.
One day after FPI, injured animals showed a decrease in many parameters of locomotor behavior. Both movement time and stereotypy time were inversely related to RRT. Conversely, rest time was positively correlated with RRT. As expected in the FPI model, motor deficits were not long lasting. Rather, at times remote from injury, locomotor indices normalized and a trend towards hyperactivity was seen in injured animals.
RRT has traditionally been used to characterize the extent of experimental TBI. We show that locomotor indices as measured by activity box can be used as an additional behavioral task to stratify the severity of injury in mice after lateral FPI.
We thank the Center for Brain and Spinal Cord Repair at The Ohio State University.
Activity box, mice, fluid percussion
ACUTE ETHANOL TREATMENT REDUCES BRAIN EDEMA AND AQUAPORIN EXPRESSION AFTER TRAUMATIC BRAIN INJURY
Yuchuan Ding, Ph.D., Wayne State University
The neuroprotective properties of ethanol were examined, including whether: 1) alcohol administration helps maintain blood brain barrier (BBB) integrity after TBI; 2 up-regulation of aquaporin 4 and 9 water channel proteins can be ameliorated by alcohol administration; and 3) acute alcohol treatment can improve functional outcomes.
A modified Marmarou model of TBI was used. Sprague-Dawley rats were randomly divided into 4 groups: 1) sham, 2) TBI without treatment, and 3) 0.5 g/kg ethanol or 4) 1.5 g/kg ethanol given 60 minutes post-TBI. Blood brain barrier (BBB) integrity after TBI was determined by amount of edema (brain water content). Aquaporin 4 and 9 mRNA and protein levels were assessed by real-time reverse transcription PCR technique and Western Blot analysis. Animals were assessed for cognitive outcome by the Radial Arm Maze test twice a day for 18 consecutive days. For motor outcome, foot-fault testing was used with animals tested for up to 21 days post-TBI.
Post-TBI ethanol administration contributed to decreased edema levels. Animals treated with 0.5 g/kg exhibited an average brain water content of 77.08%, while animals treated with 1.5 g/kg showed average brain water content of 76.95%. Both of these results represent significant (p<0.05) decreases when compared to TBI without treatment. No significant difference was found between the two ethanol treatment groups. ANOVA and post-hoc analyses indicated that post-TBI ethanol administration led to a significant down regulation of AQP 4 and AQP9 mRNA expression. Both ethanol treatment groups (0.5 g/kg, 1.5 g/kg) exhibited decreased AQP4 and AQP9 mRNA expression when compared to TBI without treatment. No significant difference was found between the 0.5 g/kg and 1.5 g/kg ethanol treatment groups. Findings in AQP protein expression largely mirrored results seen in mRNA expression except for that the 0.5 g/kg ethanol dose failed to show a significant (p<0.05) decrease in AQP4 protein expression. The 1.5 g/kg did show a significant decline in AQP4 expression. Additionally, both treatment doses significantly (p<0.05) decreased AQP9 protein expression. Cognitive studies also support the notion of ethanol induced neuroprotection. Ethanol treated animals not only demonstrated larger gains in maze completion times but also faster gains. In contrast to TBI-only animals which showed completion time improvement by day 7, both ethanol treated groups exhibited completion time improvements within the first few days of the trial. Motor testing also revealed that the ethanol treated animals committed fewer foot faults while exhibiting improved balance when compared to no treatment animals.
Post-TBI administration of ethanol reduced levels of brain edema as supported by lower water content percentages as well as decreased AQP4 and AQP9 expression. These changes were additionally associated with improved cognitive and motor functions. Taken together, the results support the possibility of ethanol-induced neuroprotection. The result of this study could lead to the future development of novel therapeutic strategies beyond the current treatment options for TBI victims.
I would like to acknowledge Dr. Yuchuan Ding for his support and guidance in this project.
Neuroprotection, blood-brain-barrier, edema, aquaporin, Marmarou
DIFFERENTIAL PROTEIN CHANGES IN PENETRATING AND NON-PENETRATING MODELS OF TBI
Min Tong, Walter Reed Army Institutes of Research
Rebecca Pedersen, Walter Reed Army Institutes fof Research
Michael Shaughness, Naval Medical Research Center
Eric Maudlin-Jeronimo, Naval Medical Research Center
Kara Schmid Maj., U.S. Army, Ph.D., Walter Reed Army Institutes of Research
Raymond Genovese, Ph.D., Walter Reed Army Institutes of Research
Stephen Ahlers, Ph.D., Naval Medical Research Center
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Jitendra Dave, Ph.D., Walter Reed Army Institute of Research
Protein biomarkers are being developed for diagnosis of traumatic brain injury (TBI), but their injury specificity is unknown. Rodent models of penetrating and non-penetrating TBI offer distinct models for translational biomarker research. Therefore, we systematically analyzed protein abundance and degradation of multiple protein biomarkers in two models of TBI.
Systems biology and immuno-based methods have been used to determine protein biomarker abundance in the penetrating ballistic-like brain injury (PBBI). For this study we employed two distinct cohorts of TBI subjects: (1) PBBI injured rats and (2) repetitive blast overpressure (BOP) injured rats. For PBBI, ipsilateral and contralateral coronal tissue sections 6 mm from bregma were collected 24 h after injury. For BOP studies, the prefrontal cortex, cortex, hippocampus and cerebellum were dissected 24 h after the last of three daily, 75 kPa blasts. Proteins extracted from brain tissues were analyzed by semi-quantitative western blotting and/or ELISA to determine specific abundance of intact proteins or their degradation fragments. Target proteins were chosen based on informatics data and reported literature.
Our analyses identified statistically significant increases in GFAP and UCH-L1, but decreases in PSD-95, in ipsilateral tissues 24 h after PBBI. In contrast, GFAP, UCH-L1, and PSD-95 showed no change in BOP. The relative abundance of α-spectrin and different spectrin breakdown products (SBDPs) were differentially expressed in PBBI compared to BOP. In ipsilateral PBBI tissue, there was a loss of full length α-spectrin, with a moderate increase in SBDP-145/150. BOP exposures lead to a small, albeit non-significant, decrease in alpha-spectrin within the cerebral cortex, hippocampus, and cerebellum, but a moderate increase in SBDP-145/150 in the prefrontal cortex and cerebellum. There was a greater increase of SBDP-120 in the prefrontal cortex, which was not present in the cerebellum. Overall, significantly increased SBDP-145/150 and decreased α-spectrin were detected after PBBI. In contrast, decreased α-spectrin after BOP exposure was not-significant, and there was a larger increase in BOP-induced SBDP-120 compared to SBDP-145/150.
TBI biomarkers such as GFAP, spectrin and its BDPs, as well as UCH-L1, were increased in PBBI; loss of PSD-95 was also observed. The expression of SBPD-145/150, a product of calpain activity, was greater in PBBI vs. BOP; whereas SBDP-120, a product of caspase activity, was increased after BOP exposure. Spectrin may be selectively cleaved by different mechanisms of cellular injury. Spatial differences in SBDP abundance may identify the prefrontal cortex as the primary injury site in blast, a region that has been implicated in TBI from BOP due to the unique pattern of symptoms. In addition, the results demonstrated the upregulation of GFAP, UCHL1 and SBDP in the PBBI model similar to the changes seen in human serum. Not only does the study confirm the use of informatics analysis but it also possibly identified different mechanisms and patterns of severe injury that differentiate penetrating from non-penetrating.
This study was supported by the Army Combat Casualty Care Research Program and the Congressionally Directed Medical Research Programs.
traumatic brain injury, biomarker, spectrin
PROGESTERONE TREATMENT IMPROVES LONG-TERM BEHAVIORAL OUTCOME AFTER TRAUMATIC BRAIN INJURY IN MIDDLE-AGED RATS
Fang Hua, M.D, Ph.D., Emory University
Jun Wang, B.S., Emory University
Seema Yousuf, Ph.D., Emory University
Fahim Atif, Ph.D., Emory University
Iqbal Sayeed, Ph.D., Emory University
Donald Gerald Stein, Ph.D., Emory University
Progesterone (PROG) has been shown to be effective in protecting the brain from the consequences of traumatic injury and is currently in Phase III clinical trials. This project sought to investigate whether PROG would also improve behavioral outcomes in middle-aged rats subjected to bilateral frontal cortex contusion injury.
Twenty-nine male Sprague-Dawley rats subjected to CCI or sham operation received PROG (16 mg/kg) or vehicle by intraperitoneal injection 1 h post-injury, subcutaneously at 6 h post-injury, and then subcutaneously every 24 h for 168 h. The doses for the final two treatments were tapered. Neurobehavioral tests were evaluated from onset of injury to 21 days after surgery.
The Morris water maze probe test showed that the PROG-treated rats spent significantly more time in the platform quadrant compared to the CCI-vehicle group (p<0.05). The sticky task showed that, compared to the Sham-vehicle group, the latency to contact and remove the sticker increased significantly over post-injury days 3 to 15 in the CCI groups (p<0.05). Contact latency in PROG-treated rats was significantly shorter than that in the CCI-vehicle group at 15 days post-injury (p<0.01). The locomotor activity test showed no significant difference in average travel distance and rest time between the CCI-vehicle and PROG-treated groups at 3, 8, 15 and 20 days post-injury. The evaluation of the necrotic cavity showed no significant reduction by the PROG treatment.
Our data indicate that PROG is also effective in middle-aged male rats in preserving spatial memory and ameliorating sensory neglect from bilateral frontal CCI.
This work was supported by NINDS U01 NSO 62676 and 5R01HD61971 to DGS and partially by a gift from BHR Pharma.
TBI; progesterone; behavior
APOLIPOPROTEIN E-MEDIATED CLEARANCE OF THE ALZHEIMER'S DISEASE PEPTIDE AMYLOID BETA IS IMPAIRED AFTER EXPERIMENTAL TRAUMATIC BRAIN INJURY IN APOE-ɛ4 GENOTYPE
Maia Parsadanian, Georgetown University
Sonya Dumanis, Georgetown University
David Zapple, Georgetown University
G. William Rebeck, PhD, Georgetown University
Mark P. Burns, PhD, Georgetown University Medical Center
Human APOE-ɛ4 carriers have increased amyloid-beta (Aβ) plaque deposition acutely after TBI, and a tenfold increased risk of developing Alzheimer's disease (AD) compared to non-ɛ4 carriers. As Aβ accumulation is central to AD pathogenesis, we hypothesized that clearance of Aβ is impaired in ɛ4 carriers after TBI.
We used a Leica Impact One controlled cortical impact (CCI) device to induce brain injury in C57/Bl6 mice, 3xTg AD mouse model, and APOE targeted replacement mice. In order to study the relationship between Aβ and soluble apoE after TBI we used a sequential multi-step extraction protocol to extract soluble (extracellular and cytosolic proteins), detergent-soluble (membrane bound proteins) and formic acid-soluble (insoluble proteins) from CCI-injured brains at 1, 3 and 7 days post-injury. We measured soluble and membrane bound apoE levels by Western blot, and Aβ40 and Aβ42 levels by ELISA. We also determined APP processing (APP, a-CTF, β-CTF and BACE1) and levels of the apoE lipidating protein ABCA1 by Western blot.
We have previously shown that CCI model causes an acute increase in Aβ40 and Aβ42 levels in the injured cortex of C57/Bl6 mice. Here, we repeat those findings and also find that CCI causes an acute increase in Aβ levels (peaking at 24h) in 3xTg AD mice. The increase in Aβ is primarily caused by increased processing of the amyloid precursor protein. However, we also find that TBI-induced Aβ is rapidly cleared (by 3d) to sham levels in both C57/Bl6 mice and in the 3xTg AD mouse. As clearance of Aβ is mediated by soluble apoE, we measured apoE levels in our mice and found that soluble, but not membrane-bound, apoE was decreased by over 70% at 24h post-injury. Soluble apoE levels recovered by 3d and were significantly elevated at 7d post-injury. Levels of soluble apoE and Aβ in the injured cortex are inversely correlated after TBI (R2=0.8894), with the trough and recovery of apoE corresponding to the peak and clearance of TBI-induced Aβ.
To investigate why human APOE-ɛ4 carriers have genotype dependent increases in Aβ deposition after TBI, we repeated our experiments in APOE targeted replacement mice. These mice have the human APOE-ɛ3 or APOE-ɛ4 gene knocked-in to replace the murine apoegene. TBI caused a significant increase in Aβ levels at 24h post-injury, which was similar in both APOE-ɛ3 and APOE-ɛ4 targeted-replacement mice and there were no differences in APP processing between injured APOE-ɛ3 and APOE-ɛ4 mice. However, while TBI-induced Aβ levels returned to normal levels by 3d in APOE-ɛ3 mice, there was a prolonged elevation of the TBI-induced Aβ peak in APOE-ɛ4 mice, which remained significantly elevated for the 7d duration of the experiment.
Our data demonstrate that soluble apoE is reduced after TBI, similar to reports of reduced CSF apoE after human TBI. As apoE mediates the clearance of Aβ, reduced levels of apoE could contribute to Aβ accumulation after injury. We find a very tight inverse-correlation between soluble apoE and Aβ levels after TBI, such that TBI-induced Aβ levels return to baseline levels as soluble apoE levels recover.
30% of acute human TBI fatalities present with Aβ plaques. However, the presence of plaque is associated with the APOE genotype, with plaque found in 10% of non-APOE-ɛ4 brains, 35% of heterozygous APOE-ɛ4 brains, and 100% of homozygous APOE-ɛ4 brains after TBI. In our study we demonstrate the rapid clearance of TBI-induced Aβ in APOE-ɛ3, but not APOE-ɛ4 mice. This work demonstrates that the apoE4 protein is dysfunctional at Aβ clearance after injury, and explains why Aβ aggregates in APOE-ɛ4 brain after TBI.
This project was supported by: grants R03NS057635 and R03NS067417 from NINDS (MPB), and Neural Injury and Plasticity Training Program (5T32NS041218-10) and Georgetown MCGSO (PMW).
TBI, Aβ, apolipoprotein-E, Alzheimer's, transgenic
BENEFICIAL EFFECT OF AMYLOID BETA AFTER CONTROLLED CORTICAL IMPACT
Jacqueline Berglass, BA, Department of Medicine, Children's Hospital Boston
Jimmy Zhang, BA, Department of Pediatrics, Massachusetts General Hospital
Jianhua Qui, MD, PhD, Harvard Medical School
Michael J. Whalen, MD, Harvard Medical School
We have previously demonstrated worse outcome after CCI in young adult Bace1−/− mice, suggesting a protective role of BACE after CCI. To determine whether the effect of BACE is specific to Aβ, we tested whether reconstitution of Bace1−/− mice with Aβ40 would improve outcome after CCI.
Young adult (2–3 mos old) Bace1 knockout (Bace1−/−)(n=33) and wild type Bace1+/+ (C57Bl/6)(n=34) mice were subjected to CCI. Immediately after CCI, mice underwent intracerebroventricular (ICV) injections of Aβ40 (200 pmol in 5 ul PBS, Anaspec, Freemont Ca) (n=18–23/group) or vehicle (PBS, n=10–16/group). Functional outcomes were assessed with wire grip (motor) and Morris water maze (spatial memory). Soluble Aβ levels were assessed at 24 hrs after CCI. Lesion volume was measured 21 d after CCI. Data were analyzed by t-test, clustered ordinal regression (motor data) or repeated measures analysis of variance (RM ANOVA) as appropriate.
At 24 hrs after injury, Aβ –treated Bace1−/− mice had Aβ40 levels similar to those of vehicle-treated WT mice (3.8 pmol/g vs. 4.7 pmol/g, p=0.3) but significantly higher Aβ levels than vehicle-treated Bace1−/− mice (3.8 pmol/g vs. 0.1 pmol/g, p=0.002). All groups demonstrated time dependent improvement in motor function (p<0.001). Vehicle-treated Bace1−/− mice performed worse than vehicle Bace1+/+ mice (p<0.001). Wire grip scores were significantly better in Aβ-treated Bace1−/− compared to vehicle-treated Bace1−/− mice (p=0.012) and were similar to scores in Aβ-and vehicle-treated Bace1+/+ mice. Hidden platform trial performance did not differ between Aβ-treated and vehicle-treated Bace1−/− (p=0.73), Aβ-treated and vehicle-treated Bace1+/+ animals (p=0.74) or Aβ-treated Bace1−/− vs. Aβ-treated Bace1+/+ (p=0.66). Bace1−/− mice had increased lesion volume compared to Bace1+/+ (p<0.001) but there was no effect of Aβ-treatment on lesion volume in either Bace1−/− (p=0.4 for Aβ-treated vs. vehicle treated) or Bace1+/+ (p=0.3 for Aβ-treated vs. vehicle treated).
Reconstitution with Aβ40 improved motor but not histopathological outcome in injured Bace1−/− mice. In contrast to current concepts of BACE as a therapeutic target, the data suggest beneficial effects of Aβ40 after CCI in young adult mice.
This work was supported by grants from Charles Hood Foundation (RM) and NINDS 5RO1NS047447 (MW).
Traumatic Brain Injury, Beta Amyloid
ALTERATION IN BDNF AND ITS RECEPTORS, FULL-LENGTH AND TRUNCATED TRKB AND P75NTR FOLLOWING PENETRATING TRAUMATIC BRAIN INJURY
Elham Rostami, MD, Karolinska Institutet
Frank Krueger, PhD, NINDS, George Mason
Stefan Plantman, PhD, Karolinska Institutet
Johan Davidsson, PhD, Chalmers University of Technology
Denes Agoston, MD, PhD, USUHS
Jordan Grafman, PhD, NINDS, Kessler foundation
BDNF exerts multiple functions during CNS development and after injury. We have recently shown that the polymorphism of human BDNF gene predicts the cognitive recovery and outcome following penetrating TBI. In this study we have examined the expression of BDNF and its receptors in a rat model for penetrating TBI.
We used new model for penetrating TBI model in rat. The injury is produced by a controlled penetration of 2 mm thick needle shaped object, which is accelerated with a bullet from air gun or a pendulum. We used in situ hybridization and investigated the mRNA expression of BDNF and its receptors, the full-length and the truncated TrkB and p75NTR, from 1 day to 8 weeks following penetrating TBI. In addition, the protein level of BDNF in cortex and hippocampus was measured following injury by reverse phase protein microarray (RPPM).
The mRNA expression of BDNF and its receptors decreased in the penumbra zone ipsilateral to the injury while there was an increase in mRNA expression at the contralateral side of injury. The increase of BDNF mRNA expression in the hippocampus sustained for 2 weeks following injury. The most affected area was CA3. Furthermore, the protein analysis by RPPM showed increased levels of BDNF in the cortex and the hippocampus up to 2 weeks after TBI. At 8 weeks following injury the there was an intense labeling of the truncated TrkB receptor in the area surrounding the cavity. Also the mRNA expression of p75NTR was continuously increased around the cavity 8 weeks after injury.
Our study is the first report on the expression of BDNF and its receptors following penetrating TBI and suggests that their expression is altered long after the acute phase of injury. Further studies are needed to investigate if these late expressions of these receptors are beneficial or deleterious. In either case it raises the possibility to influence the recovery of brain injury during the chronic phase and the development of treatments that may improve TBI patients outcome.
We thank Mrs. Maria Angeria for technical assistance.
Penetrating TBI, BDNF, TrkB receptor
NEUREGULIN-1 EFFECTS ON ENDOTHELIAL AND BLOOD-BRAIN-BARRIER PERMEABILITY AFTER EXPERIMENTAL INJURY
Song Zhao, M.D., Jilin Hospital, China
Wendy Leung, B.S., Massachusetts General Hospital
Ji Hae Seo, Ph.D., Massachusetts General Hospital
Deepti Navaratna, Ph.D., Massachusetts General Hospital
Michael J. Whalen, Massachusetts General Hospital
Xiaoying Wang, M.D., Ph.D., Massachusetts General Hospital
Eng Lo, Ph.D., Massachusetts General Hospital
Blood-brain-barrier disruption occurs with a high incidence after traumatic brain injury, and contributes to brain edema, inflammation, and neuronal death. Therefore, blood-brain-barrier integrity is an important potential therapeutic target in the treatment of the acute phase of brain trauma.
Endothelial Cell Permeability Assay: Rat brain endothelial (RBE.4) cells were grown to be a confluent monolayer on transwell inserts and treated with IL-1β+/− NRG-1 for 24h. Permeability was measured by quantifying the amount of (FITC)-labeled dextran that passed through the monolayer. Visualization of tight junctions: CSC cells were grown to be a confluent monolayer and were treated with TNFα+/− NRG-1 for 18h. Tight junction protein integrity was evaluated using immunocytochemistry. Controlled Cortical Impact: Male C57Bl/6 mice (10–12 weeks of age) were subjected to CCI. PBS or NRG-1 was administered by tail-vein injection 10 min after trauma. Determination of Evans Blue dye extravasation: Evans Blue dye was given by tail vein injection 30 min after CCI. 2h after CCI, dye was extracted from brain homogenates and absorbance was measured. Gelatin Zymography: 50ug of brain homogenate from each hemisphere was analyzed for MMP-9 and MMP-2 activity.
For in-vitro studies, rat brain endothelial cells (RBE.4) were grown on transwell membranes and treated with IL-1β. Endothelial permeability was assessed by measuring the amount of dextran-40 extravasation through the transwell membrane. IL-1β (10ng/ml) increased dextan-40 extravasation, while co-incubation with NRG-1 (12.5nM and 37.5nM) decreased it. Immunocytochemistry studies were performed using human cerebral endothelial cells (CSC) which were incubated with TNFα. TNFα treatment resulted in decreased immunoreactivity of the adherins junction protein VE-cadherin and the tight junction protein Zona Occludens-1. NRG-1 co-incubation ameliorated this decrease. For in-vivo studies, C57Bl mice were subjected to controlled cortical impact (CCI) under anesthesia, and BBB permeability was assessed by measuring the amount of Evans blue dye extravasation from the vasculature at 2h. NRG-1 administration by tail vein injection 10 min after CCI resulted in a decrease in Evans blue extravasation to 65% of the PBS group. Since Evans blue extravasation may result from bleeding due to trauma, hemoglobin ELISA was also performed at the same time point. There was a trend towards lower levels of hemoglobin extravasation in the NRG-1 group, but the results did not reach statistical significance. MMP-9 activity was measured using gel zymography since MMP-9 may disrupt membrane integrity. MMP-9 activity was not different between groups at 2h.
These data suggest that NRG-1 decreases the permeability of endothelial cells exposed to an inflammatory cytokine such as IL-1β, and reduces BBB permeability following experimental brain trauma. Through its ability to protect the cerebral endothelium and the BBB, NRG-1 may have neuroprotective potential after brain trauma.
This study was supported by NIH grants (K08NS057339 to J.L., R01NS53560 and P01NS555104 to E.H.L.).
neuregulin-1, endothelial permeability, blood-brain barrier
DETECTION OF BRAIN INJURY IN A ROTATION ACCELERATION MODEL OF MILD TRAUMATIC BRAIN INJURY USING DIFFUSION TENSOR IMAGING
Alok Shah, MS, Medical College of Wisconsin
Frank Pintar, PhD, Medical College of Wisconsin
Brian Stemper, PhD, Medical College of Wisconsin
The majority of all treated brain injuries are mild. Diffusion tensor imaging (DTI) has a high sensitivity to subtle injury and shows promise as a diagnostic tool in mild traumatic brain injury (mTBI). DTI was used to investigate a closed head animal model of mTBI using rotational acceleration.
Rats were anesthetized and placed in a head holder to deliver a rotational acceleration of 200 krad/s2 delivered over 2.7 ms (n=6), or sham procedures (n=5). Animals were euthanized seven days after injury. The fixed brains were imaged on a 9.4T MRI using a diffusion-weighted spin-echo sequence (TR=2000 ms, TE=21 ms). Twelve diffusion-weighted images (b=1200 s/m2) and two non-diffusion weighted images(b=0 s/m2) were acquired at a spatial resolution of 0.2×0.2×0.5 mm3. DTI parameter maps, including fractional anisotropy (FA) and mean diffusivity (MD) were computed and spatially registered to a common atlas. A voxelwise analysis was used to identify regional between-group effects. Additionally, an anatomical region of interest (ROI) analysis was used for within-group tests of left-right asymmetry. An unpaired Student's t-test was used to compare group differences, and a paired t-test was used for within-group comparisons.
All rats survived the rotational acceleration without skull fracture or cervical spine injury. The left lateral rotation of the head resulted in DTI abnormalities predominantly in the right hemisphere. In the ROI analysis, the right cortex of the injured animals was significantly decreased in FA (p=0.014) compared to control animals, whereas the left cortex was unaffected (p=0.58). Consequently, in the cortex there was a significant left-right asymmetry in the injured group (p<0.001) but not in the control group (p=0.16). The FA asymmetry in the hippocampus and white matter had a trend toward significance in the injured animals (p=0.078 and p=0.065, respectively), but not in the control group (p=0.51 and p=0.12). In the voxelwise analysis, there were widespread differences between the injured and control groups in the right cortex, white matter, and hippocampus, with fewer small clusters in other regions of the brain. No significant group differences in MD were observed.
In a rotational acceleration rat model of mTBI, DTI identified unilateral abnormalities in the cortex, white matter, and hippocampus. Importantly, these changes were seen in the absence of gross injury or edema using conventional MRI techniques, highlighting the utility of DTI in detecting mild injury although the current study was limited to ex vivo specimens. The decrease in FA is consistent with the pathological features of axonal injury or gliosis that are known to occur following mTBI. Efforts are currently underway to correlate the DTI observations with graded histopathological findings and post-injury behavioral deficits associated with different injury levels. Overall, the results demonstrate the potential role of DTI as a diagnostic tool and as a potential outcome measure for therapeutic interventions.
Funded through the Advancing a Healthier Wisconsin endowment at the Medical College of Wisconsin (M.D.B.), and the Veterans Health Administration Rehabilitation Research and Development Service (B.D.S.).
diffusion tensor imaging, rotation acceleration, traumatic brain injury
PENETRATING BALLISTIC-LIKE BRAIN INJURY (PBBI) INDUCES COAGULOPATHY WITH REDUCTION OF PLATELET-FIBRIN INTERACTION: A STUDY OF THROMBOELASTOGRAPHY IN A RAT MODEL
Clayton Jackson, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shyam Gajavelli, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Carlos J. Bidot, M.D., Miami Project of Cure Paralysis/University of Miami
Wenche Jy, Ph.D., Department of Medicine
Markus Spurlock, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Frank Tortella, Ph.D., Brain Trauma Neuroprotection & Neurorestoration Branch, Center of Excellence for Psychiatry & Neuroscience, WRAIR
Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Severe coagulopathy often accompanies penetrating brain injury and its incidence isreported to be significantly higher than with closed-head trauma. However the mechanisms underlying coagulopathy in brain injury are poorly understood. The aim of this study is to assess the possible mechanism of using thromboelastography (TEG) in rat PBBI model.
Sprague-Dawley rats were allocated to PBBI or sham control groups (n=7/grp). Experimental procedures were as follows: PBBI group; a burrhole was made in the frontal skull for PBBI probe insertion and inflation/deflation. Sham control group; received craniotomy only. Arterial blood samples were collected (femoral artery) with a TEG analyzer. The following variables were measured: reaction time (R), coagulation time (K), clot formation rate (α-angle), maximum amplitude (MA), shear elastic modulus (G), coagulation Index (CI), and LY30 (%lysis during 30min after MA) and compared between the PBBI and sham groups. Group differences in each parameter were evaluated and p Values <0.05 were considered significant.
CI, which represents the overall coagulation status, was lower in the PBBI vs sham group (Mean±SEM; 3.10±0.29 vs 4.34±0.58). MA value, which reflects the platelet-fibrin interaction, was significantly smaller in the PBBI vs sham (65.6±1.6 vs 73.2±3.4, p=0.0319). Moreover, the G value, which also represents the strength of the fibrin clot, was smaller in the PBBI group vs controls (17590±4798 vs 9644±615, p=0.0411).
These results suggest platelet-fibrin interaction could be reduced in PBBI. The real time “Point of Care” assessment with TEG could help to select more targeted, goal-directed therapy in PBI patients.
This study is supported by DOD Grant PTO74521-W81XWH-08-1-0419.
Blood, Thromboelastography, TEG, PBBI, TBI
GROUP DIFFERENCES IN GLOBAL SIGNAL CORRELATION BETWEEN MILITARY TBI PATIENTS AND CONTROLS
Seonjoo Lee, Ph.D., Center for Neuroscience and Regenerative Medicine, NIH/USUHS; Henry Jackson Foundation, Rockville, MD
Binquan Wang, Ph.D., Traumatic Brain Injury Image Analysis Lab, USUHS; Henry Jackson Foundation; National Capital NeuroImaging Consortium
Ping-Hong Yeh, Ph.D., Traumatic Brain Injury Image Analysis Lab, USUHS; Henry Jackson Foundation; National Capital NeuroImaging Consortium
Hai Pan, M.S., Traumatic Brain Injury Image Analysis Lab, USUHS; Henry Jackson Foundation
Wei Liu, Ph.D., National Intrepid Center of Excellence, WRNMMC; Traumatic Brain Injury Image Analysis Lab, USUHS; Henry Jackson Foundation
Dominic Nathan, Ph.D., National Intrepid Center of Excellence, WRNMMC; Traumatic Brain Injury Image Analysis Lab, USUHS; Henry Jackson Foundation
Terrence R. Oakes, Ph.D., National Intrepid Center of Excellence, WRNMMC; Traumatic Brain Injury Image Analysis Lab, USUHS; National Capital NeuroImaging Consortium
Gerard Riedy, M.D., Ph.D., National Intrepid Center of Excellence, WRNMMC; Traumatic Brain Injury Image Analysis Lab, USUHS
Resting state fMRI connectivity analysis is often used to investigate Traumatic Brain Injury (TBI). A common preprocessing step is removal of the global signal's contribution to the data. Here, we show differences between global signal correlation maps from a cohort of military TBI patients and a group of healthy controls.
Twenty-two participants with a classification of TBI were recruited from U.S. military personnel at Walter Reed National Military Medical Center (WRNMMC). Seven control subjects (active duty but not previously deployed) were also recruited from WRNMMC. Resting state fMRI data and a high-resolution T1-weighted image were collected for each subject. After removal of nuisance effects associated with CSF and white matter from the fMRI data, the average whole-brain global time-course was calculated for each subject and correlated against his/her preprocessed resting state data on a voxel-wise basis. These correlation maps were then compared between groups using a voxel-wise t-test. Locations of regions showing significantly different global time-course correlation between the two groups were compared to the default mode network (DMN) component found via ICA of resting state data from a separate control group downloaded from the Function BIRN Data Repository (BrainScape_BS002, primary reference: Fox et al., 2007).
Global Signal Correlation: Differences between the correlation maps of the military TBI patient group and the control group generally appeared to be localized to regions in the grey matter. Two significantly different regions (p<0.05 after family-wise error correction) overlapped areas of the posterior cingulate cortex and the left precuneus, and the left superior parietal lobe, respectively. A third region that trended toward significance (corrected p<0.07) overlapped the left mid frontal gyrus. In all three regions the control group showed greater global signal correlation than the TBI patient group. The two regions showing a significant difference in global signal correlation between the TBI patient group and the control group were located in areas typically associated with the default mode network (DMN). Removal of the global signal from resting state fMRI data is theoretically carried out to remove effects of non-neural, whole-brain noise. An average is taken over the entire brain volume in an attempt to capture this noise while excluding or “averaging out” connectivity-related signal from various brain networks. However, the localization of the group differences seen here suggests that the amount of DMN-related signal captured in the global time-course differs between the TBI patient group and the control group.
ROI Analysis: Given the small number of subjects in the control group relative to the TBI patient group, an ROI analysis was carried out to determine the average global signal correlation in the three identified ROIs (two significant and one trending toward significance) for each subject. As expected from the voxel-wise analysis, the average control group ROI correlation values were significantly higher than the average TBI patient group values. There were no significant correlations between z-scores and the maximum single-voxel displacement during motion correction in either the control group or the patient group.
This analysis shows differences in global time-course correlation between a TBI group and a control group. The localization of these differences overlap with regions of the DMN, which is typically investigated in resting state connectivity analysis. This suggests that BOLD signal changes associated with DMN fluctuations in the fMRI data may contribute more to the global time-courses of the control group than to those of the TBI group. In this case, removing the global signal from the fMRI data would introduce bias when comparing connectivity results between these groups. However, the specific underlying causes of these correlation differences still need to be elucidated. These results should also be verified in an analysis involving more control subjects.
Disclaimer: Views expressed in this abstract are those of the author and do not necessarily reflect the official policy or position of the Department of the Navy, Army, Department of Defense, nor the U.S. Government.
Work supported by: CDMRP to USU Grant PT074437; CNRM Grant 300606-7.01-60855. Separate control data: fBIRN Data Repository supported by grants (U24-RR021992) Testbed funded: NCRR, NIH, U.S.A.
TBI, fMRI, resting-state, global-signal
SIMILARITIES AND DISSIMILARITIES OF TWO EXPERIMENTAL MODELS OF TBI (CLOSED HEAD AND BLAST TBI) IN LONG-TERM DISABILITIES: SPASTICITY, COGNITIVE & SYNAPTIC FUNCTION, BALANCE, ANXIETY, AND PAIN RELATED BEHAVIORS
Prodip Bose, MD, PhD, North Florida/South Georgia Veterans Health System and Departments of Neurology and Physiological Sciences at the University of Florida
Jiamei Hou, MD, PhD, Dept. of Physiological Sciences, University of Florida
Rachel Nelson, BS, North Florida/South Georgia Veterans Health System
Nicole Nissim, BS, North Florida/South Georgia Veterans Health System
Shigeharu Tsuda, MS, Dept. of Physiologycal Sciences, University of Florida
Ron Parmer, MS, North Florida/South Georgia Veterans Health System
Susie Sennhauser, BS, North Florida/South Georgia Veterans Health System
Ashok Kumar, PhD, Department of Neuroscience, University of Florida
Victor Prima, MS, Banyan Laboratories, Inc.
Stanislav Svetlov, PhD, MD, Banyan Biomarkers, Inc.
Ronald Hayes, PhD, Banyan Biomarkers
Kevin Wang, PhD, Departments Psychiatry and Neuroscience at the University of Florida
Floyd Thompson, PhD, North Florida/South Georgia Veterans Health System and Departments of Physological Sciences and Neuroscience at the University of Florida
Traumatic brain injury (TBI) associated with blast (bTBI) or with conventional closed TBI (cTBI) are signature injury among soldiers. It is not known if there is significant overlap or significant differences in the systematic patterns, primary and secondary locations, and types of neuropathology induced by bTBI compared with cTBI.
In the present study, we used two well established rat models of TBI (450g×1.25m, Marmarou cTBI Model, and 33 psi, bTBI, Banyan model) to assess alterations in stretch reflex using velocity dependent ankle torque & ankle extensor muscle EMGs (spasticity, motor disability), 3D kinematic & footprints (for gait, motor disability), rotorod (for balance performance), Morris water maze (MWM) spatial learning (for cognition), elevated plus maze (for anxiety), thermal test (for pain) tested before and after each types of injury in separate groups of animals. Conventional histology, immunohistochemistry, molecular biology, and slice electrophysiology based techniques were also applied to investigate the patterns of injury that are produced by these two methods.
Interestingly, cTBI resulted more severe spasticity than bTBI animals. However, bTBI animal showed more impaired serial learning and delayed immediate acquisition of learned behavior in MWM, although baseline synaptic transmission and long-term potentiation (LTP) recorded from CA3-CA1 hippocampal slices in vitro showed similar deficit when compared to uninjured control animals. A similar moderate decrease in mature BDNF (14kDa) level was detected in hippocampus and prefrontal cortical tissue of both groups. Interestingly, a drastic decrease in pCREB was detected in hippocampus in the bTBI animal, however no such change was detected in corresponding hippocampus tissue of the cTBI animal. Moreover, bTBI animal showed more exaggerated pain-like behavior (paw licking, lick latency and face washing) than cTBI. It is note worthy that the cTBI animal showed a drastically decreased pain marker, GCH1, in the cortex and in the hippocampus (Western blot). However, in bTBI animals, a markedly lower level of GCH1 showed only in the hippocampus. Gait and balance disabilities were similar in both groups. No changes were detected in GABAB-R1A and GAD67 in the hippocampus tissue in both groups.
cTBI and bTBI produce long-term multiple disabilities. The most frequent reported disabilities include cognitive, post traumatic stress disorder (PTSD), balance, sleep disturbance and motor disabilities that not only obstruct re-deployment but substantially diminish successful re-entry into community life. A major limitation of the expansion of current therapy is the paucity of information on the pathophysiology of TBI in general and blast-related injury in particular, which may differ in significant ways from the mechanisms associated with closed impact head injury typical of civilian TBI.
Both conventional and blast TBI produced an array of overlapping neurological changes. However, several behavioral, physiological and molecular changes after bTBI showed distinct patterns compared to cTBI. This outcome reflects likely a different type of blast energy transmission to the brain, which requires further investigation.
Supported by Veterans Affairs RR&D Merit Review B5037R and B6570R.
TBI, spasticity, motor disability, cognitive
CHRONIC NEUROBEHAVIORAL AND NEUROPATHOLOGICAL CHANGES AFTER REPETITIVE VERSUS SINGLE MILD TRAUMATIC BRAIN INJURY
Helena Chaytow, BS, Cardiff University
Janice Stewart, BS, Greater Glasgow and Clyde NHS
Gogce Crynen, PhD, Roskamp Institute
Corbin Bachmeier, PhD, Roskamp Institute
Michael Mullan, MD, PhD, Roskamp Institute
William Stewart, MBChB, PhD, Greater Glasgow and Clyde NHS/ University of Glasgow
Fiona Crawford, PhD, Roskamp Institute
There are currently no published laboratory models which specifically address the long term consequences of single and repetitive mTBI (s-mTBI; r-mTBI). In this study we used a combination of neurobehavioral and neuropathological tests to address these issues in C57BL/6J mice.
Mice of 10 weeks of age were anesthetized with 3% isofluorane and received 1 mTBI or 5 mTBI 48H apart or relevant anesthesia control delivered by an electromagnetic controlled impact device (Leica) driving a 5.0mm flat tip at a 5m/s with a 1.0mm strike depth. Behavioral and pathological analyses were carried out at an acute time point, (24h post injury), and at chronic time points of 6 and 12 months post injury. Behavioral analyses consisted of tests for motor deficit (rotarod), spatial memory tasks, (Barnes maze), and anxiety-like behavior (elevated plus maze once at 12 months post mTBI). Animals subject to pathological analyses were perfused and fixed at 24 hours, 6, and 12 months after injury. The brains were analyzed for: general histological assessment (luxol fast blue & cresyl violet (LFB/CV)); traumatic axonal injury (APP), and neuroinflammation (GFAP, IBA1).
All injured animals exhibited significant acute deficits in motor function compared to their anesthesia controls (s-mTBI, p=0.005; r-mTBI, p<0.001). There were no significant motor effects evident at chronic time points. After 6 days of acquisition in the Barnes maze the probe trials examined the latency of each animal to travel to the target zone (which included the target hole and its two adjacent holes). For both injured groups, the time to reach the target zone was significantly longer than for their anesthesia controls (p<0.02 for both injury groups). The r-mTBI mice performed the worst, requiring on average 15 sec to reach the target zone, followed by the s-mTBI mice (8.3sec), as compared to the repetitive sham (4.8sec), and the single sham (2.9sec). The average velocities were found to be not significantly different across all groups (p>0.05). At 6 and 12 months post injury, performances of the single and repetitive anesthesia groups and the s-mTBI group were similar, with mice taking approximately 5 sec to reach the target zone. The r-mTBI group remained significantly impaired at 6 and 12 months, requiring on average 11sec to reach the target zone (p<0.05). At the acute time point LFB/CV revealed no overt cell loss in the sub regions of the hippocampus. However, evidence of focal contusion was observed in the cerebella of the r-mTBI animals. Intra-axonal accumulation of APP was observed in the corpus callosum of both injured groups. Reactive astrocytes assessed with GFAP immunohistochemistry were visible in the deep cortical layers below the impact site and in the CA1 region of the r-mTBI animals. In contrast, animals who received a single impact did not exhibit astrogliosis in these brain regions. Neuropathological analyses of the 6 and 12 month time points are ongoing.
The repetitively injured animals demonstrated significant and sustained neurobehavioral deficits at chronic time points post injury, whereas the behavioral effects of a single injury were resolved by 6 months. Taken together, the current findings provide support for the relevance of this model to investigate the long term consequences of TBI, particularly repetitive TBI, in humans. Further biochemical characterization of the model will identify potential targets for therapeutic intervention.
This research was supported by Department of Defense awards (W81XWH-07-1-0700 and W81XWH-10-1-0759) to Dr. Fiona Crawford and by the Roskamp Foundation.
single-mTBI pathology repetitive-mTBI chronic behavior
EFFECT OF NICOTINAMIDE ON BEHAVIORAL, NEUROPATHOLOGICAL, AND BIOMARKER OUTCOMES AFTER CONTROLLED CORTICAL IMPACT IN RATS: AN OPERATION BRAIN TRAUMA THERAPY CONSORTIUM STUDY
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Xiecheng Ma, MD, University of Pittsburgh/School of Medicine/ Neurosurgery
Jeremy Henchir, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Mu Xu, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Keri Janesko-Feldman, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Ronald Hayes, PhD, Banyan Biomarkers
C. Edward Dixon, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Operation Brain Trauma Therapy (OBTT) is a multi-center consortium evaluating therapies and biomarkers across traumatic brain injury (TBI) models. The University of Pittsburgh uses controlled cortical impact (CCI) in rats—a well-characterized model producing contusion, global damage and behavioral deficits. We tested nicotinamide's effect on neuropathology, behavior, and biomarkers.
Forty-eight male Sprague-Dawley rats were anesthetized and surgically prepared for CCI (4 m/sec, 2.4 mm deformation) or sham surgery. Rats were randomized into four groups: CCI + vehicle, CCI + nicotinamide (50 mg/kg), CCI + nicotinamide (500 mg/kg), and Sham. Nicotinamide was given IV at 15 min and 24h after injury. Blood sampling (0.7 mL whole blood) via the tail artery was obtained at 4h, 24h, and at sacrifice from each rat to measure serum levels of the neuronal biomarker UCHL-1 and the glial biomarker GFAP. Motor function (beam balance and beam walking) was evaluated on days 1–5 and Morris water maze (MWM) (acquisition and probe trial) were tested on days 14–20. The rats were sacrificed on day 21 for histopathological analysis of lesion volume and total hemispheric tissue loss—both expressed as a % of the contralateral hemisphere.
In both motor tests neither CCI+nicotinamide group differed vs. CCI+vehicle. For MWM acquisition, the TBI + nicotinamide (50mg/kg) group performed worse than sham (P<0.05). No differences were observed between 500 mg/kg and sham groups. Probe trial revealed that while the three CCI groups spent a significantly less time in the target quadrant that the three TBI groups, there were no differences between TBI groups. Lesion volumes (% contralateral hemisphere) were 8.48+1.01, 8.36±1.23, 11.18+1.16, and 0.05+0.05 for the CCI+50mg/kg, CCI+500mg/kg, CCI+vehicle, and sham groups, respectively. There was no significant difference between nicotinamide and vehicle treated groups. However, hemispheric volume loss (% contralateral hemisphere) was 18.08±1.50, 15.74±1.70, 20.73±1.29, and 0.24±0.67 for the CCI+50mg/kg, CCI+500mg/kg, CCI+vehicle, and sham, respectively–the 500 mg/kg nicotinamide group showed significantly less hemispheric tissue loss vs. CCI+vehicle. Nicotinamide at the 500 mg/kg dose (vs. vehicle) also significantly reduced the post-trauma (24h) increase in GFAP (P<0.05), but not UCHL-1 levels.
We conclude that nicotinamide did not attenuate functional deficits after CCI. The 50mg/kg dose of nicotinamide surprisingly worsened MWM performance. However, the 500 mg/kg dose produced a significant tissue sparing effect and reduced serum GFAP levels at 24h vs. vehicle. Our data are consistent with previous reports from other laboratories showing significant reductions in tissue loss by nicotinamide treatment after CCI in rats. However, this effect failed to translate into a behavioral benefit. The novel biomarker data incorporated into the OBTT study design suggests the interesting possibility that beneficial effects of nicotinamide on tissue loss after CCI are related to an effect on astrocytes rather than neurons—which could explain the lack of benefit on behavioral outcomes. Nicotinamide in CCI may merit additional studies, notably investigation of its effects when combined with other therapies targeting behavioral outcomes.
Supported by grant W81XWH-10-1-0623 from the United States Army.
traumatic brain injury, neuroprotection, GFAP
MODERATE TRAUMATIC BRAIN INJURY CAUSES ACUTE DENDRITIC AND SYNAPTIC DEGENERATION IN THE HIPPOCAMPAL DENTATE GYRUS
Ping Deng, M.D., Ph.D., SNRI/Indiana University
Zao Xu, M.D., Ph.D., SNRI/Indiana University
Jinhui Chen, M.D., Ph.D., SNRI/Indiana University
Hippocampal injury-associated learning and memory deficits are frequent hallmarks of brain trauma and are the most enduring and devastating consequences following traumatic brain injury (TBI). Several reports, including our recent paper, showed that TBI brought on by a moderate level of controlled cortical impact (CCI) induces immature newborn neuron death in the hippocampal dentate gyrus. In contrast, the majority of mature neurons are spared. Less research has been focused on these spared neurons, which may also be injured or compromised by TBI.
Here we examined the dendrite morphologies, dendritic spines, and synaptic structures using a genetic approach in combination with immunohistochemistry and Golgi staining.
We found that although most of the mature granular neurons were spared following TBI at a moderate level of impact, they exhibited dramatic dendritic beading and fragmentation, decreased number of dendritic branches, and a lower density of dendritic spines, particularly the mushroom-shaped mature spines. Further studies showed that the density of synapses in the molecular layer of the hippocampal dentate gyrus was significantly reduced. The electrophysiological activity of neurons was impaired as well.
These results indicate that TBI not only induces cell death in immature granular neurons, it also causes significant dendritic and synaptic degeneration in pathohistology. TBI also impairs the function of the spared mature granular neurons in the hippocampal dentate gyrus. These observations point to a potential anatomic substrate to explain, in part, the development of posttraumatic memory deficits. They also indicate that dendritic damage in the hippocampal dentate gyrus may serve as a therapeutic target following TBI.
This work was supported by grants from the Indiana Spinal Cord & Brain Injury Research Fund (SCBI 200-12); the Ralph W. and Grace M. Showalter Research Award, and the Indiana University Biological Research Grant to J. Chen.
TBI, hippocampus, dendrite, synapse, degeneration
ROD-MICROGLIA UNIQUE TO DIFFUSE BRAIN INJURY DIFFERENTIALLY EXPRESS IMMUNE RECEPTORS AND ALIGN PARALLEL TO NEURONAL ELEMENTS
Samuel E. Taylor, University of Bath, Bath, England. Spinal Cord & Brain Injury Research Center/University of Kentucky
Touxin Cao, BSc, Spinal Cord & Brain Injury Research Center, Department of Anatomy & Neurobiology, Department of Physical Medicine & Rehabilitation/University of Kentucky College of Medicine, Lexington, KY
Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Microglia play a pivotal role in inflammatory events following neurotrauma, with documented neurotoxic and neurotrophic roles. Nissl first described a specific rod-like morphology of microglia in conditions of infection and toxic exposure. Since then, little research has been conducted on rod-microglia and their role within the brain remains enigmatic.
To test the hypothesis that rod-microglia express a unique complement of inflammatory markers following diffuse brain injury, adult male, Sprague-Dawley rats were subjected to a single moderate severity (1.9 atm; 6–10 min righting reflex time) midline fluid percussion injury (FPI). Iba-1 (ionized calcium binding adapter protein) immunohistochemistry identified microglia in all states, with their phenotype compared between FPI brain-injured and uninjured animals. Double-labelling of Iba-1 was conducted with cell markers for neurons, astrocytes and oligodentrocytes, as well as other known markers for microglia including CD68, CD11b, CD11c, CD45 and Ox6.
In somatosensory cortex, a vast proportion of microglia show elongation and alignment after diffuse brain injury, with a rod-like morphology similar to that described by Nissl. Cells with elongated cell bodies have side branches that disappear over 1 week post-injury and principle processes protrude solely from the apical and basal ends. The cell bodies themselves become extremely long, cells couple together to form trains that exceed lengths of 400 μm and easily span multiple cortical layers. At day 7 post-FPI, some cells in these trains co-localize with some of the known markers for active microglia, including CD68 and Ox6; however these cells are not CD45, CD11b or CD11c positive. Furthermore, Ox6 staining was intermittent along trains with positive and negatively labelled microglia indiscriminately patterned throughout the train. Punctate staining along the processes and within the cytoplasm of CD68 positive rods was observed along the length of some trains. In addition, trains of rod-microglia align parallel and adjacent to neuronal elements as evidenced by double-labelling with neurofilament (NF-M) and microtubule associated protein (MAP-2). The aligned trains have been observed to consist of between 2 and 15 rod-microglia, which lay adjacent to the entire length of the observable neuronal process. No anatomical association was evident with GFAP positive astrocytes or CC1 positive oligodendrocytes.
Results thus far indicate that diffuse brain injury induces a sub-acute morphological change in microglia, such that their elongation aligns with neuronal processes in the S1BF. These structural changes are in line with previous reports that suggest microglia structure follows function, whereby ramified cells which survey the microenvironment have long thin process but amoeboid microglia that phagocytose debris look morphologically similar to macrophages. The preferential association of neuronal elements with these phenotypically distinct rod-microglia leads us to conclude that their primary role is synaptic stripping and is associated with late onset behavioral sensory sensitivity as previously described.
Amanda Lisembee for her surgical expertise. Studies were conducted at the University of Kentucky. Partly supported by NIH/NINDS R01 NS065052.
Rod-microglia, diffuse brain injury
EFFECTS OF THERAPEUTIC HYPOTHERMIA ON INFLAMMASOME SIGNALING AFTER TRAUMATIC BRAIN INJURY IN RATS
Juan Pablo de Rivero Vaccari, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Robert W. Keane, PhD, Department of Physiology and Biophysics, University of Miami Miller School of Medicine
Helen M. Bramlett, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton Dietrich, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
The inflammasome constitutes an important component of the innate CNS inflammatory response after TBI. Therapeutic hypothermia has been reported to improve outcomes in several animal models of TBI and has been successfully translated to specific patient populations. We investigated the influence of therapeutic hypothermia on inflammasome activation after TBI.
Adult male Sprague-Dawley rats (250–350 g) were subjected to parasagittal fluid percussion brain injury at a moderate severity (1.7 – 2.2 atm). Temperature manipulation (33C) was initiated 30 min after TBI and maintained using cooled air and heating lamps for a period of 4 h. The normothermia groups were maintained at 37C throughout the procedure. At the end of the temperature manipulation or at 24 h after TBI, rats were sacrificed and the samples of traumatized cortex and hippocampus were immediately excised, and prepared for immunoblot analysis or immunohistochemistry.
In the normothermic groups, caspase-1, caspase-11 and expression of the purinergic receptor P2X7 significantly increased at 24 h after TBI (p<0.05). In contrast, the hypothermic groups demonstrated a significant decrease at 24 h after trauma as compared to the normothermic groups (p<0.05). In addition, cortical neurons in culture subjected to 2 h posttraumatic hypothermia demonstrated significant decreased secretion of caspase-1 into culture medium as compared to injured neurons treated with normothermia (p<0.05) and significantly decreased caspaes-3 activation (p<0.05).
Therapeutic hypothermia decreases inflammasome signaling in neurons and reduces the innate immune response to TBI.
This study was supported by NIH/NINDS NS030291 and NS 042133.
Caspase-1, inflammasome, hypothermia, IL-1, TBI
A CANNABINOID-TYPE-2 RECEPTOR AGONIST ACTS AS A POTENT IMMUNOMODULATOR AND ATTENUATES NEUROVASCULAR AND BEHAVIORAL CONSEQUENCES OF CORTICAL INJURY IN MICE
Christine Macolino, BS, Thomas Jefferson University
Craig Hooper, PhD, Thomas Jefferson University
Larry Harshyne, PhD, Thomas Jefferson University
Jack I. Jallo, MD, PhD, Thomas Jefferson University
Melanie B. Elliott, PhD, Thomas Jefferson University
Post-traumatic inflammation participates in both the secondary injury cascades and repair of the CNS, both of which are influenced by the endocannabinoid system. The effects of repeated treatment with a cannabinoid type-2 receptor (CB2R) agonist on blood-brain-barrier integrity, neuronal degeneration, immune responses, and behavior were studied.
C57BL/6 mice underwent controlled cortical impact (CCI) and received repeated treatments with a CB2R agonist, 0-1966, or vehicle. Craniotomy and naïve mice served as controls. Following euthanasia at six hours, one, two, three or seven days post-injury, brains were removed for histochemical analysis. Blood-brain-barrier permeability changes were evaluated using sodium fluorescein (NaF). Tissue damage and neurodegeneration were assessed using fluorojade C staining and injury volumes in cresyl violet stained sections. Antibodies against ionized binding adaptor-1 protein for macrophage/microglia cell counts were quantified using stereological methods. The time course post-injury and effects of treatment on intracellular adhesion molecule (ICAM) and tumor necrosis factor-alpha (TNF-α) mRNA was determined using real-time polymerase chain reaction. Flow cytometry was utilized to assess the changes in immune cell populations after injury and treatment. Rota-rod and open-field testing were performed at seven days post-injury to evaluate motor function and natural exploratory behavior in mice.
0-1966 treatment resulted in a significant reduction in NaF uptake and the number of degenerating neurons compared to the vehicle-treated group, p<0.01. ICAM mRNA was significantly increased for up to two days post-CCI, while TNF-α remained increased in a cyclical pattern over seven days. 0-1966 significantly attenuated increases in ICAM and TNF-α mRNA at one day after injury. CCI injury results in a significant influx of immune cells in different brain regions after injury and were altered with treatment. These changes in CCI mice treated with 0-1966 were associated with a prolonged reduction in macrophage/microglia cell counts. 0-1966-treated mice demonstrated improvement on a rota-rod and open-field testing when compared to vehicle-treated mice.
In conclusion, repeated treatments with a CB2R agonist, 0-1966, results in attenuated blood-brain-barrier, neuronal damage, and significant immunomodulation. These changes were associated with improvements in behavioral outcome after injury. We propose that the neuroprotective effects of 0-1966 in our model of TBI are mediated predominantly through the CB2R, though whether stimulation of other non-CB1/CB2 receptors is also involved requires confirmation through additional genetic and pharmacological modulation of the cannabinoid system. Our results add to the multitude of data showing inhibition or modulation of microglial actiation is beneficial for neuron survival and improved function. We propose that 0-1966 stimulation of the CB2R reduces immune cell infiltration and in addition, modulate microglial responses to injury, in turn promoting an advantageous environment for neuronal survival and function.
none
cannabinoid, BBB, neuroinflammation, immunomodulation, microglia
MATERNAL DIET IS A VULNERABILITY FACTOR FOR THE REGENERATIVE CAPACITY OF THE BRAIN FOLLOWING TBI
Ethika Tyagi, Ph.D., Department of Integrative Biology and Physiology, UCLA
Rahul Agrawal, Ph.D., Department of Integrative Biology and Physiology, UCLA
Vulnerability factors are an important aspect in the pathology and treatment of TBI; however, their molecular bases are poorly understood. We have embarked to determine the influence of imbalances in maternal nutrition and its interaction with subsequent dietary habits during the phase of brain maturation on the long-term consequences in the TBI pathology.
Pregnant Sprague-Dawley rats and their offspring were kept on either n-3 adequate (1.2 %) or deficient diets. After 8 weeks, the male offspring was transitioned to a diet high in saturated fat for 6 weeks and then subjected to moderate fluid percussion injury (FPI).
The deficiency in n-3 evoked an increase in anxiety-like behavior in elevated plus maze, which found associated with an increase in the receptor for the anxiolytic molecule NPY. The disruption in BDNF signaling (decrease in phosphorylation of TrkB, CaMKII, Akt and its downstream effector CREB) with n-3 deficiency was in direct proportion to NPY1R in the frontal cortex. These effects were exacerbated by the transition of diet to high fat and the response was even more profound after FPI.
These results are novel in showing the importance of nutrition in defining the brain capacity to respond against TBI.
NIH RO1 NS050465, NIH RO1 NS 056413, the UCLA Brain Injury Research Center
diet, brain injury, anxiety, rat
PROFILING THE INJURY SIGNATURE OF ASTROCYTES FOR NEW NEUROTRAUMA BIOMARKERS
Gregg A. Czerwieniec, PhD, UCLA, Molecular Instrumentation Center
Pablo M. Paez, PhD, UCLA, Semel Institute for Neuroscience and Human Behavior
Melissa Sondej, PhD, UCLA, Molecular Instrumentation Center
W. Dalton Dietrich, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Thomas C. Glenn, PhD, UCLA, Brain Injury Research Center, Department of Neurosurgery
Joseph A. Loo, PhD, Department of Biological Chemistry, UCLA, David Geffen School of Medicine
Michael V. Sofroniew, PhD, MD, UCLA, Department of Neurobiology, Geffen School of Medicine
There is an urgent need for biofluid markers to diagnose traumatic brain injury (TBI). Astrocytes are the most abundant cells in the human cerebral cortex with key metabolic and neuroprotective functions relevant for TBI recovery. Glial pathology and reactive astrogliosis are incompletely understood with few well described molecular markers.
Rodent and human cortical astrocytes were purified and matured in vitro. Cells grown on deformable membranes were injured by a controlled abrupt pressure pulse causing reproducible and widespread strain forces (Ellis et al., 1995). Traumatized astrocytes were monitored over time on a temperature and CO2 controlled spinning disk confocal microscope stage. Transient mechanoporation was measured by uptake of 10/70kD size dextran dyes. Apoptosis onset was recorded using activated caspase3 converted fluorescent substrate. Conditioned medium from traumatized and control astrocytes was analyzed using 2D-PAGE and stretch-released proteins were identified by liquid chromatography-tandem mass spectrometry (LC-MS/MS) and confirmed by western blotting. The proteome of TBI patient cerebrospinal fluid (CSF) was determined by bottom-up MS using two abundant protein depletion strategies. Multiple reaction monitoring (MRM)-MS with labeled (heavy) peptides as internal standards was performed for quantification of individual astrocyte trauma markers along with CSF and plasma markers.
Mechanically traumatized rodent and human astrocytes displayed stellation, cell body contraction and increased levels of GFAP and other astrocyte markers (Wanner et. al., 2008; Wanner, 2011). Mechanical trauma caused transient loss of membrane integrity in 40% of astrocytes across species, whereas delayed apoptosis was higher in mouse compared to human and rat. Using proteomic 2D-PAGE analysis of mouse astrocyte conditioned medium after mechanical trauma we determined 107 significantly changed spots (>1.8 fold or <0.5 fold, P<0.05) among the routinely detected >500 spots (Sondej et al., 2011). We identified a novel set of 79 trauma associated astrocytic proteins, and confirmed >20 by immunoblotting. Among the acutely released (during and within 30 min after stretching) were mainly cytosolic proteins with average MW of 32kD and functions in energy metabolism, oxidative stress and neurotransmitter regulation. A set of proteins with delayed release (elevated 12–24 hours post-injury) was identified as predominantly cytoskeleton and associated proteins, with average MW of 60kD. For translational relevance, this in vitro mouse astrocyte traumatome was compared to proteins in brain fluid of TBI patients. First, we identified >649 proteins in brain injury CSF constituting a trauma proteome substantially larger than currently published. Surprisingly 65% of the astrocyte-released proteins were found in TBI patients' CSF. From this pool we selected new trauma fluid markers that were enriched in astrocytes but absent from blood. Six astrocyte-injury specific markers, among them GFAP and glutamine synthetase, were found in TBI CSF by immunoblotting. Using MRM-MS we detected these astrocyte release markers in TBI CSF. Internal standard heavy peptides were spiked into raw TBI CSF and were measured along with endogenous peptides to quantify the astrocytic proteins along with CSF and plasma markers as a chemical measure for astrocyte injury and extent of blood contamination after TBI.
Mechanical trauma inflicted reproducible reactivity, mechanoporation and cell death rates in rodent and human cultured astrocytes. Combined with a proteomic approach we determined a novel set of proteins released by mechanical trauma, the astrocyte traumatome. Distinct profiles of fast leaking, smaller cytosolic proteins like metabolic enzymes were associated with acute mechanoporation. Delayed released, larger cytoskeleton and associated proteins were related to secondary cell death. Our translational data revealed an astrocyte protein release signature in TBI patient CSF that can be mined for new neurotrauma biomarkers. We show that identifying protein markers associated with astrocyte damage or death is a new approach to evaluate TBI patient biofluids that has diagnostic potential. The novel use of quantitative MRM-MS on neurotrauma samples allows multiplexing of many markers in the same, limited patient sample and enables standardized measurements of injury markers in relation to blood brain barrier compromise after TBI.
Eunice Kwon, Ana Fernandez, Melania Apostolidou, David Hovda, Ross Bullock, Zsuzsanna Nemeth, Ryder and Ronald Reagen trauma centers. Buoniconti Fund; Neilsen Foundation #82776; NIH#NSO72606-01A1.
Astrocyte, culture, proteomics, CSF, biomarker
ROLE OF SHOCK WAVE, ITS CHARACTERISTICS AND OTHER FACTORS IN A RAT MODEL OF BLAST-RELATED NEUROTRAUMA
Sandor Krisztian. Kovacs, M.D., Uniformed Services University of the Health Sciences
Erin Murphy, M.S., Uniformed Services University of the Health Sciences
Eleanor Kitces, B.S., Uniformed Services University of the Health Sciences
John Magnuson, B.S., Uniformed Services University of the Health Sciences
Suhana Sarkar, B.S., Uniformed Services University of the Health Sciences
Peter Swauger, B.S., ORA Inc.
Blast-related traumatic brain injury is highly prevalent in the war theater, but the specific role of blast's shock wave remains unclear. We addressed this role by using a rat model to evaluate the neuropathological and neurobehavioral outcome of blast's shock wave in relation to several factors.
Anesthetized male Sprague-Dawley rats were immobilized and exposed to primary blast as generated either by detonation of high explosive or rupture of a diaphragm under high gas pressure within different “blast tubes.” Most animals were exposed leaning on their right side, with back facing the blast source. A subgroup had their right side facing the blast, to test therole of orientation to blast. Blast-intensity dependence (12–50psi) of neuropathological outcome was evaluated in the presence and absence of chest protection, to also test the role of chest-mediated mechanisms. The role of pulse duration is being evaluated by comparing the outcome of exposure to short (2–3msec) and longer (7–9msec) shock waves. The adequacy of gas-driven shock tubes to reproduce “real-life” explosive-driven blasts is being tested by comparing the outcome of exposure to gas- and explosive-driven shock waves with similar pressure-time profiles.
Thresholds of lung-injury-related lethality for shock pulse durations of 7–9 msec were at peak static overpressure levels of about 50 and 21 psi in the presence and absence of chest protection, respectively. Extent of lung, but not brain, injury was blast-intensity dependent. Overall, results in the presence of chest protection show a limited and highly inconsistent impact of primary blast exposure on both behavioral and neuropathological outcome from 3 hours to 28 days after exposure to blast, even at blast intensity levels approaching lethality by lung injury. Neurological scores, test of motor coordination, memory, anxiety and hyperarousal were mostly negative (except for occasional and predictable deafness). Histopathological evidence includes presence of subdural, subarachnoidal, subependimal and intraventricular hemorrhages. In the presence of chest protection, no evidence of a cellular inflammatory reaction was observed; amyloid precursor protein (APP) was occasionally seen accumulating in the neuronal body, but not in the axon; fluorojade B staining provided no evidence of neuronal degeneration; no evidence of edema was observed either in terms of water content at 6 hours, or albumin breach of the blood-brain barrier at any time between 6 hours and 28 days post-blast. However, limited evidence of inflammatory reaction (neutrophil infiltration, microglia and astrocyte activation) and axonal injury (by APP staining) were observed in areas of subarachnoidal bleeding in animal exposed without chest protection.
Both in the presence or absence of chest protection, and even at blast intensity levels approaching lethality, the neuropathological and neubehavioral impact of primary blast exposure has been mild and inconsistent, though preliminary observations suggest some qualitative difference in the outcome of blast exposure without chest protection. These results, obtained using consistent animal models and protocols, add to the heterogeneity of results reported in the study of blast neurotrauma. Ongoing experiments are aimed to address this heterogeneity by testing the possible role of the shock wavelength, evaluating the impact of blast-induced acceleration, and assessing the adequacy of the gas-driven shock tube in reproducing the impact of “real-life” explosive-driven blasts. Additional ongoing experiments are evaluating the longer-term (90-day) behavioral impact of blast exposure.
Blast Spinal Cord Injury Program Center for Neuroscience and Regenerative Medicine Defense Advanced Research Projects Agency
Blast Shock wave Neurotrauma Rat
IDENTIFYING NEURONAL PROTEINS THAT INTERACT WITH CYTOPLASMIC DYNEIN AFTER BRAIN INJURY
Xuanli Yao, Ph.D., Uniformed Services University of the Health Sciences
Xin Xiang, Ph.D., Uniformed Services University of the Health Sciences
Kevin Pfister, Uniformed Services University of the Health Sciences
Cytoplasmic dynein, a multi-subunits minus-end-directed microtubule motor required for retrograde transport, is crucial for neuronal survival. Upon injury of peripheral neurons, several signalling proteins form complex with dynein and are transported retrogradely towards the nucleus in the cell body, and this process is implicated in initiating transcription in the nucleus.
We designed a proteomic approach to identify proteins that bind to dynein upon injury of the cerebral cortex. We first created a knock-in mouse line in which a neuon-specific dynein intermediate chain is tagged at its C-terminus with GFP and 3XFLAG, an affinity tag that allows dynein-bound factors to be purified. We used the CCI (controlled cortical impact) method to injury the mouse cerebral cortex and isolate dynein-bound proteins for proteomic analysis.
So far, we have identified several proteins that bind to dynein upon injury, which include clatharin heavy chain, 14-3-3-zeta, HSP90 and myosin-10. These proteins have never been identified previously as retrograde injury factors in peripheral neurons.
The functional significance of these interactions will be studied in the future. We are particularly interested in the 14-3-3-zeta protein that may serve to protect the signaling property of phosphorylated proteins.
Funding source: CNRM-USUHS
Brain Injury Axonal transport Cytoplasmic dynein Proteomic approach 14-3-3
OUTCOMES OF MILD TRAUMATIC BRAIN INJURY IN A RODENT MODEL
Hariett Rea, B.S., Univ TX Med Br
Kathia Johnson, B.S., University of Texas Medical Branch at Galveston
Margaret Parsley, B.S., Univ TX Med Br
Geda Unabia, B.S., Univ TX Med Br
Guo Ying Xu, CMD, University of Texas Medical Branch at Galveston
Douglas S. Dewitt, PhD, UTMB
Raymond Grill, PhD, Univ TX Houston
Claire Hulsebosch, PhD, University of Texas Medical Branch at Galveston
Mild traumatic brain injury (mTBI), due to blast exposure from improvised explosive devices, motor vehicle accidents, or sport-related brain injuries, result in long term impairment of cognition and behavior. We hypothesize that blockade of the cytokine signaling response to mTBI with FDA-approved Kineret and Etanercept will be beneficial.
Adult, male, Sprague-Dawley rats were anesthetized (1.5% isoflurane) and subjeceted to mild (1.0 atm) fluid percussion injury (FPI) or sham injury (surgical preparation but no FPI). We assessed level of FPI (duration of righting reflex suppression) and acute vestibulomotor function (beam balance/beam walking, foot faults). We also assessed working memory (Morris water maze) and stress (glucocorticoid levels, fecal pellet counts). Brain cytokine levels were measured by immunoassay and immunohistochemistry while astrocytic and microglial activation, blood brain barrier function, and cell losses were measured by immunohistochemistry. Rats were treated with clinically relevant doses of Kineret, an Interleukin-1 Receptor Antagonist, (IL-1Ra) or Etanercept, an antibody to the Tumor Necrosis Factor Receptor alpha (TNFα) 30 minutes and 6 hours post-injury and then daily for 14 days. Animals were sacrificed at three or six hours or 18 days post-mTBI.
Righting reflexes and beam balance and foot faults performance were impaired acutely after mTBI. Similarly, the times to find a hidden platform were significantly longer in the mTBI than sham rats, indicating impaired working memory. Rats subjected to mTBI also showed increased stress, as measured using a fecal count assay. There was a significant increase in brain microglial-derived IL-1α/β and TNFα levels in cortex and hippocampus as early as three and sxi hours after mTBI. There was also significant astrocytic activation as reflected in GFAP, vimentin and desmin in situ levels as early as six hours after mTBI that persisted for 18 days and a parallel increase in microglia and macrophages as reflected by increased levels of IBA − 1, a calcium binding protein specifically expressed in microglia and macrophages. There was also a significant presence of IgG and albumin in parietal cortex 18 days after injury, perhaps due to a heightened blood brain barrier disruption as early as 6 hours after injury that persisted for 18 days as shown by the significant decreased levels of SMI71 antibody. We present data documenting a beneficial effect from the two proposed interventions in both neuropathological and behavioral outcomes.
- There is persistent inflammation in CNS after mild TBI. - There are significant increases in cytokine levels in brain after mild TBI. - There is disruption of the blood brain barrier after mild TBI. - There are significant behavioral impairments after mild TBI. - There is a correlation between inflammation, blood brain barrier disruption and mild TBI. - Kineret or Etanercept treatment improved outcomes after mild TBI.
Supported in part by DOD grant T074693, Mission Connect Mild TBI translational Research Consortium (W81XWH-08-2-0132) & Moody Center for TBI & SCI Research
mild TBI, inflammation, cytokines
EMERGENCE OF ANTI-SOCIAL AND AGGRESSIVE BEHAVIORS AFTER PEDIATRIC TRAUMATIC BRAIN INJURY IN MICE
Sandra Canchola, University of California, San Francisco
Prof. Linda J. Noble-Haeusslein, Ph.D., University of California San Francisco
Impaired social skills are prominent after traumatic brain injury (TBI), particularly when sustained at a younger age, and include social withdrawal, aggression and difficulty maintaining relationships. Nevertheless, no studies to date have fully profiled clinically relevant social behaviors in mice after experimental TBI to the pediatric brain.
Controlled cortical impact (or sham-operation) was applied to the parietal lobe of male C57Bl6 mice (n=9/group) at post-natal day 21 (p21), an age approximating the toddler-aged child. Social behaviors were evaluated during adolescence (p35–42) and again during early adulthood (p60–70). ‘Stimulus’ mice were unfamiliar, naïve mice matched for vendor, strain, sex, age and weight. Social tasks were as follows: Resident-Intruder test (stimulus mouse introduced to home cage of test mouse, behaviors quantified from recorded videos); Crawley 3-Chamber task (3 stage paradigm which evaluates a preference of sociability and social novelty; video-tracking of movements); Partition test (test and stimulus mice separated by a permeable partition); and a Tube-Dominance task (test and stimulus mice come face-to-face in a narrow tube; one will force the other out and is declared the ‘winner’). A Buried Food task was also performed to assess olfactory function.
Brain-injured and sham mice showed similar investigation times in the Partition test, which was reduced overall at adulthood compared to adolescence. Adolescent TBI and sham mice exhibited a comparable level of social investigation and preference for sociability in the Resident-Intruder and 3-Chamber tasks. However, by adulthood, TBI mice spent significantly less time investigating the stimulus mouse in the Resident-Intruder paradigm compared to sham mice. Similarly, while adult sham mice showed a normal preference for sociability and social novelty in the 3-Chamber task, adult TBI mice in fact preferred an empty chamber and a familiar stimulus mouse compared to a novel one. Adult TBI mice also had an increased likelihood of exerting dominance in the Tube-Dominance task compared to sham, suggesting a more aggressive phenotype. These behavioral changes were not attributed to deficits in olfaction, as TBI and sham mice shared a similar ability to rapidly detect a hidden food source.
Problems with social behavior are common after TBI in both adults and children, with widespread consequences on academic achievement, relationships with peers and re-integration into society. The development of normal social behaviors is protracted through childhood and adolescence, suggesting that injury at a young age is likely to adversely affect the acquisition and establishment of social skills. Here we show that mice subjected to TBI at a young age show deficits in social interactions, including an aggressive phenotype, which emerge across development to adulthood. This trajectory parallels other cognitive deficits previously reported in this model and seen in brain-injured children, including spatial memory and learning problems, hyperactivity and elevated anxiety. Importantly, these findings provide essential groundwork for addressing the underlying substrates that govern evolution of aberrant social interactions after trauma to the pediatric brain.
B.D.Semple is supported by a Sir Keith Murdoch Postdoctoral Fellowship from the American Australian Association.
brain injury, pediatric, social, behavior
A MILD CLOSED HEAD INJURY MODEL IN MICE PRODUCES INCREASED GLIOSIS WITH REPEATED INJURIES
Benjamin Tuttle, BS, University of Kentucky/SCoBIRC
Kathryn Saatman, PhD, Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Due to increasing concern about cumulative effects of sports and war-related mild traumatic brain injuries (mTBI) it is necessary to develop a translatable model of repetitive closed head injury (CHI) to assess the histological and behavioral outcomes associated with mTBI.
This CHI model utilizes a pneumatically controlled cortical impact device, modified with a pliant 5mm diameter tip to impact the intact mouse cranium along the dorsal-midline suture at a prescribed velocity and intended depth to produce a diffuse mTBI. Mice receiving a single 0.5mm-3.0mm CHI were examined 48hrs post-injury (n=2–4/depth) for histological effects, including astrogliosis (GFAP immunohistochemistry), axonal injury (cAPP immunohistochemistry), and neuronal degeneration (Nissl, Fluorojade C). Additionally, mice receiving a single 0.5mm-2.0mm CHI, were evaluated behaviorally for righting reflex time and apnea duration immediately after impact, along with a modified Neurological Severity Score (NSS) to evaluate motor coordination deficits 1hr post injury. To determine the effects of dual mCHI, two impacts of the same depth (0.5mm, 1.0mm, or 1.5mm; n=3/depth) were produced spaced 24hrs apart and analyzed 48hrs after the first injury by the same histological and behavioral parameters as the single impact group.
Increasing depths of impact from 0.5mm to 3.0mm produced a gradation of reactive astrocytosis in the dentate gyrus of the hippocampus and the entorhinal cortex at 48hrs post-injury compared to naive mice, without causing axonal injury. Increased depth impacts of 2.5mm or greater led to hippocampal neuronal death but were associated with an increased incidence of skull fracture/death and were therefore avoided from further experimentation. Single CHI of 1.0mm to 2.0mm, but not 0.5mm, caused only a very mild motor deficit 1hr post-injury. However, both apnea (p<0.0005) and righting reflex times (p<0.001) were significantly increased for the 2.0mm group (46.9+/− 32.8 seconds) compared to all other groups assessed. Dual impacts of 1.0mm and 1.5mm, but not 0.5mm, enhanced astrocytosis compared to a single impact, without resulting in axonal injury or neuronal degeneration. Behaviorally, the additional mCHI did not increase motor deficits, apnea, or righting reflex duration.
By utilizing the modified cortical impact injury device we successfully developed a model of CHI that will provide a tool for exploration of mechanisms underlying the pathology and behavioral impairment after mTBI as well as changes that occur with repetitive injuries. From our single impact studies we standardized the different impact depths to eliminate skull fractures and mortality. Dual CHI exacerbated astrocytosis but did not lead to axonal injury or neuronal degeneration 48hrs after the first impact. Future experiments will employ multiple impacts (>2) to determine if there is additional increases in astrocytosis, development of axonal injury or neuronal degeneration. Behavioral tasks to detect cognitive deficits will also be incorporated to assess effects of multiple CHI. In addition, further studies will investigate whether parameters such as apnea duration, the righting reflex, and initial NSS correlate with the extent of brain pathology.
NIH P01 NS058484 and P30 NS051220
Mild TBI, Repetitive, Astrocytosis
THE ROLE OF PIOGLITAZONE, mitoNEET AND MITOCHONDRIA IN TBI
Werner J. Geldenhuys, Ph.D., Northeast Ohio Medical University
Richard T. Carroll, Ph.D., Northeast Ohio Medical University
Patrick G. Sullivan, Ph.D., The University of Kentucky Chandler College of Medicine
The fundamental concept underlying this study is that TBI-induced excitotoxicity increases mitochondrial Ca2+ cycling/overload, ultimately leading to mitochondrial dysfunction and subsequent neuronal cell death. In this study, we hypothesize Pioglitazone to be neuroprotective following TBI with a mechanism directly related to interactions with the novel mitochondrial protein, mitoNEET.
Wild-type, mitoNEET null, heterozygous mice (Pioglitazone study) and Sprague Dawley rats (NL-1 study) were subjected to a sham surgery or a severe TBI. A craniotomy was performed and animals were injured using a pneumatic impactor. They received a 1.0mm (mice) or 2.0 mm (rats) depth CCI contusion injury at 3.5 m/s with a dwell time of 500 ms. Sham animals received a craniotomy but did not receive an impact to the brain. Following the injury the skull cap was replaced and secured with dental acrylic and incision was closed with surgical staples. All mice received intraperitoneal injections of vehicle or Pioglitazone (10 mg/kg) and rats received NL-1 (10mg/kg) at 15minutes and 24hours post-injury. Animals were allowed to survive for 7 days and then sacrificed. Histological studies with cresyl violet staining of coronal slices were performed and cortical sparing was measured in a blinded fashion using region of interest with ImageJ.
The results demonstrate that Pioglitazone is neuroprotective following TBI in wild-type (+/+) but not mitoNEET null (−/−) and heterozygous (+/−) mice. Adult mice (n=6 per group) were injured (1.0 mm, severe) and cortical tissue sparing assessed at 7 days post-injury. Pioglitazone treatment significantly increases tissue sparing in wild-type (+/+) but not mitoNEET null (−/−) and heterozygous (+/−) mice. Additionally, no significant differences were measured between vehicle treated animals of any genotype. We also see that NL-1, a specific mitoNEET ligand that lacks PPAR-gamma binding, is neuroprotective following TBI. Adult Sprague-Dawley rats were injured (severe) and tissue sparing assessed at 7 days post-injury. NL-1 administration increases cortical tissue sparing after TBI. Although significance was not reached in this pilot study NL-1 had a very robust effect that is trending (p=0.0585) toward significance.
We have shown here that Pioglitazone, a drug known for its PPAR agonistic properties, can increase mitochondrial bioenergetics and cortical sparing following TBI but these effects do not appear to be dependent upon PPAR interaction. We believe Pioglitazone to be a novel mitochondrial targeting drug which is able to alter mitochondria bioenergetics following TBI. We also believe that mitoNEET may be a novel therapeutic target for TBI. The results of these studies will help to shed light on the fundamental processes involved in TBI neuropathology and may pinpoint potential interventions and targets for the treatment of TBI.
This work is supported by NIH/NINDS grants R01- NS048191, NS062993, KSCHIRT (PGS) and P30 NS051220
Neuroprotection, Mitochondria, mitoNEET, Pioglitazone, TBI
ATYPICAL CALPAIN, CALPAIN 5 IS PREDOMINANTLY PRESENT IN NEURONAL MITOCHONDRIA; AND CONTRIBUTES TO NEURODEGENERATION FOLLOWING TRAUMATIC BRAIN INJURY VIA CASPASE DEPENDENT PATHWAY
Dingyuan Lou, Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Vimala Bondada, Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Sarbani Ghoshal, Ph.D., Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Adam Benson, Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Kathryn Saatman, Ph.D., Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
James Geddes, Ph.D., Department of Anatomy and Neurobiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Unlike typical calpains 1 and 2, an atypical calpain, Calpain 5 (Capn5) is largely uninvestigated. The goal of this study was to evaluate the relative expression, developmental regulation and localization of Capn5 in the CNS, followed by examining its role in cell death and TBI.
Our study materials are Capn5+/− mice, Sprague-Dawley rats, rat-B35 and human-SHSY-5Y neuroblastoma cells, and stable Capn5 knock-down (Capn5KD) SHSY-5Y cells created using lentiviral shRNA. An antibody, Abcam 28280 was used to detect Capn5 and it's specificity for calpain 5 was confirmed. Relative mRNA expression of various calpains in the rat CNS was examined based on qPCR-comparative CT method (ΔΔCT). Capn5 protein and mRNA were profiled using western blot and ΔΔCT method across late embryonic to adult rat CNS. Localization was examined through X-gal staining, immunohistochemistry, confocal microscopy, subcellular fractionation, mitochondrial subfractionation and proteinase K treatment. For TBI study, Capn5+/+ and Capn5+/− mice received severe controlled cortical impact (CCI) injury. Cortical lesion was measured using Bioquant software at 24hrs post injury. Cell death was studied in staurosporine treated Capn5 KD cells using TUNEL assay and Caspase 3 activation.
Capn5 has only one subunit, which possesses a conserved non-EF hand Ca2+ binding catalytic domain and a unique domain T. Knocking down TRA-3, an orthologue of Capn5 protects from neuronal cell death in C.elegans, indicating its importance in the CNS. We compared relative mRNA expression of ubiquitous calpains 1,2, 5, 7,10, & 13, revealing that Capn5 is the second most highly expressing calpain, after Capn2, both in brain (N=4, p<0.0001) and spinal cord (N=2). Capn5 was also detected prominently at later postnatal developmental periods only. Relative mRNA and protein levels were measured at embryonic day 18, p0 (postnatal day 0), p5, p10, p15, p20, p30 and p90, both in brain (N=4, per time point) and pooled spinal cord (N=2). Antibody Ab28280 is specific to Capn5, it does not detect purified ubiquitous Capn2; and peptide blocking with the Capn5 amino terminal peptide, abolishes immunoreactivity with the 75 kDa Capn5 band. Using Ab28280, Capn5 was found to be predominantly expressed in neurons and localized to neuronal mitochondria and nuclei. This was evaluated using lacZ staining of Capn5+/− brain sections (N=3), double-label immunohistochemistry (N=3); probing Capn5 in fractions obtained through differential centrifugation of rat cortex (N=7); and confocal microscopy against mHsp70 and Capn5 in SHSY-5Y-neuroblastoma cells (N=5). Proteinase K treatment of mitochondria and mitoplasts from rat B35-neuroblastoma cells and rat synaptic mitochondria indicates that Capn5 localization is similar to apoptosis-inducing factor protein (N=3–4). Our preliminary data show that knocking down Capn5 protects from cell death and cortical tissue damage in TBI. A decrease in caspase activation and TUNEL positive nuclei was observed in Capn5KD cells following staurosporine treatment. Also, following severe CCI injury, Capn5+/− mice showed significantly smaller cortical contusion volumes 24hrs post injury, as compared to Capn5+/+ mice (N=5, P=0.03).
Our investigation on Capn5 concludes that Capn5 is the second most highly expressing calpain in the CNS, after Capn2. It appears later in postnatal development and is present in adult CNS. Capn5 is found to be predominantly expressed in neuronal mitochondria and nuclei. In mitochondria, Capn5 localization appears to be similar to AIF protein, which is anchored on the inner membrane and projects into the inner membrane space. Our preliminary results suggest that knocking down Capn5 protects from cell death and Caspase 3 activation in human-SHSY-5Y cells, and cortical tissue damage following severe CCI injury in mice. Together, these results identify Capn5 as a major calpain protease in the CNS which may contribute to neurodegeneration following TBI via a caspase-dependent pathway in neurons.
This research was supported by predoctoral fellowship from the Kentucky Spinal Cord and Head Injury Research Trust (KSCHIRT) and NIH grant P01 NS058484.
Calpain 5, Capn5, TBI, neurodegeneration
PREOPERATIVE MILD HYPOTHERMIA ATTENUATES NEURAL DEGENERATION IN RAT SUBDURAL HEMATOMA DECOMPRESSION MODEL
M. Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine,
Shyam Gajavelli, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Christine Bomberger, University of Miami, The Miami Project to Cure Paralysis
Daniel Diaz, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Stephen S. Burks Jr., B.A., University of Miami Miller School of Medicine
Helen M. Bramlett, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton Dietrich, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Posttraumatic hypothermia is known to improve outcome of traumatic brain injury using various experimental models. However, hypothermia has not shown efficacy in clinical multi-center trials. Based on the latest trial (NABISH:II), we hypothesized hypothermia will reduce the effects of reperfusion injury and excitotoxicity following decompressive craniotomy of acute subdural
Acute subdural hematoma was induced in 28 Sprague-Dawley rats, allocated into four equal groups. Temperatures were manipulated as follows: Group1 (Normothermia) maintained at 37°C throughout. Group2 (Early hypothermia) Brain temperature (BT) 33°C 30 min prior to decompression, then continued for three hours. Group3 (Late hypothermia) BT lowered to 33°C 30 min after decompression and maintained for 3 hours. Group4 (Sham) normothermia without ASDH. Animals were sacrificed 24 hours post-injury and perfusion-fixed for quantitative histopathological evaluation. Brain sections were stained with Flurojade B (FJB) to estimate neuronal degeneration. Slides were counterstained with DAPI and select slides were counterstained for NeuN to confirm neuronal identity. Using the physical fractionator method (MicroBrightField Stereo Investigator software), FJB positive neurons were counted in bilateral hippocampus and subcortical areas in five sections spaced 360 μm apart in the region spanning the hematoma. Data were analyzed with two-way ANOVA followed by a post hoc Bonferroni test.
FJB positive neurons were predominant in the ipsilateral and subcortical regions under the hematoma. Compared to the normothermia group, the number of FJB positive cells in the early hypothermia group was significantly less (Normo: 904,218±303,358 vs Early: 319,180±82,369, p<0.05).
Early mild hypothermia may reduce reperfusional neuronal damage after decompression of focal mass in ASDH rat model, by reducing neurotoxic mechanisms including oxidative stress and free radical generation. These data support the concept that early induced hypothermia prior to decompression significantly decreases neuronal vulnerability in this model of severe focal brain injury.
These studies were made possible by funding from the Miami Project to Cure Paralysis and NIH NS042133.
ASDH, hypothermia, fluoro-jade, decompression, reperfusion
SUBCONVULSIVE EPILEPSY AND CHANGES IN PTZ-INDUCED SEIZURE THRESHOLD IN RATS 1 YEAR AFTER MODERATE FLUID PERCUSSION BRAIN INJURY
Justin Sick, BS, University of Miami Miller School of Medicine
Eric Bray, BS, University of Miami Miller School of Medicine
Alexandra Wick, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton Dietrich, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Thomas Sick, PhD, University of Miami Miller School of Medicine
Helen Bramlett, PhD, University of Miami Miller School of Medicine
The development of epilepsy is an important secondary consequence of traumatic brain injury.
In this study we monitored electrocorticographic (ECoG) activity and behavior simultaneously in awake freely moving rats 1 year after moderate (2.0 ATM) parasagittal fluid percussion brain injury (mTBI) or sham injury (n=5/group). Rats were implanted with two epidural electrodes near the injury (or sham injury) site 24 hr prior to video/ECoG recording. ECoG activity was recorded with a miniature instrumentation amplifier attached magnetically to the electrode assembly and tethered to a conditioning amplifier (CWE Inc). ECoG signals and video were collected and stored digitally with AD Instruments PowerLab. Power Spectral Analysis was conducted offline using EEGLab on the MatLab software platform.
None of the injured or sham-operated animals showed convulsive behavioral seizures. However, injured animals showed periodic electrographic epileptic discharges that were associated with stereotypic “freezing” behavior. The seizure discharges displayed characteristic “spike” and “dome” waveforms and power spectral analysis indicated increased power in the “Theta” and “Beta” bands. Intraperitoneal injections of a subthreshold dose (30mg/kg) of pentylenetetrazol (PTZ) produced electrographic seizure dischages in sham-operated animals and some mild myoclonic jerks, but few major convulsions. In contrast, PTZ caused both severe electrographic epileptic discharges and behavioral convulsions in mTBI animals and in one case caused respiratory failure and death.
These data indicate that epileptic activity can be detected electrographically at 1 year following moderate TBI in animals that do not show convulsive behavior prior to administration of a seizure inducing agent. Nonconvulsive seizures may therefore be a common secondary insult occurring in patients even after moderate levels of TBI.
Supported by the Veterans Administration and NIH NS030291.
TBI, epilepsy, pentylenetetrazol, electrocorticographic
IN VIVO ACUTE PATHOPHYSIOLOGY RODENT MODEL OF PRIMARY BLAST INJURY
Fang Wang, MD, University of Nebraska-Lincoln
Namas Chandra, PhD, University of Nebraska-Lincoln
The blast injury animal model remains an outstanding challenge - biomechanical loading requires controlled blast replication similar to field explosive conditions. Using advanced shock tube we evaluated injury severity caused by primary blast overpressure (BOP). Acute post-exposure pathophysiological response was monitored to relate injury severity to incident BOP intensities.
The 10-week old, male Sprague-Dawley rats (345±25 g) were anesthetized with ketamine/xylazine mixture in PBS (90.8 and 9.9 mg/kg ketamine and xylazine, respectively) via intraperitoneal injection. The anesthetized animal was placed on heating pad, the body temperature and heart rate (HR) was monitored for 30 minutes before and after the blast exposure. Rats were exposed to controlled blast using 9-inch shock tube with square cross-section developed at the UNL-Trauma Mechanics laboratory. We used five levels of blast intensity, characterized by peak overpressures of 130, 190, 230, 250, and 300 kPa. After 24 hours surviving animals were transcardially perfused with 4% paraformaldehyde in cacodylate buffer (pH=7.3), heads were decapitated and stored for 20 hours (storage buffer). Brains were extracted, sliced into 40 μm sections (MultiBrain™ technology), stained with amino-cupric silver (neurodegeneration) and against IgG (compromised blood-brain barrier (BBB) and plasma membrane permeability).
Our overarching research interest encompasses the study of blast induced neurotrauma (BINT) in the mild range. In the first step, we performed tolerance studies to establish the upper level ofinjury severity and corresponding lethality scale for the rat model. Biomechanical loading was gradually increased from 190 to 300 kPa peak overpressure (OP). The mortality among tested animals reached 45.5% (n=11) and 100% (n=5) at 250 and 300 kPa, respectively. At lower values of peak OP survival rates were high: 80% and 100% for 190 and 230 kPa groups, respectively (n=10). The extent of the lung damage was evaluated using Yelverton's pathology scoring system for blast injuries (J. Trauma, 1994, 40, S111-5). We found the value of 130 kPa peak OP is the threshold for lung injury in our model. The average score for this group (1.8±1.6) is statistically insignificant with respect to control group (p>0.05, power 0.9). At higher biomechanical loading conditions (230, 250 and 300 kPa) the average score increases to 13.8±6.7 for all three groups. The 190 kPa group has an intermediate injury score (10.7±6.4). Overall; the lung injury is rather mild (the maximum possible value is 64) and appears not correlated to the lethality. The heart rate monitoring in the acute phase post exposure revealed the heartbeat decreased in animals exposed to higher peak OP (ΔHR=62±8 bpm, four groups). At 130 kPa the pulse difference was significantly lower (ΔHR=40±10 bpm), indicating HR is a sensitive metric of the blast exposure in the acute phase. Examination of BBB damage revealed elevated levels of IgG in the brain parenchyma in rats exposed to blast at higher peak OP. The IgG penetrated inside the neuronal cells in the cortical and hippocampal regions indicating severe plasma membrane damage.
We established injury criteria for the rat model exposed to primary blast. In the first step, the peak overpressure-lethality relationship was established. At 300 kPa peak OP, the lethality was 100% and at 250 kPa peak OP it decreased to 45.5%. It suggests the severe injury is induced at 250 kPa, which was further corroborated by physiological response (HR changes) and anti-IgG immunostaining. The BBB damage had diffuse character and was widespread in the brain parenchyma. The IgG immunostaining integration results indicate the extent of the damage is proportional to the peak OP values. We detected plasma membrane disruption of neural cells caused by blast exposure at peak OP values as low as 190 kPa. The lung injury evaluated using dedicated scoring system revealed mild to moderate type of damage, and it suggests lung injury is not a significant factor for the survival of the test subjects in our model.
Financial support under the US Army Research Office project “Army-UNL Center of Trauma Mechanics” (Contract No. W911NF-08-10483) is gratefully acknowledged.
BINT, injury scale, plasmalemma permeability
IMMEDIATE BIOPHYSICAL RESPONSES IN NEURAL CELLS FOLLOWING CLOSED HEAD ROTATIONAL BRAIN INJURY IN SWINE
Kevin Browne, B.S., University of Pennsylvania
Laura Struzyna, B.S., University of Pennsylvania
John A. Wolf, Ph.D., University of Pennsylvania
Douglas Smith, M.D., The University of Pennsylvania
D. Kacy Cullen, Ph.D., University of Pennsylvania
The mechanisms of immediate ultrastructural damage in neural cells due to closed-head inertial TBI remain poorly understood, yet are believed to be key inceptive events in subsequent pathophysiology. We sought to characterize acute traumatic responses, particularly alterations in plasmalemmal permeability, in relation to predicted micro-injury biomechanics and evolving neuropathology.
We utilized a model of non-impact diffuse brain injury (DBI) in female Yorkshire swine. Animals were anesthetized and subjected to sham conditions or rapid head rotational velocity/acceleration using the Hyge pneumatic actuator. We utilized head rotational accelerations that would induce a range of injury severities, from mild to severe, in coronal or sagittal planes (rotational accelerations: 57,478–123,550 rad/s2 in coronal, 19,575–45,480rad/s2 in sagittal; angular velocities: 190–300 rad/s in coronal; 80–140 rad/s in sagittal). In a subset of animals, Lucifer Yellow (LY), a small (457 Da), aldehyde fixable, cell impermeant dye, was administered into the lateral ventricles to diffuse evenly throughout the interstitium prior to injury. Animals receiving LY injections were sacrificed within 15 minutes post-injury. Additional animals were euthanized at 8 hours or 7 days post-injury. Brains were processed for regional quantification of altered plasmalemmal permeability and neuropathological analyses.
There were clear differences in neurological recovery following injuries in the sagittal versus coronal plane (noted in animals not acutely sacrificed post-injury). Following coronal injury, animals regained balance and began to feed within hours, whereas sagittal injury generally resulted in prolonged coma. Sagittal injuries characteristically demonstrated extensive bleeds throughout the caudal brainstem, the supratentorium of the cerebellum, as well as extensive subdural hematomas (115–140 rad/s), except at the lowest injury levels (≤95 rad/s). Coronal injuries only demonstrated overt vascular compromise at the highest angular velocities (≥275 rad/s), but with few, and when present, minor hematomas. Following microscopic analysis, we observed that LY cells were present following injury in both the coronal or sagittal planes, but not following sham conditions, indicating that DBI caused by non-impact head rotational acceleration induced acute alterations in plasmalemmal permeability. In particular, heterogeneous neural cell permeability was observed in deep layers of the cerebral cortex. Extensive clusters of permeabilized cells were observed predominately at grey/white matter interfaces, regions predicted to experience maximal shear strains. In many cases, clear undulations were evident in permeabilized dendrites or axons, potentially proportional to the magnitude of the local strain field. Additionally, populations of LY axons were present throughout subcortical white matter tracts. These permeabilized axons were limited to cortical regions, and were not present within the thalamus. Moreover, the degree and location of LY cells differed based on plane of rotation; however, in both cases the extent of neuronal and axonal permeability increased with increasing angular velocity and acceleration. Further analyses are correlating this acute plasmalemmal damage with other ultrastructural changes such as cytoskeletal discontinuities as well as evolved changes including neuronal/axonal degeneration and neuro-inflammation.
Previous reports have demonstrated that impact-based DBI in rodents caused immediate neuronal somatic plasmalemmal and axolemmal injury. We advance these findings by demonstrating acute alterations in neuronal and axonal membrane permeability in a large animal model of non-impact head rotational injury. This model mimics forces associated with closed-head TBI common in sports, motor vehicle collisions, or military environments because both humans and swine possess relatively large, gyrencephalic brains with substantial white matter domains. We are currently correlating the initial extent of ultrastructural changes with predicted cell/tissue micro-strain fields to elucidate neuronal and axonal tolerances to injury, which may then be appropriately scaled to human TBI based on macro-injury biomechanics. Understanding these parameters may inform the links between the physical and physiological consequences of TBI, thus guiding the development of targeted medical therapeutics to address the predominantly afflicted cell populations based on the mechanisms of injury.
The authors thank Carla Hernandez, Anthony Parisi, Jill Ralston, and Melissa Byro for technical assistance.
membrane permeability, axonal injury, biomechanics
ENHANCED INFLAMMATORY REACTION AFTER CONTROLLED CORTICAL IMPACT INJURY IN HUMAN AMYLOID-BETA-PRODUCING MICE
Eric Abrahamson, PhD, University of Pittsburgh
Violetta Pivtoraiko, PhD, University of Pittsburgh
John Melick, MS, University of Pittsburgh
William Paljug, MS, University of Pittsburgh
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Traumatic brain injury (TBI) is a risk factor for Alzheimer's disease (AD); this may be due to production of inflammatory cytokines which are associated with pathological cascades in both TBI and AD, and increased concentration of amyloid-beta (Aß) peptides, which play a central role in the pathogenesis of AD.
The goal of the current study was to gain a better understanding of the global inflammatory response and its relation to Aß changes after controlled cortical impact (CCI) injury in mice. We performed a Luminex multiplex assay on brain tissue samples harvested from young C57 wild-type mice and genetically modified C57 mice that produce human Aß (hAß) peptide. The effect of brain injury on levels of inflammatory molecules was assayed in both groups of mice at 24 and 72 hours after CCI or sham surgery, and values were compared to naïve mice.
CCI injury resulted in an acute increase and sustained elevation of hAß peptide concentration in transgenic hAß mice. Compared to naïve wild type controls, injured wild type mice had increased levels of interleukin (IL)-6, IL-7, IL-9, eotaxin, granulocyte colony-stimulating factor (G-CSF), inflammatory protein-10 (IP-10), keratinocyte-derived chemokine (KC), leukaemia inhibitory factor (LIF), and monocyte chemotactic protein-1 (MCP-1) at 24 hours, but not 72 hours, after CCI. Additionally, hAß mice, but not wild type C57 mice, had increased levels of IL-1ß, IL-13, IL-15, monokine induced by gamma-interferon (MIG), and macrophage inflammatory protein-1a (MIP-1a), at 24 hours after injury, and sustained elevations in all measured inflammatory markers, except IL-1ß and G-CSF, at 72 hours after injury.
CCI injury results in increased production of a greater array of cytokines for longer time periods in human Aß producing transgenic mice compared to C57 wild-type (murine Aß producing) mice. Sustained elevations in human Aß peptide after CCI injury intensify and prolong inflammatory reactions which may contribute to an increased risk of developing AD in TBI survivors. These preliminary results also provide an initial framework for future studies designed to test the effect of anti-inflammatory therapeutics in a TBI model of preclinical AD.
Funded by: NIH NINDS NS30318
Alzheimer, inflammation, amyloid-beta, brain injury
THE EFFECT OF HYPOTHERMIA ON HIPPOCAMPAL CELLULAR PROLIFERATION FOLLOWING TRAUMATIC BRAIN INJURY
Dr. John Caltagarone, PhD, University of Pittsburgh
Glenn Gobbel, DVM, PhD, Vanderbilt University Medical Center
Wendy Fellows-Mayle, PhD, University of Pittsburgh Medical Center
Dr. P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Hypothermia reduces TBI-induced injury to the hippocampus, a brain region instrumental in learning and memory and particularly susceptible to damage following TBI. Because the proliferation of neural progenitor cell may contribute to post-TBI cognitive recovery, we examined the effect of hypothermia on hippocampal cellular proliferation following TBI.
Post-natal day (PND) 17 rats were injured using controlled cortical impact, treated with hypothermia (32–33 °C) or normothermia (37 °C) for 4 h post-TBI, and given bromodeoxyuridine (BrdU) to label dividing cells 1h before euthanasia at post-injury day (PID) 1, 7, 14, or 28. The brains were perfused, paraffin-embedded, and sectioned coronally at 5 μ thickness. Immunohistochemical staining for BrdU was performed, and numbers of positively-stained cells within the subgranular zone (SGZ) of the dentate gyrus (ipsilateral and contralateral to injury) was assessed by light microscopy. ANOVAs and post-hoc Tukey HSD tests were performed to examine injury (injury, sham), PID (1, 7, 14, 28), treatment (hypo, norm), and hemisphere (ipsilateral, contralateral) effects.
There were greater numbers of BrdU-positive cells in the SGZ for injured animals and this increased cellular proliferation was greatest at earlier PIDs (ps<0.01). Although there was no overall effect of hypothermia (p=0.09), hypothermia significantly reduced the cellular proliferation observed in injured animals at PID1 (p=0.04) such that it did not differ from sham animals at this timepoint (ps>0.11).
There was increased hippocampal cellular proliferation in immature animals post-TBI. However, hypothermia attenuated this proliferation to levels similar to those observed in sham injured animals early after injury. Considering that changes in neurogenesis and gliogenesis influence cognitive outcome post-TBI, hypothermia may ameliorate cognitive dysfunction by influencing post-TBI patterns of cellular proliferation.
None
TBI, hypothermia, hippocampus, neurogenesis, rodents
ESTRADIOL EFFECTS ON OXIDATIVE STRESS RESPONSE GENES IN THE CEREBRAL VASCULATURE AFTER TRAUMATIC BRAIN INJURY
Debbie Boone, BS, University of Texas Medical Branch
Donald S. Prough, MD, UTMB
Douglas S. Dewitt, PhD, UTMB
Helen L. Hellmich, PhD, University of Texas Medical Branch
Since no treatment for early-stage TBI exists and oxidative stress contributes to the deleterious physiological cascade after injury, this experiment investigated the effects of acute treatment with 17β-estradiol (E2), early after injury, on the expression of genes associated with oxidative stress in the hippocampus and cerebral vasculature.
Under isoflurane anesthesia, ovariectomized female rats (n=3/group, total 12 rats) received moderate fluid-percussion injury (FPI; 2 atm) or sham injury followed by treatment with E2 (33 μg/kg, s.c.) or vehicle (0.1 ml/kg sesame oil) 15 minutes post-injury. Brains were harvested 15 minutes after injection, fresh frozen and stored at −80°C. Total RNA was isolated from the hippocampus and cerebral blood vessels and reverse transcribed. Real-Time PCR was performed on an MX3000P thermal cycler (Stratagene, La Jolla, CA) using the RT2Profiler PCR Array of rat oxidative stress and antioxidant defense genes (SABiosciences, Frederick, MD). Eighty-four genes related to oxidative stress were measured, including peroxidases, peroxiredoxins, oxidative stress responsive genes such as superoxide dismutases, genes involved in reactive oxygen species metabolism and oxygen transporter genes.
Within 15 min of treatment, E2 had a significant effect on oxidative stress response gene expression in both the cerebral vasculature and hippocampus. In particular, data from 38 oxidative stress responsive genes suggests that treatment with E2 mitigated the gene expression changes induced by TBI in cerebral vascular and hippocampal tissue. For example, mRNA that had increased expression after TBI showed decreased expression when treated with E2 and vice versa. Examples of genes significantly affected in cerebral blood vessels include: Uncoupling protein 3 (Ucp3), Glutathione peroxidase 2 (Gpx2), and Thioredoxin reductase 1 (Txnrd1). Examples of genes significantly affected in the hippocampus include: Peroxiredoxin 3 (Prdx3), Glutathione peroxidase 3 (Gpx3), and Catalase (Cat).
The neuroprotective effects of E2 are yet to be utilized clinically, possibly because little is known about the genomic mechanisms of E2 that mediate its pro-survival effects on injured neuronal tissue and cerebral vasculature. This study contributes to understanding the genomic effects of E2 after brain injury and is a first step in developing a therapeutic strategy in which modulation of the expression of particular genes could lead to improved functional outcomes similar to or better than those induced by E2 with improved side-effect profiles.
John Sealy Memorial Endowment Fund (SLS), NS19355 (NIH/NINDS-DSD), Moody Center for Traumatic Brain & Spinal Cord Injury Research/Mission Connect.
gene expression, estradiol, vasculature, hippocampus
POST-TBI MINOCYCLINE TREATMENT INHIBITS INFLAMMATORY CELL RESPONSE AND ENHANCES HIPPOCAMPAL NEUROGENESIS IN BOTH YOUNG AND AGED BRAINS
Ashley Harvin, B.S., Virginia Commonwealth University
Andrew Rolfe, B.S., Virginia Commonwealth University
Christopher Shin, M.Sc., Virginia Commonwealth University
Following traumatic brain injury (TBI), there is an enhanced cell proliferative and neurogenic response in the hippocampus of young adult animals. This endogenous cellular response may be associated with the innate cognitive recovery observed following TBI in this age group. Compare with younger adults, in the aged brain an increased level of inflammatory cell response was observed following injury concomitant with a decreased hippocampal neurogenesis and cognitive recovery in this aging population. This suggests that excessive inflammation produced in the injured aging brain have a detrimental effect on neurogenesis and cognitive function. In this study, we examined the effect of anti-inflammation treatment with minocycline on cell proliferation and generation of new neurons in the dentate gyrus (DG) of the hippocampus in both young and aged rats following TBI.
Fisher 344 rats aged at 3 months and 20 months were subjected to a moderate lateral fluid percussion injury (LFPI). Minocycline was administered intraperitoneally at the dose of 45mg/kg either at 30 minutes or 4 hours post-injury and followed by the second injection at 8 hours after-injury. Thereafter reduced dose at 22.5mg/kg was given twice daily for 2 more days. BrdU was injected at 2 days post-injury to label proliferating cells. To examine the effect of minocycline on cell proliferation, generation of new neurons and inflammatory cell response at the acute stage post-injury, one group of rats was perfused at 3 days post-injury. To study the effect of minocycline on cell survival, differentiation and chronic inflammatory cell response, another group of animals was perfused at 28 days post-injury. Brain sections were immunostained for BrdU, early neuronal marker doublecortin (DCX), and other cell-type specific markers for mature neurons, microglia and astrocytes.
The results show that short-term anti-inflammatory treatment with minocycline reduces the cell proliferative response, presumably inflammatory cell responses, in young and aged rats following LFPI and CCI injury, and enhances generation of new neurons in the hippocampus in both young and aged rats following LFPI and in aged rats following CCI injury. Further studies examining minicycline on cell differentiation, long term survival of the newly generated neurons and chronic inflammatory cell response are ongoing.
Our data suggests that injury-induced inflammation, particularly microglial activation, is detrimental for neurogenesis. Therapies that control microglial activation may have potential to enhance hippocampal neurogenesis and improve cognitive recovery following TBI.
Supported by NIH/NINDS RO1-NS055086 (Sun).
minocycline, inflammation, hippocampal neurogenesis
CYCLOPHILIN D DELETION ATTENUATES AXONAL INJURY FOLLOWING DIFFUSE BRAIN INJURY
Melissa J. McGinn, PhD, Anatomy & Neurobiology/Virginia Commonwealth University SOM
Jesse A. Simms, BA, Anatomy and Neurobiology/Virgnia Commonwealth University SOM
John T. Povlishock, PhD, Anatomy & Neurobiology, Virginia Commonwealth University
Anders G. Haanell, Anatomy & Neurobiology/ Virginia Commonwealth University SOM
Mitochondrial permeability transition pore (mPTP) formation has been suggested to play a central role in the ionic and metabolic disturbances of traumatic axonal injury (TAI). Pharmacological inhibition of mPTP formation (cyclosporin A, CsA) is confounded by the modulation of several other biochemical pathways, which may contribute to axon protection.
To circumvent the issues associated with pharmacological inhibition of mPTP formation, genetic deletion of cyclophilin D (CypD), a key regulator of mPTP formation, was utilized to assess the definitive role for mPTP formation in TAI. Wild type (WT) and Ppif−/− mice, lacking the gene encoding for CypD, were subjected to central fluid percussion injury (cFPI). At various times following cFPI animals were perfused and their brains sectioned and processed for immunohistochemistry with antibodies targeting β-amyloid precursor protein (APP), a known marker of TAI following diffuse brain injury. The number of APP+ swellings was quantitatively assessed in 2 neocortical regions, a rostral region containing predominantly pyramidal neurons of the primary somatosensory and the caudal region containing pyramidal neurons of the primary visual and auditory cortices.
Brain sections from cFPI injured WT and ppif−/− animals showed no evidence of contusion or cavity formation. Intraparenchymal hemorrhage was absent, except for isolated petechial hemorrhages within the corpus callosum. Limited subarachnoid hemorrhage incident to the injury site was noted in both groups. In the mediodorsal neocortex, dorsolateral thalamus, hippocampal dentate gyrus and stratum lacunosum moleculare and the subcortical white matter, diffuse APP+ axonal swellings were observed in both WT and ppif−/− mice sustaining cFPI, while only sparse APP immunoreactive swellings were observed in sham-injured animals. Qualitatively, the number of APP+ axonal swellings within these regions from mice lacking CypD appeared dramatically reduced in comparison to WT controls. Quantitatively, the number of APP+ swellings in the rostral neocortex within WT mice equaled 32.2±2.2 mm−2, while within the same region in ppif−/− mice the average number of APP+ swellings was 14.3±2.4 mm−2; representing a 55.4% reduction in the number of damaged axons (p<0.003). A similar result was found within the more caudal neocortex. Here the number of APP+ swellings was reduced from 38.8±2.2 mm−2 in WT mice to 18.2±1.7 mm−2 in ppif−/− mice, representing a similar degree of axon protection with a 53.0% reduction relative to WT levels (p<0.0001).
The current study provides compelling evidence for CypD-dependent mPTP formation in the pathological cascade responsible for secondary injury following TAI. CypD deletion significantly reduces the burden of TAI following mTBI. These findings are consistent with the widely accepted role of CypD in regulating mPTP formation and are consistent with the neuroprotective and axon protective effects of CypD deletion observed in models of Alzheimer's disease, focal cerebral ischemia, amyotrophic lateral sclerosis and multiple sclerosis and supports previous work from our laboratory and others utilizing pharmacological inhibition of mPTP formation to attenuate the secondary cascades resulting in axonal injury following diffuse brain injury. The reduction in the number of APP+ swellings, while not complete, matches the level of axon protection provided by pharmacological inhibition of mPTP formation (CsA administration).This suggests the primary mechanism for the axon protective effects of CsA is through this inhibition and not by way of immune modulation or calcineurin inhibition.
This study was supported by NS077675 ppif−/− mice from Dr. Michael Forte OHSU.
Axons, mitochondria, neuroprotection, trauma, cyclophilin
TEMPORAL ALTERATIONS IN HIF1-ALPHA AND AQUAPORIN EXPRESSION FOLLOWING PBBI
Katie Phillips, Walter Reed Army Institute of Research
Valeta Sanders, Walter Reed Army Institute of Research
Kara Schmid, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Jitendra R. Dave, Ph.D., Walter Reed Army Institute of Research
Hif1α is a transcription factor that regulates aqp4, the main aquaporin in the brain and a bidirectional regulator of water flow over the blood brain barrier, important in development and resolution of edema. Hif1α also regulates aqp9, which controls bidirectional flow of water and many other non-charged solutes.
Here we investigated hif1α, aqp4, and aqp9 mRNA levels after penetrating ballistic-like brain injury (PBBI) over time. PBBI is model of severe traumatic brain injury, with primary injury affecting ten percent of the total brain volume. The unilateral frontal injury was induced by stereotactic insertion of a custom probe through the right frontal cortex. A temporary cavity was formed by the rapid inflation/deflation (i.e. <40 msec) of an elastic balloon attached to the end of the probe. Total RNA was isolated from a 100 g coronal slice of the ipsilateral hemisphere from animals at 4, 12, 24 hours and 3 and 7 days post-injury. cDNA was synthesized using oligodT primers and real time PCR was performed using Taqman expression assays. Relative quantities of mRNA were calculated in comparison to shams.
The levels of hif1α increased in the initial stages following PBBI (starting by 12 hours, significant at 24 hours, p<0.05) but were significantly suppressed at the later time points (3 days, p<0.01; 7 days, p<0.05, 2 factor ANOVA). However, there were no significant changes in mRNA levels of aqp4, although there was a trend for levels to be lower than sham. In addition, mRNA levels of aqp9 were significantly suppressed at 12 and 24 hours post injury (p<0.01, 2 factor ANOVA).
Previous studies of PBBI indicate edema significantly increases within 4 hrs, and continues through 7 days, peaking at 3 days. Increased ICP levels peak at 24 hrs but are upregulated through 3 days after injury. As would be expected, hif1α mRNA levels also increase after TBI with a reversal to suppressed levels at 3 days. Interestingly, increases in hif1α did not correlate with increases in aqp4 or aqp9 although this transcription factor is reported to increase expression of these genes. Aqp4 is expressed throughout the brain but its expression in astrocytes in the endfeet borders between the CNS and the periphery is critical for the regulation of edema. It may be in these specific cells that regulation of aquaporins is relevant to the observed changes in edema after PBBI. Further studies investigating changes in cell type specific protein expression of these genes, focusing particularly on astrocytic expression, are planned.
This research was funded by the Army Combat Casualty Care Research Program.
TBI, PBBI, Hypoxia, Edema
WRAIR PROJECTILE CONCUSSIVE IMPACT (PCI) MODEL: INJURY DEVICE AND HELMET ADVANCED DEVELOPMENT
Zachary Larimore, BS, U.S. Army Research Laboratory
Larry Holmes, MS, U.S. Army Research Laboratory
Shawn McLoughlin, Walter Reed Army Institute of Research
Andrea Mountney, Ph.D., Walter Reed Army Institute of Research
Kara Schmid, PhD, Walter Reed Army Institute of Research
Deborah A Shear, Ph.D., Walter Reed Army Institute of Research
Frank Tortella, Ph.D., Brain Trauma Neuroprotection & Neurorestoration Branch, Center of Excellence for Psychiatry & Neuroscience, WRAIR
The WRAIR-PCI model was established for studying concussion. The first generation PCI device used dry ice sublimation to trigger the targeted release of a small projectile and a prototype stainless steel “rat” helmet to produce mildTBI. Here, we describe recent advancements made to the PCI device and helmet.
In the modified PCI device, compressed CO2 was substituted for dry ice sublimation to serve as a trigger mechanism for the projectile. The velocity of the projectile was estimated based on high speed videos. Impact force and pressures were measured at different CO2 input pressures. Helmets were fabricated using a 3-D mold of an adult Sprague-Dawley rat. Three types of helmets were constructed of (A) woven glass/carbon, (B) woven glass, (C) woven carbon/epoxy. Strength and modulus of these composites were determined by tensile testing. To evaluate the helmet performance, rats were assigned into four groups (n=5/group): Sham control(received anesthesia only), helmet A, helmet B and helmet C. The helmet groups were subjected to PCI once daily for consecutive five days. Sensor films were used to measure pressure distribution and magnitude on the outer/inner surface of the helmets. Gross pathology and histopathology were assessed at 24 hours post-injury.
The modifications made to the PCI device produced a range of projectile velocity and impact force dependent on the CO2 input pressure. Specifically, an input pressure of 56 psi produced a velocity=39.93±1.38m/s and impact force=171.84±15.49N whereas an input pressure of 25 psi produced a velocity=22.34±0.08m/s and impact force=55.04±7.85N. In contrast, the average magnitude and duration of both reflected (2.95±0.19kPa, 7.12±0.55s) and side-on pressures (2.50±0.30kPa, 12.89±1.52s) did not vary with CO2 input pressure. The average pressure applied to the helmet's outer surface by the projectile was 5328±89kPa. All helmets effectively protected against skull fracture, subarachnoid hemorrhage and contusion. Pressure data of the inner surface and material testing demonstrated that helmet A (473±6kPa) yielded the most consistent pressure distribution. In contrast, the other 2 helmets did not yield a satisfactory pressure distribution on the inner surface indicating these helmets may not effectively transfer the load to the rat head.
The primary advantage of the modified PCI device is that the impact force and velocity are directly proportional to the input pressure, which facilitates greater manipulation of the injury parameters. In addition, the pressure wave generated from the release of compressedCO2 is of low magnitude and is not related to the input pressure. Thus, any “pressure wave” effect is minimal and can be effectively controlled. Helmet A (constructed of woven glass/carbon material) showed the most consistent load transfer efficiency owing to its fiber orientation and layer arrangement. The good load transfer is critical for producing concussions in the absence of any overt pathology. Overall, these advancements greatly improve our concussive model such that the mechanical force used to induce the injury is more controllable, reproducible and quantifiable. Moreover, the intensity of the force can be titrated, potentially producing a wide spectrum of concussive injury severities for further study.
This work is supported by CCCRP. LYL is sponsored by National Research Council Research Associateship Program. Special thanks to DMAV(WRAIR) for taking high-speed video.
mTBI, Concussion, Injury Device, Helmet
OPERATION BRAIN TRAUMA THERAPY CONSORTIUM: DOSE-RESPONSE EVALUATION OF NICOTINAMIDE IN THE WRAIR MODEL OF PENETRATING BALLISTIC-LIKE BRAIN INJURY
Rebecca Pedersen, Walter Reed Army Institutes fof Research
Justin Sun, B.S., Walter Reed Army Institute of Research
Melissa Long, B.S., Walter Reed Army Institute of Research
Kara Schmid, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Operation Brain Trauma Therapy (OBTT) is a multi-center consortium evaluating promising therapies across traumatic brain injury (TBI) models.The WRAIR penetrating ballistic-like brain injury (PBBI) model has been well-characterized and produces reliable and reproducible neurobehavioral and neuropathological profiles. This study assessed the therapeutic efficacy of nicotinamide in the PBBI model.
Unilateral frontal PBBI was produced in the right hemisphere of isoflurane anesthetized rats (10% injury severity level).Rats were randomized into 4 groups: PBBI + vehicle (PBBI; n=14), PBBI + 50 mg/kg nicotinamide (NIC_50 mg/kg; n=15), PBBI + 500 mg/kg nicotinamide (NIC_500mg/kg; n=16) and sham (n=9).Nicotinamide was delivered intravenously (IV) at 15m and 24h after injury. Blood samples (0.7 mL whole blood) were obtained via the IV catheter at 4h, 24h, and at sacrifice from each rat to measure serum levels of the neuronal biomarker UCHL-1 and the glial biomarker GFAP.Motor function was evaluated on the rotarod (7 and 10 days post) at fixed-speed increments of 10, 15, and 20 rpm.Cognitive performance (acquisition and probe trial) was evaluated on the Morris water maze (MWM) days 13–17 (4 trials/day).Histopathological analysis included lesion volume and total hemispheric tissue loss; both expressed as a % of the contralateral hemisphere.
Motor testing revealed significant deficits in all injury groups with mean rotarod latencies reduced by 60±8% (PBBI), 60±8% (NIC_50mg/kg), and 48±9% (NIC_500mg/kg) vs. sham (p<.05).PBBI rats treated with the high dose (500 mg/kg) of nicotinamide showed modest, albeit not significant, improvement on the rotarod task (p>.05).MWM results revealed significant deficits in all injury groups with the average latency to find the hidden platform (across all testing days) increased by 121±13% (PBBI), 115±12% (NIC_50mg/kg), and 132±13% (NIC_500mg/kg) vs. sham (p<.05).No significant therapeutic effect was detected in nicotinamide-treated rats on MWM parameters.Nicotinamide did not affect PBBI-induced lesion volume measured at 22 days post-injury (% contralateral hemisphere): PBBI=15±2%, NIC_50mg/kg=15±2% and NIC_500 mg/kg=13±1%.Likewise, nicotinamide treatments failed to reduce PBBI-induced hemispheric volume loss: PBBI=25±3%, NIC_50mg/kg=24±4% and NIC_500 mg/kg=24±1%.The high dose nicotinamide (500 mg/kg) significantly reduced post-injury (24h) increases in GFAP serum biomarker levels (p<0.05), but not UCHL-1 levels.
The acute (24h) beneficial effects of nicotinamide observed on PBBI biomarker (GFAP) data are supportive of research demonstrating neuroprotective efficacy on acute histopathology in other TBI animal models. However, in the absence of any corresponding neuroprotective profiles on the neurobehavioral and neuropathological outcome metrics, nicotinamide is not likely to be selected as a lead candidate for advanced OBTT studies in the PBBI model.
This research was funded by the Army Combat Casualty Care Research Program and by grant W81XWH-10-1-0623 from the United States Army.
PBBI, TBI, nicotinamide, OBTT, WRAIR
ALTERATIONS OF CEREBRAL BLOOD FLOW AND METABOLISM IN CHILDREN WITH TRAUMATIC BRAIN INJURY
Robert C. McKinstry III., MD, PhD, Washington University School of Medicine
Tammie LS. Benzinger, MD, PhD, Washington University School of Medicine
Jeffrey R. Leonard, MD, Washington University School of Medicine
Jose A. Pineda, MD, MSCI, Washington University School of Medicine
TBI causes the release of mitotoxic products. The resulting mitochondrial impairment can decrease cerebral oxygen metabolism and cause decoupling of oxygen extraction and cerebral blood flow (CBF). Using emerging magnetic resonance imaging methods we measured the whole-brain Oxygen Metabolism Index (OMI) and CBF in pediatric TBI patients.
Pediatric patients 0–17 years of age with severe TBI (n=17) were included in the study and received MRIs within the first 3 weeks after injury. Ten healthy patients served as controls. Studies were performed on a Siemens Trio 3.0 T MRI scanner. Oxygen extraction fraction (OEF) was determined by measuring the deoxyhemoglobin content in the superior sagittal sinus. CBF data was acquired using a pseudocontinous arterial spin labeling sequence. The OMI was calculated using the formula: OMI=CBF×OEF. Whole-brain CBF and OMI were compared between patients and controls using t-tests. The correlations between CBF and OEF and between CBF and OMI were used to assess the presence of hyperemia, hypoemia, and decoupling of OMI and CBF in children with TBI. Measurements were repeated three-months post-injury.
The behavior of CBF after TBI is time-dependent. No overall decrease in average CBF was observed in the patients scanned during the first week after TBI. Patients scanned in the second and third week post-injury displayed a significant decrease in CBF compared to controls (29 vs 40 mL/min/100g; p=0.01). At the three-month time point, CBF values in the TBI patients remained consistently depressed compared to controls (25 vs 40 mL/min/100g; p<0.0001). OMI was significantly decreased in TBI patients at both time points (7.6 vs 19 at 2 weeks, 8.3 vs 18 at three months; p<0.0001 for both). OMI was also decreased during the first week of recovery. OEF and CBF, had a weak correlation, supporting the uncoupling of oxygen delivery and consumption. For the subacute data, the Spearman correlation coefficient was −0.15 with a 95% interval of (−0.80, 0.62). At three months it was −0.02 (−0.67, 0.75). A scatter plot of OMI and TBI was constructed to display the relationship between oxygen consumption and CBF. TBI and control subjects were clearly separated by OMI, even though perfusion values overlapped. Even patients with CBF in the upper end of the normal rage demonstrated low cerebral metabolism.
We found that CBF displays a heterogeneous time course after TBI, which is consistent with previous reports. By 7 days post-injury a significant perfusion decrease is consistently observed, and this abnormality persisted three-months post-injury. This is consistent with adult and animal studies of CBF. OMI was consistently decreased despite the presence of normal or high CBF. This indicates that perfusion increases are incapable of producing normal metabolic levels. The absence of raised OEF indicates that there was no overall ischemia, although we cannot exclude small ischemic regions. This is reinforced by the lack of correlation between CBF and OEF, indicating that neither is significantly changing in response to the other. Our study indicates that presence of metabolic uncoupling in pediatric TBI patients, even several months post-injury.
Financial support was received from the Robert Wood Johnson Foundation, the American Heart Association and from the Washington University Institute of Clinical and Translational Sciences.
cerebral blood flow
FEASIBILITY STUDY OF AMPAR PEPTIDE ASSAY TO IMPROVE DIAGNOSTIC CERTAINTY OF CONCUSSIONS
Sara Gill, MS, ATC, LAT, KSU Club Sports Owls Nest, Kennesaw State University
Laura St. Onge, BA, KSU Sports and Recreation Park/Owls Nest, Kennesaw State University
Svetlana A. Dambinova, DSc, PhD, WellStar College of Health & Human Services, Kennesaw State University
Richard Sowell, PhD, RN, FAAN, WellStar College of Health & Human Services, Kennesaw State University
The diagnosis of concussion is complicated because many primary impacts go unidentified or are not reported, especially when they are sports-related and there is often no loss of consciousness. In this preliminary study, we present data concerning AMPAR peptide assessment in club sport athletes with semi-acute and chronic concussions.
The study protocol was reviewed and approved by the KSU Institutional Review Board. A total 84 club sport athletes (56 male, 28 female; ages 20.8±1.8 years) and 40 volunteers (nonathletes, 21 male, 19 female; ages 22.0±4.1 years) were enrolled in the study between September and October 2011. Participating athletes represented the majority of the KSU Sports & Recreation Department, where rugby, soccer, lacrosse, and cheerleading occupied about 75% of the athletic curricula. All study participants had baseline ImPACT testing performed on enrollment.
AMPAR peptide values were assessed using samples from healthy population (n=40) and reference values of 0.1–0.3 ng/ml for AMPAR peptide assay have been established. A pilot study concerning AMPAR peptide assay values in club athletes (n=84) with semi-acute and chronic concussions in conjunction with ImPACT testing were assessed. The biomarker assay results correlated with ImPACT data differentiating semi-acute concussions. Three athletes with neurological symptoms who suffered from multiple concussions and drastically high levels of AMPAR peptide have been assessed using MRI.
The neurotoxicity biomarker assays allowed differentiate healthy non-TBI volunteers from athletes with history of concussions. AMPAR peptide values showed tendency to correlate with ImPACT data. More data will be needed to proof the ability of neurotoxicity biomarkers to speed up assessment of mild TBI and assist in determining when concussed athletes should return to play.
The study was supported in part by US Army Medical Research and Materiel Command grant # W81XWH-11-2-0081.
atheletes, concussions, AMPAR-peptide, ImPACT, MRI
REPEATED POST-INJURY DOSING WITH AMANTADINE IMPROVES COGNITIVE OUTCOME AFTER FLUID PERCUSSION TBI IN RATS
Tao Wang, M.D., U.C. Davis
Ken Van, M.S., U.C. Davis
Gregory Went, Ph.D., Adamas Pharmaceuticals
Jack Nguyen, Ph.D., Adamas Pharmaceuticals
Dopaminergic pathway dysfunction may contribute to the cognitive sequelae of TBI. Amantadine can enhance arousal and cognition after human TBI, while only modest cognitive improvement is reported with low dose amantadine treatment after experimental TBI. The effect of clinically relevant doses of amantadine is evaluated in experimental TBI.
Forty-two male Sprague-Dawley rats (∼325 grams) were subjected to moderate lateral fluid percussion injury (2.12–2.15 atm) and randomly assigned to treatment groups of saline (n=14) or amantadine (15 (n=9), 45 (n=12), or 135 (n=7) mg/kg/day i.p.). Drug administration occurred every 8 hours in equal doses. Based on oral pharmacokinetic data, doses of 45 and 145 mg/kg/day are projected to be comparable to the low and high doses used for the treatment of Parkinson's disease in humans. Drug administration was initiated one hour post-TBI, and occurred on consecutive days for 16 days post injury. Spatial learning/memory performance was evaluated with the Morris water maze task on days 12–16 post-TBI with probe trial and visible platform trials on day 16. Rats were transcardially perfused on day 16, brains were coronally sectioned at 40um, stained with cresyl violet, and CA3 pyramidal neurons quantified using unbiased stereology.
The primary endpoint was overall performance (mean latency to find the hidden platform on days 12–16), and the secondary endpoints were terminal performance (mean latency to platform on days 15–16), the slope of the learning curve on days 12–14, and the number of surviving neurons in each group.
For the primary endpoint, treatment with amantadine (AMT) at 45 mg/kg/day or higher resulted in improved overall performance in the MWM compared to the saline control group (p=0.02 and 0.06 for AMT at 45 mg/kg/day and 135 mg/kg/day, respectively). Similar results were obtained for the terminal performance, where improved performance was obtained with AMT treatment at 45 mg/kg/day or higher (p=0.07 and 0.04 for AMT at 45 mg/kg/day and 135 mg/kg/day, respectively).
The rate at which information was acquired was evaluated by determining the slope of the latency curve on days 12–14, which represents the linear phase of the curve. Treatment with AMT at 135 mg/kg/day resulted in a faster reduction in latency on consecutive days relative to control (p=0.02), indicative of faster learning.
To determine the effects of AMT on neuroprotection, the number of CA3 pyramidal neurons was determined. CA3 neuronal cell counts for the saline, 45 mg/kg/day and 135 mg/kg/day groups were 79,738+/− 1295, 83,120+/− 1307, and 85,407+/− 1747, respectively. Differences in number of surviving neurons between the 135 mg/kg/day amantadine group and the saline group were significant (p=0.02).
AMT at 15 mg/kg/day was not effective relative to saline by any measure. Pharmacokinetic studies with AMT from intraperitoneal administration are ongoing to confirm drug levels in plasma.
Amantadine is the only drug with clinical evidence demonstrating a benefit for the treatment of TBI (Giacino et al. NEJM, 2012, 366:819). Here, we show that amantadine produced a dose-responsive improvement in cognitive performance after moderate TBI in rats. Cognitive performance was significantly improved with the higher doses (45 and 135 mg/kg/day) while the highest dose also resulted in significantly more surviving hippocampal neurons. A previous study (Dixon et al., Restor Neurol Neurosci. 1999;14:285) reported only modest attenuation of water maze performance deficits and no effects on CA3 cell death with a single 10 mg/kg daily dosing of amantadine post-TBI. The present study employed doses and dosing intervals projected to produce plasma exposures of amantadine similar to those in humans. These results demonstrate the therapeutic potential of amantadine for the post-injury treatment of TBI, providing a rationale for further evaluation of amantadine for the treatment of TBI.
Supported in part by Adamas Pharmaceuticals (Emeryville, CA).
Dopamine, agonist, cognition, cell death
MICROGLIAL ACTIVATION IN A RAT MODEL OF PEDIATRIC TRAUMATIC BRAIN INJURY
Manda Saraswati, PhD, Johns Hopkins School of Medicine
Bindu Balakrishnan, PhD, Johsn Hopkins School of Medicine
Sujatha Kannan, MD, Johns Hopkins School of Medicine
TBI triggers neuroinflammation, mediated by activated microglia. In the developing brain, microglia are present in white matter, with a role in remodeling. The temporospatial pattern of microglial activation after pediatric TBI is not known. We hypothesized that pediatric TBI would increase microglial activation, with postnatal age influencing their regional expression.
Immature rats at 2 different developmental ages (postnatal day, PND 10–11 and 17–18) underwent controlled cortical impact (CCI) to the left parietal cortex. Sham rats underwent craniotomy only. At 3 and 7 days after CCI, rats were sacrificed and brains were prepared for immunohistochemical analysis. Microglial localization and morphology were assessed using Iba-1 staining.
Compared to sham rats, TBI rats had increased numbers of microglia present in the pericontusional cortex, hippocampus, thalamus and the white matter tracts of the corona radiata. This was seen 3 days after CCI, and persisted to 7 days in some brain regions. In addition, microglia exhibited a more amoeboid morphology after TBI, with short, stubby branches indicating an activated state. Younger rats (PND 10–11) with TBI had a greater expression of activated microglia than older rats (PND 17–18). Furthermore, the spatial distribution of microglial staining was age-dependent, with more pronounced staining in the white matter tracts of younger rats.
Microglial activation occurs early (3d) after TBI in the immature rat and persists for at least 7d in some brain regions. Younger developmental ages (PND 10–11) were associated with greater microglial activation, and more persistent microglial presence in the white matter tracts. This may lead to a greater propensity for white matter injury and hypomyelination at the youngest ages. A better understanding of the role of microglial activation and its regulation at different stages of brain development will help in determining the optimal time for targeting therapies for attenuation of secondary inflammatory injury while promoting repair and regeneration after pediatric TBI.
Department of Anesthesiology and Critical Care Medicine, Johns Hopkins SOM
pediatric, neuroinflammation, microglia, development
IL-1 RECEPTOR ANTAGONIST ATTENUATES MECHANICAL ALLODYNIA, THERMAL AND PRESSURE HYPERALGESIA, CENTRAL SENSITIZATION AND BOTH MICROGLIAL AND ASTROCYTIC ACTIVATION AFTER CHRONIC SPINAL CORD INJURY (SCI) IN RATS
Guo Ying Xu, CMD, University of Texas Medical Branch at Galveston
Young Sob Gwak, PhD., University of Texas Medical Branch at Galveston
Kathia Johnson, B.S., University of Texas Medical Branch at Galveston
Olivera Nesic-Taylor, PhD., University of Texas Medical Branch at Galveston
Regino Perez-Polo, PhD., University of Texas Medical Branch at Galveston
Current treatment for SCI includes early intervention with steroid and non-steroidal therapies, both controversial. Few interventions are available for chronic SCI treatment. One promising intervention is to target specific inflammatory pathways, since inflammation is significantly increased and persists after CNS injury.
We tested whether intrathecal treatment (750 ng/ml, 1 μl/hr delivery rate, Alzet pump 1003D) for 3 days or 7 days, 40 days after SCI, with an FDA approved agent used in rheumatoid arthritis, IL-1 Receptor Antagonist (IL-1ra, Kineret®, would attenuate mechanical allodynia and central sensitization of dorsal horn neurons; and if mechanisms of action included inhibition of microglial and/or astrocytic activation. SCI is modeled in our laboratory by using a contusion injury (IH Impactor: 150 kdynes, 1 sec dwell) which results in consistent development of mechanical allodynia (non-noxious stimuli becomes noxious) that persists the life of the animal. Male Sprague-Dawley rats (225–240 g) were injured by contusion at spinal segment T10. At least 40 days after SCI, intrathecal deliver of either IL-1ra or vehicle treatment commenced for 3 or 7 days. Behavioral, immunocytochemical and electrophysiological studies were performed. IL-1ra treated were compared to vehicle treated, SCI untreated groups and sham.
We report statistically significant reductions in mechanical allodynia, dorsal horn neuronal hyperexcitability to tactile stimuli (central sensitization), decreased GFAP and decreased OX-42.
Thus, pro-inflammatory receptor antagonists could be useful therapeutic agents for chronic central neuropathic pain syndromes such as that after SCI.
Supported by NIH grants NS39161, NS 11255, Mission Connect of TIRR Foundation, Mr. Liddell and the West and Dunn Foundations.
spinal cord, contusion injury, pain
APPLIED ELECTRIC FIELD DIFFERENTIALLY GUIDES HUMAN ASTROCYTES AND NEURAL STEM CELLS MIGRATION IN A SCRATCH WOUND MODEL
Jing Liu, PhD, Institute for Regenerative Cures, University of California Davis; Stem Cell Program, Department of Internal Medicine, University of California Davis
Natalie Grace, Institute for Regenerative Cures, University of California Davis; Stem Cell Program, Department of Internal Medicine, University of California Davis
Michelle So, Institute for Regenerative Cures, University of California Davis; Department of Dermatology, University of California Davis
Dr. Jan Nolta, PhD, Institute for Regenerative Cures, University of California Davis; Stem Cell Program, Department of Internal Medicine, University of California Davis
Min Zhao, MD, PhD, Institute for Regenerative Cures, University of California Davis; Department of Dermatology, University of California Davis
Applied electric field (EF) demonstrates robust guiding effects on neural stem cells while not on astrocyte. However, this may be different after injury. It is therefore critical to investigate the EF guiding effects on co-cultured cells in a pathological process before any potential in vivoapplications in wound healing.
Human neural stem cells (hNSCs) were derived from H9 embryonic stem cells and then stained with fluorescent probe (dye). With a fluorescent time-lapse imaging system, the migration of these stained hNSCs was investigated when direct current EFs applied (field strength: 0, 50, 100, 300mV/mm). The migration of fetal astrocytes with or without a scratch wound was also characterized when exposed to EFs. Additionally, fetal astrocytes and the stained hNSCs were co-cultured and scratch wounds were made 24h post cells seeding. With EF application, both the migrations of two types of cells were examined.
Similar to the wide type hNSCs, stained hNSCs with fluorescent signal showed directional migration toward the cathode in EFs. Human fetal astrocytes reoriented their processes, aligned perpendicular to the voltage gradient while no obvious directional migration was found. When a scratch wound was made, however, applied EFs guided astrocytes migrate directly toward the anode. In the co-culture wound model, preliminary results revealed that fluorescence traced hNSCs migrated to the cathode while fetal astrocytes migrated to the anode simultaneously in an applied EF.
We conclude that scratch wound enables the fetal astrocyte to response to an applied EF with directional migration. Applied EF differentially guides the migration of human astrocytes (to the anode) and hNSCs (to the cathode) in a scratch wound. Therefore, applying EF could be a novel and effective cue to guide hNSCs to the injury sites to facilitate neural repair while drive astrocytes away to prevent a scar formation. The differential guiding effects essentially demonstrate an integrated beneficial effect on the wound healing in central nervous system.
Supported by: California Institute of Regenerative Medicine (RB1-01417 to M.Z, TR1-01257 to J.N.).
Electric Field; Human Neural Stem Cells; Astrocyte; Directional Cell Migration; Scratch Wound_Model
GENIPIN PROTECTS AGAINST GLUTAMATE-MEDIATED EXCITOTOXICITY IN ORGANOTYPIC HIPPOCAMPAL SLICE CULTURES
Ijaz Ahmed, Ph.D., Rutgers University
Rebecca H. Hughes, B.S., Columbia University
Barclay Morrison III., Ph.D., Columbia University
Elizabeth Stucky, Rutgers University
Genipin is an extract from the Gardenia Jasminoides that may be a multi-potent neuroprotective agent but has not been tested in relevant models of TBI. Herein, genipin reduced cell loss in a model of glutamate-mediated excitotoxicity using organotypic hippocampal slice cultures.
Hippocampi of post natal day 9–10 rat pups were removed aseptically, cut into 400 um thick sections, and plated on Millipore membranes. Cultures were maintained for 14 days under standard conditions (37°C and 5% CO2) prior to injury. Slices were first exposed to a range of genipin concentrations to establish a toxicity curve. Separate slices were then incubated in 10 mM glutamate for 3 hours. Cultures were treated with 5–50 uM genipin or vehicle (0.01% DMSO). Treatment began either concurrent with or immediately after incubation in glutamate. Slices were imaged with Sytox Green at 0 and 24h to calculate cell death as the percent area staining above a threshold within a given anatomical region.
Negligible genipin toxicity was observed up to 50 uM. Exposure to 10 mM glutamate for 3h induced widespread death in the hippocampal slices at 24h. Maintaining the organotypic cultures in 5–50 uM genipin during the 3h exposure to glutamate and for the remainder of the culture period after removal of the glutamate significantly decreased cell death in the CA1, CA3, and dentate gyrus regions (all P<0.0001). Introducing genipin after the 3h exposure to glutamate also significantly reduced cell death in the CA3 region (P=0.00026) and the dentate gyrus (P=0.0006). Cell death in the CA1 region was substantially reduced, but the results were not significant (P=0.07). Treating cultures with 1 uM or lower concentrations of genipin was less effective.
Genipin is a natural plant extract that in the biomaterials field is used at millimolar concentrations as a crosslinking agent, but that has also been used in lower doses in Chinese medicine for a variety of maladies. In this report, we demonstrated that genipin can protect organtoypic hippocampal slices against glutamate-mediated excitotoxicity. Genipin offered protection when delivered during the insult and after the insult. Future work will examine mechanisms of genipin-mediated protection and test the natural compound in vivo.
This research was supported by the New Jersey Commission on Brain Injury Research Award #10-3215-BIR-E-0.
excitotoxicity, organotypic, Chinese medicine, genipin
ADMINISTRATION OF THE NRF2-ARE ACTIVATORS SULFORAPHANE AND CARNOSIC ACID ATTENUATE 4-HYDROXY-2-NONENAL INDUCED MITOCHONDRIAL DYSFUNCTION EX VIVO
Indrapal N. Singh, Ph.D., University of Kentucky College of Medicine
Edward D. Hall, Ph.D., University of Kentucky College of Medicine
Traumatic brain injury (TBI) currently signifies a substantial health and socioeconomic dilemma in the United States. The role of oxidative damage after TBI has been extensively demonstrated. The transcription factor Nrf2 mediates transcription of antioxidant/cytoprotective genes by binding to the antioxidant response element (ARE) within DNA.
Upregulation of these genes constitutes a pleiotropic cytoprotective-defense pathway. As neuronal mitochondria have previously been shown to be susceptible to oxidative damage, we sought to mechanistically investigate whether Nrf2-ARE activation in vivo could protect mitochondria under conditions of oxidative stress ex vivo. Young adult male CF-1 mice were administered one of two known Nrf2-ARE activators I.P. – sulforaphane (5.0mg/kg) or carnosic acid (1.0mg/kg) – or their respective vehicle 48 hours prior to Ficoll isolation of cortical mitochondria. Purified mitochondria were then exposed in vitro to 30uM of 4-hydroxy-2-nonenal (4-HNE) for 15 minutes at 37 degrees Celsius.
Previously, we demonstrated the in vivo post-injury time-course of Nrf2-ARE mediated gene expression in the cortex and hippocampus of male CF-1 mice utilizing a unilateral controlled cortical impact (CCI) injury model. Interestingly, increased Nrf2-ARE mediated expression was not observed until 24 hours, whereas our recent work showed oxidative damage also occurring 24 hours post-TBI. Mitochondrial bioenergetics was then assayed on the XF-24 Bioanalyzer (Seahorse Bioscience, USA) in this study. The administration of sulforaphane (SFN) and carnosic acid (CA) significantly (p<0.05) attenuated 4-HNE induced inhibition of mitochondrial respiration for both Complex I and II. Furthermore, CA and SFN both significantly (p<0.05) reduced 4-HNE bound mitochondria protein as determined by Western blot.
These results demonstrate the capability of Nrf2-ARE induction in vivo to protect from 4-HNE toxicity to cortical mitochondria ex vivo. Ongoing studies will determine the therapeutic efficacy of Nrf2-ARE activators to attenuate post-TBI pathophysiology.
Supported by Grants NIH-NIDA 1T32 DA022738, NIH-NINDS 2P30 NS051220-01 and funds from the Kentucky Spinal Cord & Head Injury Research Trust.
mitochondria, 4-hydroxy-2-nonenal, Nrf2, bioenergetics, antioxidants
AN IN VITRO INJURY MODEL FOR SH-SY5Y NEUROBLASTOMA CELLS: EFFECT OF STRAIN AND STRAIN RATE
Maciej Skotak, PhD, University of Nebraksa-Lincoln
Namas Chandra, PhD, University of Nebraska-Lincoln
There is a great need to have an in vitro cell injury model wherein a wide range of strain and strain rate can be precisely and independently applied. It will enable exploration of various biomechanical loading conditions pertinent to cells under either blunt or blast impact-induced TBI.
SH-SY5Y neuroblastoma cells were cultured on an elastic, transparent silicon membrane and stretched using modified cultured axonal injury (CAI) device in a wide range of strain and strain rates. The strain mapping was performed using ARAMIS system equipped with high-speed cameras to allow 2D and 3D strain calibration. Viability and membrane permeability (injury levels) of cells subjected to different levels of strain and strain rates was evaluated via Live/Dead® assay. The live, injured and dead cells were counted automatically by Laser Scanning Cytometer. Mitochondrial activity was measured by JC-1 staining.
The modified CAI device achieved a wide range of strain (0–140%) and strain rate (15-68-1). Following injury, we observed a sharp decrease in the number of viable cells with concomitant increment of injured and dead cells' populations with increasing substrate strain. The effect of strain rate on viability was studied when the strain was fixed at a specific level of 50%. Despite the increase in strain rate, there was no statistical difference between samples tested using strain rate from 20 to 68 s-1. By plotting stretch-mediated dose response curves for live and injured population, we found the population of injured cells increases sharply in the narrow strain range (30–55%) with corresponding decrease in live cell numbers. From these results, we identified three regions with different modes of injury: (I) mild, with predominantly live cells (ɛ=0–30%), (II) moderate, where the transition between live and injured cells takes place (ɛ=30–55%) and (III) severe with mainly injured and dead cells (ɛ=55% and more). The model further shows that time-after-injury plays a vital role in the determination of recovery-deterioration pathways and the biological selection depends on the severity of initial injury. To examine whether the lethality of stretch correlated with mitochondrial dysfunction (gauged by mitochondrial potential measurements), we used JC-1, a plasma membrane-permeant dye. Cultures exposed to stretch exhibited a rapid loss of mitochondrial membrane potential in a strain-dependent manner. Partial recovery was observed within a few hours for cells in the moderate (II), but not in the severe (III) injury group.
The extended in vitro TBI paradigm presented in this paper, allows application of accurate and repeatable levels of strain and strain rate. The Laser Scanning Cytometry permitted fully automated data acquisition and precise analysis of injury levels, with excellent spatial resolution. This technique helped us thoroughly inspect a cellular injury at predefined sets of experimental conditions. Dose-response analysis revealed there are three regions characterized by mild, moderate and severe injury levels. This finding was further correlated with mitochondrial dysfunction via in-situ mitochondrial membrane potential measurements. We believe our in vitro model will be useful in precise evaluation of injury criteria and injury risk curves, which can then be used in a computational model. Such a model will ultimately aid in the understanding of TBI, and development of relevant therapeutics.
Financial support under the US Army Research Office project 'Army-UNL Center of Trauma Mechanics' (Contract No. W911NF-08-10483) is gratefully acknowledged.
stretch strain rate viability mitochondria
CONTINUOUS MONITORING REVEALS A WIDE RANGE OF CEREBRAL BLOOD FLOW VALUES AT A GIVEN CEREBRAL PERFUSION PRESSURE AFTER SAH
Suguna Pappu, M.D., Ph.D., Department of Neurosurgery, University of New Mexico School of Medicine
Edwin M. Nemoto, Ph.D., Department of Neurosurgery, University of New Mexico School of Medicine
Howard Yonas, M.D., Department of Neurosurgery, University of New Mexico School of Medicine
Despite the advent of continuous multimodal monitoring in the neurocritical care unit, characterizing cerebral auto-regulation remains a significant challenge. However, by visualizing the distributions of continuously observed neuromonitoring parameters, we may elucidate the potentially dynamic and unique autoregulatory behavior of each patient during recovery.
Thirteen SAH patients were included in this retrospective study. Each patient had intracerebral monitoring via the Hummingbird SynergyDuo(Innerspace) bolt system with continuous capture of data via the CNS system (Component Neuromonitoring System, CNS Technologies LLC) for at least 1 day (maximum 7 days). Intracranial data captured include ventricular intracranial pressure (ICP), parenchymal ICP, focal brain tissue oxygenation (tO2) (Licox, Integra) and thermodilution cerebral blood flow (CBF) (Boxman Perfusion, Hemedex); systemic variables are also continuously captured. The oxygen and perfusion probes have a fixed spatial relationship between them, with placement into the deep frontal white matter adjacent to the ventriculostomy. Data points were digitized at one-second (CBF) and two-second intervals (all other parameters) for the entire duration of each patient's stay in the intensive care unit. Distributions of CBF and ICP values corresponding to each observed CPP and MAP, respectively, were calculated.
For a given CPP (=MAP–ICP), we observed hundreds to thousands of CBF values within a range of 60 ml/100g/min in width. Similarly, for a given MAP, we observed a range of ICP values of approximately 20 mmHg in width. These results were typical of patients who were monitored four or more days, including one of the four patients with poor outcomes, who died during monitoring. Upon visualizing the distributions of observed CBF and ICP measurements at each observed CPP and MAP, respectively, we find that the middle 50% of measured CBF values are within plus or minus 15 mg/100g/min above and below the median CBF for a given CPP. Similarly, the middle 50% of observed ICP values are within approximately plus or minus 5 mmHg above and below the median ICP for a given MAP. We find that the range and frequency of observed MAP values were unique in each individual's case. Distributions of CBF and ICP values across CPP and MAP values, respectively, were also unique for each patient. These relationships differ also in their behavior between groups of patients who lived and those who died. Preliminary inspection of CBF vs. CPP plots for each patient, with the above distributions also visualized, suggest that a curve through the median CBF at each observed MAP may help elucidate “classic” autoregulatory dynamics between CBF and CPP in a way that is unique for each patient.
Our analyses reveal a significant amount of noise in our data collected while continuously monitoring intracerebral dynamics in patients in recovery from neurosurgical treatment for SAH. Indeed, visualizing the distributions of observed CBF measurements for each observed CPP, rather than simply visualizing the scatter of CBF values at each CPP, reveals a significantly narrower range of CBF for each CPP than a simple scatter of the data would indicate. Further analyses are needed to elucidate the range of CPP or MAP values at which CBF or ICP are well regulated. Lastly, analyzing the dynamic relationships between CPP and CBF, and also ICP and MAP, during the entire duration of monitoring for each patient are needed in order to reveal how the unique autoregulatory relationships between CPP/CBF and MAP/ICP evolve as each patient's condition improves or declines.
The authors thank the Post-Baccalaureate Education and Research Program (PREP) at the University of New Mexico for funding.
Multimodal neuromonitoring, cerebral autoregulation
ISCHEMIC NEURONAL DEATH IS INCREASED BY INCREASED TEMPERATURE THROUGH ER STRESS IN VITRO AND IN VIVO
Chunyan He, MD, Illinois State University
Christina Pegg, MS, Illinois State University
Amy Wang, Illinois State University
Ann Stroink, MD, Central IL Neuroscience Foundation
Previously, we have shown that hyperthermia increases neuronal death and worsens stroke outcome. In the present study, we have examined the mechanisms about how temperature worsens the neuronal death, using both in vitro and in vivo models of ischemic injury. More specifically, we studied whether temperature increase sensitizes neuronal cells to ischemia through ER stress-apoptotic death signal transduction pathways.
In in vitro model, ischemia was induced by treating human SH-SY5Y neuroblastoma cells with sodium azide. The cells then received treatment of 37°C or 41°C, for a predetermined period of time and collected for analyses. Cell death was assayed with trypan blue exclusion. Cell size was measured by BD FACScan flow cytometer. For analyses of protein levels with Western blot, the cells were harvested and lysed in a cold lysis buffer. Supernatants were collected and protein concentrations in the supernatants were determined using Bio-Rad protein assay. The lysates were subjected to SDS-polyacrylamide electrophoresis and transferred to polyvinylidene difluoride (PVDF) membrane. The membranes were blotted with various primary antibodies and were then incubated with horseradish peroxidase conjugated secondary antibodies. The membranes were developed by chemiluminescence. Caspase activities were determined using EnzChek assay kit. In in vivo model, ischemic brain injury was induced by occluding the middle cerebral artery (MCA) with a filament suture in anesthetized rats. The animals received either normothermic or hyperthermic treatment. The rectal temperature in the normothermic group was kept at 37.5°C, while the hyperthermic group at 39.5°C with a feed-back controlled YSI heating system. The body temperature was raised to desired level immediately before the MCA occlusion, and maintained at this level for a period of 3 hours after the occlusion. The protein changes in the injured brain were detected with Western blot analyses.
In vitro study, ischemic death was induced by treatment of SH-SY5Y cells with different concentration of sodium azide. Treatment with sodium azide caused a dose-dependent reduction in cell viability at 37°C. Combination treatment of increased temperature and sodium azide was more effective for inducing cell death than either treatment alone, indicating increased temperature renders the cells more vulnerable to ischemic induction. The increased temperature worsened ischemia-induced shrinkage of cellular volume (a characteristic of apoptotic death), enhanced ischemia-induced ER stress as demonstrated by elevated phosphorylation of eIF2α (a translation initiation factor) and elevated levels of CHOP (a proapoptotic transcription factor). The increased temperature also intensified the activation of caspase-3 (an apoptotic effector protease). Inhibition of caspase-3 with DEVD-FMK significantly reduced the cell death at 37°C and also reduced the cell death promoted by temperature increase. In in vivo studies, elevated phosphorylation of eIF2α and elevated levels of CHOP were observed in the ischemic injured brain and hyperthermia further enhanced these elevations.
Similar to previous published reports, we found that SH-SY5Y cells underwent apoptosis after chemically induced ischemia in a dose-dependent manner. We also found that combination treatment of increased temperature and ischemia resulted in more cell death than either treatment alone. Concomitantly, combination treatment with ischemia plus increased temperature also worsened decrease of cellular volume, a major characteristic of apoptosis. Previous study showed that phosphorylation of eIF2α by PERK and increased CHOP was involved in hypoxic neuronal apoptosis. We have further shown in the present study that temperature increase and ischemia synergistically elevated eIF2α phosporylation and CHOP levels, suggesting this combination treatment synergistically fostered the phosphorylation activities of PERK. In vivo data are also in agreement with the findings form the in vitro studies. Cleaved caspase-3 increased following ischemic induction and treatment with increased temperature, suggesting the activation of caspase-3. Combination treatment with ischemia and increased temperature further increased the cleavage of caspase-3. Additionally, inhibition of caspase-3 activities ameliorates the temperature increase-exacerbated apoptotic death. Collectively, these findings suggest that temperature increase exacerbates the neuronal death by regulating ER stress-apoptotic signal transduction. Thus these findings unveiled a potential therapeutic target that could ameliorate hyperthermic ischemia by modulating a linkage of ER stress response to apoptotic cell death.
These studies were carried out in Illinois State University, and the work and supported by Kattner scholarship and National Institutes of health, National Institute of Neurological Disorders and Stroke (R15NS072858).
Apoptosis, ER stress, hyperthermia, ischemia
DETERMINATION OF SPINAL NERVE INJURY TOLERANCE TO STRETCH
John Cavanaugh, MD, Professor, Wayne State University;
Chaoyang Chen, MD, ; Gurjiwan Virk, MS, ; Joseph Yaldo, MS, ; Katsumasa Tanimoto, MD, PhD, ; Srinivasu Kallakuri, PhD, ; John M. Cavanaugh, MD,
Tensile loading to axons causes traumatic axonal injury (TAI) in the brain, spinal cord, and peripheral nerves. Injury tolerance of nerves to strain and strain rate is not yet clear. The current study investigated functional changes of lumbar nerve roots subjected to dynamic tensile strains using neurophysiologic techniques.
Sixty-five L5 dorsal nerve roots in anesthetized adult male Sprague-Dawley rats were subjected to predetermined strain ranges (<10%, 10–20% and >20%) at a specified displacement rate (20, 200 or 800 mm/sec) once per root. Image analysis was used to determine actual strains on the roots during the pull. Neurophysiologic recordings were performed on the nerve root before and after stretch (over 6 hours) to determine functional changes in response to stretch, focusing on the magnitude of evoked compound action potential (CAP) determined by the amplitude of CAPs. Functional loss and recovery was analyzed using repeated measures ANOVA. Binary logistic regression methods were used to determine the strain threshold that caused 50% probability of nerve conduction dysfunction. Cut-off for mild dysfunction was set at the decrease of 10.30% CAP amplitude (lower limit with 95% confidence interval), and decrease of 24.73% (mean −1SD) CAP amplitude for moderate dysfunction.
Overall, the actuator displacement produced a designated average strain of each L5 nerve. However, the segmental strain along each root varied and the actual strain distribution along the root was not uniform. In most of groups, the strain at the proximal end of root (RE2) was greater than at the distal end of root (RE1), (t test, p < 0.05). In sham group, nerve conduction over the 6 hours was not significantly less than baseline (0 time point) (One Way ANOVA, p > 0.5). There was no difference of amplitude at different time points in both distal and proximal recording segments in the sham group. Effects of stretch rates on nerve dysfunction: In all strain groups, the 800 mm/s rate caused greater decrease of CAP amplitude than 20 mm/s and 200 mm/s rates. There was no statistical difference of CAP amplitude between 20 mm/s and 200 mm/s rates (GLM, Univariate, p < 0.05). CAP amplitude decreased after stretch in all groups. Effects of stretch rates on nerve dysfunction: Amplitude of the CAP decreased as strain and displacement rate increased. In low strain (<10) and low stretch rate (20 mm/s) groups, nerves showed recovery over 6 hours, based on amplitude percentage change compared to the baseline. The strain that caused 50% probability of mild nerve dysfunction was 8.67% strain in 20 mm/s group; 7.32% strain in 200 mm/s group; and 6.27% strain in 800 mm/s group. The strain that caused 50% probability of moderate nerve dysfunction was 31.25% strain in the 20 mm/s stretch speed group; 22.14% strain in the 200 mm/s group; and 9.55% strain in the 800 mm/s group.
In control and sham groups, CAP amplitude change over time was not significantly less than baseline, while in 20, 200, and 800 mm/s groups, CAP amplitude over the time was significantly less than baseline. Higher stretch rates produced more deficits in nerve conduction function. This was obvious in 800 mm/s stretch rate. In this group, the proximal nerve segment near the clamped end demonstrated more deficit than the distal end. This was reflected by the higher strain at the proximal end during high rate loading compared to the distal end. In low strain (<10%) and low stretch rate (20 mm/s) groups, the spinal nerve roots showed recovery over 6 hours.
This research was support by a NHTSA grant (JMC) (DTNH22-08-C-00082).
A SINGLE INSTITUTIONAL EXPERIENCE OF 42 CASES OF PEDIATRIC CERVICAL SPINE TRAUMA FROM INDIA: ARE THEY DIFFERENT FROM ADULT CERVICAL TRAUMA
G. Laxmi Prasad, MCh, AIIMS
Ashok Kumar Mahapatra, MBBS MS MCh, AIIMS
Pediatric cervical spine injuries are relatively uncommon and are associated with significant morbidity and mortality. Pediatric anatomy and physiology predispose to upper cervical spine injury and spinal cord injury without radiologic abnormality in contrast to lower cervical spine injury seen in adults. Falls are the most common cause of spinal trauma in Indian pediatric population. Children have relatively high incidence of upper cervical injuries due to their mobile anatomy, ligamentous laxity and cephalocervical disproportion.
Objectives: This study was carried out to analyse the epidemiologic profile of pediatric cervical spine injuries and to evaluate the clinico-radiological characteristics of cervical spine injuries in children, to correlate the outcome with the clinico-radiological features. Methodology: Prospective cum retrospective data analysis (2008–2011) at JPN apex trauma centre, New Delhi, India. 42 children (0–18 yrs) were operated during this period. Fall from height was noted in 27 cases, upper cervical spine injuries in 12/42 cases (odontoid fractures in 6 cases). Associated injuries were noted in 8 cases, preop ASIA A score was noted in 16 cases. Two cases had prep systemic hypotension and required ventilator support. Methylprednisolone was not given in this series. All patients had CT of whole spine (as protocol) and MRI imaging study. Patients with whiplash/sciwora were excluded.
Postoperative VAP was noted in 28 cases, 38 required tracheostomy, perioperative mortality noted in 2 cases. Implant failure on long term follow up seen in 2 cases requiring refixation in one patient. 13 of 16 children who came for follow up had become independent for ADL (81) at a follow up range of 4–40 months thereby showing significant neurological recovery in these group of patients.
Ligamentous injuries are commoner in pediatric patients, fall from height most commonly noted in Indian population, sub axial trauma more common than upper cervical trauma. 81 had good outcome at last follow up. Such observation warrants aggressive surgical approach in pediatric population to give them better neurological outcome.
cervical spine injury, pediatric, sciwora
TOLL-LIKE RECEPTOR 4 (TLR4) IS BIPHASICALLY UPREGULATED AFTER SPINAL CORD INJURY AND CONTRIBUTES TO NEURODEGENERATION AND FUNCTIONAL LOSS
John Vender, MD, Georgia Health Sciences University
Krishnan Dhandapani, PhD, Georgia Health Sciences University
Clinically efficacious treatment options for spinal cord injury (SCI) are lacking due, at least in part, to the poorly defined mechanisms underlying the secondary phase of the injury. In this study, we hypothesize that Toll-like Receptor 4 (TLR4) promotes secondary injury using a compression model of SCI.
Functional recovery was quantified using two distinct rating systems (Basso Mouse Scale (BMS) and modified Tarlov scale (mTarlov)) to compare wild type and TLR4 signaling incompetent mutant mice. Protein expression of TLR4 was quantified by western blot densitometry and localized with immunohistochemical analysis. Neurodegeneration was qualitatively assessed using fluoro-jade B to identify dead/dying cells.
Improved motor function recovery was observed in TLR4 mutant mice compared to wild type on both the BMS and mTarlov scales. TLR4 expression was biphasically upregulated after SCI with a rapid acute upregulation lasting less than 12 hours and a delayed upregulation 2 days post-injury. TLR4 expression was observed in neurons during both upregulation phases, but not in astrocytes or microglia, and fluoro-jade B staining revealed diminished neurodegeneration in TLR4 mutant mice compared to wild type.
Together, these data suggest an important role for TLR4 signaling in the secondary pathology of SCI and identify TLR4 as a potential therapeutic target.
Dr. Krishnan Dhandapani
Inlammation, TLR4, SCI
CHONDROITIN SULFATE PROTEOGLYCANS INHIBIT OLIGODENDROCYTE MYELINATION THROUGH PTP-SIGMA
Michael J. Shamblott, Ph.D., University of South Florida College of Medicine/Department of Pediatrics
Devin S. Gary, Ph.D., Johns Hopkins University School of Medicine/ Department of Neurology/Hugo W. Moser Research Insitute at Kennedy Krieger/ International Center for Spinal Cord injury
Visar Belegu, Ph.D., Johns Hopkins University School of Medicine/ Department of Neurology/Hugo W. Moser Research Insitute at Kennedy Krieger/International Center for Spinal Cord injury
Andres Hurtado, M.Sc.B.E., M.D., Johns Hopkins University School of Medicine/Department of Neurology/Hugo W. Moser Research Insitute at Kennedy Krieger/International Center for Spinal Cord injury
Misti Malone, M.S., Ph.D., Johns Hopkins University School of Medicine/Department of Neurology/Hugo W. Moser Research Insitute at Kennedy Krieger/International Center for Spinal Cord injury
John W. McDonald, M.D., Ph.D., Johns Hopkins University School of Medicine/Depts of Neurology and PM&R/Hugo W. Moser Research Insitute at Kennedy Krieger/International Center for Spinal Cord injury
The process of gliosis following spinal cord injury (SCI) results in upregulation of chondroitin sulfate proteoglycans (CSPGs), inhibitors of axonal regeneration. Protein tyrosine phosphatase sigma (PTPsigma) mediates this inhibitory effect in neurons. In this study we sought to determine if CSPGs inhibit oligodendrocyte process outgrowth and myelination through PTPsigma.
Oligodendrocyte precursor cells (OPCs) and dorsal root ganglion (DRG) neurons were derived from the spinal cords of either post-natal day 4 (P4) Sprague-Dawley rats or P2 PTPsigma +/+ and −/− mice. OPCs were treated with laminin (control) or laminin + CSPG, and then cultured alone or over a layer of DRGs prior to fixation. OL process outgrowth was determined by measuring the minimal elliptical area of formed by processes extending from OLs (n=3–6). Myelin segments were defined as continuous linear co-localization of MBP and neurofilament of at least 10 microns in length. Differentiation was characterized on the basis of morphology and expression of OL-specific lineage markers. MBP levels in OLs were assessed using quantitative RTPCR and Western blot (n=3–4). Values are expressed as mean (+/−SEM) as a percentage of control. Statistical significance was determined using ANOVA analysis with Bonferroni-Holm Post-Hoc testing: *p<0.05, **p<0.01.
Elliptical area of OL processes decreased in the presence of aggrecan (63.7%+/−7.7; p<0.01), neurocan (64%+/−6.1; p<0.01), and NG2 (62.2%+/−5.5; p<0.01). Myelination of DRGs was decreased following pre-exposure of OLs to aggrecan (48.6%+/−4.9; p<0.01), neurocan (61.4+/−7.45; p<0.01), and NG2 (62.3%+/−8.2; p<0.01). Chondroitinase ABC (chABC) treatment of aggrecan prior to cell plating reversed its ability to suppress process outgrowth (93.5%+/−10.0; p>0.05) and myelination segment formation (77.8%+/−8.9; p>0.05). There was no statistical difference in the expression of several OL-specific lineage markers between laminin and aggrecan groups. Additionally, there was no statistical difference in the distribution of OL phenotypes between the laminin and aggrecan groups. In the presence of aggrecan, the MBP mRNA level was unchanged at 104.5%+/−11.3 (p>0.05, n=4). We found no statistical difference in total MBP levels from cells plated on aggrecan (124%+/−14.3; p>0.05). PTPsigma-targeting shRNA (PTPsigma-shRNA) reduced PTPsigma mRNA expression levels in OLs to (19.3%+/−0.2; p<0.001) and (13.6%+/−0.1; p<0.001) in the presence of laminin and aggrecan, respectively. In the presence of aggrecan, process outgrowth of OLs expressing the PTPsigma-shRNA was unchanged (77.0%+/−8.7; p>0.05). OLs expressing the NS-shRNA exposed to aggrecan had significantly fewer myelin segments (42%+/−6.1; p<0.001). Aggrecan-exposed OLs expressing PTPsigma-shRNA exhibited no deficiency in forming myelin segments (133.9%+/−17.1; p>0.05). The process outgrowth of aggrecan-treated PTPsigma −/− derived OLs was unchanged (117.5%+/−17.3; p>0.05). Myelin segment formation of aggrecan-treated PTPsigma −/− derived OLs was (105.8%+/−3.9; p>0.05). ROCKi-treated OLs exhibited a reduction in process outgrowth (53.2%+/−8.9; p<0.001) and myelin segment formation (60.2%+/−8.0; p<0.01), in the presence of aggrecan.
We demonstrate that while the CSPG, aggrecan, had no effect on the expression of OL-specific lineage markers, it significantly inhibited OL process outgrowth and reduced the ability of OLs to form myelin segments. Ablation of PTPsigma rendered OLs blind to the inhibitory effects of aggrecan. These findings indicate that the effects of CSPGs extend beyond inhibition of axon regeneration to include suppresion OL outgrowth and myelination, and identify PTPsigma as a target to mitigate CSPG-based OL dysfunction in demyelinating injury and disease.
US Department of Defense USAMRMC/TATRC/USAMRAA contracts W81XWH-08-2-0192 and W81XWH-09-2-0186
Spinal cord Injury Protein tyrosine phosphatase demyelination oligodendrocyte CSPG
POST-TRANSLATIONAL MODIFICATIONS OF AGGRECAN AND INHIBITION OF NEURITE OUTGROWTH
Christopher M. Calulot, BS, Spinal Cord and Brain Injury Research Center, Department of Anatomy and Neurobiology, University of Kentucky
Diane M. Snow, PhD, Spinal Cord and Brain Injury Research Center, Department of Anatomy and Neurobiology, University of Kentucky
Thomas M. Hering, PhD, Spinal Cord and Brain Injury Research Center, Department of Anatomy and Neurobiology, University of Kentucky
Aggrecan, a chondroitin sulfate proteoglycan (CSPG), inhibits neurite outgrowth. Post-translational modifications (PTMs), include N- and O-linked glycosylation, and substitution with keratan sulfate (KS), and chondroitin sulfate (CS). A novel assay was used to understand the contribution of each PTM to neurite outgrowth inhibition.
Steer articular cartilage aggrecan was placed in separate wells of a collagen-coated 96-well plate at 0, 150, and 300μg/mL overnight at 4°C. Following adsorption, the aggrecan was subjected to glycosidase digestion. Either chondroitinase ABC (chABC; removes chondroitin sulfate chains), chABC + keratanase II (k'ase; removes keratan sulfate chains) + endo-β-galactosidase (EβG; removes keratan sulfate chains), or chABC + k'ase + EβG + Peptide N-glycosidase F (PNGase F; removes N-linked oligosaccharides) were added to the wells. All enzyme digestions were 2 hours and conducted sequentially as listed above. NS-1 (a subclone of PC12) cells were added to the wells, allowed to attach overnight, and increasing concentrations of NGF were added to separate wells (0–500ng/mL). After 72 hours, cells were fixed, stained, and imaged using an automated method. Images were then analyzed using Neurite Tracer, an Image J plug-in.
Following overnight treatment of collagen-coated culture dishes with aggrecan over a range of 0 to 300 mg/ml, an aggrecan ELISA assay performed on an identically treated plate confirmed adsorption of aggrecan in proportion to the concentration of the coating solution. In preliminary work, NS-1 cells grown on an aggrecan-collagen substrate showed an aggrecan dose-dependent inhibition of neurite outgrowth. We tested 150μg/mL aggrecan in the coating solution, since this concentration resulted in a midrange inhibition effect, permitting the observation of further inhibition or stimulation after glycosidase treatment. NGF-1 concentration was varied between 0 and 500 mg/ml. Treatment of surface-bound aggrecan with chABC to remove CS chains resulted in an ∼4–11% increase in outgrowth at doses of 125 to 500 mg/ml NGF-1. Unexpectedly, sequential CS and KS degradation resulted in an ∼4–15% decrease in neurite length. Sequential enzymatic removal of CS, KS, and N-linked oligosaccharides resulted in a 49–57% decrease in neurite length.
We have developed a novel outgrowth assay that can be modified to measure the effect of surface-adsorbed proteoglycans on neurite outgrowth. We examined the effect of sequentially removing components of the aggrecan molecule to determine the relative contributions of CS, KS and N-linked oligosaccarides to neurite outgrowth inhibition. In agreement with previous studies, we found that the chondroitin sulfate chains of aggrecan inhibit outgrowth. Novel to the study is the result that further removal of KS enhanced inhibition, and that removal of CS, KS and N-linked oligosaccharides further enhanced inhibition, which will be studied further. It is possible that N-glycosylation promotes neurite outgrowth. Alternatively, removal of N-linked oligosaccharides may reveal a region of the G1-domain of aggrecan that signals neurite outgrowth inhibition.
This work was supported by R01 NS054370, and KSCHIRT #10-11A. * Thomas M. Hering and Diane M. Snow contributed equally to this work.
In vitro, neurite outgrowth, CSPGs
VISUALIZING TRANSLATIONAL SPINAL CORD INJURY RESEARCH
Jesse Paquette, MS, Ayasdi Inc.
Jennifer Kloke, PhD, Ayasdi Inc.
Aiwen W. Liu, BS, UCSF Brain and Spinal Injury Center
Cristian F. Guandique, BA, UCSF Brain and Spinal Injury Center
John C. Gensel, PhD, Ohio State University
Karen-Amanda Irvine, PhD, Stanford University
Michael S. Beattie, PhD, UCSF Brain and Spinal Injury Center
Jacqueline C. Bresnahan, PhD, UCSF Brain and Spinal Injury Center
Pek Y. Lum, PhD, Ayasdi Inc
Adam R. Ferguson, PhD, UCSF Brain and Spinal Injury Center
Spinal cord injury (SCI) produces a complex constellation of biological and functional changes. An under-appreciated hurdle for translational SCI testing is the need to manage and integrate massive amounts of data produced by basic research. Here, we demonstrate a novel data-driven approach for SCI data integration and visualization.
Visualization tools combined with multivariate statistical analysis has the potential to aid translation. Our goal is to develop user-friendly analytical tools that can interface with large databases for novel discoveries by members of the general SCI research community. The current project reports on a collaborative effort between an academic medical center (UCSF-BASIC) and a Silicon Valley startup (Ayasdi Inc.) to develop a bioinformatic data analysis/visualization system to assist bench-to-bedside translation. We applied a novel computational approach called topological data analysis (TDA), wherein complex datasets are computed and rendered into geometric visual representations. Clusters of subjects appear as nodes and relations among clusters are represented as interconnections (‘edges’) among nodes. We tuned the analysis parameters for reliable translation across different injury severities and types (IH vs. NYU contusion vs. hemisection) using a database subset of rats (N=159) with unilateral cervical SCI collected at OSU and UCSF.
Proof-of-concept TDA on our large preclinical cervical SCI dataset produced fruitful rapid visualization and large-scale data-driven mining of the preclinical SCI syndrome. Subjects were compared simultaneously on 120 biomechanical, histological, and functional outcome measures using a principal components ‘lens’ that clusters subjects into nodes based on shared variability, and into flares based on similarities between nodes. Node size reflects the number of subjects within a given node, and node color represents the values of the specified variable or outcome measure and where that falls along a spectral continuum of highest (red) to lowest (blue). For this specific cervical dataset, after the network topology is derived, these node colors are determined using a colored density map to assess the importance of 4 key variables often used in univariate analyses of SCI datasets: 1) spinal cord tissue compression, 2) white matter sparing, 3) grooming scores, and 4) motoneuron sparing. Clusters of subjects with values for each of these variables were graphed onto the spectral continuum from highest to lowest. Subjects receiving the most severe tissue compression during injury clustered together in a single flare. Subjects with less severe injuries clustered in another flare, and intermediate levels clustered in a third flare. The same essential pattern was observed in the outcomes: white matter sparing, grooming score, motoneuron sparing. However, none of these standard outcomes taken in isolation could fully account subject clustering across the syndromic network, highlighting the value of multivariate integration. In future work, we will cross-validate the tuned analytic approach with the additional data contributions from other centers to discover a set of analytic parameters with the goal of enabling translational comparisons across mouse, rat, and primate datasets.
Early proof-of-concept work is demonstrating that TDA can describe a range of SCI syndrome patterns in preclinical models using a user-friendly graphical interface. Further work is needed, however, to demonstrate whether a common set of analysis parameters can yield stable networks across a wide diversity of injury types, different laboratories, and across species. Our long-term goal is to facilitate these critical translational comparisons to provide a common analytic platform for integrative multivariate data analysis and data-mining of SCI research. The current collaborative data analysis system shows great promise for providing an integrated system for translational bioinformatics research in the SCI field.
Supported by NS031193, NS038079, NS069537, NS067092
Bioinformatics, datamining, data visualization, biomarker
ENHANCING ENDOGENOUS PROTECTIVE MECHANISMS FOLLOWING SPINAL CORD INJRY
Chen-Guang Yu, Ph.D, University of Kentucky/SCoBIRC
Ranjana Singh, Department of Anatomy and Neurobiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Ronan Power, Ph.D, Alltech, Inc.
Jignesh Pandya, University of Kentucky, SCoBIRC
Samir Patel, University of Kentucky, SCoBIRC
Patrick Sullivan, University of Kentucky, SCoBIRC
Alexander Rabchevsky, University of Kentucky, SCoBIRC
James Geddes, Ph.D, University of Kentucky/SCoBIRC
Spinal cord injury (SCI) continues to be a prevalent clinical problem and dietary selenium supplementation represents a potential strategy to attenuate SCI damage. We hypothesize that supplementation with selenized yeast will intervene in the secondary injury cascade following SCI by acting as an essential co-factor for antioxidant enzymes.
Selenized yeast, added on top of normal levels dietary levels, was formulated within rat chow and supplemented to female Sprague-Dawley rats for four months prior to receiving a moderate (150kdyn) contusive spinal cord injury or a sham laminectomy. One group of animals was monitored for locomotor functional recovery using the Basso, Beattie, and Bresnahan (BBB) scoring system for six weeks after the injury. Additionally, within this group of animals, bladder expression was performed as part of normal post-surgical care. The number of days until bladder function was recovered in each rat was recorded and examined as a marker for functional recovery. A separate group of animals was also subjected to a moderate (150kdyn) contusive spinal cord injury and mitochondria were isolated from the injured and sham spinal cords. Mitochondrial respiration was measured by looking at oxygen consumption rate using the Seahorse.
Selenium supplementation resulted in no significant changes in locomotor function recovery based on the BBB scale. Both groups followed a similar pattern of gradual recovery and eventual plateau. However, within this experimental group we did observe a significant improvement in recovery of bladder function after SCI with selenium supplementation. Rats receiving selenium supplementation recovered bladder function in an average of 3 days±0.36 as compared to rats on a control diet, which recovered bladder function in 5 days±0.47 (p-value: 0.014). Recovery of bladder expression is an additional functional marker for improvement. This area is of significant interest for patients suffering from SCI. Additionally, preliminary data in a small, separate subset of animals showed improved oxygen consumption rate in mitochondria isolated from rats maintained on the selenium-supplemented diet as compared to animals on the control diet.
While selenium supplementation did not improve locomotor functional recovery after SCI, supplementation did accelerate bladder function recovery and improve mitochondrial respiration in injured animals. The improvement in bladder function recovery is an area of high importance for SCI patients. The mitochondrial functional improvement provides a preliminary look at the mechanism of action for selenium as a neuroprotective agent. For high-risk populations, an effective dietary supplement could serve as a pre-treatment approach to attenuate the damage that follows neurotrauma.
Funding provided by Alltech, Inc.
spinal cord injury, antioxidant, selenium
DELAYED EXPRESSION OF CELL CYCLE PATHWAY CONTRIBUTES TO ASTROGLIAL SCAR FORMATION AND CHRONIC INFLAMMATION AFTER RAT SPINAL CORD CONTUSION
Ahdeah Pajoohesh-Ganji, PhD, George Washington University
Bogdan A. Stoica, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Michael Dinizo, MS, University of Maryland, School of Medicine
Kelsey Guanciale, BS, University of Maryland, School of Medicine
Alan I. Faden, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
SCI induces secondary tissue damage that lasts for days, weeks and even months after the initial insult and is associated with astrogliosis and inflammation. We have previously demonstrated that acutely up-regulation of a cluster of cell cycle genes contribute to post-mitotic cell death and secondary damage after SCI.
Adult male rats were subjected to a moderate contusion SCI at T8 using a well-characterized weight-drop model. Tissue from the lesion epicenter was obtained 28 days and 4 months post-injury and processed for protein expression by immunoblot analysis and immunohistochemistry. Roscovitine administered centrally by intrathecal injection 30 min post-injury and continuing for 7 days. At 28 days post-injury, lesion cavity was assessed using histological analysis; stereological assessment of GFAP+ astrocytes was also performed. In vitro, primary cultured astrocytes and microglia were stimulated by TGFb1 and LPS. Co-culture of astrocytes/microglia and neurons were applied for evaluation of neurite outgrowth.
Here we examined chronic expression of cell cycle-related proteins up to 4 months following SCI and effects of central administration of the selective cyclin-dependent kinases (CDKs) inhibitor roscovitine on astrogliosis and inflammation in a rat SCI contusion model. Immunoblot analysis demonstrated a marked chronic up-regulation of cell cycle-related proteins through 4 months post-injury, including cyclin E and D, CDK4 and E2F5. These proteins were highly expressed by hypertrophic GFAP+ astrocyes and concentrated at the boundary between residual spinal cord tissue and the central lesion by 4 months after injury. Further, delayed expression of inflammatory proteins was also colocalized with the cell cycle-related proteins through 4 months post-injury. Roscovitine administered centrally by intrathecal injection 30 min post-injury and continuing for 7 days can significantly reduce chronic induction of cell cycle proteins and increased p27 expression in the injured spinal cord. Treatment remarkably attenuated astroglial activation evidenced by reduction of GFAP expression and accumulation of CSPG and CHL1. Treatment also reduced GFAP+-lesion scar volume and the number of GFAP+ astrocytes in the preserved tissue. Furthermore, roscovitine attenuated the induction of factors associated with microglial activation. In addition, reactive astrocytes and microglia stimulate expression of cell cycle pathway, and cell cycle inhibition suppresses activation of astrocytes and microglia and neurite outgrowth in vitro.
These data demonstrate that cell cycle-related proteins are chronically up-regulated after SCI and may contribute to astroglial scar formation and further tissue loss. Our current study also provides further understanding of the role of cell cycle activation in the secondary injury that is important to the development of SCI therapeutics.
This work was supported by a National Institute of Health contract NIH-R01NS054221-06.
Chronic expression of cell cycle
A CATALYTIC ANTIOXIDANT MnTBAP PROTECTS AGAINST NEURONAL AND GLIAL OXIDATIVE STRESS AND DEATH AFTER EXPERIMETAL SPINAL CORD INJURY
Lokanatha Valluru, Ph.D., University of Texas Medical Branch
Antioxidant therapy reduces oxidative damage in secondary spinal cord injury (SCI). To evaluate the potential for treating SCI, this study examined in vivo the beneficial effects of a catalytic antioxidant Mn (III) tetrakis (4-benzoic acid) porphyrin (MnTBAP) to attenuate neuronal and glial oxidative damage and death following SCI.
Following laminectomy, the exposed spinal cord of rats was contused using the NYU-MASCIS weight-drop impactor (10-g rod dropped 1.25 cm). Immediately following injury, MnTBAP (4 mg/kg) or saline (vehicle control) was administered through the catheter into the intrathecal space of the cord. The cord was removed at 24 h post-administration following transcardial perfusion, and processed for immuno-staining. One set of sections was immuno-stained with cellular marker antibodies anti-NSE (for neurons), anti-ChAT (motoneurons), anti-GFAP (astrocytes), and anti-CC1 (oligodendrocytes). Another set of sections was double immuno-fluorescence-stained with above cellular marker antibodies and with the stress marker antibodies anti-DNP which against 2,4-dinitrophenyl hydrazine (DNPH)-labeled protein carbonyls – a marker of protein oxidation, and anti-nitrotyrosine - an indicator of protein nitration to immuno-colocalize oxidative stress in different types of cells. The positively stained cells in the cord sections at different distances from the injury epicenter were counted for quantitative comparison.
The spatial profiles of MnTBAP protection against oxidative stress for different each types of cells were established by counting the immuno-colocalized cells and comparing the counts between MnTBAP- and vehicle-treated groups. Our profiles demonstrated that MnTBAP statistically significantly reduced numbers of nitrotyrosine- and DNP-positive neurons, astrocytes, and oligodendrocytes in ranges of 1.6 to 3.2 mm and 1.8 to 2.85 mm rostral to the epicenter respectively. Comparison of the counts between the two treatments in the sections immuno-stained with cellular markers demonstrated that MnTBAP statistically significantly increased the numbers of neurons 1.5 to 4.0 mm, motoneurons 1.55 to 2.55 mm, astrocytes 1.7 and 2.2 mm, and oligodendrocytes 1.8 – 2.8 mm rostral to the epicenter. Comparing the percentages of MnTBAP rescued cells at the maximum protective distances for different type cells, we found that MnTBAP more effectively reduces neuron death (118%) compared to deaths of motoneuron and glial cells (29% – 51%).
This study established the profiles of MnTBAP protection against neuronal and glial oxidative stress and death after SCI. These profiles provide a database of the effective areas from the epicenter for therapeutic intervention by MnTBAP. MnTBAP significantly reduced oxidation and nitration of proteins in neurons and glial cells and increased the numbers of surviving neurons, motoneurons, astrocytes and oligodendrocytes. This suggests that MnTBAP - a broad spectrum scavenger of RS – combats oxidative stress by scavenging RS, thereby attenuating RS-initiated neuronal and glial cell death after SCI. Therefore antioxidant therapy by MnTBAP is potentially valuable for the treatment of SCI. It also provides evidence that generation of RS during the secondary damage cascade after SCI is a cardinal factor leading to secondary neuronal and glial death. The finding that MnTBAP more effectively reduced neuron death relative to that of glial cells merits further exploration.
The authors thank the National Institutes of Health (National Institute of Neurological Disorders and Stroke, RO1 NS 44324 to Danxia Liu) for financial support.
MnTBAP, cell death, oxidation, SCI
DECOMPRESSIVE CRANIECTOMY FOLLOWING TRAUMATIC BRAIN INJURY: RESULTS OF AN INTERNATIONAL SURVEY AND UPDATE ON THE RESCUEICP TRIAL (WWW.RESCUEICP.COM)
Lucia M. Li, MD, Addenbrooke's Hospital & University of Cambridge
Ivan Timofeev, MD, FRCS (SN), Addenbrooke's Hospital & University of Cambridge
Elizabeth Corteen, BSc (Hons), Addenbrooke's Hospital & University of Cambridge
Sian Ingham, Addenbrooke's Hospital & University of Cambridge
Marek Czosnyka, PhD, Addenbrooke's Hospital & University of Cambridge
John D. Pickard, MD, FRCS (SN), FMedSci, Addenbrooke's Hospital & University of Cambridge
David K. Menon, MD, PhD, FRCP, FRCA, FFICM, FMedSci, University of Cambridge
Peter J. Kirkpatrick, MD, FRCS (SN), Addenbrooke's Hospital & University of Cambridge
Peter J. Hutchinson, MD, PhD, FRCS (SN), Addenbrooke's Hospital & University of Cambridge
Even though the last 15 years have seen a resurgence of interest in the use of decompressive craniectomy (DC) following TBI, there is still no consensus on if and when to proceed with the operation.
From 2002 to 2004, two multi-centre randomised trials started recruiting patients: the DECRA trial and the RESCUEicp trial. The DECRA trial finished recruitment in 2010 (total recruitment 155 patients) and failed to show a clinical benefit with early/neuroprotective DC for patients with diffuse TBI. The RESCUEicp trial is investigating the use of DC as a last-tier therapy for post-traumatic refractory intracranial hypertension. Six months after the publication of the DECRA results (October-November 2011), we undertook a survey of members of the European Association of Neurosurgical Societies, Neurocritical Care Society, NeuroCritical Care Network, and Society of British Neurological Surgeons with the objective of examining opinion and practice patterns of DC after TBI. We used a secure online survey tool to disseminate the questionnaires. The questionnaire survey was approved by the Academic Committee of the Society of British Neurological Surgeons (project number NE0026).
The survey was completed by 503 individuals. Most of the respondents (225/503; 45%) were neutral/undecided regarding the long-term benefits of DC for post-traumatic refractory intracranial hypertension. Similar proportions of neurosurgeons (128/290; 44%) and intensivists (93/207; 45%) were neutral/undecided. However, a higher proportion of UK/Irish (87/142; 61%), as compared with other European (34/110; 31%), neurosurgeons were neutral/undecided (p<0.001). A higher proportion of those who had recruited a patient in the RESCUEicp trial were neutral/undecided (62/109; 57%) compared to those who had not (160/384; 42%; p=0.006). Only 8% of the respondents changed their practice following the DECRA results (42/503). The majority of those changed their practice from DC to medical therapy with barbiturates (28/42). RESCUEicp differs from DECRA in terms of threshold for ICP (25 mmHg vs. 20 mmHg), duration of refractory intracranial hypertension (1 hour vs. 15 minutes), timing of randomisation (any time when inclusion criteria are met vs. within first 72 hours only), acceptance of contusions/evacuated mass lesions and longer follow-up (2 years). Outcome is assessed at 6 months after injury using the eGOS (primary endpoint). Secondary endpoints include the assessment of outcome at 1 and 2 years post-injury with the use of the eGOS and the SF-36 questionnaire, ICP control, health-economic analysis, length of stay in ICU/neurosurgical unit and serious adverse events. The required sample size is 400 participants to detect a 15% difference in dichotomised outcome (power 80%, alpha 5%). The study is ongoing and has recruited 339 patients from more than 40 hospitals in 17 different countries. UK centres have recruited 69 of the patients to date. We are very keen to increase international contribution to RESCUEicp. A trial extension until December 2014 was recently awarded to us; therefore, new sites are still welcome to join RESCUEicp.
There remains significant uncertainty in the practice of DC following TBI. The hypotheses and inclusion criteria between the DECRA and RESCUEicp are different. Hence, the results from the DECRA study should not deter recruitment into surgical evaluation studies where thresholds for treatment are so different. On the other hand, the DECRA results emphasise the fact that decompressive craniectomy remains an unproven therapy, which should ideally be undertaken in the context of randomised trials. We believe that the ‘jury is still out’ as to whether decompressive craniectomy can improve outcomes after TBI despite the results of the DECRA study. Further patients should be studied both in terms of multicentre randomisation and multi-modality studies (including focussed imaging) to increase our understanding of the effects decompressive craniectomy has on the injured brain.
The RESCUEicp study is funded by a Medical Research Council/National Institute of Health Research Clinical Trials grant.
TBI, decompressive craniectomy, clinical trial
PENETRATING BRAIN INJURY FROM NAIL GUNS: EPIDEMIOLOGICAL, CLINICAL AND FORENSIC CHARACTERISTICS
Glenn Sandberg, MD, Harris County Institute of Forensic Sciences and Baylor College of Medicine
Jennifer Ross, MD, The Methodist Hospital Dept of Pathology and Genomic Medicine
Shankar Gopinath, MD, Baylor College of Medicine
Claudia Robertson, MD, Baylor College of Medicine
Nail guns, essential tools in modern construction, propel nails using compressed air, electromagnets, explosive gases or gunpowder. 37,000 nail gun injuries, including cranial injuries, occur annually. There are similarities and differences between penetrating head injury from nail guns versus firearms.
We ascertained and reviewed cases of cranial nail gun injuries at the Ben Taub General Hospital (one of two Level I Trauma Centers in Houston, Texas) and the Harris County Institute of Forensic Sciences. The site of injury, patient demographics, clinical outcome and nature of the projectiles were recorded. Nails ejected from nail guns are classified by size ranging from finishing nails (25.4 mm long×1.47 mm weighing 336 mg) to framing nails (152.4 mm long×6.68 mm weighing 45.36 gm).
We identified six cases of cranial nail gun injury. All were in men below age 45 years. Two cases were fatal and four were non-lethal with four accidents, one assault and one suicide attempt. The penetration was supratentorial in five and infratentorial in one. Large framing nails were the solitary projectiles in 5 cases and in the one suicide attempt, multiple (4) finishing nails penetrated the skull.
In the lethal cases, one involved the nail traveling through the thin bones of the orbit into the brain, and the other caused hemorrhage following removal of the nail at the construction site where the injury occurred. It appears that venous sinuses were penetrated and tamponaded by the nail, and upon removal of the projectile, hemorrhage and clinical deterioration ensued. In the survivable cases, nails were removed under controlled conditions. The most serious morbidity was ataxia and cranial nerve dysfunction from pontine and cerebellar damage in the case of infratentorial penetration.
The severity of tissue damage inflicted by a penetrating missile depends on the amount of kinetic energy that the projectile brings to the transaction. Nail gun injuries differ from firearm injuries because they possess less kinetic energy. Nails emerge from nail guns at a velocity of between 100–150 meters/second; therefore, even the heaviest framing nail will possess modest kinetic energy compared to firearm bullets. Additionally, framing nails possess a flat head that affixes the nail to the skull, preventing the projectile from completely entering the cranial vault. Lethality occurs when vascular structures such as venous sinuses are lacerated or when the projectile travels relatively unimpeded through thin bone.
Nail gun brain injuries are increasing in frequency with more widespread use of these tools largely among young men. Mechanistically the tissue injury reflects parenchymal laceration without large kinetic energy transfer with cavitation and shock waves. Survival is likely unless vascular structures are lacerated or the projectile penetrates vital parenchymal structures. These injuries may be preventable and public health campaigns are underway to educate end users about the correct and safe use of nail guns. If injury with cranial penetration occurs, it is critical that the nail not be removed at the construction site, during transport or in the emergency center, but under controlled conditions in the operating room. The ballistics of nail gun injuries differ fundamentally from firearm injuries; therefore, they are not an appropriate model of high energy penetrating brain injury.
None
TBI, nail gun
DETECTION OF TRAUMATIC BRAIN INJURY (TBI) BY LONGITUDINAL DIFFERENCE IMAGING
Yiyu Chou, M.S, Image Processing Core, Center for Neuroscience and Regenerative Medicine
Dzung Pham, Ph.D, Image Processing Core, Center for Neuroscience and Regenerative Medicine
Lawrence L. Latour, Ph.D, Stroke Branch, NINDS, NIH
John Butman, M.D, Ph.D, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
In many cases, MRI performed in head injured patients exhibit little clinical evidence of TBI. To help detect subtle evidence of TBI in head injured patients, we developed an image processing pipeline to emphasize differences between images obtained at two time points following injury.
As part of the THINC (Traumatic Head Injury Neuroimaging Classification) study, MRI was obtained from 20 head injury patients on a 1.5T GE system acutely (within 48 h of injury) and at 3 month follow-up. MRI included a 3D spoiled gradient T1-weighted sequence with 0.94×0.94×1.6mm voxel size, flip angle 15°, TR 12.4ms, and TE 4.2ms. To compute difference images from the two 3D T1 volumes, the second time point was registered to the first time using a 6 parameter rigid body transformation (MIPAV). Intensity inhomogeneities were then corrected using the n4 algorithm, followed by intensity normalization between the two datasets based on matching the cumulative histograms. Finally, difference images were computed by subtracting the co-registered, corrected and normalized T1-weighted images. Images were sent to a clinical PACS for interpretation by neuroradiologist.
Of all 20 dataset, over half of the subtraction results show thin borders around ventricles, indicating that ventricle size increased after 90 days of injury. These results may imply brain swelling after acute injury resolving by the second time point. Only one case shows decreased ventricle size.
In two cases, subtle changes in gray matter were revealed by inspection of the difference images. These foci of cortical encephalomalacia were difficult to detect by direct visual comparison of the two datasets independently.
Although difference images can highlight ventricle or cortex changes, accuracy is dependent upon accurate co-registrations. In two cases of subtle subdural hematomas, slight mass effect and subtle brain shift was present on the acute scan and absent on the follow-up. Thus, the difference images largely demonstrate this gross change in anatomy, rather than more subtle changes.
To compare longitudinal images of TBI patients, a pipeline is developed to convert images at different time points into the same coordinate and grayscale systems. Simple subtraction of corrected images highlights changes in ventricle and cortex shape. This technique shows promise for facilitating the detection of subtle longitudinal changes in the brain that may indicate TBI in patients with head injury. Other techniques, possibly relying on deformable models, may be required to identify such changes when there is a greater degree of brain deformation as from extra-axial or large parenchymal hematomas.
This work was supported by CNRM and the NIH Intramural Program. The authors gratefully acknowledge the investigators of the CNRM THINC study.
TBI, MRI, registration, inhomogeneity-correction, histogram-matching
ASSESSING TRAUMATIC BRAIN INJURY WITH ACCELERATED SUSCEPTIBILITY WEIGHTED IMAGING USING SEGMENTED ECHO-PLANAR IMAGING
Ningzhi Li, PhD, Image Processing Core, Center for Neuroscience and Regenerative Medicine
Leighton Chan, MD, MPH, Rehabilitation Medicine Department, NIH Clinical Center
Dzung Pham, PhD, Henry Jackson Foundation/CNRM Center for Neural and Regenerative Medicine
John Butman, MD, PhD, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
3D gradient recalled-echo (GRE) susceptibility weighted imaging (SWI) is increasingly used in assessing traumatic brain injury (TBI). The goal of this work is to determine whether accelerated SWI using segmented EPI can substitute traditional GRE SWI, which has a relatively long imaging time, for the detection of traumatic intracranial microhemorrhage.
Images from six patients enrolled under a longitudinal natural history trial of TBI were acquired with a product SWI sequence based on 3D GRE and with a segmented 3D EPI sequence (segEPI) that we developed. With TE/TR=25/40 ms, image matrix 448×439×72, voxel size=0.5×0.4×2 mm3, and parallel imaging GRAPPA×2, the imaging time for 3D GRE SWI was 9 minutes 47 seconds. For segEPI SWI, the imaging time was 1 minute 52 seconds with the same imaging parameters except TR=80 ms, echo-train length (ETL) of 15, and without parallel imaging. The images were acquired with a Siemens mMR 3T MRI scanner and visually evaluated by a neuroradiologist comparing hypointensity indicative of microhemorrhage.
Multiple (3–36) parenchymal microhemorrhages were identified in 3 of the 6 patients. Both GRE and segEPI were positive for intracranial hemorrhage in each of these 3 cases, and negative in the 3 others. Thus, no patient was misclassified by the segEPI as compared to the GRE. In total, 70 microhemorrhages were identified both on GRE and segEPI, 1 only on GRE and 1 only on segEPI.
While the overall visualization of the hemorrhages was very similar, there were significant differences in the image contrast of the vessels. Because flow compensation was used for the 3D GRE, arterial signal was relatively enhanced as compared to venous signal, minimizing the T2* susceptibility induced signal loss in the arteries as compared to the veins, which exhibited a greater degree of signal loss.
For the 3D segEPI, flow compensation was not used, so there was a comparable degree of signal loss in the arteries and veins. Consequently the vessels can be better visualized in EPI SWI. However, the long ETL led to T2* modulation of k-space data, causing blurring of the vessels in the resultant images. Qualitatively, there was a slightly greater degree of image distortion at the skull base for the segEPI.
3D segmented EPI can dramatically reduce imaging time for susceptibility weighted imaging with little or no loss in the ability to detect hemorrhages.In the context of MR imaging in the acute setting where time is of the essence, implementation and use of the 3D segmented EPI sequence is warranted.
Supported by the Center for Neuroscience and Regenerative Medicine and the Intramural Program of the National Institutes of Health.
TBI, SWI, GRE, EPI, Hemorrhage
INCIDENCE AND OUTCOME OF INTRACRANIAL HEMATOMA EXPANSION IN CHILDREN AFTER MODERATE AND SEVERE TBI
Michael Morriss, MD, Children's Medical Center Dallas
Rong Huang, MS, Children's Medical Center Dallas
Ana Hernandez, MS, Children's Medical Center Dallas
Darryl K. Miles, MD, Children's Medical Center Dallas
Computed tomography (CT) is integral in the early diagnosis and management of TBI. Follow-up head CT imaging is common in the management of acute intracranial hematomas after closed head injury. This study aims to quantify the temporal profile and outcome of intracranial hematoma expansion in children after TBI.
Patients were screened from a prospective cohort database of long-term TBI at a large level 1 pediatric trauma center. All patients were ages 0 to 13 years with a moderate or severe TBI resulting in either an epidural (EDH), subdural (SDH), intraparenchymal hematoma (IPH) and/or cortical contusion (CC). Patients with spontaneous hemorrhages, penetrating injury, non-accidental trauma or craniotomy prior to a second CT were excluded. All patients had an initial head CT scan within 12 hours of injury and at least one repeat CT scan within 72 hours from injury. An attending pediatric neuroradiologist reviewed each CT scan for lesion type, the volume of each hematoma was then determined using the ABC/2 method and compared to the baseline measurement. Hematoma progression was considered significant if the volume percent change from baseline was greater than 33%.
58 children met inclusion criteria, from which 181 hematomas were identified, Overall, 55/118 (46%) of hematomas increased to the threshold significance value of >33% from baseline. There was a minor difference in the percentage of hematoma types which increased >33% with cortical contusions increasing the most frequently and epidural hematomas the least, CC 24/42 (57%), IPH 11/23 (47%), SDH 13/37 (44%), EDH 7/16 (35%). We next examined whether the presence of hypotension or hypoxia (within 8 hours of injury) or skull fracture increased the risk for hematoma expansion. Combining all hematoma types, there was no association between hypotension, hypoxia or skull fracture with increase in hematoma size. However, Fisher's Exact Test showed a significant association between skull fracture and increase in the size of cortical contusion hematomas (p=0.004). There are 31cortical contusion patients with a skull fracture, 71% of those patients had a >33% increase in size within 72 hours. To examine the effect of hematoma expansion on functional outcome, we examined candidate predictors and 12–18 month neurologic outcome using the Glasgow Outcome Scale Extended-Pediatrics (GOSEP). Candidate predictors were field GCS, hypotension, hypoxia and >33% increase in hematoma size. Both forwards and backwards Stepwise methods showed that hypotension (p=0.004) and >33% increase in hematoma size (p=0.03) were significant predictors for outcome.
In this study of children with moderate and severe TBI we found a surprisingly high incidence of hematoma expansion after TBI. Our findings are summarized by the following: 1) cranial hematomas increased in size >33% in 35–57% of lesions examined, 2) hypoxia and hypotension within 8 hours of injury did not appear to increase the risk for hematoma expansion 3) cortical contusions associated with skull fracture had the highest risk for expansion and 4) hypotension and >33% increase in hematoma size are significant predictors for neurologic outcome. These findings need to be supported in larger samples, however this study adds to the evidence based practice to support safe and judicious use of head CT scans in children after TBI.
This work was supported by the Perot Brain and Nerve Injury Center at Children's Medical Center Dallas.
TBI, Hematoma, Expansion, Children, Outcome
TIME DEPENDENT CHANGES IN SERUM BDNF LEVELS IN TRAUMATIC BRAIN INJURY PATIENTS
Hani Seoudi, MD, Department of Trauma Services, Inova Fairfax Hospital
Mr. Sameer Yousuf, BS, George Mason University
Anne Rizzo, MD, Department of Trauma Services, Inova Fairfax Hospital
Margaret Griffen, MD, Department of Trauma Services, Inova Fairfax Hospital
James Ecklund, MD, Department of Neurosciences, Inova Fairfax Hospital, Inova Health System
New molecular targets for treating traumatic brain injury (TBI) need to be identified. Preclinical models of TBI suggest a reduction in brain-derived neurotrophic factor (BDNF) following injury. In healthy humans a diurnal variation in plasma BDNF has been reported. This study examines time-dependent changes in BDNF following TBI.
Subjects: Seven adults (4 males, 3 females) were recruited for this pilot study. Ages ranged between 39 to 73 years. Five individuals met inclusion criteria for the TBI group having evidence of intracranial injury on computed tomographic (CT), and two met inclusion criteria as injury controls having no intracranial abnormality on CT. Enrollment occurred within 24 hr of injury. Protocol: Blood samples from each subject were collected at time of enrollment and every 6 hrs afterward for a total of eight samples in 48 hrs. To ensure accurate assessment of diurnal changes, the time of day was recorded for each sample. Blood samples were drawn and processed immediately, aliquoted, and stored at −80 °C. BDNF assay: Plasma and serum levels of BDNF were determined by ELISA. BDNF concentrations were determined from a BDNF standard curve.
Consistent with previous reports, BDNF levels in our serum samples were approximately 50-fold higher than plasma levels. This likely reflects differences in the cellular pool of BDNF released during coagulation (most probably arising from platelets). Because of higher serum BDNF values, our results were reported using serum samples. Four of seven clinically enrolled patients had successful blood collection every six hrs over 48 hrs. The other three enrolled patients had at least one of their collections missed because of competing clinical priorities, and were excluded from analysis. Two of the four patients analyzed had TBI and two were control subjects. The samples were matched based on age and sex (set 1, females, ages 53 and 71 yrs; set 2, males, ages 39 and 45 yrs). There were four major findings: 1.) Serum BDNF levels showed diurnal variation in both TBI and control groups. 2.) Consistent with previously published studies, serum BDNF tended to be lower in older women than younger males. 3.) Aberrant diurnal regulation of serum BDNF levels compared to control was noted with TBI in the female patients; however, diurnal regulation of BDNF was altered in the control but not the TBI patient in the males. 4.) For both male and female subjects, serum BDNF levels were decreased in the patients with TBI relative to control. For the female patients the time of initial sampling was identical, thus eliminating diurnal variation as a confounding variable. In the female the reduction in BDNF levels in the TBI patient was 3.1 fold at 6AM which is a time when BDNF has been reported to increase in normal individuals. This finding was statistically significant (p=0.003) despite the small sample size.
Our findings show, for the first time, a time course of serum BDNF levels for a sample of TBI patients and non-TBI injury controls. Because BDNF accumulation is under diurnal regulation, similar to cortisol, serum BDNF levels need to be monitored at specific times in the day-night cycle. These preliminary findings suggest that serum BDNF levels may be reduced in TBI. The explanation for the gender differences seen in diurnal cycling of BDNF levels in our samples is unclear but may be explained by the small sample size. Further research is needed to determine the utility of serum BDNF as a surrogate marker for TBI.
Sponsor: Inova grant (Dr. Seoudi) and intramural funds from the Department of Neurosciences. We thank Abigail Novak for her assistance with the BDNF assays.
BDNF, biomarker, human, TBI, serum
COAGULOPATHY AS A PREDICTOR OF EXACERBATION IN PATIENTS WITH MILD-TO-MODERATE TRAUMATIC BRAIN INJURY
Hiroyasu Koizumi, M.D., Ph.D., Department of Neurosurgery, Yamaguchi University School of Medicine
Hirosuke Fujisawa, M.D., Ph.D., Department of Neurosurgery, Yamaguchi University School of Medicine
Michiyasu Suzuki, M.D., Ph.D., Department of Neurosurgery, Yamaguchi University School of Medicine
We have reported previously that an acute operation due to exacerbation was required in 8.3% of patients with mild-to-moderate traumatic brain injury (TBI) and that the outcome was poor in 11.5%. Therefore, careful observation is required in patients with mild-to-moderate TBI after admission.
In this study, we examined the factors associated with aggravation in 96 patients with a single mild-to-moderate TBI (Glasgow Coma Scale score at admission 9–15) admitted to our hospital from September 2008 to October 2010. Patients were divided into a surgery group (n=8, 8.3%) and a conservative group (n=88, 91.7%) based on the performance of acute surgery due to exacerbation, and into groups with favorable (Glasgow Outcome Scale (GOS) score 4–5; n=85, 88.5%) and poor (GOS score 1–3; n=11, 11.5%) outcomes based on GOS at discharge. Age and blood test results (AST, ALT, γ-GTP, blood platelets, PT-INR, APTT, fibrinogen, FDP, D-dimer) at admission were compared between these groups.
Patients in the surgery group had significantly higher AST (74.9 vs. 32.1 U/l), APTT (38.0 vs. 28.9 sec), FDP (112.9 vs. 34.9 μg/ml), and D-dimer (69.6 vs. 18.9 μg/ml) compared to those in the conservative group. Patients with a poor outcome were significantly older (73.5 vs. 56.7 years old) and had significantly higher APTT (36.3 vs. 28.9 sec), FDP (112.5 vs. 30.8 μg/ml) and D-dimer (67.4 vs. 17.5 μg/ml) and a significantly lower blood platelet count (18.8 vs. 25.0 104/μl) compared to patients with a favorable outcome.
These results indicate that careful observation is required for patients with mild-to moderate TBI who are elderly and have coagulopathy and a low platelet count.
nothing
coagulopathy, exacerbation, mild, moderate, TBI
LETHAL CONSEQUENCES OF MISSED TBI DIAGNOSIS: EXECUTION BY WORLD WAR I FIRING SQUAD FOR TBI SEQUELAE
Bruce Capehart, MD, MBA, Durham VA Medical Center and Duke Univ Medical Center
Garrett Wood, BS, Duke University
Cameron R. Dale Bass, PhD, Duke University
Blast neurotrauma has become a principal source of long-term morbidity in US service members. Owing to the nature of improvised blast exposures, unexpected with unknown charge size and placement, current military field epidemiology is difficult to analyze to provide guidance on the etiology of injury, biomechanics and clinical practice.
Case reports from World War I offer a potential epidemiological laboratory with relatively controlled exposure conditions: known artillery and mortar types and firing rates per battle, personnel generally confined to trenches with overhead blasts. WWI presented a unique challenge to military medical clinicians and commanders for its introduction of high explosives and widespread static trench warfare that likely contributed to frequent occurrence of blast injuries. A case report review was conducted of blast brain injuries to understand the historical role of blast in neurotrauma and its implications for modern clinical diagnosis and TBI research. The current study focuses on a series of British army soldiers who were executed for various offenses including desertion and cowardice using contemporary understanding of clinical and biomechanical knowledge on blast induced TBI, including a newly published blast neurotrauma risk assessment in gyrencephalic animals, to assess the presence of blast induced TBI.
Using modern TBI diagnostic criteria and engineering analysis, we determined that WWI-era physicians overlooked blast injury in general and that there is clear clinical and biomechanical evidence to suggest that blast TBI was a common factor contributing to atypical behavior following primary blast exposure. Estimated blast exposures in the case population ranges from less than 100 kPa to 1000 kPa peak overpressure with positive phase durations from 1 ms to approximately 8 ms for the largest German mines and mortar charges. Neurotrauma and pulmonary trauma risk assessments combined with clinical results suggest that at least some British Expeditionary Force (BEF) soldiers had moderate/severe blast exposure. These results indicate a strong probability that British army soldiers during WWI were executed for wartime criminal acts that, using our present understanding of clinical neuroscience, might have been attributed to behavioral sequelae of TBI. Comparison of these WWI case histories with current military scenarios demonstrates that personnel are exposed to injury mechanisms that are similar to those seen in WWI with similar physical blast exposures and similar behavioral outcomes. When resulting injuries are compared with neurotrauma risk assessments developed in a gyrencephalic animal model that have been scaled to humans, the occurrence of injuries is consistent with the scaled risk functions.
The results of this study have significant implications in the assessment of blast injury risk and in the treatment of service members following blasts. The case series emphasizes the potential for extensive behavioral changes after blast exposures that are similar to anecdotal findings in current US service members. This study also provides epidemiological support and a link to human scaling for a new blast neurotrauma risk assessment developed in ferrets. These findings emphasize the strong need for interdisciplinary teams to investigate blast pathophysiology including collaborative efforts between clinicians and scientists and engineers including clinical psychiatrists, psychologists, biomechanical engineers, blast biomechanists and military historians.
We gratefully acknowledge support, in part, by the Multidisciplinary Research Initiative (MURI) program (W911MF-10-1-526; University of Pennsylvania as prime institution) through the Army Research Office.
blast, epidemiology, traumatic brain injury
QUANTIFICATION OF LONGITUDINAL BRAIN CHANGES IN HEAD INJURY PATIENTS USING NONLINEAR REGISTRATION
Navid Shiee, PhD, Henry M. Jackson Foundation
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
John Butman, MD, PhD, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
Dzung Pham, PhD, Henry Jackson Foundation/CNRM Center for Neural and Regenerative Medicine
To detect and quantify regional brain volume changes following traumatic head injury, we computed volume changes within gray matter, white matter and the ventricles between volumetric MRI acquired within 48 hours and 1 week after head injury, using two methods: nonlinear registration and segmentation.
Thirty head injury patients from the Center for Neuroscience and Regenerative Medicine (CNRM) THINC (Traumatic Head Injury Neurioimaging Classification) study received MRI within 48 h and 1 w following injury.
For each subject, 3D T1-weighted IR SPGR MRI images (∼1mm cubic resolution) at the two time points were corrected for intensity inhomogeneity, segmented (gray matter (GM), white matter (WM), sulcal CSF, ventricles, skull) and rigidly registered using MIPAV (http://mipav.cit.nih.gov/) and in-house developed software. Images were then nonlinearly registered using the ANTS toolkit (http://www.picsl.upenn.edu/ANTS/), with cross-correlation (radius=5) and symmetric normalization diffeomorphism (gradient radius=3) used for the similarity and transformation models. A local measure of volume change (determinant of the Jacobian of the transformation) was calculated between the registered images and averaged within GM, WM and ventricles. Volume changes were also calculated using only the segmentation. Comparison was made with 5 control datasets from the Multi-Modal MRI Reproducibility Resource (http://www.nitrc.org/projects/multimodal/).
Using the nonlinear registration approach, the ventricles exhibited an average local expansion of 6.3% (s.d. +/− 5.5%; max/min +18.5/-3.5%), WM increased by 0.5% (+/- 0.8%; +2.4/-1.8%) and GM decreased by 1.1% (+/- 0.5%; +0.2/-1.2%) in the head injury group. In controls, the average volume of the ventricles increased by 0.2% (+/- 0.5%; +1.1/-0.2%), WM decreased by 0.2% (+/- 0.3%; +0.3/-0.6%) and GM decreased by 0.1% (+/- 0.3%; +0.2/-0.6%). A preliminary analysis examining a subset of 24 subjects with Neurobehavioral Symptom Inventory (NBSI) score at 90 days post-injury yielded a significant correlation between the change in ventricular volume and NBSI using Spearman's rank correlation (Spearman r=0.4, p=0.05).
As a comparison, volume changes computed using segmentation alone was less stable across patients. The average volume of the ventricles decreased by 1.1% (+/− 6.4%; +14.2/−13.1%), WM increased by 5.2% (+/− 5.1%; +15.0/−8.1%) and GM decreased by 0.4% (+/− 4.2%; +11.4/−7.5%).
Changes in volume computed from nonlinear registration of anatomical images identified expansion of the ventricles together with a slight contraction of GM in many of the patients. White matter volume showed little change. This pattern was not identified when volumes were computed by performing a tissue class segmentation at both time points. Therefore, a combination of nonlinear registration and segmentation may be more sensitive in identifying local volume changes as compared with segmentation alone.
The changes in brain volume following traumatic head injury demonstrated by the nonlinear registration approach may be interpreted as resolution of brain edema, resolution of space occupying lesions (e.g. hematomas) and/or loss of brain tissue. Preliminary results suggest a correlation with clinical measures.
This work was supported by CNRM and the NIH Intramural Program. The authors gratefully acknowledge the investigators of the THINC study.
Longitudinal volume change, MRI, TBI
NEUROTOXICITY BIOMARKERS ASSAYS FOR MILD TRAUMATIC BRAIN INJURY
Neurological complications after brain injury are initiated by metabolic disturbances, one of which is glutamate-induced neurotoxicity (excitotoxicity). The term excitotoxicity is related to increased release and impaired uptake of excitatory amino acids, such as glutamate or aspartate leading to over-stimulation of glutamate receptors, which causes neuronal injury.
How, when, and which specific glutamate receptor biomarker is released from the CNS into peripheral fluids in response to concussions or mild injury is explained using a neurodegradomics approach. Multi-systemic mechanisms of neurotoxic coupling, microvascular dysfunction, and immune responses that result in primary (acute) and secondary (chronic) events are hypothesized. The release of specific glutamate receptor biomarkers from the subtle brain injury into peripheral fluids is demonstrated.
Several clinical research data concerning the possibility of post-concussion effect assessment in club sport athletes with semi-acute and chronic concussions, persons with mild, moderate, and severe TBI are presented. The focus of the presentation is on detection of brain-based biomarkers to improve diagnostic certainty of TBI in conjunction with radiological and clinical findings in civil and military settings.
Biomarkers of neurotoxicity may assist in the assessment of subtle, or asymptomatic, ischemic, hemorrhagic, or axonal injuries associated with abnormal spiking activity. Timely assessment of concussions is essential for optimal management of athletes and prevention of neurological complications. Clinical use of blood tests that can reliably predict consequences, or forecast recovery and outcome, might help in selection of the preventive therapy for improving acute or subacute health outcomes.
R. Sowell, A. Shikuev, S. Gill, L. St. Onge (Kennesaw, GA, USA), A. Skoromets, T. Skoromets, M. Odinak, D. Skuliabin (St. Petersburg, Russia)
Concussions, neurotoxicity biomarkers, blood assays
DEVELOPING A TBI CLINICAL TRIAL INFRASTRUCTURE AT UNIVERSITY OF MIAMI/JACKSON MEMORIAL HOSPITAL: REASON FOR ENROLLMENT FAILURE
Zsuzsanne Nemeth, Clinical Research Coordinator, University of Miami
Jose Sanchez-Chavez, MD, University of Miami/Jackson Memorial Hospital
Rebecca Safon, RNS, University of Miami
Vianca Cabrera, AA, University of Miami
Tony Perez, AA, University of Miami
Annette Amoros, AA, University of Miami
Johanatan R Jagid MD, MD, University of Miami/Jackson Memorial Hospital
Kristine O'Phelan, MD, Illinois State University
Leo Harris, PA, University of Miami/Jackson Memorial Hospital
Stephen Olvey, University of Miami/ Jackson Memorial Hospital
Professor Ross Bullock, MD, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Traumatic brain injury (TBI) is the major cause of death and disability in the USA and all over the world. Clinical care improvements have reduced mortality and morbidity. However, no new therapy has been FDA approved for TBI. The Ryder Trauma Center of JMH, is the only level I trauma center, for South East Florida, a compact urban area, with around a population of 3.5 million population, and should contribute to Clinical TBI trials. Only 5.3% of TBI admissions have been successfully enrolled into trials, and the reason are presented here.
Of a total of 849 TBI patients screened, we enrolled 46 patients (5.3 %) in our clinical trials (Fig. 1): The enrollment by studying is shown in Fig. 2; Cosbid (17; 36.9%), Banyan (12; 26%); BOOST (10; 21%) and INTREPID (7; 15.2%). In this group of TBI patients adults male were predominant (Fig. 3 and 4). We observed that the three major causes of the TBI were: motor vehicle and traffic related accidents (52%), fall (30%), and assault (16%) (Fig. 5). Ranked by the severity of TBI the majority of patients had either mild (43.5 %) or severe injuries (41%), while only 15% had moderate injuries (Fig. 6). In Fig. 7 is shown the distribution by brain damage, subdural hematoma being the most frequent, following by subarachnoid hemorrhage and the combination of both.
Factors that made the enrollment of patients difficult in our studies are ranked below: 1.- Improvement in patients'neurological clinical condition; 2.- No proxy available within the window for enrollment; (About 12% of patients were from outside the United States and arrived >12 hours post injury); 3.- Admission/screening outside time window (7 and 12 hours) for randomization; 4.- Notification failure about patients after hours and on weekends; 5.- Refused consent was not a significant factor of not enrollment in our studies (<1%).
We gratefully acknoledge the contributions and support that were made by all the personnel of Ryder Center and Neurocritical Care Unit.
TBI, neurocritical care, clinical trial
TEMPORAL CHANGES IN HYPOINTENSE LESIONS ON T2* WEIGHTED IMAGING
Katherine J. Williams, ScB, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Marcus T. Dean, BA, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
In patients with mild TBI the presence of hypointense regions on SWI is a common pathology. In some subjects, it is unclear if microbleeds are analogous to stroke or incidental findings. In this study, T2* weighted imaging is used to identify and track microbleeds in patients who have head injury.
Subjects enrolled in the CNRM THINC study who had positive findings on T2* were included provided subjects had at least one follow-up scan and less than 5 hypointense lesions on imaging. Images from T2*, FLAIR, and 3DT1 were reviewed for the presence of two types of foci: microbleeds and/or what we term “linear hypointensities” (defined below). Only those patients who had more than one time point were included. For this study, microbleeds are defined as hypointense lesions less than 0.5 cm in diameter and what we term “linear hypointensity” is defined as a non-spherical hypointense area on T2* with a major axis at least twice as long as the smallest minor axis. To eliminate voxel averaging, if the major axis falls along the axial direction, the major axis must be longer than the axial voxel dimension, in this case 3.5 mm.
Microbleeds and/or linear hypointensities were found in 31 of 149 (21%) subjects, of which 20 of 149 (13%) were included, who presented with head injury between October 2010 and February 2012 at Suburban Hospital (Bethesda, MD) and Washington Hospital Center (Washington, DC). Microbleeds were seen in 15 (75%) patients, 14 (70%) have linear hypointensities, and 9 (45%) have both.Volume measurements of the lesions were taken using the xyz/2 method where the cardinal axes were aligned along the principal x, y, z directions for measuring the microbleeds and aligned with the major axis for measurement of the linear hypointensity. There were a total of 16 microbleeds ranging in volume from 1.57 mm3to 45.7 mm3 while the total of 28 linear hypointensities ranged from 5.3 mm3 to 566.4 mm3. For both types of lesion there is a general trend to reduce in size with increasing time. In particular, the larger the hemorrhage, the more quickly it will reduce in volume. However, 25% of microbleeds and 11% of linear hypointensities have a later time point volume greater than that of the initial time point. In patients with at least one linear hypointensity, 60% of those also had at least one microbleed. The decrease in volume over time for both microbleeds and linear hypointensities suggests that the appearance of these features are due to the initial trauma and not a prior pathology of the brain. Microbleeds have been reported in elderly patient population, stroke, and those with amyloid angiopathy on conventional T2* weighted GRE. More sensitive susceptibility weighted imaging has been used to detect similar appearing microhemorrhages in children and adults with traumatic brain injury. It is not clear if the two pathologies differ. Our findings suggest that there exists a temporal association in the head injured patients based on the change in size and linear appearance over time.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
hypointense lesion, T2*WI
EFFECTIVE AND EARLY PREDICTIVE INDICATORS OF TALK AND DETERIORATE FOLLOWING TRAUMATIC BRAIN INJURY
Koichi Hayakawa, M.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Tadahiko Shiozaki, M.D., Ph.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Jiro Iba, M.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Hiroshi Ogura, M.D., Ph.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Yasuyuki Kuwagata, M.D., Ph.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Osamu Tasaki, M.D., Ph.D., Emergency Medical Center, Nagasaki University Hospital
Takeshi Shimazu, M.D., Ph.D., Department of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine
Expansion of traumatic intracranial hemorrhage is one of the most important causes of talk and deteriorate in patients with TBI. To establish effective and early predictive indicators of hematoma expansion and talk and deteriorate, we investigated extravasation on multidetector-CT angiography (MD-CTA) and hemostatic abnormalities in patients with TBI.
One hundred fourteen patients with traumatic intracranial hemorrhage were included in this study, who were admitted to our Critical Care Medical Center from January 2003 to March 2012, could talk at admission (with their best verbal response 3 or more on the Glasgow Coma Scale (GCS)), and underwent MD-CTA within 3 hours after admission. We evaluated extravasation on MD-CTA, and coagulation and fibrinolysis as predictors of hematoma expansion and talk and deteriorate following TBI. We defined talk and deteriorate as the case in which the best verbal response on GCS score was 3 or more at admission, but required surgical treatment due to deterioration of consciousness.
Out of 114 patients, extravasation was observed in 23 patients (20%). Twenty-one of 23 patients (91%) showed hematoma expansion and 16 of 23 patients (70%) deteriorated due to increased hematoma. Out of the other 91 patients without extravasation on MD-CTA, 7 patients (8%) deteriorated due to hematoma expansion (Group 1), 17 patients (19%) had expanded hematoma without deterioration (Group 2). Hematoma did not increase in other 75 patients (82%, Group 3). Patients, who deteriorated without extravasation (Group 1), had significantly higher plasma d-dimmer level on admission, compared to patients who did not deteriorate (Group 1 and 2). (Group 1 vs 2 vs 3: 86.3±52.0 μg/mL vs 33.0±33.9 μg/mL vs 13.3±14.8 μg/mL).
Extravasation on MD-CTA is an effective and early predictive indicator of hematoma expansion and ‘talk and deteriorate’ following TBI. D-dimmer is also an effective and early predictor of hematoma expansion and ‘talk and deteriorate’ in patients without extravasation following TBI.
No sponsor
extravasation, multidetector-CTA, d-dimmer, talk-and-deteriorate
DATA COLLECTED FOR RESEARCH PURPOSES PROVIDES SUPPORT FOR CLINICAL PATHWAY FOR CONCUSSION SCREENING IN ED
Genevieve Jacobs, BS, Suburban Hospital Johns Hopkins Medicine
Jessica L. DeStefano, BS, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Martin R. Cota, BA, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
Melissa Meyers, MSN, Suburban Hospital Johns Hopkins Medicine
Veronica Sudekum, MA, VCU
During the course of a research study on acute head trauma, it was noted that some subjects did not receive a diagnosis of concussion in the emergency department, despite meeting clinical criteria. Using data gathered for research purposes, we estimate the disparity between presentation and clinical diagnosis.
Patients with acute head injury presenting to Suburban Hospital (Bethesda, MD) trauma/ED were enrolled in the THINC study. Data obtained as part of standard care included Glasgow Coma Scale (GCS), findings on CT, and diagnosis. Free-text clinical diagnosis was dichotomized into concussion yes/no and hemorrhage yes/no. Data collected as part of study procedures from subject and clinical staff interviews included Loss of Consciousness (LOC) and Post Traumatic Amnesia (PTA). Altered mental status was not available. As a comparison, a “research diagnosis” of concussion was imputed based on positive LOC or PTA, of any duration, or GCS<15.A “research diagnosis” of hemorrhage was inferred from the free-text radiology report.Values reported are median (IQR).
During the 14-month period from October 2010 to March 2012, 108 subjects were enrolled in the study. The age was 45(30–61), 37 (34%) were female. Time from injury to triage was 41(19–75) minutes, time from triage to ED discharge was 4.8(3.0–6.9) hrs. 88(81.5%) were admitted to the hospital. 58 (53.7%) were admitted to the floor, 26(24.1%) were admitted to ICU, 2 (1.9%) went direct to OR from ER and 17 (15.7%) were discharged home. 106 (98.2%) received CT during ED admission. Of the 108 subjects, 32 (29.5%) had a diagnosis of concussion. Of the 32, 27 (84.4%) were assigned a research diagnosis of concussion based on +LOC (n=25), +PTA (n=15), or GCS<15 (n=6). Of the 76 (70.5%) without the diagnosis concussion, 52 (68.4%) were assigned a research diagnosis of concussion based on +LOC (n=43), +PTA (n=31), or GCS<15 (n=10).
Adding clinical CT findings to the criteria, 63 (58.3%) of 108 subjects received neither a diagnosis of concussion nor diagnosis of intracranial hemorrhage. Of these subjects, 10 (16%) had positive CT findings while 42(67%) had concussion based on +LOC (n=34), +PTA (n=23), or GCS<15 (n=8). The majority of the subjects studied had some degree of polytrauma, not isolated head injury, which contributed to the decision to admit to the hospital or discharge home from the ED. Of 17 subjects that were discharged directly home without a clinical diagnosis of concussion, 7 (41% of 17, or 6% of 108 studied) subjects met criteria.
While awareness of concussion has rapidly increased in recent years, the optimal clinical pathways for triaging and diagnosing head injury in the ED that is not life threatening continues to slowly evolve. Data collected for research purposes indicates that more than half the patient who met the criteria for concussion did not have a clinical diagnosis of concussion in their charts. Many of these patients were not immediately discharged but rather admitted and monitored, some for injuries other than concussion. Regardless, anecdotal cases indicate that some patients with concussion are discharged from the hospital without an appreciation for their injury or the potential sequel, later return with worsening symptoms resulting in re-evaluation. Recognition of these issues has lead to a quality improvement effort and implementation of a Concussion Screening Tool to identify the injury, document the diagnosis, and provide appropriate discharge instructions to the patient.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
THINC TBI, Screening, Concussion, Diagnosis
EXTERNAL VALIDATION AND COMPARISON OF PREDICTION MODELS FOR TRAUMATIC BRAIN INJURY USING A LARGE SINGLE-CENTER TBI DATABASE
Erin Dienes, MS, UC Davis
Nancy Rudisill, RN, MSN, CN IV, UC Davis Medical Center
Karen Smith, RN, UC Davis Medical Center
Marike Zwienenberg-Lee, MD, UC Davis Medical Center
Professor Paul Muizelaar, MD, PhD, UC Davis
Kiarash Shahlaie, MD, PhD, UC Davis Medical Center
TBI outcome prediction models have been developed from combinations of clinical and/or radiographic data on admission. Some models are based on small sample sizes and lack external validation, while larger studies frequently draw from clinical trials. These heterogeneous populations limit the utility of a side-by-side comparison of accuracy.
We compared existing models that have been published using a single center independent database to determine which previously developed TBI prediction model provides the most accurate outcome prediction in the setting of a tertiary care Level 1 Trauma Center. Data was prospectively gathered from all patients with head injury that required neurosurgical consultation from October 2008 until May 2011. demographic, clinical, laboratory and radiographic data points were collected prospectively on 1332 TBI patients. These data points were used to externally validate 10 previously published prognostic models. Both decision tree and logistic regression analyses were included. For each model, Receiver Operating Characteristic curves were generated using our database and the area under the curve was compared between each to determine relative accuracy. Predictive accuracy was compared at 3 and 6 month time intervals, both for survival and dichotomized outcome scores (“good prognosis” versus “poor prognosis”).
In both types of outcome prediction (survival and “good” versus “poor” prognosis), and at both 3 and 6 months, models from Hukkelhoven et al, Combes et al, Signorini et al and Steyerberg et al had higher predictive accuracy (78–90%) in severe, moderate, and mild TBI. All of these models were logistic regression models, suggesting that this may be a more accurate way of predicting outcome. Of the decision tree prediction models, Choi et al was significantly more accurate than Maas et al.
While most of these models were developed for severe TBI, they are also fairly accurate at predicting outcome in moderate and mild TBI. Of the numerous TBI prediction models available, logistic regression may provide a more accurate means of providing prognosis in TBI. Acknowledgements: We would like to acknowledge the UC Davis Clinical and Translational Science Center for their assistance with statistical analysis.
TBI, prognosis, prediction model
THE ALTERATION OF INTRACRANIAL PRESSURE AND IMAGING FEATURES AFTER DECOMPRESSIVE CRANIECTOMY WITH LATTICE DURAPLASTY
John Diaz Day, M.D., University of Arkansas For Medical Science
To investigate the alteration of intracranial pressure and imaging features after decompressive craniectomy with lattice duraplasty in patients with severe head injury.
Fifty patients suffered from severe head injury with brain swelling were operated using Amercian standard large trauma craniotomy, the lattice duraplasty technique was applied intraoperatively. The pre- and post-operative ICP and imaging features were observed and recorded, followed by a statistical comparative study.
The preoperative ICP was 37.6±7.9mmHg, the midline shift was 11.7±3.8mm, the patients with open ambient cistern were 3 cases. The postoperative ICP reduced to 14.1±6.3mmHg,the midline shift decreased to 4.6±2.7mm, and the patients with open ambient cistern were 31cases. Compared with preoperative data all postoperative data were improved significantly (P<0.01).
The technique of lattice duraplasty used in decompressive craniectomy could reduce ICP and midline shift, and alleviatee the ambient cistern compression.
We thank all the staff members in the Department of Neurosurgery, The affiliated Jiangyin Hospital southeast university school of medicine, China, for their outstanding contribution.
duroplasty, decompressive cranioectomy, intracranial pressure
ASSESSING A NECROSIS AND APOPTOSIS INDEX USING CSF αII-SPECTRIN BREAKDOWN PRODUCT LEVELS IN SEVERE TBI
Steven Robicsek, MD, PhD, University of Florida
Claudia Robertson, MD, Baylor College of Medicine
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Andrea Gabrielli, MD, University of Florida
H. Julia Hannay, PhD, University of Houston
Shelley Heaton, PhD, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Linda Papa, MD, Orlando Regional Medical Center
αII-spectrin breakdown products SBDP150 and SBDP145 produced bycalpain are indicative of early necrosis phase of TBI whereasSBDP120 generated by caspase-3 is indicative of delayed apoptosis. Examination of whether the sum of total SBDPs could be used to assess the proportioncell death from necrosis versus apoptosis.
This prospective observational study recruited adults with severe TBI from 2 Level I Trauma Centers. Patients were included if they had a blunt head injury with a GCS score of ≤8 and required intracranial pressure monitoring. Ventricular CSF was sampled from each patient over 240 hours after injury and analyzed for spectrin breakdown products SBDP150, 145 and 120 using sandwich ELISA and measured in ng/ml. The proportion of apoptotic and necrotic biomarkers were calculated for each time point post-injury.
There were 131 patients enrolled in the study. Mean age of patients was 38 years and 78 were male; 68 had CT Marshall Classification of Diffuse Injury I-IV and 32 were mass lesions. Over the whole time course, there was a possible correlation between SBDP150 and SBDP145 levels (r 2=0.58 (p<0.001)). In contrast, there is an inversecorrelation between SBDP150 and SBDP120 levels (r 2=0.27(p<0.001)), and between SBDP145 and SBDP120 levels (r 2=0.23(p<0.001)). The percent of total SBDP from SBPD120 over time was 5 at 6 h, 3 at 12 h, 3 at 18 h, 3 at 24 h, 4 at 48 h, 5 at 72 h, 9 at 96 h, 8 at 120 h, 11 at 144 h, 13 at 168 h, 14 at 192 h, 16 at 216 h, and 13 at 240h. This data suggest that cell death mode starts switching towards apoptosisbeginning at about 96 h and sustained for at least 240 h.
The total SBDPs levels in CSF after severe TBI can be used as an index for total cell death load over time. In addition, The total SBDP from (SBDP150 SBDP145) vs. that from SBDP120 might be used as a index for necrosis vs. apoptosis at a given time point after TBI.
This study was funded by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
biomarker, TBI, necrosis, apoptosis
PREDICTING POOR 6-MONTH OUTCOME IN SEVERE TBI USING THE IMPACT SCORE AND EARLY CSF BIOMARKERS
Linda Papa, MD, Orlando Regional Medical Center
Shelley Heaton, PhD, University of Florida
H. Julia Hannay, PhD, University of Houston
Andrea Gabrielli, MD, University of Florida
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Ronald Hayes, PhD, Banyan Biomarkers
Claudia Robertson, MD, Baylor College of Medicine
The value of early levels of neuronal biomarkers as measured in ventricular CSF within 24 hours of injury was assessed to determine the clinical prediction of poor outcome at 6 months in patients with severe TBI.
This study was designed as prospective observational study that enrolled adults with severe TBI presenting to two Level 1 trauma centers. Patients were included if they had a blunt head injury with a GCS score of ≤8 and required a ventriculostomy. Ventricular CSF was sampled from each patient within 24 hours of injury andanalyzed for candidate neuronal biomarkers using ELISA. Levels of S100B, MAP-2, UCH-L1, SBDP-120, SBDP-145 and SBDP-150 were measuredin ng/ml. An IMPACT score (www.impact-net.org) was calculated for each patient to determine their risk of poor outcome at 6 months. Outcome was defined by the traditionaldichotomy of the GOS score (death, vegetative state and severe disability versus moderate disability and good recovery). ROC curves were constructed to determine if the biomarkers added prognostic value to the Core IMPACT score (age, pupil reactivity and GCS motor score), Extended IMPACT score (core hypoxia, hypotension, CT findings), Lab IMPACTscore (extended glucose and hemoglobin).
There were 131 patients enrolled in the study. Mean age of patients was 38 years and78 were male; 68 had CT Marshall Classification of Diffuse Injury I-IV and32 were mass lesions. At 6-months, 104 patients had follow-up data and, ofthese, 84 had CSF collected for biomarker analysis within 24 hours of injury. Fifty-five (66) patients had poor outcome per GOS at 6 months. The area underthe ROC curve (AUC) for the Core, Extended and Lab IMPACT models for predictingpoor outcome at 6 months were 0.73 (0.62–0.85) for Core, 0.76 (0.65–0.87) forExtended, 0.77 (0.65–0.88) for Lab. The AUC for UCH-L1 drawn within 24 hours ofinjury was 0.74 (0.63–0.85). However, when the UCH-L1 was added to the LabIMPACT model, the AUC for predicting mortality increased to 0.80 (0.71–0.90).
These data suggest that early levels of neuronal biomarkers provide added prognosticinformation about 6-month risk of poor outcome in patients with severe TBI.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
TBI, biomarker, predictive model, IMPACT
A TIME SENSITIVE TRIAL OF PROGESTERONE FOR CHILDREN WITH MODERATE-TO-SEVERE TRAUMATIC BRAIN INJURY (TBI): A FEASIBILITY STUDY IN THE PEDIATRIC EMERGENCY CARE APPLIED RESEARCH NETWORK (PECARN)
Mike Johnson, MD, University of Utah
Cheryl Vance, MD, University of California Davis
Lalit Bajaj, MD, University of Colorado, Children's Hospital Colorado
Lynn Babcock, MD, Cincinnati Childrens Hospital
Shireen Atabaki, MD, Children's National Medical Center
Danny Thomas, MD, Medical College of Wisconsin
Doug Nelson, MD, University of Utah
Angie Webster, MStat, University of Utah
Harold K. Simon, MD, MBA, Emory University School of Medicine
Bema Bonsu, MB ChB, Nationwide Children's Hospital
Daniel Rubacalva, MD, Texas Children's Hospital
P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Blake Bulloch, MD, Phoenix Children's Hospital/University of Arizona
Alexander J. Rogers, MD, University of Michigan
Prashant Mahajan, MD, MPH, MBA, Children's Hospital of Michigan, Wayne State School of Medicine
Jill Baren, MD, University of Pennsylvania
Lois Lee, MD, Children's Hospital Boston
John Hoyle, MD, Helen DeVos Childrens Hospital, Michigan State University
Quayle Kimberley, MD, Washington University School of Medicine
Nathan Kuppermann, MD, MPH, University of California, Davis School of Medicine
In preparation for a future clinical trial of progesterone for TBI, the PECARN and other collaborative institutions conducted a prospective observational study of children with moderate-to-severe TBI to evaluate time of patient arrival after injury, patient accrual, and time of arrival of legal guardians for consent purposes.
Children (ages 0- <18 years) with blunt head trauma and Glasgow Coma Scale (GCS) scores of 3–12 were prospectively enrolled over 6 months (2011–12) at 16 level 1 pediatric trauma center emergency departments (ED). This study was conducted using a waiver of informed consent. We collected data on variables thought to be important to determine eligibility and accrual for a future trial of progesterone for moderate-to-severe pediatric TBI, including: patient demographics; ED GCS scores; presence of hypotension (age-adjusted hypotension for >15 minutes) and/or hypoxia (oxygen saturation <90% for >15 minutes); history of prehospital cardiac arrest with resuscitation; non survivable injury; and death in the ED. In addition data were collected to determine time of arrival of patient and legal guardian at the treating ED.
262 children with blunt head trauma and GCS scores of 3–12 were enrolled over a 6 month period with a range of 5–34 children enrolled per site. 171(65%) had GCS scores of 3–8. Cumulative total ED patient volume during the study period was approximately 1 million visits per year. Median age was 6.5 years (range 0.1–17.9 years). 172 (66%) were male. 128 (49%) were transferred from another ED. Enrolled patients had the following potential exclusion criteria for a future therapeutic trial: 23 (9%) had age-adjusted hypotension; 10 (4%) had hypoxia; 31(12%) had prehospital cardiac arrest with resuscitation; 29 (11%) had non-survivable injuries and 14 (5%) died in the ED. The majority of patients arrived within 2–3 hours of injury; 97 (37%) patients arrived within 1 hour of the time of their injury; 152 (58%) arrived within 2 hours; 188 (72%) arrived within 3 hours; 215 (82%) arrived within 4 hours; 231 (88%) arrived within 5 hours; 244 (93%) arrived within 6 hours and 248 (95%) arrived within 7 hours of injury. Arrival time of legal guardians to the treating ED after the time of injury was: 56 (21%) arrived within 1 hour; 107 (41%) arrived within 2 hours; 140 (53%) arrived within 3 hours; 171(65%) arrived within 4 hours; 196 (75%) arrived within 5 hours; 214 (82%) arrived within 6 hours; and 221 (84%) arrived within 7 hours. Nineteen (7%) took longer than 9 hours after the time of injury to arrive and 9 (3%) of legal guardians never arrived at the treating ED.
We identified important characteristics of potentially eligible patients for a future multicenter, time-sensitive trial of progesterone for pediatric TBI. One half of guardians arrived in the ED within 3 hours of injury; most were present by 7 hours. Guardian arrival time defines the window in which parental permission for a clinical trial could be obtained at the study ED and may be an important consideration when defining enrollment window for a future trial of progesterone for TBI and to determine the need for potential Exception From Informed Consent (EFIC). Enrolling children into a future time-sensitive trial of progesterone for TBI poses challenges regarding timing of patient and guardian arrival.
PECARN is supported by cooperative agreements U03MC00001, U03MC00003, U03MC00006, U03MC00007, and U03MC00008 from the EMSC program of the MCHB and HRSA.
Progesterone, pediatric trial, head injury
INTERACTIONS WITH BDNF GENE VARIATION AND AGE IN MORTALITY FOLLOWING SEVERE TBI
Christian Niyonkuru, MS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Joelle Scanlon, PhD, University of Pittsburgh
Sandra Deslouches, BS, University of Pittsburgh
Robert Ferrell, PhD, University of Pittsburgh
Yvette Conley, PhD, University of Pittsburgh
Amy Wagner, MD, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Brain-derived neurotrophic factor (BDNF), a ubiquitous neurotrophin important for synaptic plasticity and neurogenesis, likely plays a role in recovery post-traumatic brain injury (TBI). Variation within the BDNF gene has been associated with mortality in other clinical populations and may inform predictions of mortality following severe TBI.
This study evaluated genetic variation within the BDNF gene, specifically the functional variant RS6265 (val66met) and the tagging single nucleotide polymorphism (tSNP) RS7124442 in relation to mortality in 349 persons with moderate to severe TBI. All patients were genotyped and mortality was examined during two time-epochs, acute (within the first week post-injury) and post-acute (post-1 week following injury). The predictive ability of genetic variation was examined in association with demographic variables such as age (both continuous and as a categorical variable, split at age 45), gender, and pre-existing conditions (e.g. diabetes, cardiac) as well as with injury severity (Glasgow Coma Score (GCS)), mechanism of injury, and post-injury complications.
Age was a significant predictor of mortality in both acute and post-acute mortality analysis. Variation within the tSNP RS7124442 predicted mortality in both the acute or post-acute phase (p=0.001), such that carriers of the tSNP variant were more likely to die during the acute phase compared to non carriers. The age interaction with the genetic variant showed unique associations in younger (n=232) compared to older (n=79) cohorts within this population. Also, in the younger population, there is a significant effect of the val66met polymorphism in predicting overall mortality (acute and post acute collapsed, p=0.037), such that Met-carriers have a higher rate of mortality (31% compared to 19%).
Genetic variation within the BDNF gene plays a role in mortality following severe TBI. Interestingly, associations with the functional variant and the tSNP are stratified by age, suggesting interactions between age-related secondary injury cascades and BDNF genotype. Evidence of age-related declines in BDNF expression may explain variation in genetic associations and post-injury mortality. BDNF function in relation to normal aging may further the understanding of specialized care in older, brain-injured populations. This study demonstrates again, unique interactions between genetic associations and TBI pathology, with similar associations seen in stroke populations and mortality. This study further demonstrates that genetic factors and demographic variables may aid in stratification of mortality predictions within similar injury parameters. Further studies will investigate interactions between BDNF genotype and function in specific injury types and other covariates in relation to mortality and other outcomes.
This research was supported by DODW81XWH-071-0701 and R01HD048162-02.
BDNF, mortality, genetics, TBI
PRELIMINARY RESULTS AND EARLY EXPERIENCE OF DECOMPRESSIVE CRANIECTOMY IN PATIENTS SUFFERING SEVERE TRAUMATIC BRAIN INJURY
Gatos Haralampos, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Mpakopoulou Maria, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Anagnostopoulos Vasilis, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Soultogiannis Konstantinos, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Paterakis Konstantinos, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Fountas Konstantinos, Department of Neurosurgery, University hospital of Larissa, School of Medicine, University of Thessaly
Decompressive craniectomy can be life saving for patients with critically raised ICP secondary to TBI. The indications for DC, the extent of decompression and the appropriate time for surgical intervention still remain controversial. In this observational study we report the clinical results in a series of patients sustaining severe TBI.
Forty craniectomies were performed on TBI patients (admitting GCS score <8, one at least episode of intracranial pressure >20 mmHg lasting more than 5 min). Clinical data were collected at admission, as were data on the indications for craniectomy, timing and surgical technique. Their GCS scores, head CT findings and perioperative intracranial pressure (ICP) measurements were recorded as outcome variables. The extent of the performed DC was calculated on the postoperative CT scan. Unilateral fronto-temporo-parietal DC was performed in 33 patients and bilateral in 6 patients. In the majority of our cases (33/40 pts) DC was performed within 48 hours from their admission, while in the remaining 6 patients it was performed from 48–72 hours. Outcome was measured using the Glasgow Outcome Scale (GOS).
Nine patients (22,5%) died during their initial hospital stay (GOS 1),8 (20%) patients remained in persistent vegetative state (GOS 2), 15 patients (37,5%) were discharged with severe and moderate disability (GOS 3–4) and 8 patients (20%) had a good outcome (GOS 5). The extent of craniotomy and the timing of intervention seemed to show a trend for earlier and more radical intervention although our results did not reach the levels of statistical significance.
Our results support the significant effectiveness of DC in patients with TBI. An appropriately timed surgical intervention and extensive decompression seemed beneficial for our series of patients.
none
Decompressive craniectomy; Intracranial pressure
TELEPHONE AND IN-PERSON COGNITIVE BEHAVIORAL THERAPY FOR DEPRESSION AFTER TRAUMATIC BRAIN INJURY
Jesse Fann, MD, MPH, University of Washington
Charles Bombardier, PhD, University of Washington
Steven Vannoy, MD, University of Washington
Sureyya Dikmen, PhD, University of Washington
Evette Ludman, MD, Group Health Research Institute
Objective: To describe design, progress, and pilot data from a randomized clinical trial of cognitive behavioral therapy (CBT) for patients with complicated mild to severe traumatic brain injury (TBI) and major depressive disorder (MDD).
Design: Randomized controlled trial with three arms: 1) in-person CBT, 2) telephone CBT, and 3) Usual Care. Randomization is via choice-stratified randomization. The CBT intervention consists of 12 sessions over 16 weeks. The main outcome is change in depression severity measured by the Hamilton Rating Scale for Depression assessed at 8, 16, and 24 weeks. Participants: We are recruiting 90 persons who have MDD within 10 years following complicated mild, moderate, or severe TBI. Participants are recruited from hospital and community settings throughout the United States. Inclusion criteria: 1) at least 18 years old, 2) speaks English fluently, 3) has a telephone, 4) has a stable home, and 5) had a complicated mild to severe TBI within the past 10 years. Exclusion criteria include: 1) refusal to participate, 2) comorbid bipolar or psychotic disorders or current substance dependence, and 3) cognitive impairment on neuropsychological testing that would preclude adequate participation.
We modified an empirically-supported telephone-based CBT protocol to accommodate cognitive impairments common in persons with TBI. We began recruiting participants in 2008 and have accessioned 80. We describe the rationale for telephone-based CBT and the challenges associated with modifying CBT for persons with TBI, engaging support persons to help the subject carry out treatment, monitoring and maintaining treatment fidelity, recruiting, treating, and retaining depressed patients with TBI using both in-person and telephone-based CBT. We compare and contrast the clinical and research implications of working with depressed TBI patients with more traditional psychotherapy studies. Outcomes from two pilot patients who completed telephone-based CBT and preliminary data from Patient Health Questionanire-9 (PHQ-9) depression scores from intervention sessions will be presented. Thus far, of 42 participants enrolled in the CBT intervention, 22 (52%) responded (>50% drop in PHQ-9). The telephone group demonstrated greater retention and response rates than the in-person group.
In-person and telephone CBT for major depressive disorder appears feasible and acceptable in persons with complicated mild to severe TBI. Telephone CBT holds promise for enhancing access and adherence to treatment, as well as possibly increased efficacy.
Supported by NIH/NICHD grant R21HD53736 and DOE/NIDRR grant H133G070016
Depression Cognitive Behavioral Therapy Clinical Trial
NEW APPROACH TO CHRONIC TBI REHABILITATION: CRANIAL NERVE NONINVASIVE NEUROMODULATION (CN-NINM TECHNOLOGY)
Mitchell Tyler, MS, UW-Madison, Biomedical Engineering Dept.
Kurt Kazcmarek, PhD, UW-Madison, Biomedical Engineering Dept.
Cranial-Nerve Non-Invasive NeuroModulation (CN-NINM) is a primary and complimentary multi-targeted rehabilitation therapy that initiates the recovery of damaged or suppressed brain functions that are most often affected by traumatic brain injury (TBI).
Four female subjects (A, B, C, D) were selected for the pilot study (mean age: 48.3, SD: 5.9 yrs). All subjects had sustained and significant balance and gait deficits due to moderate closed-head, non-penetrating, concussive TBI (9–11 on Glasgow Coma Scale) at initial diagnosis. All were approximately 5 years post acute injury and had previously completed rehabilitative therapy programs at their respective primary care facilities. CN-NINM therapy was administered twice daily, 5 days. Daily sessions comprised functional evaluation, flexibility and conditioning exercises, and simultaneous CN-NINM training. Tongue stimulation was provided by a PoNS (portable neurostimulator), an experimental device developed in our laboratory at the University of Wisconsin-Madison. The electrotactile stimulation is delivered to 144 gold-plated electrodes in a 3 cm×3 cm square matrix placed on the anterior portion of the tongue and held lightly in place by the lips.
Gait improvement. For assessment we used the Dynamic Gait Index (DGI) - a clinician-scored index of 8 facets of gait. Scores range from 0 (worst) to 24 points (normal), with scores of 19 or lower predictive of falls in the elderly. A score change of 3.0 points is generally considered clinically significant. Subjects score change (or improvement) was 13.5, 14, 10 and 21.5 points correspondingly. The DGI scores indicate significant improvements in stability and gait that are retained for as much as 6 hours after completion of the second intervention session in the day. Balance improvement. Additionally our subjects were tested on the NeuroCom CPD Sensory Organization Test (SOT) before and after the week of twice-daily interventions. The SOT is an objective, automated measure of sensory-motor integration that evaluates the functional contribution of the somatosensory, visual, and vestibular components of balance. A composite score (range from 0 to 100) is calculated and compared with a database normalized for age and height. Changes in the Composite score of 5 points or greater are considered clinically significant. In our study subjects balance score improved in 62, 10, 22 and 47 points correspondingly. We are continuing to monitor the changes in performance in these subjects. All have anecdotally reported progressive improvements in eye movement, short-term memory, executive function, and mood elevation. Additionally, Subjects C and D have noted significant reductions in both expressive aphasia and social anxiety. Taken on the whole, the results clearly indicate that the CN-NINM intervention is changing brain activity that is apparently producing concomitant improvements in functional behavior, and on perceived magnitude of symptoms, as measured on standardized metrics.
CN-NINM represents a synthesis of a new non-invasive brain stimulation technique with applications in physical medicine, cognitive, and affective neurosciences. Our new stimulation method appears promising for treatment of a full spectrum of movement disorders, and for both attention and memory dysfunction associated with traumatic brain injury. The integrated CN-NINM therapy aims to restore function beyond traditionally expected limits by employing both newly-developed therapeutic mechanisms for progressive physical and cognitive training - while simultaneously applying brain stimulation through a portable neurostimulation device. Based on our previous research and recent pilot data, we believe a rigorous in-clinic CN-NINM training program, followed by regular at-home exercises that will also be performed with CN-NINM, will simultaneously enhance, accelerate, and extend recovery from multiple impairments (e.g. movement, vision, speech, memory, attention, and mood), based on divergent, but deeply interconnected neurophysiological mechanisms.
The authors gratefully acknowledge the collaboration by Kathy Rust and Alla Subbotin in patient testing and training.
neurorehabilitation, neuromodulation, gait, balance, tongue
TRANSIENT CHANGE OF MODULAR STRUCTURE OF RESTING-STATE FUNCTIONAL CONNECTIVITY IN U.S. MILITARY PERSONNEL FOLLOWING MILD BLAST-RELATED TRAUMATIC BRAIN INJURY
Christine Mac Donald, PhD, Washington University in St. Louis
David Brody, MD, PhD, Washington University in St. Louis
Blast-related TBI has been called the ‘signature injury’ in the wars in Iraq and Afghanistan. Most of these blast-related TBIs are mild, uncomplicated TBI. To investigate modular-structural change following mild blast-related TBI, we applied graph theoretical analyses to resting-state functional connectivity of U.S. military personnel with TBI.
41 U.S. military personnel with clinically diagnosed mild TBI participated in this study. These subjects were compared with 16 military personnel who had blast exposure but no clinical diagnosis of TBI. Within 90 days after exposure, both groups underwent initial resting-state functional MRI scans at Landstuhl Regional Medical Center, Landstuhl, Germany. 6 to 12 months later, follow-up scans were performed at Washington University in St. Louis. We first identified modules of each brain network using the resting-state functional connectivity scans. With these identified modules, we assessed two module-based graph theoretic measures of the networks: within-module degree z-score and participation coefficients, which quantify within-module and between-module connectivity, respectively. Each of these measurements at each node between TBI patients and controls was compared using permutation tests. Each node-specific measurement was then averaged over entire nodes in each subject for group analysis on global network properties.
Overall, within-module degree z-score and participation coefficient in TBI patients decreased (p=0.035, 0.058, respectively) in comparison to controls at the time of the first scan. This was not observed at the follow up scan (p=0.072, 0.689, respectively). At the early time point, within-module degree z-score in TBI patients showed marginal decreases versus controls in transverse temporal cortex and insula. In contrast, these parameters were increased in the superior and middle frontal gyri (p<0.05, uncorrected for multiple comparisons). At the later time point, decreases in superior frontal gyrus and left insula as well as a marginal increase in right middle-posterior cingulate cortex were observed. Following scan 1, the participation coefficient in TBI patients decreased in comparison to controls in the superior and middle frontal gyrus, angular gyrus and left anterior and bilateral posterior cingulate gyri. These decreases resolved at the second scan.
Graph theoretical analyses revealed temporal dynamics of module-based connections in patients with mild blasted-related TBI. Most of these changes resolved at follow-up, supporting the use of early scans for assessment of mild TBI. Further studies are required to explain the underlying mechanisms and consequences of these phenomena.
TBI, graph, connectivity, fMRI, MRI
USE OF DEXTROMETHORPHAN/QUINIDINE SIGMA-1 RECEPTOR BLOCKADE IMPROVES AGITATION AND IRRITABILITY FROM TRAUMATIC BRAIN INJURY
Parmod K. Mukhi, MD, Wayne State University School of Medicine
Jean D. Peduzzi, PhD, Wayne State University School of Medicine
Kristina Freese, PA-C, Wayne State University School of Medicine
Behavioral dyscontrol and agitation commonly seen in the post-acute traumatic brain injury (TBI). It has many of the same attributes of pseudobulbar affect, which also is characterized by uncontrollable affect that is out of all proportion to the emotions felt by patients with other neurological disorders.
Dextromethorphan/quinidine 20mg/10mg twice daily has been associated with a greater decrease in the rate of affect episodes per week in multiple sclerosis and amyotrophic lateral sclerosis. Reported is a case of a mild TBI patient who was involved in a motor vehicle crash in 2007 with severe pseudobulbar affect issues, primarily agitation and irritability. CT scan was normal. He was having two episodes per day documented by his wife in a diary daily. He also had developed some antegrade memory issues and action tremors. Patient had treated without success utilizing sertraline and propranolol. Patient was weaned off these medications and started Dextromethorphan/quinidine 20mg/10mg twice daily.
Within one month his outbursts had diminished to three episodes per week in a diary. He then was taken off the medication and the episodes returned to two per day for one month. Patient was then restarted Dextromethorphan/quinidine and again by diary dropped to three episodes per week. He also noted a decrease in intention tremors.
Sigma-1 receptor blockade at the mid-brain and brain stem does have an effect on behavioral dyscontrol syndrome following TBI. Clearly, the next step is to perform a phase II clinical trial to evaluate primarily for safety and effect on cognition as well as to obtain preliminary information on the use of this medication for behavioral dyscontrol and agitation from TBI.
We acknowledge Oakwood Hospital and Wayne State University School of Medicine.
Sigma-1, agitation, dextromethorphan, pseudobulbar, tremor
MASS SPECTROMETRY-BASED IDENTIFICATION AND QUANTITATION OF PROTEINS RELEASED IN TRAUMATIC BRAIN INJURY CELL CULTURE MODEL
Kimberly Newsom, PhD, Banyan Biomarkers Inc
Anatoliy Vakulenko, PhD, Banyan Biomarkers, Inc
Firas Kobeissy, PhD, Banyan Biomarkers, Inc
Emilio Cagmat, MS, Banyan Biomarkers, Inc
Kevin Wang, PhD, Department of Psychiatry, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
John Anagli, PhD, Banyan Biomarkers
Mass spectrometry (MS)-based proteomics has been the method-of-choice for biomarker discovery. However, MS is yet to be fully exploited for biomarker quantitative validation in tissue and biofluids. Here, we demonstrate the use of MS in qualitative protein identification and targeted detection of TBI biomarkers.
Astrocytes and mixed neuronal cells isolated from day-old rats were grown in separate plates and incubated in serum-free media with or without maitotoxin (MTX) for 1, 4, 8 and 24 hours. Separate plates of mixed neuronal culture were subjected to NMDA and STS. At harvesting, 1 mL conditioned media was collected from the glial cell culture and 300 μl from the neuronal cell culture. Aliquots of the media were trypsinized and proteins released into the media were identified by LC-MS/MS. Isotope dilution approach (IDA) was developed for targeted peptide-based MS monitoring of the selective release of GFAP into the conditioned culture media following MTX challenge. Native and isotope-labeled versions of GFAP and UCH-L1 proteotypic peptide were synthesized and absolute concentrations were determined by amino acid analysis. The native peptide was used for method development and the heavy isotope labeled peptide served as internal standard. Western blotting was used for confirmation.
In both cell cultures, selective release of GFAP-BDPs was quantified by LC-MS/MS post MTX challenge at 2, 4, 8 and 24 h. The immunoblots confirmed the progressive release of GFAP- BDPs in a time dependent manner as indicated by the increase in the 38kDa GFAP-BDP. The MS-quantitation showed good correlation and with linear increase in GFAP peptide concentration from 0 to 24 h MTX incubation. UCH-L1 was detected by Western blot, however, the levels were below the detection limit for MS. Label-free proteomic analysis was conducted on the conditioned media from the glial and mixed neuronal cell cultures after challenge with maitotoxin, NMDA and STS. There were 13 common proteins identified in the all treatments including CRMP-2, creatine kinase B-type and alpha enolase. The total number of proteins identified in the media challenged with MTX, NMDA and STS; were 58, 49 and 53, respectively.
The developed MS-based method demonstrated sensitivity and selectivity for absolute quantitation of the release of calpain-dependent proteolytic product of GFAP (38kDa) in both rat glial and mixed neuronal cell cultures. The presence of GFAP was further confirmed by Western blotting and data dependent LC-MS/MS identification. UCH-L1, on the other hand, may require immuno-based enrichment prior to MS analysis to be detectable. In addition to these 2 proteins, other differentially expressed proteins were identified in the glial and mixed neuronal cell culture models and select proteins were analyzed by bioinformatics tools. Pathway analysis revealed association of MTX-specific proteins with necrotic dominant mechanisms while STS/NMDA-specific proteins with apoptosis dominant mechanisms.
N/A
Biomarkers, TBI, MS-quantitation, GFAP, proteomics
PRIOR FEBRILE SEIZURES INCREASE MORTALITY AFTER EXPERIMENTAL TRAUMATIC BRAIN INJURY IN IMMATURE RATS
Roxy Perez, BS, Barrow Neurological Institute
P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Traumatic brain injury (TBI) is a leading cause of death/disability in children. Five to 50% of people with TBI develop post-traumatic epilepsy (PTE). Factors influencing the development of PTE are poorly understood. We hypothesized that prior insult (febrile seizures) increases mortality associated with TBI and probability of developing PTE.
For these experiments, litters from Sprague-Dawley rats were culled to 8 pups (4 males, 4 females) on post-natal day (PND) 1. On PND 10, febrile seizures (FS) were induced in half of each litter using a hyperthermia chamber. One week later (PND 17), an experimental frontoparietal TBI was produced using controlled cortical impact (CCI) (6 mm tip; 4 m/s; 2.0 mm/depth). Sham animals were subjected to all procedures except impact. There were four experimental groups: Group 1 – FS plus CCI; Group 2 – FS plus craniotomy only/ sham injury; Group 3 – no-FS plus CCI; and Group 4 – no-FS plus craniotomy only/ sham injury. Death within 72 hours of TBI-induction was included in the determination of mortality. Four weeks post injury cortical screw electrodes were implanted into the skull for EEG recording monthly through six months post TBI. Epileptiform activity and seizure frequency were determined by visual inspection of the EEG. In addition, some animals were sacrificed at 2 weeks and 6 weeks for histology.
Our findings showed increased mortality associated with pre-injury FS with or without later CCI. Mortality rates were Group 1: 29.4% (10/34), Group 2: 19.2% (5/26), Group 3: 13.3% (4/30), and Group 4: 6.1% (2/33). Overall, mortality was greater for FS animals: 25% (15/60) compared to no-FS animals: 9.5% (6/63). While PTE after CCI compared to sham was not significantly increased with or without FS: Group 1: 28.6% (2/7), Group 2: 20% (1/5), Group 3: 14.2% (1/7) and Group 4: 0% (0/5), the addition of FS increased the incidence of PTE when the groups were pooled: 25% (3/12) than for no-FS animals: 8.3% (1/12). As well, a number of animals exhibited epileptiform discharges without definite seizures. Histologic studies for extent of injury (hematoxylin and eosin) and mossy fiber sprouting (Timm's stain) are underway and will also be presented.
Febrile seizures one week before CCI in immature rats markedly increases mortality associated with experimental TBI and increases the likelihood of PTE developing after injury. These observations support the hypothesis that prior insult (i.e.) FS increase the likelihood of mortality associated with TBI and the probability of subsequently developing PTE.
This study was supported by a grant from Baxter and by support from Phoenix Children's Hospital and Barrow Neurological Institute.
TBI, PTE, febrile seizures, rats
IL-12 TREATMENT IN A MOUSE MODEL OF TRAUMATIC BRAIN INJURY RESTORES IFN-GAMMA PRODUCTION AND DRIVES PROTECTIVE TH-1 IMMUNITY
Akinori Osuka, M.D., Brigham and Women's Hospital/Harvard Medical School
James Lederer, Ph.D., Brigham and Women's Hospital/Harvard Medical School
Traumatic brain injury (TBI) induces immune deficiency leading to increased infection rates among TBI patients. One of the immune defects following TBI is the suppressed production of Interferon-gamma (IFN-g). We hypothesized that IL-12, an upstream inducer of IFN-g might improve immune function after TBI.
Groups of six male CD-1 mice were subjected to either closed skull mild TBI by a weight-drop device (falling height 2 cm, weight 281g) or sham operation. All animal procedures were approved by the Harvard Medical School Standing Committee on Animal Research and were performed in accordance with the guidelines of the National Institutes of Health. 25 ng of IL-12 or saline were injected i.p. two hours after the injury. Mice were sacrificed one day after injury and spleens were harvested. Immunophenotyping was conducted by FACS staining of splenocytes and assessment of LPS and antiCD3-antibody (aCD3) stimulated cytokine secretion capacity of splenocytes.
aCD3 stimulated secretion of IFN-g, a central cytokine of the Th-1 immune phenotype which increases antimicrobial immunity, was found to be restored to sham levels by IL-12 treatment in TBI animals. Furthermore, we found that the production of the immunosuppressive Th-2 cytokines IL-4, IL-5 and IL-13 was decreased by IL-12 treatment as compared to untreated animals, further indicating a shift towards protective Th-1 immunity with IL-12 treatment. Additionally, production of IL-10, a negative regulator of IFN-g production, was decreased in IL-12 treated mice. FACS staining revealed no differences between groups in the percentages of CD4+ T cells, regulatory T cells, B cells, B-1 B cells, dendritic cells and dendritic cell subsets, macrophages, PMNs and NK cells. However, we found an increase in CD8+ T cells, which are known to produce IFN-g, in IL-12 treated mice.
IL-12 treatment restores production of IFN-g and drives Th-1 immunity following TBI which might enhance antimicrobial immunity. IL-12 treatment might be a promising immunomodulatory approach to treating TBI-induced immune dysfunction and might contribute to increased host defense against secondary infections.
None
TBI, immune function, IL-12, Interferon-gamma
DETERMINING RECOVERY TRAJECTORIES AND TERMINAL REHABILITATION POTENTIAL VARIATIONS FOLLOWING TRAUMATIC BRAIN INJURY
Mark Ashley, Sc.D., CCC-SLP, CCM, CBIST, Centre for Neuro Skills
Lisa Kreber, Ph.D., CBIS, Centre for Neuro Skills
Philip Seneca, Ph.D., Centre for Neuro Skills
Disease management and treatment trajectories for traumatic brain injury (TBI) have been less well-defined in comparison to other diseases. Management of disease is highly specific in diagnoses that are well understood. Treatment trajectories are developed and refined as understanding and knowledge of disease matures. However, we have yet to articulate the terminal disease outcome (rehabilitative outcome) for TBI. A patient's rate and extent of recovery must be considered when creating an objective approach to rehabilitation length of stay (LOS) and access to the treatment continuum of care. Concern has been expressed regarding arbitrarily contrived approaches to the duration of access to treatment such as a defined period of days for rehabilitation following brain injury.
The current study examined several aspects of recovery in 906 patients admitted to one of three post-acute rehabilitation facilities between the years 1990–2010. Survival analysis using Disability Rating Scale (DRS) admission and discharge disability categories was used to conceptualize the time required for a patient to transition during post-acute rehabilitation to a disability level specified as one level lower than that recorded at admission. Separate analysis was done based upon injury severity upon admission.
For patients admitted with a DRS rating of moderate, the terminal event was defined as achieving a rating of partial disability. The median length of stay was 153 days. For patients admitted with a DRS rating of moderately severe, the terminal event was defined as achieving a rating of moderate disability. The median length of stay was 151 days. For patients admitted with a DRS rating of severe, the terminal event was defined as achieving a rating of moderately severe disability. The median length of stay to reach was 213 days. For patients admitted with a DRS rating of extremely severe, the terminal event was defined as achieving a rating of severe. The median length of stay was 234 days.
The data suggests time to reach the terminal DRS score varies with level of disability at admission. This suggests that in establishing standards for rehabilitation benefits, the degree of disability upon admission to post-acute rehabilitation would be considered as a crucial factor prior to establishing LOS restrictions.
n/a
Post-Acute Rehabilitation
QUANTIFICATION AND PREDICTION OF ELECTROPHYSIOLOGICAL FUNCTION IN THE HIPPOCAMPUS AFTER IN VITRO TRAUMATIC BRAIN INJURY
Zhe Yu, Columbia University
Barclay Morrison III., PhD, Columbia University
Currently, finite element models are only capable of predicting mechanical responses to traumatic brain injury (TBI) and not biological responses. To address this deficiency, we are developing functional tolerance criteria based on changes in neuronal network function in brain tissue in response to controlled mechanical stimuli.
Organotypic hippocampal slice cultures were subjected to a single, biaxial deformation with our well-characterized in vitro model of TBI at specific combinations of strain (0.05 to 0.35) and strain rate (0.1/s to 20/s). All injuries were verified via high-speed video.
At 4–6 days post-injury, electrophysiological activity was quantified throughout the hippocampus with a 60-electrode multielectrode array. Stimulus-response curves were generated via stimulation at the mossy fibers to determine the maximum evoked response (Rmax) and the stimulus intensity necessary to generate a half-maximal response (I50).
Injury altered the electrophysiological function of the CA1, CA3 and DG hippocampal regions in a complex manner. Both Rmax and I50 were found to be significantly dependent on strain, strain rate and region by multivariate ANOVA. Rmax significantly decreased after slow, moderate injuries (0.20 strain at 0.1/s and 1/s) and fast, mild injuries (0.05 and 0.10 strain at 10/s and 20/s). However, after fast, severe injuries (0.35 strain at 10/s and 20/s) Rmax increased relative to uninjured controls. Rmax was also slightly elevated relative to uninjured controls after slow, mild injuries (0.05 and 0.10 strain at 0.1/s).
I50 was increased relative to uninjured controls after slow, moderate injuries (0.20 strain at 0.1/s and 1/s). Fast, mild injuries (0.05 strain at 20/s) also induced an increased I50 relative to uninjured controls. However, I50 was either not altered significantly or decreased relative to uninjured controls after slow, mild injuries (0.05 and 0.10 strain at 0.1/s and 1/s) and fast, severe injuries (0.35 strain at 10/s and 20/s).
Synaptic function, indicated by Rmax, generally decreased after injury, except after slow, mild injuries (0.05 and 0.10 strain at 0.1/s) and, paradoxically, after fast, severe injuries (0.35 strain at 10/s and 20/s). These results may be partially explained by the concomitant decrease in I50, suggesting an alteration in inhibitory tone of the neural network, possibly due to preferential effects of injury on inhibitory interneurons. Decreased inhibition may result in relative increases in excitatory drive, indicated by decreases in I50, resulting in the hyperexcitability observed. Our ongoing study suggests that alterations in the hippocampal neural network may be related to injury severity and rate in a more complex fashion than previously understood. We hypothesize that this effect may be caused by differential injury susceptibility of inhibitory and excitatory neurons, disrupting inhibitory feedback and feed-forward mechanisms necessary for complex interactions between neural networks in the hippocampus.
This study was funded by the National Highway Traffic Safety Administration (NHTSA) project # DTNH22-08-C-00088.
TBI, in vitro, electrophysiology, function
LONGITUDINAL ANALYSIS OF BEHAVIORAL OUTCOME FOLLOWING CLOSED SKULL INJURY IN POST-NATAL DAY 35 RATS
Gene Gurkoff, Ph.D., UC Davis
Tomas Tesfasilassie, University of California, Davis
Hang Yin, University of California, Davis
Professor Bruce Lyeth, Ph.D., UC Davis
Marike Zwienenberg-Lee, M.D., UC Davis Medical Center
In the pediatric population, traumatic brain injury is superimposed on the development of the nervous system. We examined the effects of closed skull cortical injury on the post-natal day (PND) 35 rat brain by assessing motor and cognitive function over a 6-week period following injury.
14 male PND 35 Sprague-Dawley rats (90–120 grams) underwent either a sham (n=6) or midline closed skull cortical injury (n=8, 5 mm depth, 3.5 m/s velocity, 100 ms dwell time). Animals were tested longitudinally for motor outcome using the rotarod and ladder walk twice per week for 6-weeks following injury. In addition, animals were tested on each of four behavioral tasks: metric, topological, temporal ordering, and Morris water maze. The metric, topological and temporal tasks are low stress, cognitive paradigms that take advantage of an animal's innate exploratory behavior to test spatial working memory (metric and topological) and temporal sequencing of odors. At the end of the sixth week animals were perfused and tissue will be analyzed for changes in neurons (cresyl violet), microglia (Iba-1), and astrocytes (GFAP).
We observed no overt motor changes between sham and closed skull cortical injured rats over the 6-week period; however, there was a consistent subtle deficit in both the rotarod and ladder walk tasks over the first 4-weeks post-injury. Pilot data indicates that sham and injured animals were unable to perform the metric task at PND42. At PND49 there was no difference in cognitive performance in the topological task between sham and injured animals as both re-explore two objects after they were transposed in space (sham=12.0±4.7sec and TBI=12.3±3.6 sec). At PND56 there was a trend toward a deficit in temporal ordering memory (p=0.07) with shams preferentially re-exploring the initial presented odor (60.1±7.9) while injured animals showed no preference (46.1±4.2). Analysis of odor exploration during the acquisition period did not detect differences between sham and injured rats suggesting that impaired performance is not related to changes in olfaction. At PND 77, in a limited number of animals (n=4 per group), there was a trend toward a decrease in Morris water maze performance in injured animals (p=0.2).
Preliminary data indicates that at PND42, animals were unable to perform the metric task, a spatial working memory paradigm. Future studies will determine whether TBI delays or prevents the development of this cognitive function. Injury did not affect performance in the topological task when animals were tested at PND49. There was a strong trend in injured animals toward diminished performance in the temporal ordering task. Finally, there was evidence of a deficit in MWM performance in injured animals. Therefore midline TBI in PND35 animals appears to affect animals on a spectrum of motor and cognitive tasks. Histological analysis will examine changes in neurons, astrocytes, and microglia at 6 weeks post-injury in the injured cortex and hippocampus and will be compared with behavioral outcome. Such studies are critical to increase our understanding of how TBI affects the development of neuronal connectivity and cognitive dysfunction in this vulnerable population.
Dept. of Neurological Surgery, University of California, Davis
pediatric, behavioral, cognitive, longitudinal
ENVIRONMENTAL ENRICHMENT CONFERS ROBUST FUNCTIONAL AND HISTOLOGICAL BENEFITS IN ADULT FEMALE RATS AFTER EXPERIMENTAL TRAUMATIC BRAIN INJURY
Christina M. Monaco, University of Pittsburgh
Vincent V. Mattiola, University of Pittsburgh
Kaitlin A. Folweiler, University of Pittsburgh
Justin K. Tay, University of Pittsburgh
Narayana K. Yelleswarapu, University of Pittsburgh
Lauren M. Curatolo, University of Pittsburgh
Ashley M. Matter, University of Pittsburgh
Anthony E. Kline, PhD, University of Pittsburgh
Environmental enrichment (EE) has been shown to produce marked benefits after TBI and therefore may be considered a rodent model of preclinical rehabilitation. However, the EE focus has been primarily in male rats. In order to validate the model, demonstration of similar efficacy in female rats is necessary.
Normal cycling females were evaluated for estrous cycle stages on the morning of surgery via vaginal cytologies and then randomly assigned to a controlled cortical impact (2.8 mm tissue deformation at 4 m/sec) or sham injury. Following surgery they were assigned to either TBI + STD (n=8), TBI + EE (n=8), Sham + STD (n=4), or Sham + EE (n=4) groups. Motor recovery was tested on a beam-balance/walk task (days 1–5) or rotarod (every other day for 19 days). Cognitive function was assessed with a Morris water maze task (days 14–19). Hippocampal cell survival was quantified at 3 weeks.
No differences were observed in pro-estrous, estrous, or di-estrous stages and therefore the data were pooled. Additionally, no differences were observed between the sham groups and thus these data were also pooled. The analyses revealed that EE robustly improved motor performance, facilitated spatial learning, and attenuated hippocampal cell loss vs. STD housing (p<0.05).
These findings, which are in contrast to a previous report, indicate that females do benefit cognitively from EE and suggest that this preclinical rehabilitation paradigm has widespread utility.
Supported, in part, by NIH grants HD046700 and NS060005 (AEK)
EE, females, rehabilitation, recovery, TBI
SYSTEMS BIOLOGY META-ANALYSES OF GENOMIC DATASETS TO IDENTIFY CONSERVED MECHANISMS AND NOVEL BIOMARKERS OF TRAUMATIC BRAIN INJURY
Chenggang Yu, Ph.D., DoD Biotechnology HPC Software Application Institute U.S. Army Medical Research and Materiel Command BHSAI/MRMC
Xueping Yu, Ph.D., DoD Biotechnology HPC Software Application Institute U.S. Army Medical Research and Materiel Command BHSAI/MRMC
Bhaskar Dutta, Ph.D., DoD Biotechnology HPC Software Application Institute U.S. Army Medical Research and Materiel Command BHSAI/MRMC
Jacob Feala, Ph.D., DoD Biotechnology HPC Software Application Institute U.S. Army Medical Research and Materiel Command BHSAI/MRMC
Kara Schmid, Ph.D., Walter Reed Army Institute of Research
Jitendra Dave, Ph.D., Walter Reed Army Institute of Research
Frank C Tortella, Ph.D., Walter Reed Army Institute of Research
Jaques Reifman, Ph.D., DoD Biotechnology HPC Software Application Institute U.S. Army Medical Research and Materiel Command BHSAI/MRMC
High-throughput technologies quantify expression of thousands of genes, but the plethora data hinders unambiguous gene identification associated with the condition of interest. We exploited systems biology to systematically analyze multiple high-throughput genomic studies of traumatic brain injury (TBI) to gain insights into conserved mechanisms and to infer biomarker candidates.
We employed systems biology approaches to incorporate multiple gene expression datasets with biological knowledge in the form of pathways and protein-protein interaction (PPI) networks. We selected four datasets from TBI experiments using different animals and TBI models, integrating the data with over 130 non-metabolic canonical pathways in the KEGG database and a PPI network consisting of over 11,000 proteins and 80,000 interactions. We separately overlaid each set of data onto the pathways and the PPI network to identify a set of pathways and PPI modules that are statistically significant for each dataset. We then inferred TBI biomarker candidates from those specific pathways and biologically relevant PPI modules that are conserved in two or more datasets and validated them by protein abundance changes in rat brain subjected to a penetrating ballistic-like brain injury (PBBI).
Our analyses identified statistically significant pathways and PPI modules that are conserved in two or more datasets. Functional analyses showed that only down-regulated pathways and PPI modules were specifically associated with the nervous system; those that were up-regulated were associated with immune responses. The 19 conserved down-regulated pathways included 7 out of 9 KEGG nervous system pathways, which are all associated with synaptic functions; the 6 conserved down-regulated PPIs formed a subnetwork of synaptic proteins. Thus, we predicted that PSD-95, DISC1, and NOS1 would be down-regulated in the ipsilateral brain tissues which were confirmed in subsequent in vivo experiments using the PBBI model. We observed a statistically significant decrease (69%, 47%, and 50%, respectively, p<0.05) in protein abundance in the ipsilateral brain tissues compared to sham controls. We also observed a 50% increase for PSD-95, a 46% decrease for NOS1, but no change for DISC1 in contralateral tissues.
We employed systems biology approaches to integrate molecular evidence from distinct TBI studies with biological pathways and networks to systematically identify conserved and TBI-specific molecular processes and biomarkers. We found that only down-regulated pathways and PPI modules are associated with the nervous system, especially synaptic functions. This effect is likely due to neuronal loss coupled with increased inflammation caused by TBI. The in vivo experimental results confirmed the TBI biomarker candidates hypothesized using our systems biology approach. Our results suggest that systems biology can provide a high-yield approach to generate testable hypotheses that can be experimentally validated to identify potential TBI biomarkers.
Support, in part, provided by the DoD Defense Medical Research and Development Program, U.S. Army MRMC, Ft. Detrick, Maryland.
Brain Injury, Biomarker, Systems Biology
THE EXPRESSION OF INFLAMMATORY CYTOKINES IN RAT BRAINS AFTER DIFFUSE AXONAL INJURY
Hao Wang, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Yu Lin, MD, Zhejiang University City College
Quan-Cheng Li, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Xin Huang, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Jiang-Biao Gong, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Lin Su, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Xiao-Feng Yang, MD, First Affiliated Hospital, College of Medicine, Zhejiang University
Inflammation reaction is an important mechanism leading to cerebral injuries after TBI (traumatic brain injury), and it is once reported that microglial/macrophage were activated in injured loci of DAI (diffuse axonal injury). In further, in the present study, we tried to detect the expression of pro-inflammatory cytokines.
Adult, male Sprague-Dawley rats, subjected to DAI according to Marmarou's method, were sacrificed 30min, 1 hour, 6 hours, 12 hours, 24 hours, 2 days and 7 days after injury. Hippocampus, thalamus and part of neocortex and brain stem were isolated from both left and right cerebral hemisphere, in which, the left parts were to detect the transcriptional level (mRNA), and the right parts, expression level (protein) of TNF-α, IL-1β and IL-6, respectively. Additionally, in prepared brain tissue sections, HE staining and immunohistochemistry staining were also performed, employing antibodies targeting TNF-α, IL-1β, IL-6 and β-APP.
Through these approaches, we found that, in the above injury loci, the expression of TNF-α, IL-1β and IL-6 in both mRNA and protein levels elevated significantly 30min after injury. The expression of TNF-α and IL-1β peaked 6∼12 hours after injury, while IL-6 approximately 1 day after injury. Comparatively, these cytokines expressed more in hippocampus and brain stem than those in thalamus and neocortex. Although the expression of mRNA returned to normal level 7 days after injury, the protein level of these cytokines were still higher than that before head trauma.
Our findings revealed that rapid elevated expression of pro-inflammatory cytokines was involved in DAI, which indicates that they participate in secondary axonal injury. And it may imply that inflammation reaction play a role in the DAI. It is our major mission to explore how these pro-inflammatory cytokines participating in DAI in our future study.
There is no specific sponsor for this study.
Axonal Injury, Inflammation
BALANCE DEFICITS FOLLOWING CLOSED HEAD MILD TRAUMATIC BRAIN INJURY IN THE RAT
Michelle C. LaPlaca, PhD, Georgia Institute of Technology
Martin Park, Georgia Institute of Technology
Mild traumatic brain injury (mTBI) remains a serious health and socioeconomic problem, despite increased public awareness over the past decade. Of the approximately two million TBIs per year in the US, 80% are classified as mild, with acute symptoms, such as confusion, balance problems, headache, nausea, and memory loss.
Sprague Dawley rats (n=7, n=5 shams) were anesthetized with isoflurane, the skull exposed, and injured using CCI (mid-line suture, 3mm diameter tip, 5 m/s, depth of 1mm, dwell time of 0.5s). Balance ability was assessed using angle board both before and 4 hours after the injury by placing the rats on a board at a starting angle of 35°. If an animal remained in place for 5 seconds, it passes that trial and the angle increases one degree until the animal fails. To assess short-term memory, novel object recognition (NOR) was used. The animals were placed in a cage with two identically interesting objects and were allowed to explore them for 5 minutes prior to injury. Five hours post-injury, rats were placed in the same cage with one of the original objects and a new object. The time that the animals spent exploring the each object was recorded.
No skull fractures were produced as a result of the mTBI, and therefore the force produced by the CCI transferred to the brain was consistent from animal to animal.In addition, following perfusion, the brains were examined and there was no visible macroscopic damage in either the sham or injured groups. The angle board data showed a significant difference between the injured group (p<0.05) and the sham group, demonstrating a clear deficit in balance as a result of the mTBI. The average angle difference before and after the injury for the injured group was 4° and sham group was 0.8 degrees. However, the NOR data did not show significant memory deficits as both the injured and sham rats both spent significantly more time on the novel objects.
In conclusion, the closed head mTBI model produced significant deficits in balance ability using angle board behavioral test. However the NOR test indicates that both the injured rats and the shams showed equal amount of curiosity on the identical objects and more curiosity to the new object after the injury, suggesting that the level of injury did not affect short term memory. These results show that an impact speed of 5m/s to the skull produced diffusive damage to the brain that impaired balance ability, while not showing any sign of direct physical insult to the brain. In order to develop a clinically relevant mTBI model, a closed head impact is preferred over craniectomy or craniotomy. This study showed subacute balance deficits, which is a symptom of human mTBI. Future studies will examine more time points, as well as examine the brain for microscopic level damage.
GTEC Regenerative Medicine Seed Grant, Emory/Georgia Tech
mTBI Traumatic Brain Inury
AGE DEPENDENT DIFFERENCES IN THE EFFECT OF PROGESTERONE AFTER CCI
Jacqueline Berglass, BA, Department of Medicine, Children's Hospital Boston
Jimmy Zhang, BA, Department of Pediatrics, Massachusetts General Hospital
Rachel Stanley, MD, MHSA, University of Michigan
Michael J. Whalen, MD, Harvard Medical School
Prior studies demonstrated that progesterone improves outcomes after controlled cortical impact in adult rats and mice and there is a phase III clinical trial of progesterone in adult TBI patients. Here, we test the effect of age on outcome after CCI in progesterone treated mice.
3 week old, 4 week old and 12 week old male mice were subjected to CCI. Immediately after CCI, mice underwent ip injections of 8–16 mg/kg progesterone or vehicle and then daily sc for 4 days. Edema was assessed at 24 hr in 3 week old and 12 week old mice (n=6–8/group). Functional outcomes were assessed with wire grip (motor) and Morris water maze (spatial memory)(n=4–10/group). Data were analyzed by t-test, clustered ordinal regression (motor data) or repeated measures analysis of variance (RM ANOVA) as appropriate.
24 hrs after CCI, there were no differences in edema between progesterone vs. vehicle-treated 3 week old or 12 week old mice. After CCI, all groups demonstrated time dependent improvement in motor function (p<0.001). 4 week old progesterone-treated mice had better motor performance compared to vehicle treated mice (p=0.013). In contrast, 3 week old progesterone-treated mice had worse performance than vehicle-treated mice (p=0.008) and there was no difference in motor performance in progesterone vs. vehicle treated 12 week old mice. There were no differences in MWM performance in progesterone vs. vehicle treated mice in 3 week old mice but progesterone-treated 12 week mice performed better on probe trials comapred to vehicle-treated mice (p=0.03).
These data suggest that there may be age-dependent differences in the effect of progesterone on outcome after CCI. Further studies are needed prior to clinical trials in children.
Maternal and Child Health Bureau, Health Resources and Services Administration, Department of Health and Human Services Grant U03MC0003 The Baby Alex Foundation
Traumatic Brain Injury, progesterone, neuroprotection
GLOBAL FUNCTIONAL CONNECTIVITY IN THE VISCEROMOTOR NETWORK IS NEGATIVELY CORRELATED WITH IL-1β SERUM LEVELS IN CONCUSSED ATHLETES
Rashmi Singh, Ph.D., Laureate Institute for Brain Research
Rayus Kuplicki, M.S., University of Tulsa
Ashlee A. Taylor, B.S., M.B.A, University of Oklahoma Health and Sciences Center
David Polanski, M.S., LAT, ATC, University of Tulsa
Thomas W. Allen, D.O., MPH, FACP, FCCP, University of Oklahoma Health and Sciences Center
Lamont Cavanagh, M.D., University of Oklahoma Health and Sciences Center
Jerzy Bodurka, Ph.D., Laureate Institute for Brain Research
Wayne C. Drevets, M.D., Laureate Institute for Brain Research
T. Kent Teague, Ph.D., University of Oklahoma College of Medicine
Sports-related concussion is associated with increased levels of pro-inflammatory cytokines and mood dysregulation. Recently, IL-1β levels have been linked with symptomsof “sickness behavior” (anhedonia, fatigue, psychomotor slowing, irritability) and depression. We studied the relationship between fMRI connectivity and IL-1β levels in sports-related concussion.
Functional magnetic resonance imaging (fMRI), cognitive testing, psychiatric interviewsand serum cytokine levels were collected in a prospective longitudinal designfrom 9 collegiate athletes before, and 24–48 hours, 1-week and 1-month after concussion. Gradient-echo echoplanar imaging was collected on a GE 3-Tesla MR750.To enhance the fMRI signal and reduce signal dropout subjects were scanned using a 32-channel arrayed head coil combined with parallel imaging methods (SENSE factor=2). To avoid task-related effects of concussion task-independent resting-state fMRI was collected. Global functional connectivity measures were calculated from these resting-state data for each athlete at each session. Ratings of depression and anxiety symptomseverity were collected immediately prior to each scanning session. Cognitive performance was measured using the ImPACT assessment tool. Finally, serum cytokine levels were analyzed using BD Cytometric Bead Array Enhanced Sensitivity Flex Sets from BD Biosciences.
The number of days “held out” from practicing was used as the primary indicator of concussion severity and was determined using neurological and SCAT2 criteria. The number of days held out of practice ranged from 4 to 31 days with and average of 16.5 9.73 days. Subjects showed a significant increase in cognitive efficiency across post concussion time points as demonstrated by paired t-test comparisons of 24-hour to 1-week post concussion (t(7)=−2.49; p<0.03) and 24-hour to 1-month post concussion (t(7)=−5.30; p<0.01). No participants showed mood ratings that approached the depressed range at baseline. Both depressive and anxiety ratings were significantly higher at 24–48 hours and 1-week (t[7]=−3.59; p<0.01; t[7]=−2.51; p<0.05; respectively) but not 1-month after concussion. Interestingly, depression severity 24 – 48 hours after injury was negatively correlated with our concussion severity measure. Similar to the depression ratings, IL-1β levels were significantly elevated 24 – 48 hours compared to either 1-week(t[7]=−4.5; p<0.01) and 1-month after concussion (t[7]=−2.62; p<0.05). There was no difference in IL-1β levels between 1-week and 1-month post concussion.Global functional connectivity was calculated for each resting-state scan by correlating the average correlation coefficient at each voxel with every other voxel in the brain. This measure is proposed to be a gross measure of voxelwise functional connectivity. Global functionalconnectivity measures were then correlated with IL-1β levels across all scanningsessions. Results, corrected for multiple comparisons, demonstrate a significant negative correlation between IL-1β and global functional connectivity bilaterally in the dorsal anterior cingulate cortex, left subgenual prefrontal cortex, left middle temporal gyrus, leftprecuneus, right inferior frontal gyrus, left posterior cingulate cortex, andleft middle frontal gyrus. No brain regions were significantly positively correlated with IL-1β levels.
The negative correlation between self-reported depression ratings and our outcome measure implies that full recovery of affective regulation is not properly accounted for in the criteria used for making return to play decisions. Conceivably, this may increase the risk of long-term emotional disturbances in players returned to play prior to recovery ofaffective regulation. Further, our data demonstrate that increased levels of the pro-inflammatory cytokine IL-1β negatively correlate with dysregulation in the visceromotor network that has been implicated in affective processing. This finding suggests that changes in functional connectivity within the affective processing visceromotor network underlie the “sickness behavior” associated with increasing IL-1β levels. Improved return to play decisions following concussion should include establishing stable connectivity within this neural network. In summary, our results suggest that global connectivity fMRI data from the visceromotor pathway may serve as an objective biomarker for mood dysregulationfor assisting with return to play decisions.
Supported by The William K. Warren Foundation
Inflammation, fMRI, Depression, Concussion, Connectivity
MECHANISMS OF COGNITIVE DYSFUNCTION AFTER CONCUSSION TRAUMATIC BRAIN INJURY IN MICE: ROLE OF SURVIVAL AND DEATH PATHWAY KINASES
Xiaoxia Zhu, PhD, Massachusetts General Hospital
Juyeon Park, BA, MAssachusetts General Hospital
Jimmy Zhang, BA, Department of Pediatrics, Massachusetts General Hospital
Jianhua Qiu, MD, Massachusetts General Hospital
Michael J. Whalen, MD, Harvard Medical School
Concussive traumatic brain injury (TBI) induces cognitive deficits by unknown mechanisms. Antagonists of receptor interacting protein kinase-1 (RIPK1) or Akt/mTOR improve cognitive outcome after cerebral contusion. We tested the hypothesis that RIPK1 and Akt/mTOR mediate cognitive deficits in a concussion model devoid of acute and chronic cell death.
Adult male C57/BL6 mice were anesthetized with isoflurane and subjected to weight drop closed head injury (CHI; 53 g, 66 in) or sham injury (anesthesia without weight drop) using a model that results in cognitive deficits in the absence of acute cell death. At 6 or 24 h, phosphoproteins from cortical and hippocampal lysates were analyzed by Western blot and densitometry or by immunohistochemistry using validated antibodies against Akt, mTOR, S6, or GSK3-beta. In separate experiments, mice were administered Akt inhibitor VIII (12.5 uM) or rapamycin (mTOR inhibitor, 12.5 uM) in the presence or absence of necrostatin-1 (4 mM), a specific inhibitor of receptor interacting protein kinase-1 (RIPK1), or vehicle ICV. Cognitive function was tested in a Morris water maze beginning 3 d after injury. Densitometry data were analyzed by t-test. Morris water maze data were analyzed by RM ANOVA. p<0.05 was considreed significant.
Robust phosphorylation of Akt-473 and its direct substrate GSK3-beta, as well as mTOR and its direct substrate S-6 was detected in cortex and hippocampus by 6–24 h after CHI (p<0.05 vs. sham injured, n=3–6/group). Immunohistochemical analyses demonstrated neuronal S-6 phosphorylation in sham-injured and injured mice, and markedly increased S-6 phosphorylatiopn in astrocytes and microglia in cortex and hippocampus of injured mice compared to sham at 24 h. Akt or mTOR inhibitors reduced phosphorylation of downstream substrates GSK3 beta and S-6, respectively in hippocampus (p<0.05 vs. vehicle treated mice) and inhibited MWM performance (p<0.05 for group, hidden platform trials and probe). Treatment with necrostatin-1 prior to CHI improved postinjury probe trial performance to that of sham-injured mice (p<0.05 vs. vehicle) and increased phosphorylation of Akt-473 and S-6 in injured hippocampus (p<0.05 by densitometry), effects that were abolished by co-administration of Akt inhibitor VIII or the mTOR inhibitor rapamycin.
The data suggest beneficial roles for Akt and mTOR signaling, in neurons as well as glial cells, in limiting cognitive deficits after concussion TBI. Necrostatin-1 improves cognitive outcome after CHI at least in part by augmenting Akt/mTOR signaling. To our knowledge, the data are the first to suggest a detrimental role for RIPK1 in a non-cell death TBI model. Necrostatin-1 has therapeutic potential for patients with concussion TBI, which is extremely common and may lead to short and long term cognitive sequelae especially in war fighters and athletes. Further studies are needed to better define the molecular events associated with RIPK1 and Akt/mTOR activation in concussion TBI so that rational therapies may be developed to limit cognitive dysfunction and improve outcome in patients.
Support: NIH RO1 NS064545
necrostatin, Akt, mTOR, mice, concussion
SHOCKWAVE ATTENUATION DURING PASSAGE THROUGH THE PMHS HEAD: DIFFERING CHARACTERISTICS DEPENDING ON BRAIN SIMULANT
Brian Stemper, PhD, Medical College of Wisconsin
Narayan Yoganandan, PhD, Medical College of Wisconsin
Barry Shender, PhD, Naval Air Warfare Center
Mechanisms of mTBI due to open field blast have not been well defined. To quantify injury thresholds and develop effective safety enhancements, it is essential to understand the transmission of shockwaves through human head. This study quantified shockwave transmission through the post-mortem human head using brain simulants.
A compressed-helium driven open-ended shock tube generated shockwaves by accumulating driven section pressure until exceeding bursting pressure of a Mylar diaphragm. Peak overpressure was modulated through variation of diaphragm thickness. A single PMHS head was used, with scalp and hair remaining intact. The intracranial contents were removed and the vault was filled with Sylgard or Ballistic gelatin brain simulants. Pressure transducers (PCB 113b28 & Kulite LQ-125-50SG) were mounted in different positions around the skull to record external and intracranial pressures. The head was attached to a surrogate neck from an anthropomorphic testing device and exposed to shockwaves of different characteristics and in frontal and lateral orientations. Pressure transducer data were digitally recorded at 10MHz to capture the fast rising peak overpressure.
Attenuation of shockwave peak overpressure varied with transducer location on the head and type of brain simulant (i.e. Sylgard or Ballistic gelatin). Incident pressures measured external to the cranium on the side of the head facing the shock tube were between 150 and 370 kPa with rise times to the peak pressure less than 10 ms. Pressure attenuations are reported as percentage decrease from the incident shockwave. Peak intracranial overpressure was attenuated 71% for ballistic and 94% for Sylgard gelatin from the incident to the measurement at the top of the head in the frontal orientation. In the lateral orientation the shockwave attenuated 38% for ballistic gel from the incident to the intracranial measurement at the top of the head. This indicates that the shockwave remains somewhat intact at the mid-point of the head for both frontal and lateral orientations with ballistic gel, whereas the shockwave has completely dissipated at the mid-point for Sylgard gel. Additionally, for both orientations and brain simulants, shockwave overpressure was minimized at the distal aspect of the head. The shockwave attenuated 60% for ballistic and 70% for Sylgard gel from the incident side to the transducer mounted externally to the mid-point of the head in frontal orientation. Additionally, the shockwave attenuated 83% in ballistic and 69% in Sylgard gel from incident side to the transducer mounted externally to mid-point of the head in the lateral orientation.
This preliminary study demonstrated shockwave attenuation as it passes through the human skull, indicating that brain pathology may be more evident on the ipsilateral side during blast scenarios as energy is attenuated by the cranium mid-point. Attenuation differences based on head orientation has implications for injured brain regions and patient outcomes following blast TBI. Additionally, given that peak overpressure attenuation and shockwave profile varied considerably between two commonly used brain simulants, it is important to obtain accurate material properties for physical or computational modeling of the head for blast injury simulations. Currently, a paucity of information exists regarding human brain material properties at rates applicable to the blast environment. Continued investigation of high-rate brain properties under multiple loading modes is required. Ongoing testing using this model will permit better understanding of shockwave transmission through the human skull.
This research was supported by the Office of Naval Research through Naval Air Warfare Center Aircraft Division Contract N00421-10-C-0049 and the Department of Veterans Affairs.
mTBI, blast, shockwave, post-mortem, shock tube
INFLAMMATION FOLLOWING EXPERIMENTAL TRAUMATIC BRAIN INJURY IS MEDIATED BY NFKB AND TOLL-LIKE RECEPTORS
Todd E. White, Ph.D., Morehouse School of Medicine
Gregory D. Ford, Ph.D., Morehouse College
Alicia S. Gates, B.S., Morehouse School of Medicine
Michelle C. LaPlaca, Ph.D., Georgia Institute of Technology
Byron D. Ford, Ph.D., Morehouse School of Medicine
Traumatic brain injury (TBI) is a major cause of morbidity and mortality in the United States. While several pathways are activated following TBI, the inflammatory pathway is documented to be associated neuronal death. A complete understanding of the role of inflammatory pathway in TBI may ultimately lead to effective therapies.
Sprague-Dawley rats (Charles River) weighing 290–300 grams were either placed into naïve (n=3) or injured (n=3) groups. The injured group were anesthetized and subjected to a 15° unilateral left cortical controlled impact (Pittsburg Precision Instruments, Inc) velocity of 3m/s to a depth of 2mm (n=3). Rats were sacrificed 24 hours following enrollment and tissue was prepared for microarray analysis.
Naïve brains (what areas of these brains were collected for experiment). Injuredbrains were divided into two sections, contralateral (non-injury) and ipsilateral (injured). Total RNA was extracted with TRIzol Reagent and samples were processed for Affymetrix microarray analysis.
Affymetrix GCOS software normalized and analyzed data for each chip. This data was then imported and analyzed in Microsoft Excel for calculation of fold change and confirmation of presence in tissue. Data were then analyzed using Genesis software and Ingenuity Pathway Analysis (Ingenuity Systems, www.ingenuity.com).
The top biological functional pathways identified by IPA in naïve and ipsilateral brains were the Inflammatory Response. STAT3, HTT, STAT1 and NFkB (complex) were predicted to be the major transcription regulators for the dataset. Pathway and gene network analysis showed thatthe Toll-like signaling pathway was a key in the propagation of inflammatory response. There were 16 molecules in this pathway that showed increased expression in the injured brain. NFkB activity was predicted to mediate the induction of the genes in the data set.
Our data indicated that increased toll-like receptor signaling mediated through NFkB results in an increase in the production of inflammatory molecules. The activity of NfkB was predicted to play a major role in mediating the induction of genes in the dataset. Potential therapies to reduce the detrimental effects of traumatic brain injury may involve the modulation of molecules in this pathway.
This work was partially supported by NIH grants U01 NS057993, C06 RR-07571, G12-RR03034; MBRS/Rise Program; DoD Contract W81XWH-10-2-0055 and the W.M. Keck Foundation.
Brain, Injury, Inflammation, Toll-like
HSP70 INDUCTION BY 17-AAG AS A POTENTIAL THERAPY FOR TRAUMATIC BRAIN INJURY
JongYoul Kim, PhD, UCSF
Estelle Kim, Wellesley College
Midori A. Yenari, MD, UCSF
The 70kDa heat shock protein (Hsp70) is known to protect the brain from injury through multiple mechanisms. Pharmacological means of inducing Hsp70 as a potential therapy for traumatic brain injury (TBI) is explored here.
3 month old male C57/B6 mice were studied. Naïve animals were given 17-N-allylamino-17-demethoxygeldanamycin (17-AAG) intraperitoneally (IP, 2 mg/kg) or intracerebroventricularly (ICV, 1 ug/kg) to determine whether Hsp70 could be induced in the brain. Another group of mice were subjected to TBI via cortical controlled impact, and were treated with 17-AAG (or vehicle) IP according to one of two treatment regimens: (1) 2 mg/kg at the time of injury, (2) a total of three doses (4 mg/kg) at 2 and1d prior to TBI and again at the time of injury. Brains were assessed for Hsp70 induction, hemorrhage volume at 3d and lesion size at 14d post-injury. The swing test and corner test were performed at 3d and 7d to assess motor function recovery. Immunofluorescent staining and Western blots of Hsp70 were performed from brain tissue.
Both IP and ICV administration of 17-AAG induced HSP70 expression primarily in microglia and in a few neurons by 24h but not in astrocytes. 17-AAG induced Hsp70 in injured brain tissue as early as 6h, peaking at 48h and largely subsiding by 72h after IP injection. Both treatment groups showed decreased hemorrhage volume relative to untreated mice (p<0.1) as well as improved neurobehavioral outcomes in swing test at 3d (p=0.05). There was a trend towards lesion size reduction with treatment as well.
17-AAG-treatment improved neurobehavioral deficits following experimental TBI, and attenuated brain hemorrhage and injury. Further, peripheral administration could lead to Hsp70 induction in the brain. These observations indicate that 17-AAG may prove to be a promising, non-invasive treatment for TBI.
This research was funded by grants from the DoD, NIH, and VA Merit.
tbi, hsp70, 17aag, neuroprotection
ETAZOLATE REDUCES NEUROINFLAMMATION, RESTORES sAPPalpha LEVELS AND OFFERS PERSISTENT NEUROPROTECTION FOLLOWING TRAUMATIC BRAIN INJURY IN MICE
Gemma Llufriu-Dabén, Pharm.D student, laboratoire de la circulation cérébrale, EA 4475, Université Paris Descartes, Sorbonne Paris cité
Angelo H. Cho, Pharm.D, laboratoire de la circulation cérébrale, EA 4475, Université Paris Descartes, Sorbonne Paris cité
Sandra Vidal-Lletjós, Pharm.D student, laboratoire de la circulation cérébrale, EA 4475, Université Paris Descartes, Sorbonne Paris cité
Michel Plotkine, Ph.D, laboratoire de la circulation cérébrale, EA 4475, Université Paris Descartes, Sorbonne Paris cité
Catherine Marchand-Leroux, Ph.D, laboratoire de la circulation cérébrale, EA 4475, Université Paris Descartes, Sorbonne Paris cité
Mehrnaz Jafarian-Tehrani, Ph.D, CNRS UMR 8194 (CESEM), Université Paris Descartes, UFR Biomédicale
Traumatic brain injury (TBI) provokes an intense neuroinflammatory response that promotes neuronal death and that leads to the reduction of an endogenous neuroprotector, the sAPPα. Etazolate is a phosphodiesterase-4 inhibitor that stimulates sAPPα production. In this study, the therapeutic interest of etazolate was evaluated on acute and belated TBI consequences.
The mouse model of TBI by mechanical percussion was applied as previously described. Injured mice received (i.p.) either the vehicle (PBS 0.01 M) or etazolate at the dose of 1(E1), 3 (E3) or 10 mg/kg (E10) at 2h post-TBI. Neurological score was evaluated by The Irwin/Circle exit test battery, where sensorimotor reflexes and seeking behavior, such as the animal's ability to exit a circle upon a given time (2 min), were examined. Cerebral edema was calculated using the dry-wet weight method and IL-1beta and sAPPα levels were evaluated by ELISAs. Lesion size was measured using cresyl violet staining and microglial activation was determined by CD11b immunolabeling. Locomotor activity was assessed using an actimeter and memory function was examined in the Novel Object Recognition Test (NORT). At 13 weeks post-TBI, mice were sacrificied, and olfactory bulb lesions were measured using digital images of the whole brains.
All three doses of etazolate were able to reduce neurological impairment at 6h (E1: P<0.05; E3, E10: P<0.001) and 24h post-TBI (E1: P<0.05; E3, E10: P<0.001) in a dose-dependent manner. Moreover, E10 restored sAPPα levels (P<0.01) and significantly reduced IL-1beta production (P<0.05) and cerebral edema (P<0.01). These effects were also associated with a reduction of CD11b immunolabeling (P<0.01) and ventricular enlargement (P<0.01) at 24h post-TBI. E3 and E10 reduced locomotor hyperactivity in most time-points of the test, from 48h to 12 weeks post-TBI. At 5 weeks post-TBI, memory impairment was attenuated by both E3 and E10 (E3: P<0.05, E10: P<0.01). Olfactory bulb lesions were also reduced by both E3 and E10 at 13 weeks post-TBI (E3, E10: P<0.001).
A single administration of etazolate within the acute and critical TBI aftermath was able to exert anti-inflammatory, anti-edematous and neuroprotective effects, with a therapeutic window of at least 2 hours. These effects were associated with the restoration of the levels of the sAPPα protein, known to exert neuroprotective and neurotrophic effects in TBI. Interestingly, the same prompt treatment with etazolate offered lasting neuroprotection, namely memory improvement, reduction of locomotor hyperactivity and olfactory bulb tissue protection. Taken together, these results demonstrate the neuroprotective potential of compounds that exert pleiotropic effects in the complex pathology of TBI and highlight the therapeutic interest of etazolate.
We would like to express our gratitude to the nonprofit organization, Fondation des Gueules Cassées, for supporting this work financially.
neuroinflammation, edema, etazolate, sAPPα, neuroprotection
REPETITIVE MILD TRAUMATIC BRAIN INJURY AUGMENTS TAU PATHOLOGY AND GLIAL ACTIVATION IN AGED H-TAU MICE: A PATHOLOGICAL STUDY
Benoit Mouzon, MS, Roskamp Institute
Scott Ferguson, Bsc, MS, Roskamp Institute
Corbin Bachmeier, PhD, Roskamp Institute
Michael Mullan, MD, PhD, Roskamp Institute
Fiona Crawford, PhD, Roskamp Institute
Post-mortem analyses of concussive injury victims in many cases reveal that the brain is extensively burdened with neurofibrillary tangles composed of phosphorylated-tau filaments. Until recently, very little preclinical research has investigated pathological consequences of repetitive-TBI in appropriate animal models; few of these have exploited such models to explore tau pathogenesis.
We address this shortfall in this study by exploring human Tau effects using a recently developed model of concussive injury, inflicted to transgenic h-Tau mice (at 18 months of age). This is an appealing novel model, because the mice lack familial genetic mutations, but have the relevant human tau background profile, expressing six isoforms of human wild type tau, with both three and four microtubule binding repeat regions on a null murine-tau background.
Experimental TBI was induced as a closed head injury above the sagittal-suture midway, using a 5mm blunt metal impactor tip attached to an electromagnetic motorized device. Animals received either a single injury or five injuries, each 48h apart, or the appropriate number of control anesthesias. A series of 5–6μm paraformaldehyde-fixed sagittal sections taken throughout the cortex/hippocampus were immunostained to detect a range of specific phospho-tau epitopes and analysed using quantitative morphometric analyses/optical segmentation (Image-Pro Plus, Media Cybernetics).
Although by this age (18 months) h-Tau mice demonstrate significant tau pathology, repetitive injury results in a dramatic increase in phospho-tau immunoreactivity at three weeks post injury, compared to aged-matched single concussive injury, as determined by a range of phospho-specific antibodies (CP13, PHF1, RZ3, MC1). Comparison of results of single injury with either control group showed no significant differences in tau pathobiology.
We also found that repetitive injury resulted in a marked concomitant increase in neuroglial (astrocyte and microglia) activation notably in the superficial layer of the cortex underlying the pia matter of the injury site, and the corpus callosum (demonstrated by GFAP and CD45 immunoreactivity). By contrast there was a paucity in neuroglial activation markers after single concussive injury and control anesthesia.
Another distinct hallmark of this model were abnormally shaped (pyknotic) pyramidal nuclei in the cortex and CA3 region of the hippocampus, and pronounced loss of neurons as demonstrated by TUNEL positive staining. This pattern was noticeable following repetitive injury and repetitive anesthesia.
Overall these results demonstrates that the h-Tau mouse / mTBI model represents a useful tool for further investigation into the link between TBI and tau pathobiology in humans, and an avenue for development of (tau-specific) therapeutic agents for TBI.
This work was supported by Department of Defense award W81XWH-10-1-0759 (Crawford) and by the Roskamp Foundation.
mTBI, Htau mice, Tau pathology
LONG-TERM EEG DYNAMICS FOLLOWING TBI IN A RAT MODEL OF PTE
Jamie White-James, B.S., Barrow Neurolgical Institute
Ioannis Vlachos, Ph.D., School of Biological and Health Systems Engineering, Arizona State University
Balu Krishnan, M.S., School of Electrical, Computer and Energy Engineering, Arizona State University
Vinay Venkataraman, B.S., School of Electrical, Computer and Energy Engineering, Arizona State University
Leonidas Lasemidis, Ph.D., School of Biological and Health Systems Engineering, Arizona State University and Department of Neurology, Mayo Clinic
David Treiman, M.D., Barrow Neurolgical Institute
Development of post-traumatic epilepsy (PTE) after traumatic brain injury (TBI) is a major health concern (5% – 50% of TBI cases). A significant problem in TBI management is the inability to predict which patients will develop PTE. Such prediction, followed by timely treatment, could be highly beneficial to TBI patients.
Four male Sprague-Dawley rats, weighing approximately 375 grams each, were subjected to a controlled cortical impact (CCI). A 6mm piston was pneumatically driven 3mm into the right parietal cortex with velocity of 5.5m/s. The rats were subsequently implanted with 6 intracranial electroencephalographic (EEG) electrodes, in the left thalamus, hippocampus, parietal and frontal cortex, and right occipital and frontal cortex (no electrodes in the ipsilateral to the impact hippocampus and temporal cortex). Following a 2-week recovery period, long-term (14-week) continuous EEG recordings were conducted. Using linear (coherence) and non-linear (Lyapunov exponents) measures of EEG dynamics in conjunction with measures of network connectivity, we studied the evolution over time (10 sec resolution) of the global as well as the local functional connectivity between brain sites in order to identify early precursors of development of epilepsy, that is, long before the possible occurrence of spontaneous seizures.
Three of the four TBI rats developed PTE 6 to 10 weeks after the initial insult to the brain. Analysis of the continuous EEG from these rats showed a gradual increase of the connectivity between critical brain sites in terms of their EEG dynamics, starting at least 2 weeks prior to their first spontaneous seizure. In contrast, for the rat that did not develop epilepsy, connectivity levels remained low, unchanged, or decreased during the whole course of the experiment across pairs of brain sites. The linear measure (coherence) provided consistent evidence of an increased level of connectivity between the site closest to the impact (right occipital cortex) and three more distant brain sites, namely the ispsi- and contralateral frontal cortices and the contralateral parietal cortex. These changes were observed mostly in the 3 traditional EEG frequency bands (alpha, beta, gamma) between 8–30Hz. The non-linear measure (Lyapunov exponent entrainment) provided evidence of an increased level of connectivity over weeks similarly between the right occipital cortex (impact site) and the contralateral frontal and parietal cortices. In addition, the Lyapunov exponents suggested an increase in functional connectivity between the right occipital cortex (impact site) and the contralateral hippocampus. The results above were consistent across all three rats that developed PTE with a variation in the rate of increase of functional connectivity between brain sites and the “focus” (site of impact).
This study provided us with the opportunity to quantitatively investigate several aspects of epileptogenesis following traumatic brain injury. Our preliminary results strongly support a network pathology that worsens with time, that is, long-term trends of increase of connectivity between the impaired cortex and the rest of the brain in TBI rats prior to development of epilepsy. We currently conduct studies in a larger animal population to further validate these initial findings. It is conceivable that the observed changes in spatiotemporal dynamics after an initial brain insult, and long before the development of epilepsy, could constitute a basis for predictors of epileptogenesis in TBI patients. Such predictors would facilitate a) early selection of TBI patients for epilepsy treatment, well prior to manifestation of clinical symptoms (e.g. seizures) and hence increase the probability of success for subsequent seizure control, and b) the evaluation over time of the anti-epileptic treatment itself.
This research is supported by Department of Defense (DOD) PH/TBI Research Program, Office of Congressionally Directed Medical Research Programs (CDMRP) grant PT090712.
Epilepsy, TBI, EEG, Dynamics, Entrainment
BEHAVIORAL CHARACTERIZATION OF FLUID PERCUSSION INJURY TO THE SENSORIMOTOR CORTEX
Lynn Moore, B.S., Southern Illinois University-Carbondale
Michael R. Hoane, Ph.D., Southern Illinois University-Carbondale
William Maass, Southern Illinois University-Carbondale
The primary goal of this study was to obtain a behavioral and histological characterization of FPI to the sensorimotor cortex to provide a good injury model for the evaluation of neuroprotective drugs. The SMC has been used to model neuroplasticity following CCI, but has not been extensively examined with FPI.
This study used a unilateral FPI pulse over the rat SMC. A plastic leur, 4 mm in diameter was fastened over the craniotomy (AP +0.5, ML +3.5 relative to bregma) and a fluid pulse was administered. Following injury, a battery of behavioral tasks was used to assess sensory (Tactile adhesive removal task), cognitive (Morris water maze), and both fine (Locomotor placing task, Forelimb asymmetry task) and gross (Rotorod) motor deficits. Additionally, a histological assessment of this injury model was completed at 24 hours post-injury concentrating on neuronal and glial responses of the secondary injury. Appropriate statistical procedures were utilized to evaluate the behavioral and histological assessments.
Gross behavioral deficits were found on the sensory task (Tactile adhesive removal task) and multiple motor assessments (Forelimb asymmetry task, Forelimb placing task, and Rotorod). These sensorimotor deficits occurred in the absence of cognitive deficits in the Morris water maze. Results from the histological analysis 24 hours post-injury identified the neuronal and glial response to the injury. Our results indicate that even a mild FPI injury to the SMC creates severe and enduring behavioral deficits, which are ideal for multiple drug comparisons.
Following TBI, neuroplastic changes within the SMC that increase functional recovery can be stimulated with behavioral rehabilitation. The neuroplastic changes and ease at which damage to this area can be measured behaviorally make this location an excellent model to assess TBI treatments. No injury model can completely mimic the complete spectrum of human TBI and any potential treatments should be validated across multiple models. CCI, is a focal injury, primarily to the ipsilateral hemisphere while FPI, is a more diffuse injury, affecting both hemispheres. FPI is a commonly used and well characterized preclinical model of TBI and has been applied to multiple species. Our results show that even mild FPI to the SMC results in debilitating behavioral deficits, making this an optimal model for potential treatments. Targeting multiple components of secondary injury with a combination drug treatment is necessary to improve functional recovery. Behavioral and histological characterization of this FPI model allows for the evaluation of the effectiveness of these multicomponent polytreatments.
William Mass
FPI, animal model, sensorimotor cortex
MINOCYCLINE AND N-ACETYLCYSTEINE MODULATES NEUROINFLAMMATION AND PRODUCES REMYELINATION FOLLOWING CONTROLLED CORTICAL IMPACT
Margalit Haber, BS, SUNY-Downstate Medical Center
Pramod Dash, PhD, The University of Texas Health Science Center at Houston Medical School
Raymond Grill, PhD, The University of Texas Health Science Center at Houston Medical School
Natalia Grin'kina, SUNY-Downstate Medical Center
Samah Abdel-Baki, SUNY-Downstate Medical Center
Following controlled cortical impact (CCI), MINO plus NAC (MINO/NAC) remyelinates white matter and improves both cognition and memory. We are testing whether changing the pattern of microglial activation by MINO/NAC induces remyelination. Improvement in cognition and memory may result from white matter repair.
Sprague-Dawley rats received either sham- or mild CCI (Abdel-Baki, et al., 2009) and treated at one hour and one day with MINO/NAC. Parasagittal sections were prepared 2 days after surgery that were stained with MHCII or CD68 that stains proinflammatory M1 microglia, arginase-1 that stains anti-inflammatory M2 microglia, or Iba-1 that stains both M1 and M2. M1 and M2 microglia were assessed in corpus callosum, dorsal or ventral hippocampal commissure (VHC) and anterior commissure. Cerebellar white matter acted as a control. MINO/NAC also improved performance on an active place avoidance task. The hippocampal commissures are likely needed since task acquisition requires both hippocamppi. Demyelination following CCI may impair or block commissural neural transmission. Stereotaxic lysolecithin injection produces a 2-week cycle of localized demyelination-remyelination. The VHC was injected with lysolecithin or saline (0.3ul, 1% w/v) followed by behavioral assessment.
Iba-1+ cells increased in corpus callosum in saline-treated rats two days after CCI. MINO/NAC treatment further increased the number of Iba-1+ cells. An increase in MHCII or CD68 expressing cells in injured saline-treated rats was not seen in rats treated with MINO/NAC. Examination of M2 markers is ongoing. Saline-treated rats acquired the active place avoidance task four days after VHC injection. In contrast, rats injected with lysolecithin were completely impaired. After training, rats were sacrificed and myelin content assessed in the VHC. Saline-injected rats showed no change in VHC myelin content as seen by luxol fast blue staining. Lysolecithin-injected rats showed less myelin suggesting a localized demyelination. Examination of the VHC 2 weeks after injection showed no change in myelin content in saline-injected rats. In contrast, the reduced myelin content seen at 4 days after lysolecithin injection rats was significantly increased 2 weeks after injection. These data suggest a cycle of demyelination-remyelination induced by lysolecithin injection. Behavioral assessment is ongoing of rats 2 weeks after saline or lysolecithin injection.
These data suggest that MINO/NAC increases the number of activated microglia in damaged white matter regions after CCI. These microglia do not express M1 markers suggesting that they may have a M2 phenotype. Increased M2 microglia have been implicated in remyelination suggesting that MINO/NAC may repair white matter, in part, by altering the pattern of microglial activation following CCI.
Stereotaxic injections of saline and lysolecithin have suggested a key role of the VHC in the acquisition of the active place avoidance task. This is consistent with previous observations showing a need for both hippocamppi. Commissural white matter is particularly vulnerable to TBI and in animal models of TBI. Taken together, these studies show the importance of white matter damage and repair. Drugs that target white matter such as MINO/NAC in rodents may also show efficacy in clinical TBI that also display white matter damage.
This work was supported by contract W81XWH-10-2-0171 from the USAMRMC to P.J.B.
therapeutics, white matter, repair, microglia
TBI PROMOTES LONG-LASTING MICROGLIAL PRIMING AND DEPRESSIVE-LIKE BEHAVIOR
John Gensel, PhD, University of Kentucky/Department of Physiology/Spinal Cord and Brain Injury Research Center
Eric Wohleb, BA, The Ohio State University/Department of Oral Biology/Department of Neuroscience
Yan Huang, MS, The Ohio State University/Institute for Behavioral Medicine Research
Phillip Popovich, PhD, The Ohio State University/Department of Neuroscience/Institute for Behavioral Medicine Research/Director of Center for Brain and Spinal Cord Repair
Jonathan Lifshitz, PhD, University of Arizona, College of Medicine/Department of Child Health/Director of Translational Neurotrauma Research Program
Jonathan Godbout, PhD, The Ohio State University/Department of Neuroscience/Institute for Behavioral Medicine Research/Center for Brain and Spinal Cord Repair
Traumatic brain injury (TBI) affects over 1.7 million people each year. While TBI elicits multiple acute complications, other long-term behavioral deficits, including depression, can persist and worsen in the years following injury. Few studies have investigated chronic neuroinflammation following TBI and its relation to the development of TBI-associated depression.
Young adult, male BALB/c mice received a sham operation or a moderate and diffuse TBI by fluid percussion injury. Sections of the hippocampus and cortex were isolated 4 and 72 h after injury to assess levels of the pro-inflammatory cytokines interleukin (IL)-1β and transforming growth factor (TNF)-α. Levels of glial fibrillary acidic protein (GFAP) were also evaluated to determine astrocyte reactivity. Furthermore, microglia were isolated and evaluated for the activation marker, CD14. In a related study, edema and blood brain barrier permeability were assessed 4 or 24 h after injury. One week and one month following injury mice were tested for depressive-like behavior. Mice were then euthanized and astrocyte and microglial phenotypes were assessed by immunohistochemistry and flow cytometry. Finally, one month after injury mice were given a peripheral immune stimulus with lipopolysaccharide (LPS) and sickness behavior and microglial reactivity were determined.
As expected, inflammatory cytokines, edema, blood brain barrier permeability, and markers of glial activation, were increased 4 h after injury. For example, there was a time and brain-region dependent increase in IL-1β and TNF-α associated with increased CD14 expression on isolated microglia. Moreover, IL-1β levels remained elevated in the hippocampus for at least 72 h after injury corresponding with significant upregulation of GFAP in both the cortex and hippocampus at this time-point. Microglia from TBI mice also maintained elevated CD14 expression for at least 72 h after injury. Blood brain barrier permeability was primarily noted in the prefrontal cortex, site of injury, and cerebellum 4 h after injury but was restricted to the site of injury within 24 h. One month after injury, staining of GFAP and Iba-1 by immunohistochemistry was performed to evaluate prolonged glial activation. GFAP expression was significantly increased in both the cortex and hippocampus of TBI mice and Iba-1 showed a tendency to be elevated in the hippocampus. In association with this prolonged glial activation, depressive-like behavior was significantly increased in mice that received a TBI both one week, and one month, following injury. Notably, TBI mice showed no deficits in motor function at these times. Because depressive complications have correlated with increased markers of microglial priming (i.e., major histocompatibility complex (MHC) II) in models of aging, microglia were evaluated for MHC II expression one month after injury. Indeed, microglia from TBI mice had increased MHC II mRNA and protein expression. Higher levels of MHC II corresponded with exaggerated induction of toll-like receptor (TLR)2, IL-1β, TNF-α, and inducible nitric oxide synthase (iNOS) following a peripheral injection of LPS. Moreover, these markers of exaggerated microglial activation coincided with prolonged sickness behavior in TBI mice.
Together these data indicate that a moderate TBI induces both acute (4 h) and prolonged (>1 month) neuroinflammation associated with a more reactive glial population. Notably, MHC II expression was elevated on microglia one month after injury corresponding to exaggerated and prolonged neuroinflammation and sickness behavior following a peripheral immune stimulus. This is important because more reactive and inflammatory macro and microglia have been associated with neurological and behavioral complications. Indeed, mice that received a TBI had prolonged depressive-like behavior. Our hypothesis is that as microglial priming continues with age, TBI-associated depressive complications will worsen. Because the majority of TBIs are in juveniles and young adults, these data indicate that a moderate TBI can cause long-term inflammatory changes in brain glia resulting in a life-time of depressive complications.
Research was supported by NIH:R01-AG-033028-01 and CBSCR. A.M.F. was supported by the HHMI. Thanks to Todd Lash, Rachel Rowe, and Kelley Hall for technical assistance.
TBI, depression, microglia, neuroinflammation
TBI LEADS TO ALTERATIONS IN HIPPOCAMPAL THETA, BETA AND GAMMA FREQUENCIES DURING REST AND EXPLORATION IN RATS
Darrin Lee, M.D., UC Davis
Gene Gurkoff, Ph.D., UC Davis
Angela Echeverri, B.S., UC Davis
Philip Schwartzkroin, Ph.D., UC Davis
Robert Berman, Ph.D., UC Davis
Paul Muizelaar, M.D., Ph.D., UC Davis
Bruce Lyeth, Ph.D., UC Davis
Kiarash Shahlaie, M.D., Ph.D., UC Davis Medical Center
Hippocampal theta rhythmicity is essential for learning and memory, and is attenuated after lateral fluid percussion (LFP) injury. We examine the effects of LFP injury on the power, percent time and peak frequencies of hippocampal oscillations (theta, beta and gamma frequencies).
Fourteen male Sprague-Dawley rats (300–350 grams) underwent either a sham (n=7) or LFP injury (n=7, 2.12–2.15 atm), immediately followed by stereotactic implantation of ipsilateral bipolar, twisted hippocampal recording electrodes. Each rat underwent 10-minute daily video-EEG recordings in a transparent Plexiglas box (post-injury days 1–7). EEG was analyzed separately during voluntary movement, rearing, grooming and while the animal was stationary. Additionally, on post-injury day 8, animals were observed during a 15-minute open field task including the presentation of two novel objects. EEG was analyzed specifically during object interaction. All analyses of EEG power were performed using custom MATLAB code in conjunction with the “EEG toolbox” software. An oscillatory detection method (“P_episode”) was utilized to determine the percentage of time theta oscillations occurred in the CA3 region of the hippocampus.
Power spectral analyses demonstrated a 39.4±7.2% decrease in theta power (6–10Hz) and a 50.9±3.5% decrease in beta power (15–30Hz) in the first week following LFP injuries (p<0.05). There was no difference in power between sham or LFP rats in the low gamma (30–80 Hz) or high gamma (80–150 Hz) frequency ranges. Theta oscillations predominated over beta and gamma oscillations during movement and at rest. There was a significant reduction in the percent time of theta recorded in the ipsilateral CA3 in LFP-injured animals as compared to shams (20.5±3.2% and 11.5±2.0%, respectively, p<0.05). There was no difference between percent time in beta (11.7±0.6% and 10.6±0.6%, respectively, p>0.05) or gamma (60.6±8.9% and 69.5±10.9%, p>0.05).
In addition to reduced theta power, peak theta frequency was also significantly reduced following LFP as compared to sham injury (6.41±0.12 Hz and 6.95±0.12 Hz respectively, p<0.05). There was a trend towards decreased theta power following LFP during movement (7.03±0.19 Hz and 6.67±0.11 Hz, p<0.05) and rest (6.65±0.39 Hz and 6.04±0.10 Hz, p<0.05).
Theta oscillations in the CA3 of the hippocampus play an essential role in learning and memory. Our results demonstrate a frequency-specific depression in hippocampal oscillations following LFP injury. LFP results in decreased theta power, decreased percent time in theta, and decreased peak oscillation frequency within the theta band in the CA3. Although LFP also results in decreased beta power, there is no effect on percent time in beta or the peak frequency within the beta range. LFP does not affect low or high gamma band oscillations. Therefore, theta band oscillations are specifically affected following LFP TBI, and may be a prudent target for electrical neuromodulation to improve functional recovery.
N/A
Theta, rodent, TBI, hippocampus
THE EFFECTS OF REPEAT TBI ON JUVENILE PITUITARY
David A. Hovda, Ph.D., University of California, Los Angeles
Mayumi Prins, Ph.D., UCLA
Significant growth hormone disruption in adults following TBI has been described, but is just beginning to be explored in children. GH plays a crucial role in puberty, growth and cognitive development. We hypothesize pituitary function is disrupted following pediatric repeat TBI and are likely to negatively affect development.
Postnatal day 35 male rats were given either sham, single or 4 repeat mild closed-head TBI and trans-cardially perfused with 4% paraformaldehyde after a recovery period of one week or 6 months. Free-floating 40μM horizontal sections of the anterior pituitary were labeled with a primary antibody against GH by rinsing multiple times in 0.05 M KPBS buffer to remove any cryoprotectant before exposure to primary antibody diluted 1:100K in 0.05M KPBS+0.4% Triton-X for 48 hr, then secondary antibody diluted 1:1K in 0.05M KPBS+0.4% Triton-X 1:600 for 1 hr, and incubation in A/B solution for 1 hr. Tissue was rinsed before and after a 1.5 min incubation in Ni-DAB solution with 0.175M sodium acetate buffer. After the final rinse in KPBS buffer, the slices were mounted, dehydrated and cover-slipped. Image J was used to calculate % area stained and count vessels.
Upon visual inspection of tissue from animals 1 week or 6 months after injury, those with 4 RTBI have areas of decreased staining surrounding dilated vasculature. Total staining of growth hormone decreases 42% 1 week following a single TBI and decreases 56% in animals receiving 4 RTBI. At 6 months growth hormone staining is decreased 25% in single TBI and 20% in 4 rTBI animals. In addition to decreased staining, there was an increase in vessel diameter in injured animals. Single injured animals at 1 week showed a 108% increase compared to sham and those receiving 4 RTBI show a 122% increase in vessel diameter. At 6 months vessel diameter in injured animals is comparable to shams.
Despite TBI being the #1 cause of morbidity and mortality among children, there are currently NO experimental studies addressing the occurrence of pituitary dysfunction in adolescents. Based on CDC and census statistics, an estimated 1 in 150 adolescents could be living with pituitary dysfunction, potentially causing significant deficits in normal brain and physical development, highlighting the need to study this under addressed issue. Sports related concussions and repeat concussions are one of the most common causes of TBI among children, adolescents and young adults. RTBI results in greater learning and memory deficits in juveniles compared those who only sustained a single TBI. Cumulative damage to the pituitary during this critical period of development could result in lifelong impairment.
NFL Charities, UCLA Brain Injury Research Center, Marilyn and Austin Anderson Fellowship, NS058489
Pituitary, juvenile, repeat TBI, mild
POST-TRAUMATIC INHIBITRORY PLASTICITY AUGMENTS EXCITABILITY OF ATYPICAL DENTATE GUTAMATERGIC PROJECTION NEURONS
Akshay Gupta, MD, UMDNJ-NJMS
Fatima S. Elgammal, BS, UMDNJ-NJMS
Archana Proddutur, MS, UMDNJ-NJMS
Semilunar Granule Cells (SGCs), novel projection neurons located in the dentate molecular layer, are distinguished from granule cells by the presence of molecular layer axon collaterals and persistent firing. We examined whether concussive brain injury leads to changes in SGC physiology that could impact dentate function and excitability.
Male Wistar rats were subject to moderate Lateral Fluid Percussion Injury (FPI). Age-matched sham-injured rats served as controls. Reproducibility of the injuries was verified by immunostaining for Fluoro-Jade C in sections from rats perfused 4 hours after sham- or head-injury and NeuN staining for neuronal nuclei in sections from rats one week after injury. One week after FPI or sham-injury, whole cell recordings from granule cells and SGCs were obtained in horizontal brain slices from the side of injury. Cells were filled with biocytin during recordings and processed for immunostaining for post-hoc morphological identification of SGCs. Voltage-clamp recordings from a holding potential of −70 mV using a cesium-chloride based internal solution was used record synaptic and tonic GABA currents. Tonic GABA currents were measured as the baseline currents blocked by GABAA receptor antagonists. Current-clamp recordings obtained using a K-gluconate based internal solution was used to determine active and passive properties.
One week after FPI, granule cell tonic GABA currents were significantly increased compared to controls. The post-traumatic increase in granule cell tonic GABA currents persisted for one month before returning to control levels 3 months after injury. Unlike granule cells and molecular layer interneurons, tonic GABA currents in SGCs were markedly decreased one week after brain injury. Moreover, there was a post-injury decrease in the frequency of spontaneous inhibitory synaptic currents (sIPSCs) in SGCs rather than the increase in sIPSC frequency observed in granule cells. Following FPI, SGCs showed an increase in input resistance, and enhanced firing in response to depolarizing current injections indicating increases in SGC excitability after brain injury. The GABAA receptor antagonist, gabazine, enhanced the input resistance of control SGCs but failed to alter the input resistance of SGCs in injured rats demonstrating that post-injury decreases in GABAergic tone may augment SGC excitability after brain injury.
We find that brain injury leads to early and selective decrease in SGC tonic GABA currents resulting in post-traumatic enhancement in SGC excitability suggesting that inhibitory plasticity in SGCs may contribute to abnormal dentate circuit function after brain injury. Our data extend our fundamental understanding of the dentate circuit involved in working memory, by demonstrating that intrinsic differences in inhibitory control of granule cells and semilunar granule cells that may underlie the functional distinctions between these two dentate projection neurons. Our results show post-injury increase in SGC excitability and suggest that, by virtue of their direct projections to CA3 and molecular layer collaterals, SGCs may provide a focus for early dentate hyperexcitability after brain trauma. Our findings indicate that SGCs are a potential source of increase in dentate excitability after brain injury, and suggests that SGCs may play a causal role in post-traumatic epileptogenesis and memory dysfunction.
This study was funded by the New Jersey Commission for Brain Injury Research Individual Research Grant to V.S and Postdoctoral Fellowship to A.G
Dentate, inhibition, FPI, excitability
BIOMARKERS OF TRAUMATIC BRAIN INJURY IN RATS EXPOSED TO EXPLOSIVE-DRIVEN PRIMARY BLAST
Suhana Sarkar, B.S., Uniformed Services University of the Health Sciences
John E. Buonora, M.S., Uniformed Services University of the Health Sciences
Erin Murphy, M.S., Uniformed Services University of the Health Sciences
Eleanor Kitces, B.S., Uniformed Services University of the Health Sciences
Sandor Krisztian. Kovacs, M.D., Uniformed Services University of the Health Sciences
John Magnuson, B.S., Uniformed Services University of the Health Sciences
Gregory P. Mueller, Ph.D., Uniformed Services University of the Health Sciences
The role of shock wave in blast-related traumatic brain injury (TBI) remains largely to be clarified. The objective of this study was to assess whether the exposure of rats to blast's shock wave is reflected in a rise of known and potential biomarkers of TBI in plasma.
Plasma samples were collected at euthanasia from 95 male Sprague Dawley rats exposed to primary blast as generated by the detonation of high explosives (liquid nitromethane, C4). Rats were exposed under different conditions of blast intensity (12–50 psi peak incident overpressure), chest protection and plasma collection time (6 hours-28 days). Control rats (n=47) underwent the same procedures (including transportation, exposure to the blast site environment, anesthesia), except for blast exposure. Plasma aliquots, separated by centrifugation ten minutes after blood collection, were kept frozen in liquid nitrogen until they were evaluated for levels of neuron specific enolase (NSE), S100B, creatine kinase B (CK-B) and brain-derived neurotrophic factor (BDNF) by sandwich ELISA.
Plasma levels of NSE, S100B, CK-B and BDNF were raised in both control and blast-exposed animals 6 hours after anesthesia. Blast further increased plasma levels of NSE (1.6-fold) and BDNF (2-fold), 6 and 24 hours after exposure, respectively. A dependence of plasma levels on blast intensity could be tested and confirmed only for BDNF. After reaching a peak in animals exposed at 14–18 psi blasts (around 0.5 ng/ml) levels could not be further increased at blast intensities reaching 50 psi (in protected animals). No significant blast-related changes were observed for S100B and CK-B (two astrocytic markers).
Blast exposure appears to affect plasma levels of both NSE and BDNF, but not S100B and CK-B, possibly suggesting a higher vulnerability to blast of the neuronal as compared to astrocytic subpopulation, under the test conditions.
Supported by funds from: Blast Spinal Cord Injury Program, Center for Neuroscience and Regenerative Medicine, Defense Advanced Research Projects Agency
Blast Traumatic brain injury biomarkers rat
BEHAVIORAL CONSEQUENCES OF TRAUMATIC BRAIN INJURY: A COMPARISON OF MULTIPLE INJURY MODELS IN THE LABORATORY MOUSE
Amanda Fu, M.D., Uniformed Services University and the Center for Neuroscience and Regenerative Medicine
Lyndsey Fortin, M.Eng., Henry M. Jackson Foundation and the Center for Neuroscience and Regenerative Medicine
Oz Malkesman, Ph.D., Henry M. Jackson Foundation and the Center for Neuroscience and Regenerative Medicine
Joseph T. McCabe, Ph.D., Uniformed Services University and the Center for Neuroscience and Regenerative Medicine
Basic laboratory research on traumatic brain injury (TBI) employs a wide variety of injury models to study the molecular, cellular, and behavioral consequences of TBI in an attempt to identify recovery mechanisms and therapeutic targets. Behavioral recovery after TBI is a critical determining factor in evaluating the efficacy of potential therapies, and the usefulness of a specific behavioral test depends on the nature, severity, and location of the injury; time after injury at which the test is performed; and specific protocol or procedure of testing.
Mice (male C57Bl/6J) sustained a TBI by controlled cortical impact (CCI; a penetrating injury) at a parietal or frontal location, weight-drop (a concussive injury), or high-intensity-focused ultrasound (HIFU; a novel blast-like paradigm), and post-injury behaviors on multiple measures of motor activity, cognition, and anxiety were compared. Sham-operated (craniectomy) and non-operated mice were used as controls.
Mice that received a CCI brain injury showed greater differences or deficits on all behavioral measures than mice that received concussive or blast-like injuries, although the functional differences were location-dependent, with greater differences observed in animals that sustained hippocampal damage. Motor deficits after injury, if present, typically resolved within a few days whereas cognitive deficits persisted for several weeks.
The results from this direct comparison of brain injury models in the laboratory mouse emphasize the importance of carefully considering the nature of the TBI when choosing the specific tests, protocols, and timing of behavioral assessments after injury.
Sponsor: Center for Neuroscience and Regenerative Medicine (Henry M. Jackson Foundation for the Advancement of Military Medicine)
mouse, behavior, models
ENHANCED CORTICAL EXCITABILITY IN THE PERI-INJURY ZONE OF IMMATURE RATS AFTER EXPERIMENTAL TBI
Corey Goddeyne, BA, University of Arizona - College of Medicine Phoenix
Joshua Nichols, BA, University of Arizona - College of Medicine Phoenix
Anna Yoshihiro, BA, University of Arizona - College of Medicine Phoenix
Roxy Perez, BA, Barrow Neurological Institute at Phoenix Children's Hospital
Lucy Treiman, PhD, PsyD., St. Joseph's Hospital and Medical Center
P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Following traumatic brain injury (TBI) between 5 and 50% of patients will develop posttraumatic epilepsy (PTE), characterized by spontaneous recurrent seizures. The peri-injury zone surrounding the contusion undergoes marked cellular and synaptic reorganization. We hypothesized that neurons in this region may undergo unique electrophysiological changes that promote epileptiform activity.
A frontoparietal controlled cortical impact (CCI)(6mm tip, 4m/s, 2.0 mm depth) was performed on post-natal day 17 Sprague-Dawley rats to experimentally model TBI. Following CCI, animals were allowed to recover for 14 days prior to further experimentation. Age-matched control or CCI animals were sacrificed under deep anesthesia, brains removed and coronal cortical slices (350 uM) cut on a vibratome in 4 0C carboxygenated artificial cerebrospinal fluid (aCSF). Slices were incubated for 1 hour at 32 0C in aCSF prior to electrophysiological recording in a submerged chamber. Whole cell patch clamp recordings were performed on cortical pyramidal neurons under IR-DIC visual guidance using a standard physiological internal solution. A Multiclamp 700A patch-clamp amplifier was used in either current- or voltage-clamp. The membrane potential was clamped at −70 mV for all voltage clamp recordings. Recordings were sampled at 20 kHz, filtered at 10 kHz, captured on an A–D interface and stored on a computer. For convulsant challenge the potassium channel blocker 4-AP (100 uM) was bath applied.
Pyramidal neurons recorded from controlled cortical impact (CCI) animals were examined for changes in excitability at the intrinsic, synaptic and network levels. Compared to age-matched controls no significant change in intrinsic properties measured as mean firing rate, input resistance or resting membrane potential was observed. Synaptically, spontaneous post-synaptic currents (sPSCs) were monitored and similarly no difference in the amplitude or inter-event interval of events was observed. However, a marked increase in the probability of burst events and the spontaneous development of paroxysmal discharges was observed. Paroxysmal discharges were characterized by large amplitude, multi-vesicular release events and preferentially occurred in close proximity to the injury site. Similar discharges were not observed in control cortex suggesting that peri-injury cortex is hyperexcitable and spontaneously epileptiform. To further assess the seizure susceptibility of CCI animals we pharmacologically challenged cortical slices with a known convulsant, 4-AP. Under 4-AP challenge, CCI animals were hyper-responsive as compared to controls with an increase in the amplitude of sPSCs and probability of paroxysmal discharges.
Our results demonstrate that CCI induces changes in the peri-injury cortex that has profound effects on enhancing the excitability of cortical pyramidal neurons. Specifically, while CCI does not alter the examined intrinsic properties of individual neurons it significantly promotes synchrony in the network at rest and under convulsant challenge. This hyperexcitability is more pronounced in the peri-injury zone and further work will help to characterize the spatial extent of these changes. Localized changes in cortical excitability and network synchrony may play important roles in the underlying pathophysiology of TBI induced PTE and understanding these changes is a critical first step in the development of new targeted therapeutics.
This study was supported by the University of Arizona and the Barrow Neurological Institute.
electrophsyiology, cortex, post-traumatic epilepsy, hyperexcitable
VENTRICULOSTOMY ASSOCIATED INFECTION: A NEW STANDARDIZED DEFINITION, REPORTING FORMAT, AND INSTITUTIONAL EXPERIENCE
Daniel Harwell, MD, University of Cincinnati
Mark Magner, MD, University of Cincinnati
Norberto Andaluz, MD, University of Cincinnati
Lori Shutter, MD, University of Cincinnati
Published literature cite multiple definitions of ventriculostomy associated infections (VAI) subsequently leading to wide ranging rates of infection of uncertain significance. The authors propose a standardized definition of VAI, and a standardized reporting format compliant with CDC device related infections. Our five year institutional VAI rates are also reported.
A prospective infection control surveillance program for VAI is reviewed for the last five years of ventriculostomy utilization (Oct 2006 through December 2011). Defining VAI was based on literature review in addition to our institutional analysis and definition. The definition of VAI requires a positive CSF culture in a patient with ventriculostomy catheter, plus at least one of the following: Fever (defined as a recorded temperature greater than 101.5oF), CSF glucose level either <50mg/dL or <50% of a serum glucose level drawn within 24 hours of the CSF glucose. CDC compliant reporting format is used. Rates of VAI is calculated as follows: total number of VAIs as the numerator; total number of EVD device days as the denominator. This ratio will be multiplied by 1000, to report total VAIs per 1000 EVD device days.
The CDC compliant infection rate at our institution is 2.09 per 1000 ventriculostomy days. There was a total of 5740 ventriculostomy days charted for the recorded time with 12 total infections during that period. 525 patients had ventriculostomies inserted. This correlates with a 2.29% infection rate per patient compared to older literature reporting rates of VAI with a composite rate of 8.8%. The average number of days per ventriculostomy usage was 10.93. Neither antibiotic impregnated catheters nor peri-procedural or prophylactic antibiotics were used.
A standardized definition and reporting format of VAI is described. A standardized definition of VAI, and consistent reporting of VAI in a CDC style format will facilitate future study and guideline development. Our institutional rate of infection is 2.09 per 1000 ventriculostomy days. The authors propose a benchmark measure of infection to be less than or equal to 5 per 1000 device days. Antibiotics or antibiotic impregnated catheters would have little justification or impact on our pre-existing, low institutional infection rate.
none
Ventriculostomy, extraventricular drain, infection
DEVELOPING BIOMARKERS FOR TRAUMATIC BRAIN INJURY USING PHAGE DISPLAY
James Geddes, Ph.D., Department of Anatomy and Neurobiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Kathryn Saatman, Ph.D., Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Rodney Guttmann, Ph.D., University of West Florida
Traumatic brain injury (TBI) results in more than 50,000 deaths each year with many suffering long-term disability. In addition to efforts to prevent TBI, there has been an increased interest in identifying biomarkers for TBI to aid in diagnosis and evaluation of treatment efficacy.
Phage display is a powerful tool for selecting peptides, proteins or antibodies with specific binding properties from a large number of variants. The purpose is to screen for proteins with specific properties selected from libraries of DNA variants expressed into populations of protein variants on phage surfaces. The present study was undertaken to use phage display to pan serum samples from mice 6h and 24h after controlled cortical impact brain injury for the identification of protein markers unique to brain injury. To isolate a library of phage with high affinity for proteins in serum from brain-injured as compared to sham-injured mice, a factory produced phage library was purchased from New England Biolabs. Phage from the stock library was applied to the uninjured and injured samples sequentially. Unbound phages were washed off and the phages bound to the injured samples were eluted and amplified.
Individual colonies selected after three rounds of panning and amplification were used for ELISA testing of injured and uninjured serum samples. Our preliminary data identified one unique phage clone/sequence, which detected a statistically significant difference between the injured and the uninjured serum samples. The phage display protocol that we designed is capable of identifying likely unique markers of TBI.
This method can be used to develop an ELISA-based high-throughput assay to successfully differentiate not only injured from uninjured subjects, but potentially mild from severe TBIs.
Supported by NIH P01 NS058484.
TBI, Phage Display, Biomarker
DEVOLPMENT OF A NOVEL NEUROMONITOR FOR CLINICAL TBI RESEARCH: ENZYME-BASED MICROELECTRODE ARRAY FOR REAL-TIME IN VIVO DETECTION OF NEUROCHEMICALS
Jorge Quintero, PhD, University of Kentucky
Jason Burmeister, PhD, Quanteon
Michelle Stephens, MD/PhD, Ohio State University
Zhming Zhang, MD, University of Kentucky
Jed Hartings, PhD, University of Cincinatti
Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's Hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Greg Gerhardt, PhD, University of Kentucky, Department of Anatomy and Neurobiology
We hypothesize that enzyme-based microelectrode arrays (MEAs) can provide insight into the pahtophysiological mechanisms of TBI and detect potential biomarkers to guide care of TBI patients. Modern neurocritical care uses a multitude of neuromonitoring devices in an attempt to detect secondary insults to guide care of TBI survivors.
The goal of this study was to examine the translational potential and capabilities of enzyme-based microelectrode arrays (MEAs) as a novel neuromonitor for clinical TBI research. First, the performance of MEAs was examined after sterilization. Second, the ability of MEAs to continuously record for the initial 72 hours after implantation was examined by implanting glutamate-sensitive MEAs into the rat striatum and recording for multiple days. Third, the ability of MEAs to detect combinations of glutamate, lactate, pyruvate, and glucose on a single device was examined using in vitro calibrations. Fourth, the damage to the central nervous system from implantation of the MEAs was examined in a non-human primate. Lastly, the ability to transfer this methodology to an FDA-approved device was examined by coating an AD-TECH depth electrode for glutamate detection and testing the performance of the electrode both in vitro and in vivo.
88 of the MEAs (14/16) were functional after sterilization for the detection of glutamate. MEAs continuously recorded extracellular glutamate levels in the striatum of freely moving rats during the initial 72 hours after implantation (n=2). Using selective coatings, MEAs were able simultaneously measure combinations of (glutamate, lactate, pyruvate, and glucose) on a single device. MEAs produced minimal damage to the central nervous system of a non-human primate in comparison to the damage produced from the guide cannula. Modifications to an AD-TECH depth electrode permitted selective measures of glutamate both in vitro and in vivo, detecting 9.0 uM extracellular glutamate in the rat striatum with local application of 500 uM glutamate producing a glutamate signal with a maximum amplitude of 109.3 uM. Also, we were able to modulate extracellular glutamate levels with an i.p. injection of urethane, reducing extracellular glutamate by ∼75 from 1.8 uM to 0.4 uM glutamate.
MEAs exhibit the capabilities to be a useful tool for clinical TBI research. MEAs remain functional after sterilization with a high success rate. MEAs can continuously measure glutamate during the initial 72 hours after implantation, the temporal window for disruptions in extracellular glutamate detected in TBI survivors. Selective coatings applied to MEAs allow for simultaneous detection of multiple neurochemicals (glutamate, lactate, pyruvate, and glucose), molecules identified as potential biomarkers in severe TBI patients. MEAs produce minimal damage to the central nervous system. We were able to transfer the methodology of MEAs onto an FDA-approved device for neurochemical detection producing easier access into the clinic. In conclusion, MEA technology has tremendous potential as a clinical tool to help understand the pathophysiology of TBI and possibly detect a reliable biomarker(s) to aid in the diagnosis of TBI severity, predict functional outcomes, and test the efficacy of therapeutics.
Support provided F31NS067899. GAG is the owner of Quanteon LLC. JEQ has served as consultant to Quanteon LLC. AD-TECH Medical Corporation provided probes.
neuromonitors, glutamate
THE NEUROPROTECTIVE EFFECT OF GERANYLGERANYLACETON IN THE EXPERIMENTAL TRAUMATIC BRAIN INJURY
David J. Loane, PhD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Bogdan A. Stoica, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Stephanie Custer, University of Maryland
Titilola Akintola, University of Maryland
Alan I. Faden, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
David J. Loane, University of Maryland
Geranylgeranylaceton (GGA) induces heat shock protein 70 expression and is neuroprotective in various brain injury models by reducing post-traumatic neuroinflammation and apoptosis mechanisms. Little work has been done to assess the effects of GGA treatment on long-term functional recovery and histological outcomes following experimental traumatic brain injury (TBI).
GGA (800mg/kg) or vehicle was administered (p.o.) to C57Bl/6 mice either at 48 hours prior to moderate-level controlled cortical impact injury (CCI; study 1) or 3 hours after TBI (study 2). A battery of behavioral tests including beam walk test, Morris water maze, reversal Morris water maze, novel object recognition, tail-suspension test, and open field test were evaluated in studies 1 and 2. For study 1, histological outcomes such as lesion volume and neuronal cell numbers in the hippocampus (CA1, CA2/3, and DG sub-regions) as well as ipsilateral cortex and thalamus were quantified at 28 days post injury (PID) by unbiased stereological methods. In addition, HSP70 expression and α-fodrin cleavage was measured in the ipsilateral cortex at 6 hours post-injury by Western immunoblotting.
Study 1: GGA pre-treatment improved sensorimotor function in the beam walk task at 21 PID, as well as cognitive/affective functions in the Morris water maze (PID14-23), novel object recognition (PID24, 25), and tail-suspension tests (PID24). A composite behavioral score including the standard MWM probe test, reversal MWM probe test, tail-suspension test, and beam walk task, showed better separation between the GGA and vehicle groups as compared to any single behavioral test alone. GGA pre-treatment also reduced lesion volume as well as neuronal cell loss in the CA2/3 and DG sub-regions of hippocampus, ipsilateral cortex and thalamus when compared with the vehicle-treated group. In addition, GGA pre-treatment attenuated post-traumatic programmed cell death mechanism as demonstrated by reduced α-fodrin cleavage after TBI. Study 2: GGA post-treatment at 3 hours after TBI improved performance in the beam walk task at PID21 and the tail-suspension test at PID24. There were trends to improved cognitive performance in the Morris water maze in GGA-treated animals in study 2, but they failed to reach statistical significance. However, in search strategy analysis GGA-treated animals exhibited significantly higher reliance on spatial search strategies and lower reliance on looping search strategies than vehicle treated animals.
GGA pre-treatment attenuated sensorimotor, cognitive and affective deficits after TBI in mice. These behavioral improvements were correlated with decreased cortical lesion volume and improved neuronal survival in the cortex, thalamus and the CA2/3 and DG sub-regions in the hippocampus. Futhermore, GGA pre-treatment attenuated post-traumatic programmed cell death mechanisms, which may contribute to the protective effects of pre-treatment. Similarly, GGA post-treatment at 3h after TBI significantly improved sensorimotor, affective and cognitive functions after TBI. These findings indicate that pre- or post- treatment with the HSP70 inducer, GGA, promotes significant neuroprotective effects in a well-characterized rodent TBI model.
This work was supported by NIH Grant s R01 (NS052568) and R01 (NS061839) to A.I.F.
Geranylgeranylaceton, TBI, neuroprotection, functional recovery
EFFECTS OF ROLIPRAM ON OUTCOME AFTER CONTROLLED CORTICAL IMPACT IN MICE
Maria L. Cepero, B.S., University of Miami Miller School of Medicine
David Sequeira, B.S., University of Miami Miller School of Medicine
Daniel J. Liebl, Ph.D., University of Miami Miller School of Medicine
W. Dalton. Dietrich, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Traumatic brain injury (TBI) pathology includes contusions, cavitation, cell death, all of which are exacerbated by inflammation. We hypothesized that an anti-inflammatory drug, rolipram, may reduce pathology after TBI. In several CNS injury models, rolipram has been found to reduce inflammation and improve cell survival and functional recovery.
Adult male C57BL/6 mice (25–30 gm, 2 mos) received a 5 mm craniotomy over the right parietotemporal cortex. Vertically directed CCI was delivered with a pneumatic cylinder with a 3 mm impounder at a velocity of 6 m/sec and a depth of 0.5 mm. Naïve controls were used for comparison. At 30 min post-surgery, animals were treated with vehicle (5% DMSO in saline, 2 ml/kg) or rolipram (1 mg/kg, intraperitoneally), and then once per day for 3 days (n=6 mice/treatment). At day 3, animals were anesthetized, perfused, and brains were systematically sectioned (10 μm, 150 μm apart) to visualize the resulting pathology using hematoxylin and eosin (H&E) staining. Using Neurolucida 7.50.1 software, the contusion and cavity in the parietal cortex were traced over the entire extent of the injury at 4× magnification.
Total parietotemporal cortical contusion and cavity volume was significantly increased in rolipram-treated CCI animals as compared to vehicle-treated CCI animals (2.99±0.17 mm3 CCI+vehicle versus 4.63±0.34 mm3 CCI+rolipram, n=6/group, p<0.01). Contusion areas at specific bregma levels (−1.1, −1.6, −2.7, −3.3 mm) were analyzed and indicated a significant effect of drug across bregma levels (p<0.05). Total cavity volume excluding the surrounding contused cortex was not significantly different between treatment groups.
These results indicate that rolipram treatment significantly worsened contusion volume after CCI. Although rolipram is well known to reduce pathology and inflammation in several other CNS injury models such as spinal cord injury and cerebral ischemia, the pathology resulting from TBI was worsened with rolipram treatment at this particular dose and administration schedule. Further studies are underway to determine if neuronal survival is similarly affected by rolipram treatment after CCI. These studies suggest that consideration of the unique characteristics of TBI pathology is important in the extrapolation of promising therapeutic interventions from other CNS injury models.
Supported by NIH/NINDS NS069721 and NS056072 and The Miami Project to Cure Paralysis.
contusion, CCI, inflammation, mice, rolipram
TIMELINE OF NEURONAL AND ASTROCYTIC CELL DEATH FOLLOWING CRANIOTOMY FOR ACUTE SUBDURAL HEMATOMA
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Markus Spurlock, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Professor Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shyam Gajavelli, PhD., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine,
Ishna Sharma, B.S., University of Miami Miller School of Medicine
Daniel Diaz, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Helen M. Bramlett, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton Dietrich, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Acute subdural hematoma (ASDH) significantly contributes to mortality following traumatic brain injury. Extracellular ubiquitin C-terminal hydrolase (UCH-L1) and glial fibrillary acidic protein (GFAP) are biomarkers for neuronal and astrocytic cell death respectively. We hypothesize that after ASDH these markers will exhibit unique elevation profiles, measured by cerebral microdialysis (MD).
Acute subdural hematoma was induced in Sprague-Dawley rats. Rats were allocated to either ASDH or sham groups with 7 rats in each. The ASDH group received an infusion of 350 μL of autologous blood into the subdural space over a 7-minute period. Craniotomy was performed in both groups; 2.5 hours after initiation of hematoma in the ASDH group. Temperature was held constant at 37C. Microdialysate (MD) samples were collected through a small burr hole at 4 time points: early ischemic phase (0 to 2 hours after ASDH induction), craniotomy phase (2 to 3 hours after ASDH induction), early reperfusion phase (3 to 5 hours after ASDH induction), and late reperfusion phase (5 to 6 hours after ASDH induction). Microdialysate samples were sent to Banyan Biomarkers to determine [UCH-L1] and [GFAP] in samples. Concentrations were analyzed based on two-way ANOVA followed by a post hoc Bonferroni test.
In the ASDH group UCH-L1 concentration was not significantly different than that of the sham group until the early reperfusion phase, 0.5 to 2.5 hours following decompressive craniotomy (ASDH vs Sham; 50.2±28.3 ng/dl vs 3.1±1.3ng/dl; p<0.05). GFAP concentration was not significantly different between either group until the late reperfusion phase, 2.5 to 3.5 hours following craniotomy (ASDH vs Sham; 3.3±1.0 vs 15.3±8.0 ng/dl; p<0.05). Sham animals demonstrated no significant changes in the concentrations of either biomarker at any time point.
The timeline of neuronal and astrocytic cell death, as evidenced by cell-specific proteins, sheds light on the pathophysiology underlying ASDH. From these measurements it can be concluded that the majority of neuronal death, and astrocyte activation takes place during reperfusion following decompressive craniotomy, likely representing an ischemia-reperfusion type of injury. Furthermore, astrocyte activation/lysis follows neuronal cell death sequentially, and this is consistent with their different vulnerability to ischemic/reperfusion damage. This experiment offers novel information regarding the mechanism underlying injury with subdural hematoma. Importantly, the study elucidates the period when neurons are the most vulnerable, and at which point neuroprotective therapies might be applied to improve outcomes.
I am sincerely grateful to Dr. Shoji Yokobori, Dr. Shyam Gajavelli, and Dr. Ross Bullock for their guidance and contributions.
ASDH reperfusion injury GFAP UCHL1
DYNAMIC INTRACRANIAL PRESSURE (PRX) AND CEREBROVASCULAR (CVRX) REACTIVITY REFLECT IMPAIRED CEREBROVASCULAR AUTOREGULATION AT HIGH INTRACRANIAL PRESSURE (ICP)
Gloria Statom, B.S., Department of Neurosurgery, University of New Mexico School of Medicine
Edwin M. Nemoto, Ph.D., Department of Neurosurgery, University of New Mexico School of Medicine
We previously showed that determination of critical cerebral perfusion pressure (CPP) by static cerebral blood flow (CBF) autoregulation fails at high ICP due to microvascular shunting (MVS). This study determines whether critical CPP measured at high ICP by dynamic ICP and CBF reactivity will accurately detect impaired CBF autoregulation.
In vivo 2-photon laser scanning microscopy (2PLSM) over the rat parietal cortex was used to measure microvascular red blood cells flow velocity (fluorescein dextran), NADH autofluorescence (tissue oxygenation) and blood brain barrier (BBB) integrity (fluorescein dextran leakage). Step changes in CPP from 70 to 50 and 30 mmHg were made by increasing ICP with vertical positioning of an artificial cerebrospinal fluid reservoir connected to the cisterna magna. Doppler flow, rectal and cranial temperatures, ICP, arterial pressure and arterial blood gases were monitored. At each CPP, a transient 10 mmHg rise in mean arterial pressure (MAP) was induced by i.v. dopamine bolus. Intracranial pressure reactivity (PRx) was calculated as the ratio of the change in ICP in response to a 10 mmHg MAP increase (PRx=ΔICP/ΔMAP). Cerebrovascular reactivity (CVRx) was defined as a ratio of the change in CBF with a 10 mmHg MAP change (CVRx=ΔCBF/ΔMAP).
Static CBF autoregulation curves indicated a critical CPP of 30 mmHg when CPP was reduced by increasing ICP. However, using in vivo 2PLSM we observed that reduction of CPP to 50 mmHg, achieved by increasing ICP, caused a transition from low velocity capillary to high velocity microvascular shunt flow associated with tissue hypoxia, brain edema and BBB opening. At a normal CPP of 70 mmHg the acute blood pressure challenge caused either a decrease or no change in ICP (PRx=-0.04±0.08, n=4), i.e., intact pressure reactivity. When CPP was decreased to 50 and 30 mmHg by ICP elevation, PRx increased to 0.08±0.14 and 0.26±0.13, respectively. Similarly at a normal CPP of 70 mmHg, MAP challenge induces no change or a decrease in CVRx (-0.03±0.06), reflecting intact cerebral autoregulation. When CPP was decreased to 50 and 30 mmHg by raising ICP, CVRx increased to 0.09±0.11 and 0.27±0.15, respectively, reflecting impaired autoregulation.
Our studies show that at high ICP the critical CPP is 50 mmHg as shown by measurements of PRx and CVRx and where microvascular shunting, brain edema, tissue hypoxia and BBB leakage begin to occur. This critical CPP at high ICP is higher than the 30 mmHg determined by static CBF autoregulation because high velocity MVS flow mimics falsely preserved autoregulation. Dynamic cerebrovascular reactivity and intracranial pressure reactivity by dopamine-induced MAP challenge revealed a failure in cerebrovascular autoregulation during intracranial hypertension at a CPP of 50 mmHg which corresponds to a transition from capillary to MVS flow as measured by 2PLSM.
This work was supported by the NIH CoBRE P30 Pilot Project (8P30GM103400-01) and Pilot Project Grant from the UNM SOM Dedicated Health Research Funds.
intracranial hypertension, autoregulation, CVRx, PRx
A MODEL OF BLAST-INDUCED TRAUMATIC BRAIN INJURY IN MICE
YungChia Chen, BS, MS, University of Pennsylvania
Tapan Patel, BS, University of Pennsylvania
Nicolas Jaumard, PhD, University of Pennsylvania
Tanya Merdiushev, University of Pennsylvania, Department of Bioengineering
Mr. Matthew B. Panzer, MASc, Duke University
Beth Winkelstien, PhD, University of Pennsylvania, Department of Bioengineering
Barclay Morrison, PhD, Columbia University
Cameron R. Dale Bass, PhD, Duke University
David F. Meaney, PhD, University of Pennsylvania
Increased focus on blast-induced traumatic brain injuries (bTBI) emphasizes the need to develop and characterize models of bTBI. This study presents a new methodology to investigate murine blast-induced brain injury, using biomechanical measures to compare this model to previous models of acceleration-induced traumatic brain injury.
Simulated blast waves were created using a shock tube designed to produce realistic shock tube overpressure time histories similar to that near an exploding 105 mm mortar. The overpressure wave was directed at the mouse's head, positioned roughly 0.75” away from the exit of the aluminum tube, in a dorsal to ventral direction. The mouse torso was protected by an aluminum tube. After blast exposure, the mouse was removed and righting time was recorded. High-speed video was recorded (22,099 frames/second) to analyze the motion of the mouse head (displacement, velocity, acceleration) resulting from exposure to the shock wave. Pressure transducers at the exit of the shock tube recorded the magnitude and timing of the shock wave input. Animal position was adjusted to align the head center with the center of the shock tube or the periphery, and different viscoelastic materials were placed under the mouse to produce different acceleration/deceleration histories.
A series of in vivo tests showed a sharp lethal exposure limit to a shock wave. We adjusted the magnitude of shock wave input injury by varying the thickness of the polyester membrane and found that a thickness of .02” produced a shock wave (average peak overpressure: 207±3.5 kPa, average duration: 0.617±.005 ms standard deviation) that would administer the upper limit of injury severity without causing gross anatomical disfigurations or death.
Below the lethal exposure limit, we found that blast injured mice have significantly longer righting time relative to shams (332±83 s compared to 35±3 s standard error, p<.05) when exposed to a narrow range of blast input conditions (average peak overpressure: 207±3.5 kPa, average duration: 0.617±.005 ms standard deviation). Animals showed no obvious signs of pulmonary distress or subsequent weight loss at this loading level. However, animal position in the shock tube was an important influence on some external signs of damage. Damage to the external ear surface is common when the head is aligned with the center of the tube, while damage does not appear when the head is positioned along the periphery.
From the high speed video analysis (n=10 total tests), the average maximum head accelerations (az) on the 40 durometer sorbothane base were −15,945±5,398 m/s2 and +15,691±4,978 m/s2 (standard deviation), while the maximum accelerations along the horizontal axis (ax) were less than 10% of each value. Using a stiffer sorbothane material under the animal, the average maximum accelerations (az) were −15,698±1,067 m/s2 and +16,353±2,129 m/s2 (standard deviation).
In this study we utilized a shock tube to study the exposure limit and biomechanics of blast-induced traumatic brain injury in mice with protected torsos. We found a sharp threshold for survivability to the shock wave input. Animals exposed to a shock wave input below this threshold had immediate, significant impairment. Mouse position was an important variable to control in achieving consistent, minimal signs of gross external damage while maintaining some neurological impairment. The resulting head accelerations across different surfaces were unexpectedly high, with a significant level of fluctuation occurring for a relatively narrow range of input blast conditions. These biomechanical data are similar in magnitude to accelerations reported for impact acceleration models in rodents. Together, these data show that models of blast induced traumatic brain injury can include significant inertial loading components, and this factor needs consideration when interpreting model designs to study the unique components of bTBI.
Funding provided by the Department of the Army grant W911F-10- 1-0526
blast, tube, mouse, shock wave
MITOCHONDRIA-TARGETED NITROXIDE THERAPY IMPROVES LONG TERM NEUROBEHAVIORAL OUTCOME IN PEDIATRIC TRAUMATIC BRAIN INJURY
Anthony E. Kline, Ph.D., University of Pittsburgh
Andrew Amoscato, Ph.D., University of Pittsburgh
Louis Sparvero, Ph.D., University of Pittsburgh
Mioara D. Manole, M.D., University of Pittsburgh
Jeffrey P. Cheng, B.S., University of Pittsburgh
Jesse Lewis, B.S., University of Pittsburgh
Henry L. Alexander, B.S., University of Pittsburgh
Bruno Fink, M.D., Noxygen Science Transfer & Diagnostics
Dr. Patrick M. Kochanek, M.D., Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Peter Wipf, Ph.D., University of Pittsburgh
Valerian E. Kagan, Ph.D., D.Sc., University of Pittsburgh
Hülya Bayir, M.D., Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Stable nitroxide radicals targeted to mitochondria combine electron and free radical scavenging actions with superoxide dismutase mimicking and recycling capacities. We previously showed a novel mitochondria-targeted conjugate of hemi-gramicidin-S with a nitroxide radical, XJB-5-131, partitioned into neuronal mitochondria, and decreased neuronal death after in vitro and in vivo TBI.
In the current study, we applied novel imaging tools to detect the distribution of XJB-5-131 in the brain and evaluated the effect of XJB-5-131 on TBI-induced oxidation of cardiolipin (CL), a mitochondria specific phospholipid, long-term neurobehavioral outcome, and cortical lesion volume after CCI (4±0.2 m/sec, depth=2.5mm) in PND17 rats.
By employing micro computerized tomography and L-Band in vivo EPR imaging, we documented a time- and dose-dependent distribution of XJB-5-131 in naïve PND 17 rat brains after i.v. injection (10 mg/kg). We further utilized mass spectrometry imaging and confirmed the presence of XJB-5-131 in the brain tissue after a single i.v. dose (10 mg/kg). Direct quantification of XJB-5-131 in the brain tissue by liquid chromatography mass spectrometry (LC-MS) also demonstrated its accumulation to a level of 16.5±4.3 pmole/gram of tissue (mean±s.d.) at 3 h after a single 10 mg/kg i.v. injection. 2-Dimensional LC-MS analysis revealed almost 190 individual molecular species of CL in normal brain, of which only 10 were oxygenated. CCI induced oxidation of a majority of polyunsaturated molecular species of CL; the number of non-oxidized CL species decreased to ∼100, while that of the oxygenated species increased to 166. Quantitatively, the content of oxidized CL species increased 20-fold at the expense of decreased amounts of non-oxidized CL. CL oxidation was almost completely blocked by XJB-5-131 administered 10 min after CCI. Novel object recognition (NOR) was utilized to test memory retention. XJB-5-131 increased the frequency of investigating a novel object v.s. vehicle (59.8±2.9% vs. 46.7±1.5%; mean±s.e., n=10/group; p<0.01). And XJB-5-131 enhanced acquisition of spatial learning assessed by Morris water maze on post-injury days 30–34 v.s. vehicle treatment (p<0.01, n=10/group). Furthermore, we assessed the cortical lesion volume by unbiased stereology. XJB-5-131 reduced the lesion volume at 5 weeks after CCI v.s. vehicle treatment (27.41±1.26 v.s. 31.97±1.59 mm3; mean±s.e., n=7–8/group; p<0.05).
Taken together with previous findings, our data indicate that XJB-5-131 is preferentially located in mitochondria, penetrates into CNS, decreases CL oxidation, improves long term neurobehavioral outcome, and decreases the cortical lesion volume after CCI.
Support: NS061817, U19A1068021, W81XWH-09-2-0187
nitroxide, blood-brain-barrier, cardiolipin, mitochondria, neurobehavioral
PROBENECID INCREASES N-ACETYLCYSTEINE BRAIN PENETRATION FOLLOWING EXPERIMENTAL PEDIATRIC TRAUMATIC BRAIN INJURY
Henry L. Alexander, BS, University of Pittsburgh
A. Jacob Ocque, MS, School of Pharmacy, University of Pittsburgh
Patrick J. Oberly, BS, School of Pharmacy, University of Pittsburgh
M. Beth Minnigh, PhD, School of Pharmacy, University of Pittsburgh
Thomas D. Nolin, PharmD, PhD, School of Pharmacy, University of Pittsburgh
Hülya Bayır, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Samuel M. Poloyac, PharmD, PhD, School of Pharmacy, University of Pittsburgh
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Robert S.B. Clark, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Oxidative stress plays an important role in TBI pathophysiology but antioxidants have not shown clinical efficacy possibly due to poor brain penetration. Xenobiotic transporters limit drug entry into the brain. We hypothesized that the transporter inhibitor, probenecid, would increase brain concentrations of N-acetylcysteine in a preclinical model of pediatric TBI.
The effects of probenecid on N-acetylcysteine plasma concentrations and brain penetration were measured in juvenile rats (16–18 days old; n=4/group) under control conditions and following a controlled cortical injury (2.5mm deformation at 4 m/s; n=5–6/group) in two separate studies. In each investigation (10 min post CCI in that study), rats received intraperitoneal doses of N-acetylcysteine (163 mg/kg), probenecid (150 mg/kg), or the combination. Serial plasma concentrations and terminal total brain hemispheric levels of both drugs were quantified using novel ultra-high performance liquid chromatography tandem mass spectrometry assays (UPLC-MS/MS). All data was reported as mean±SD and compared by t-tests.
N-acetylcysteine and probenecid UPLC-MS/MS assays were validated for both rat plasma and brain homogenates (R2>0.99, inter-/intra-assay variation <13% for both). In control animals, co-administration of probenecid increased overall plasma N-acetylcysteine exposure 1.84-fold (AUC1–6h: 1.16±0.21×105 vs 2.13±0.40×105 ng/mL*h; p=0.006); whereas probenecid levels were unchanged by the N-acetylcysteine combination. At 6h, probenecid levels achieved 19.2±4.8 μg/g tissue in brain tissue and increased N-acetylcysteine levels from undetectable levels (limit of quantification for the assay was 6.8 μg/g tissue) to 57.4±21.5 μg/g tissue (brain:blood ratio of 1.6±0.3%). Similarly, following injury at 1h, N-acetylcysteine plasma concentrations increased 1.76-fold (59.0±7.5 to 104.1±11.1 μg/mL; p<0.0001) and probenecid plasma levels were not altered by N-acetylcysteine. Probenecid achieved high concentrations rapidly in the brain following TBI (∼300 μg/g tissue in brain tissue at 1h). N-acetylcysteine brain penetration was low at this early time point after injury when given alone (0.14% of plasma), but was significantly increased 2-3-fold by probenecid in both the ipsilateral (0.50±0.19 to 1.0±0.29 μg/g tissue; p=0.0088) and contralateral hemispheres (0.08±0.03 to 0.25±0.02 μg/g tissue; p<0.0001).
Co-administration of the transporter inhibitor, probenecid, significantly increases N-acetylcysteine systemic exposure and brain penetration in juvenile rats suggesting that N-acetylcysteine is a transporter substrate. In a preclinical model of pediatric TBI, N-acetylcysteine penetration was increased 2–3 fold by probenecid as early as 1h after injury. Transporter inhibition is a promising therapeutic strategy to improve the delivery of N-acetylcysteine (and perhaps other pharmacologically similar therapies) to the brain after TBI. Further, as both probenecid and N-acetylcysteine are currently FDA-approved for other indications, this combination has the potential for timely clinical application for efficacy in TBI patients. An NIH-supported Phase I pharmacokinetic study in children after TBI is underway.
This work is supported by NIH grants: KL2RR024154, NS069247, NS030318, S10RR023461.
TBI, transporter, pharmacokinetics, probenecid, N-acetylcysteine
L-NAME REDUCES LEVELS OF REACTIVE OXYGEN SPECIES & IMPROVES GAP JUNCTION COUPLING AFTER RAPID STRETCH INJURY IN VASCULAR SMOOTH MUSCLE CELLS, IN VITRO
Douglas S Dewitt, PhD, UTMB
Donald S Prough, MD, UTMB
L-NAME improves vascular function and outcome after experimental traumatic brain injury. Gap junctions (GJ), low-resistance channels between adjacent cells, contribute to normal cerebral-vascular function. The effects of L-NAME on reactive-oxygen-species (ROS) levels and GJ communication were examined in vascular-smooth-muscle cells subjected to rapid stretch injury (RSI) in vitro.
Rat vascular smooth muscle cells were cultured on collagen-coated Flex Plate wells in DMEM + 10% inactivated FBS overnight and subjected to rapid stretch injury (0, 30, 40, 50 psi for 50 ms) using a Model 94A Cell Injury System or sham injury. Immediately after injury, the cells were treated with L-NAME (100 mM) or a vehicle for 30 min post-injury. Intracellular ROS levels were measured in cells loaded 5,6-Chloromethyl-2′,7-dichlorodihydrofluorescein diacetate,acetyl, mixed isomers(CM-H2DCFDA). Gap junction coupling was assessed by measuring fluorescence recovery after photobleaching (FRAP) in cells loaded with 5,6-CFDA/AM. ROS levels and GJ coupling were determined 24 hrs post-injury by measuring fluorescence intensities in randomly selected fields using confocal microscopy (Ziess LSM-510 confocal microscope, 488-nm wavelength).
In the cells subjected to RSI and treated with L-NAME, intracellular ROS levels were reduced by 22–45% (P<0.01, RSI vs. vehicle RSI). Levels of FRAP In L-NAME-treated smooth muscle cells subjected to RSI, were 31–78% higher after RSI (P<0.01, moderate RSI vs. vehicle RSI; P<0.01, severe RSI vs. vehicle RSI)) compared to vehicle-treated RSI group.
These results that L-NAME reduced the levels of intracellular ROS and improved GJ communication after RSI in vascular smooth muscle cells suggest that L-NAME may improve cerebral vascular function after traumatic brain injury.
Optical Imaging Laboratory of The University of Texas Medical Branch Support from grant NS19355 from the NINDS/NIH and The-Moody-Center-for-Traumatic-Brain-&-Spinal-Cord-Injury-Research/Mission Connect
L-NAME, rapid-stretch injury, FRAP, TBI
DELAYED LOSS OF DENDRITIC SPINE DENSITY IN COLATERAL HIPPOCAMPAL NEURONS FOLLOWING LATERAL FLUID PERCUSSION INJURY IN THE RAT
Zaid Nawaz, BS, Virginia Commonwealth University
John Campbell, PhD, Virginia Commonwealth University
Previous studies have demonstrated that TBI results in acute loss, then significantly increased spine density. This study investigated whether or not the significant increased spine density was a transient or permanent phenomenon.
Rats were subjected to sham (n=5) or lateral fluid percussion injury (n=3), as routinely performed in the laboratory. At 2 wks post-injury, animals were perfused and brains stained using a rapid golgi-cox method (FD Rapid Golgistain Kit, FD Neurotechnologies, Inc., Baltimore, MD, U.S.A.). Dendrites were traced through the x-, y-, and z-planes using a digitizing microscope stage under control of Neurolucida (MicroBrightField, Inc., Williston, VT, U.S.A.). The lengths of tracings were computed by Neurolucida Explorer software (MicroBrightField, Inc.).
TBI resulted in significant modulation of dendritic spine density at 2 wks post-injury. Total spine density in CA1 pyramidal neurons was reduced in both apical and basal dendrites relative to sham-treated animals. The reduction in spine density appeared to be selective for immature projections, suggesting an underlying impairment of new spine formation. Of interest, the spine loss was limited to the contra-lateral hemisphere, and spine density in the ispilateral hippocampus was not significantly different from control levels. In contrast to CA1 neurons, total spine density in CA3 pyramidal neurons was not significantly affected at 2 wks. However there was a decreased density in mature spines in the ispilateral hippocampus.
Overall, the data demonstrate that TBI results in altered dendritic spine density up to two weeks post-injury. The data support the hypothesis that altered spine density may be a contributing factor for ongoing neuronal dysfunction following TBI.
Commonwealth Neurotrauma Initiative (#07-302E) and an ARDRAF grant to S.B.C.
Spine Density, histochemistry, pyramidal neuron
SELECTIVE VASOPRESSIN-1A RECEPTOR INHIBITION REDUCES BRAIN EDEMA, ASTROCYTIC CELL SWELLING AND V1AR AND AQP4 EXPRESSION FOLLOWING FOCAL TRAUMATIC BRAIN INJURY IN RATS
Xiuyin Liang, MS, Virginia Commonwealth University, Department of Neurosurgery
Naqeeb Abidi, MD, Virginia Commonwealth University, Department of Neurosurgery
Shanaz Parveen, MD, Virginia Commonwealth University, Department of Neurosurgery
Andrew McCall, Virginia Commonwealth University, Department of Neurosurgery
Scott C. Henderson, PhD, Virginia Commonwealth University, Department of Anatomy and Neurobiology
Keisuke Taya, MD, Virginia Commonwealth University, Department of Neurosurgery
Professor Clive M. Baumgarten, PhD, Virginia Commonwealth University, Department of Physiology and Biophysics
Aristotelis S. Filippidis, MD, PhD, Virginia Commonwealth University, Department of Neurosurgery
Traumatic brain injury results predominantly in astrocytic swelling resulting in cellular edema. Development of cellular edema requires transmembrane water fluxes that are augmented by vasopressin-regulated aquaporin-4 (AQP4). We tested whether selective inhibition of vasopressin V1a receptors(V1aR) would suppress AQP4 in astrocytes, reduce astrocytic edema, and thereby diminish TBI-induced edematous change.
TBI was produced in Sprague-Dawley adult rats (350 – 400 g) by lateral controlled cortical (CCI) impact (6.0 m/sec, 3 mm depth). Rats were randomly assigned to sham-vehicle, CCI-vehicle or CCI-treated with SR49059 (2.76mg/kg infused over 5 h) and sacrificed at 5 h post injury. Brain edema was determined by wet-dry methods; changes in astrocytic volume were quantified with GFAP immunohistochemistry (IHC) and routine light microscopy. GFAP, AQP4 and V1aR expression was assessed by immunoblotting and quantified by densitometry; morphometric microscopic assessment using multiple label fluorescent IHC and confocal microscopy was used to evaluate the injury-induced effect of astrocytic cell swelling, GFAP, AQP4 and V1aR up-regulation, and the potential effect of SR49059 to attenuate these changes. Data are expressed as mean±standard error of mean (SEM). Data was analyzed by one-way analysis of variance (ANOVA) followed by Tukey's post-hoc test. Values of P<0.05 were considered significant using SPSS software.
Selective V1aR inhibition significantly reduced brain edema after CCI. The brain water content of the right (injured) hemisphere was significantly higher in the CCI-vehicle treated group (n=6, 80.46±0.33 %, P<.05) compared to the sham animal group (n=6, 78.26±0.14 %), and the CCI-SR49059 treated group (n=6, 78.96±0.18 %).
Analysis of astrocytic area (μm2) in sham animals (9.52±0.87) was significantly less than that in injured animals (18.03±1.47). Treatment with SR49059 significantly reduced astrocytic area (9.40±0.77; P<.05). Fluorescence intensity analysis expressed in arbitrary fluorescence units of the expression of GFAP, AQP4 and V1aR showed an injury increase in GFAP (348.5±38.4) relative to sham values (55.7±4.8) which was reduced following SR49059 treatment (244.3±30.3) (P<0.05). Parallel assessments of V1aR demonstrated an injury induced up-regulation (601.2±70.6) compared to sham (117.6±13.5), which was reduced by treatment (390.3±75.7) (P<0.05). Intensity analysis of AQP4 fluorescence showed a significant increase from sham values (158.3±5.5) following CCI (818.4±117.3), which was reduced by inhibition of V1aR (457.8±55.1) (P<0.05).
Qualitative evaluation of triple labeled images using GFAP, AQP4 and V1aR show an injury induced swelling of GFAP positive astrocytes that demonstrate an up-regulation in the distribution of AQP4 within the astrocytic end-feet. V1aR expression seemed to co-localized within the astrocytic cell body. These changes were prevalent within the cortical penumbra and surrounding blood vessels Treatment with SR49059 reduced the volume of astrocytic cell swelling and V1aR distribution as well as the distribution of AQP4.
The results of these studies support the notion that TBI is predominantly cellular and can be effectively modulated pharmacologically following focal CCI in rat. Further, we show that there is a synergistic interplay between astrocytes, AQP4 and V1AR following TBI that lead to cellular swelling. Vasopressin receptor inhibition with SR49059 reduces injury induced up regulation of GFAP, V1A and AQP4 and therefore blunts edematous change. The tripartite distribution of AQP4 and V1a receptors within astrocytes suggests a significant pathway for the entry of water in the brain that is “turned on” following injury. Our findings suggest that SR49059 and V1aR inhibitors may be potential therapeutic targets blocking this pathway, thus preventing cellular swelling and providing treatment for brain edema.
This work was supported by NIH 5R01NS019235 and NIH-NINDS center core grant (5P30NS047463-02) for Microscopy
edema, astrocyte, AQP4, V1aR, CCI
COMPARISON OF THE BLOOD-BRAIN BARRIER DISRUPTION TIMELINE AND NEUROINFLAMMATION POST-PENETRATING BALLISTIC-LIKE BRAIN INJURY IN RATS
Casandra Cartagena, Ph.D., Walter Reed Army Institute of Research
Deborah A. Shear, Ph.D., Walter Reed Army Institute of Research
Xi-Chun M. Lu, Ph.D., Walter Reed Army Institute of Research
Melissa Kolpoko, Walter Reed Army Institute of Research
Jitendra Dave, Ph.D., Walter Reed Army Institute of Research
Frank C Tortella, Ph.D., Walter Reed Army Institute of Research
The purpose of this study was to elucidate the timeline of blood-brain barrier (BBB) disruption in a rat model of penetrating ballistic-like brain injury (PBBI) using small and large molecular tracers and compare it with neuroinflammatory markers, cellular infiltration, and activation of native central nervous system (CNS) inflammatory cells.
Rats received a 10% unilateral frontal PBBI. Horseradish peroxidase (HRP; 44 kDA) or biotin-dextran amine (BDA; 3 kDA) were injected intravenously and allowed to circulate 30 minutes prior to perfusion at 4h, 24h, 48h, 72h, and 7 days post-PBBI. Experimental groups included PBBI+HRP (n=36) or PBBI+BDA (n=36), with appropriate injured and non-injured control groups (n=6/group/timepoint). Each group was stained for the appropriate tracer. Immunostaining for GFAP, MPO, and OX18 was conducted on serial sections of the PBBI+BDA group and corresponding control brains. Threshold measurements were obtained on digitized images of tissue sections to quantify tracer extravasation and degree of immunostaining.
Significant increases in extravasation of both BDA and HRP tracers were evident in the injured hemisphere as early as 4h and were completely resolved by 72h post-PBBI. Both BDA and HRP extravasation profiles peaked at 24h. However, average extravasation of BDA was consistently higher than HRP across all time points and was ∼2×the magnitude of HRP at 24h post-PBBI. Significant immunohistochemical results (i.e. GFAP, MPO, OX18) were detected in the injured hemisphere only. A significant increase in the presence of MPO-labeled cells (i.e. polymorphonuclear cells and monocytes) was detected as early as 4h post-injury and peaked at 24–48h, but did not resolve completely until 7d post-injury. Significant GFAP activation was evident as early as 4h post-PBBI and sustained out to 7d post-injury in both perilesional and subcortical regions. In contrast, OX18 expression appeared to be biphasic with significant upregulation detected first at 48h and again at 7d post-injury only.
The study successfully added to our knowledge of the timeline of increased BBB permeability post-PBBI. Early but differential extravasation of the small and large molecular tracers suggests that BBB opens in a markedly rapid and gradient fashion following injury. Timelines of increased BBB permeability and infiltration of peripheral inflammatory cells supports a strong association between level of BBB disruption and the ability of circulating cells to quickly relocate to the site of CNS injury. In contrast, the differential profiles of GFAP and OX18 upregulation suggest diverse post-injury roles for the two cell types, or that differing factors are triggering astrocytic and glial activation. The more delayed and biphasic upregulation of OX18 in microglia imply stronger responsiveness to secondary inflammatory mediators compared to direct injury-related factors.
This research was funded by the Army Combat Casualty Care Research Program.
PBBI, Neuroinflammation, Blood-brain barrier, TBI
DOSE-RESPONSE EFFECTS OF C-10068 ON ATTENUATION OF NONCONVULSIVE SEIZURES INDUCED BY PENETRATING BALLISTIC-LIKE BRAIN INJURY IN RATS
Philip B. Graham, Ph.D., CoNCERT Pharmaceuticals. Inc
Ying Cao, B.S., Walter Reed Army Institute of Research
Jessie Harris, B.S., Walter Reed Army Institute of Research
Deborah Shear, Walter Reed Army Institute of Research
Kara Schmid, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
C-10068 is a deuterium-substituted analog of dextroethorphan, a member of the (+)-morphinan class which also includes dextromethorphan. C-10068 has similar pharmacology to dextromethorphan, with substantially increased metabolic stability. Because dextromethorphan possesses anti-seizure properties, we tested C-10068 in a clinically relevant rat model of nonconvulsive seizures induced by penetrating brain injury.
All rats received 10% unilateral frontal penetrating ballistic-like brain injury (PBBI). Continuous (72 h duration) intravenous infusions of C-10068 (1.0, 2.5, or 5.0 mg/kg/h) or vehicle were initiated at 30 min post-PBBI. Nonconvulsive seizures (NCS) were detected by continuous video-EEG recordings over the 72 h infusion period. At the end of the experiment, rats were perfused and the brain tissues harvested for H&E staining for lesion volume analysis. Effects of C-10068 on NCS were evaluated using the following parameters: NCS incidence (number of animals exhibiting seizures per group), NCS frequency (number of seizure events per animal), NCS duration (in seconds), and NCS onset latency (in hours).
Seizure profile analysis revealed that 75% of vehicle-treated PBBI animals experienced NCS which occurred sporadically during the 72 h recording period. Treatment with C-10068 reduced NCS incidence by 40–50% independent of dose. On average, vehicle-treated animals had 8.3 NCS events/rat with total seizure duration of 379 sec/rat. C-10068 treatments dose-dependently (low-to-high dose) reduced NCS frequency by 20%, 42%, and 70% (p<0.05) and NCS duration by 30%, 41%, and 82% (p<0.05). C-10068 treatment also delayed the onset of NCS from 22 h (vehicle group) to 43–46 h (across C-10068 treated groups, p<0.05). No significant reductions in gross lesion volume were detected as the result of the C-10068 treatments.
Previous studies have shown that C-10068 attenuated electrical shock or chemically induced convulsive seizures in animal models. The results of this study confirmed its anti-seizure property in a post-traumatic NCS model. C-10068 binds to several potentially important CNS targets, including the sigma-1 receptor, with an overall pharmacological profile similar to dextroethorphan and dextromethorphan, both of which have established anti-seizure activities. In general, animal and human post-traumatic NCS is relatively more difficult to treat than convulsive seizures; therefore, the success of C-10068 in this study may in part be due to its improved metabolic stability. Future studies will focus on elucidating the mechanism of C-10068 as an anti-seizure drug candidate.
This research was funded by the Army Combat Casualty Care Research Program and CRADA with Concert Pharmaceuticals, Inc. (W81XWH-11-0098). C-10068 was provided by Concert Pharmaceuticals, Inc.
PBBI, nonconvulsive seizures, EEG, C-10068
DOSE-RESPONSE PROFILES FOR PROGESTERONE, DEXTROMETHORPHAN, AND SIMVASTATIN IN A MODEL OF PENETRATING BALLISTIC-LIKE BRAIN INJURY
Andrea Mountney, Ph.D., Walter Reed Army Institute of Research
Rebecca Pedersen, Walter Reed Army Institutes fof Research
Justin Sun, B.S., Walter Reed Army Institute of Research
Melissa Long, B.S., Walter Reed Army Institute of Research
Xi-Chun M. Lu, Ph.D., Walter Reed Army Institute of Research
Frank C Tortella, Ph.D., Walter Reed Army Institute of Research
The aim of this study was to identify the individual dose-response (D-R) curves of promising neuroprotective agents in the penetrating ballistic-like brain injury (PBBI) model for consideration as potential candidates for combination therapy drug development studies. Here we report the current D-R results for Progesterone, Dextromethorphan, and Simvastatin in the PBBI model.
Unilateral frontal PBBI was produced in the right hemisphere of isoflurane anesthetized rats (10% injury severity level). Intravenous (IV) 10m bolus infusions of each drug were delivered at the following doses/post-injury time points: Progesterone (PROG: 0.2, 1.0, 5.0, 10.0, or 25.0 mg/kg) at 0.5h and 6h post-injury and continued once/day out to 5 days post-PBBI. Dextromethorphan (DM: 1.0, 5.0, 10.0 or 20.0 mg/kg) at 0.5h, 2h, 4h, and 6h post-injury and continued once/day out to 72h post-PBBI. Simvastatin (SIM: 0.001, 0.01, 0.1, or 1.0 mg/kg) at 0.5h and 6h post-injury and continued once/day out to 10 days post-PBBI. Motor coordination and balance were evaluated on the fixed-speed rotarod task (7 and 10 days post) at speeds of 10, 15, and 20 rpm. Cognitive performance (acquisition, attention and retention) was evaluated on post-injury days 13–17 in the Morris water maze (MWM: 4 trials/day; 30m intertrial interval).
Motor testing revealed significant deficits in all injury groups with mean rotarod latencies reduced 63±5% (PBBI) and from 36–64% (PROG), 34–61% (DM), or 56–62% (SIM) vs. sham (p<.05). Post-injury administration of PROG or DM (but not SIM) protected against PBBI-induced motor abnormalities on the rotarod task, with the 5.0mg/kg (PROG) and the 5.0–10.0mg/kg (DM) doses producing significant improvement in motor performance. Mean rotarod latencies: Sham=54±1s; PBBI=20±3s; *PROG(5.0mg/kg)=35±5s; DM(5.0mg/kg)=35±5s; *DM(10.0mg/kg)=36±5s (*p<.05 compared to PBBI). MWM results revealed significant deficits in all injury groups with the average daily latency to locate the hidden platform increased 115±9% (PBBI) and from 78–127% (PROG), 68–107% (DM), or 71–106% (SIM) vs. sham (p<.05). Post-injury administration of PROG, DM or SIM resulted in significant improvement in MWM performance with mean overall latencies: Sham=30±1s; PBBI=65±3s; *PROG(1.0mg/kg)=54±5s; *DM(10.0mg/kg)=51±4s; *SIM(0.1mg/kg)=52±6s (*p<.05 compared to PBBI).
The D-R results for PROG indicate a differential neuroprotective profile on motor and cognitive outcome where one dose (5.0mg/kg) improved motor but not cognitive outcome and another dose (1.0mg/kg) improved cognitive outcome but not motor performance following PBBI. In contrast, DM showed similar D-R profiles across motor and cognitive parameters whereas SIM only improved cognitive outcome in the PBBI model. Overall, the current results are supportive of research demonstrating therapeutic efficacy of PROG, DM, and SIM in other TBI models. While additional compounds are currently being evaluated, the neuroprotective D-R profiles of PROG, DM, and SIM in the PBBI model support their consideration as potential candidates in advanced combination drug therapy development studies.
This research was funded by the Army Combat Casualty Care Research Program and Defense Medical Research and Development Program Grant D10_I_AR_J6_414.
PBBI, dextromethorphan, progesterone, simvastatin, dose-response
BIOMECHANICAL ANALYSIS OF BLAST INDUCED TRAUMATIC BRIAN INJURY WITH AND WITHOUT COMBAT HELMET
Mr. Sumit Sharma, M.S., Ford Motor Company
LiYing Zhang, Ph.D., Wayne State University
In the recent wars, soldiers are frequently exposed to blast threats. The combat helmets are designed primarily to protect against ballistic and blunt threats. The blast wave interaction with the helmet and subsequently biomechanical responses in the human brain resulted from traumatic events have not been delineated at tissue level.
A geometrically accurate FE model of the Advanced Combat Helmet (ACH) was developed and defined with actual material property data of the foam used in the helmet. The FE ACH was validated against standard Army helmet impact tests. The helmet model was then integrated with an anatomically detailed FE human head model that has been extensively validated against cadaveric brain responses in blunt impact tests and intracranial pressure data in shock tube experiments. A variety of blast waves with overpressures (0.27–0.66 MPa) from the Bowen's lung iso damage threshold curves were used to simulate blast insults. The resulting biomechanical responses (intracranial pressure, strain, product of strain and strain rate) of the brain to blast threats with and without helmet were compared to quantify the blast mitigating capability of the current combat helmet. Effectiveness of the helmet with respect to a variety of head orientations was also investigated.
The peak intracranial pressures in the head ranged from 0.68–1.8 MPa in the coup cortical region within the helmeted head. In comparison to the head model without helmet, ACH was found to mitigate intracranial pressures (0.62–1.1 MPa) in the head by 10–35% for a variety of loading severities. Helmeted head resulted in 30% lower average peak brain strains and product of strain and strain rate. Among three blast wave directions with the use of ACH, the highest reduction in peak intracranial pressure (44%) was due to backward blast whereas the lowest reduction in peak intracranial pressures and brain strains was due to forward blast (27%).
Blast induced intracranial pressure levels in the helmeted head generally exceeded thresholds proposed for contusive brain injury induced by blunt impact.The brain strain and product of strain and strain rate induced by blast loadings were below the thresholds for mild TBI produced by helmeted blunt trauma. The human brain exhibited directional vulnerability to the primary blast with highest intracranial pressure-related damage from sideways blast and highest strain-related damage from frontal blast. The degree of protection offered by the current helmet design was related to the coverage of the helmet construction and inherent asymmetric anatomy of the human head. Thus, direction-specific tolerances are needed in helmet design in order to offer omni-directional protection for the human head.
The study was partially funded by the Department of Defense Grant W81XWH-08-1-067.
Blast, TBI, FE head, Helmet
MODULATION OF INNATE-ADAPTIVE IMMUNE SYSTEM CROSS-TALK FOR NEUROPROTECTION: THE ROLE OF LIPOPOLYSACCHARIDE IN DIFFUSE AXONAL INJURY
Reyna L. VanGilder, PhD, West Virginia University
Stephanie L. Rellick, PhD, West Virginia University
Charles L. Rosen, MD, PhD, West Virginia University
Taura L. Barr, PhD, RN, West Virginia University
Jason D. Huber, PhD, West Virginia University
Preconditioning using lipopolysaccharide is neuroprotective in traumatic brain injury. A differential innate immune response has been found to be partially responsible for neuroprotective effects, the complete mechanism has not been fully elucidated, particularly in relation to models of diffuse axonal injury (DAI) and the role of macrophage phenotype.
We measured mRNA levels of orosomucoid 1 (ORM1), arginase 1 (ARG1) and chemokine receptor 7 (CCR7) in the peripheral blood following DAI.ORM1 and ARG1 have been associated with M1 (inflammatory) and M2 (anti-inflammatory) macrophages, respectively, and CCR7 has been associated with adaptive immunity.
Our results indicate that preconditioning before DAI leads to a different adaptive immune response, which may be differentially mediated by the M1 or M2 macrophage.
This study identifies a clear effect of DAI on the immune system and provides insight into how manipulation of adaptive immunity may lead to promising advances in the search for improved therapeutics for brain injury.
This work was supported by the National Institutes of Health and National Institute of Neurologic Disorders and Stroke.
Traumatic Brain Injury, Inflammation, Macrophage
EFFECT OF REPEAT TBI ON MEMORY IN THE JUVENILE RAT
Nicole Salame, none, UCLA
Tiffany Greco, Ph.D., UCLA
Daya Alexander, M.S., UCLA
Alejandra Rodriguez, M.A., UCLA
Mayumi Prins, Ph.D., UCLA
Previous studies showed that the magnitude of cognitive deficits increases with number of repeat mTBI at 24hrs post injury in the juvenile rat. In the current study, we hypothesize that the rats receiving repeat TBI (RTBI) will demonstrate more prolonged deficits as the number of mTBI increases.
Postnatal Day 35 (PND35) rats were given either sham (n=6), single (n=6), 2RTBI (n=6) or 4RTBI (n=6) and then tested in the novel object recognition (NOR) task one week and six months after injury to assess cognitive deficits. On day 1 all rats were habituated (10 min/rat) in the open chamber. On day 2 they were exposed to two similar objects (5 min/rat) and the following day were exposed to one novel object (5 min/rat). The object first approached, number of interactions and total duration of interaction with each object was recorded with timer and video analysis.
One week post injury sham animals showed 70.3% interaction time with the novel object visiting the novel object first 60% of the time. The single, 2 RTBI and 4RTBI groups showed 46.32%, 60.25%, and 57.21% novel object interaction durations, respectively. Animals visited the novel object first 66.7% (single), 62.5%(2RTBI) and only 42.9% (4RTBI) of the time. At 6 months post injury, the duration of time spent with the novel object did not differ between the groups with sham, single, 2RTBI and 4RTBI showing 58.7, 66.8, 67.6 and 52.9%, respectively. However, sham and single injured animals approached the novel object first 100% of the time. 2RTBI and 4RTBI animals acknowledged the new object only 40% and 20%, respectively.
While this preliminary data does not reveal increased impairments in the interaction time with the novel object with increasing numbers of mTBIs, there is a decrease in the initial approach to the novel object with injury numbers. This data suggests that learning deficits observed acutely after multiple mTBIs continue to show cognitive difficulties at 1 week and 6 months post injury and that the magnitude of the deficits increase with number of RTBIs.
NFL Charities, UCLA Brain Injury Research Center, Lind Lawrence Foundation, NS058489.
RepeatTBI, mildTBI, cognitive, juvenile
POSSIBLE EFFECTS OF PROGESTERONE IN BILATERAL CONTROLLED CORTICAL IMPACT (BCCI) INJURY IN RATS
Jeremiah Lacsina, BS, Psychogenics
Taleen Hanania, PhD, Psychogenics
Traumatic brain injury (TBI) affects 1.5 million people per year in the United States. Edema plays a crtical role in pathology of TBI. Progesterone treatment is known to reduce tissue damage after injury. We examined the effects of progesterone on cognitive and gait deficits induced by BCCI injury.
In this study we examined the effects of progesterone (4, 8 and 16 mg/kg administered 1 hour post injury followed by once daily for 5 days) on cognitive and gait deficits induced by a bilateral controlled cortical impact injury (BCCI) in male SD rats. Cognitive assesments were carried out by Plus maze and Morris water Maze and one and two weeks respectively after injury. Gait was analyzed using Neurocube NeuroCube™ 1 week after injury.
Rats that CCI injury showed a significant deficits in special learning and memory. In a plus-maze, these rats show a significant decrease in the number of spontaneous alternations compared to sham-injury rats. In addition in the Morris Water Maze (MWM) test, these rats show an increase in the latency to reach the platform during acquisition. Treatment with progesterone (16 mg/kg) reversed the cognitive deficits seen in both tests. One week after injury gait dynamics and geometry were evaluated using NeuroCube™, PsychoGenics' propriety technology for gait analysis. Significant differences in gait measures were found between CCI and Sham rats. Progesterone dose dependently recovered these gait deficits
The results of our study further support the possible use of Progesterone in management of TBI in clinics.
PSYCHOGENICS
TBI, BCCI, progesterone, neurocube, edema
EFFECTS OF THE POLY(ADP-RIBOSE) POLYMERASE (PARP-1) INHIBITOR, PJ-34, ON LOCOMOTOR-LIKE ACTIVITY IN RAT SPINAL CORD
Anujaianthi Kuzhandaivel, PhD, International School for Advanced Studies (SISSA)
Athena Akrami, PhD, International School for Advanced Studies (SISSA)
Elena Bianchetti, International School for Advanced Studies (SISSA)
Marco Milanese, PhD, University of Genova, Section of Pharmacology and Toxicology, Department of Experimental Medicine
Giambattista Bonanno, PhD, University of Genoa Department of Experimental Medicine (Di.Me.S.) Unit of Pharmacology and Toxicology
Andrea Nistri, MD, PhD, International School for Advanced Studies (SISSA)
Poly (ADP-ribose) polymerase-1 overactivity plays a prominent role in the neurological damage following brain or spinal cord injury. Although PARP-1 is normally activated to repair DNA, its excessive activation leads to build-up of the metabolite PAR leading to cell energy collapse within hours from the primary event.
Some benefits of using PARP-1 inhibitors as a neuroprotective drug in CNS ischemic and traumatic injuries have been reported before. Thus, it was interesting to explore the effect of a PARP-1 inhibitor on locomotor central pattern generator, which is an intrinsic spinal network for setting the timing and pattern of locomotion. We investigated if the PARP-1 inhibitor PJ-34 [N-6-Oxo-5,6-dihydrophenanthridin-2-yl)N,N-dimethylacetamide] applied 30 min after the excitotoxic glutamate analog kainate could protect spinal networks 24 h later and if the drug per se could change the excitability of the neuronal network. Network activity was measured by recording spontaneous discharges, fictive locomotion (evoked by dorsal root stimulation or neurochemicals application) and reflexes from lumbar ventral roots of the neonatal (0–2 days) rat isolated spinal cord.
Using our in vitro model of spinal cord injury showed that, when the damage was moderate, PJ-34 could protect locomotor pattern activity with good recovery in reflex amplitude and protection of number of motoneurons and neurons against excitotoxicity. Interestingly, we discovered that, beside the positive effects of this agent on the cell survival and locomotor activity, under control conditions, PJ-34 per se strongly increased spontaneous network discharges. These discharges occurred synchronously on lumbar ventral roots, continued for 24 h and persisted even after PJ-34 washout. Glutamate ionotropic receptor blockers could reversibly suppress this phenomenon. The main pattern of locomotor activity (fictive locomotion) evoked by NMDA and serotonin or by dorsal root stimulation could be elicited even one day after PJ-34 application. Neurochemical experiments showed that PJ-34 produced 33 % inhibition of synaptosomal glutamate uptake with no effect on GABA uptake.
Our data suggest that pharmacological inhibition of PARP-1 could prevent damage to the locomotor networks if this procedure had been implemented early after the initial lesion. The inhibition of glutamate uptake by PJ-34 suggested that this effect may compound tests for its neuroprotective activity which cannot be merely attributed to PARP-1 block. Furthermore, our data indicate that the neonatal rat spinal cord could withstand a strong, long-lasting rise in network excitability without compromising locomotor pattern generation or circuit structure.
Supported by grant of Friuli Venezia Giulia government.
excitotoxicity, motoneuron, fictive locomotion
PORCINE EYE ENUCLEATION WHILE PRESERVING CIRCULATORY STRUCTURE AND CANNULATION
Thivakorn Kasemsri, M.D. and A.B., Texas Tech University Health Sciences Center Department of Pediatrics
Alan Barhorst, Ph. D., M.S., and B.S., Texas Tech University, Whitacre College of Engineering
Kelly Mitchell, M.D. and B.S., Texas Tech University Health Sciences Center Ophthalmology and Visual Sciences
Research involving porcine eyes with vasculature has abundant applications to detecting traumatic brain injury–specifically in abusive head trauma. Yet the base level documentation to enucleate eyes for research is lacking. The following methodology to enucleate and cannulate porcine eyes was developed to preserve vasculature.
To develop a viable enucleation method, porcine heads were obtained from a local slaughterhouse. All swine utilized passed USDA inspection for human consumption and were harvested with approved methods. Because of the circulatory differences between human and porcine eye, the normal methods of enucleation destroys the blood vessels. Using trial and error, the presented methodology was developed to preserve the vasculature system of the porcine eye when dissecting it from the head.
The result of this endeavor was the development of a method to enucleate porcine eyes while keeping the circulatory system intact for cannulation. • An area around the eye orbit was fleshed with a scalpel blade revealing bone and muscle. • The zygomatic arch was cut through posterior and anterior to the orbit and extracted. • Then temporalis muscle was removed to give lateral access to the eye. • Exsecting the Zygomaticomandibular and Masseter muscles permitted access to the inferior portion of the eye. • Medially anteriorly to the eye, a scalpel was used to start extracting the eye from the orbit by cutting along the orbit wall. • To preserve as much tissue as possible the connective tissue along the orbital wall was cut. • Along the superior portion of the orbit, then inferiorly, medially, and laterally the connective tissue that attached the eye to the orbital wall was cut. • To cannulate the eyes, the most posterior portion of the severed optic nerve was located and marked. • The muscle tissue was then separated from the eye using a hemostat while looking for the internal ophthalmic artery at the posterior end. • Upon locating the internal ophthalmic artery, the tissue surrounding the Ophthalmic artery was carefully dissected away. • The remaining connective tissue and muscle surrounding the optic nerve and eye was removed. • At this point, the eye globe, optic nerve, and the associated vasculature is all that remains. • The eye was secured in place by clamping the base of the optic nerve in the ‘Helping Hands Station.’ • The External Ophthalmic Artery was catheterized with an angiocath which was sealed around the artery with two silk sutures and supported with clamps. • Krebs-Ringer Solution was injected into the angiocath and any leaking vessels sealed by cauterizing. • From this point the desired experimentation can commenced.
By following the described method, the isolation of porcine eyes with intact and exposed vasculature was accomplished. With these isolated eyes, experiments will be conducted to determine the pressure necessary to cause retinal hemorrhaging. This hemorrhaging is an indication of traumatic brain injuries in infants. With practice, eyes can be enucleated in approximately 15 minutes. Also, under proper storage in Krebs-Ringer solution at 3° C, the eye can be stoted for about a week; after a week, the tissue integrity of the eye is lost. Developing the enucleation and cannulation process was fraught with frustration, and perfecting it took time and patience; but it has been found that this method yields desirable results and clean, consistent eyes. In the words of Albert Einstein, “If we knew what it was we were doing, it would not be called research, would it?”
Texas Tech Univeristy Provost Fellowship and the Office of the Dean's Seed Grant for Scientific Exploration from Texas Tech University Health Sciences Center
Enucleation, porcine eye, cannulation, retinal
A NOVEL NLRP2 INFLAMMASOME EXPRESSED IN HUMAN ASTROCYTES IS ACTIVATED BY EXTRACELLULAR ATP
Juan Pablo de Rivero Vaccari, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Robert W. Keane, PhD, Department of Physiology and Biophysics, University of Miami Miller School of Medicine
CNS trauma involves extensive cellular damage that is due, in part, to an innate inflammatory response induced by ATP. The innate immune response is regulated by pattern recognition receptors (PRRs), which include NOD-like receptors (NLRs). The PRRs and signaling cascades regulating innate glial responses to CNS injury remain largely undefined.
Human astrocytes in culture and rat spinal cords were used in our studies. To induce inflammasome activation, cells were stimulated with 1 mM ATP (1 h). To inhibit inflammasome activation, cells were treated with 1 mM probenecid and/or 5 mM Brilliant Blue G (BBG) (30 min). To knockdown NLRP2 in vitro, siRNA (40 nM) against NLRP2 was used. In western blot and coimmunoprecipitation experiments, astrocytes were stimulated with ATP and/or treated with probenecid and BBG. Then, cells were lysed, and cell lysates were resolved and blotted for the NLRP2 inflammasome proteins. The band densities were quantified with UN-SCAN-IT software and data was statistically analyzed with one tail Student's t-test. statistically analyzed with one tail Student's t-test. In immunocyto- and immuno-histochemistry experiments, astrocytes in culture and rat spinal cord sections were double stained with NLRP2 inflammasome proteins and the astrocyte-cell-type-specific marker, anti-GFAP. Sections were analyzed with an Olympus laser scanning confocal microscope.
In untreated astrocytes, NLRP2 was expressed in a patchy punctuate pattern throughout the cell. ASC staining (speck-like) was present in the cytoplasm of GFAP positive astrocytes, whereas caspase-1 staining was seen predominantly in the perinuclear region and some staining was observed in the cytoplasm. ATP stimulation caused increased NLRP2 staining in the plasma membrane and in the perinuclear compartment, and more intense ASC staining was seen close to the cell membrane and in the cell processes. Increased caspase-1 immunoreactivity was observed in the cytoplasm of ATP-stimulated human astrocytes and in the cell processes. Moreover, rat astrocytes in vivo express NLRP2. Unstimulated and ATP-stimulated astrocytes that were immunoprecipitated with anti-NLRP2 and anti-ASC. Both anti-NLRP2 and anti-ASC antibodies coimmunoprecipitated all the NLRP2 inflammasome proteins. Caspase-8 and IgG served as negative controls. Immunopeptide depleted anti-NLRP2 also served as a control. ATP activated caspase-1 over a wide range of dosages. Next astrocytes were stimulated with ATP and inflammasome activation was blocked with probenecid and/or BBG. Astrocytes stimulated with ATP showed increased caspase-1 activation. Pretreatment with 1 mM probenecid or 5 mM BBG suppressed caspase-1 activation. ATP application significantly increased NLRP2, ASC, caspase-1, pannexin 1, P2X7, IL-1β and IL-18 in astrocytes. Pretreatment with probenecid blocked the ATP-induced increases in all proteins examined. siRNA knockdown of NLRP2 significantly decreased NLRP2 levels by about 40% as compared to scramble siRNA control. Moreover, NLRP2 knockdown led to significant decrease in caspase-1 as compared to scramble siRNA (40 nM). ATP significantly increased NLRP2 and caspase-1 protein levels in human astrocytes. However, cell treatment with siRNA against NLRP2 significantly blocked ATP-induced increases in NLRP2 and caspase-1.
Here we show for the first time that human astrocytes express a novel NLRP2 inflammasome complex composed of NLRP2, ASC and caspase-1. We also show that Pannexin 1 and P2X7 are involved in the regulation of the NLRP2 inflammasome. Stimulation of human astrocytes with the danger associated molecular pattern (DAMP) ATP, resulted in a dose-dependent activation of the NLRP2 inflammasome leading to inflammatory caspase-1 cleavage and mature interleukin-1β (IL-1β) production. ATP induction of the NLRP2 inflammasome was blocked by the pannexin 1 channel inhibitor, probenecid and by treatment with the P2X7 receptor antagonist, BBG. siRNA knockdown of NLRP2 significantly decreased NLRP2 inflammasome activation and caspase-1 activation in response to ATP stimulation. Our findings suggest that the NLRP2 inflammasome in astrocytes is an important component of the CNS inflammatory response and may be a therapeutic target to inhibit inflammation induced by CNS injury.
This work was supported by NIH/NINDS NS059836 to RWK
Inflammasome, astrocytes, CNS, NLRP2, caspase-1
CARBOXYFULLERENE NANOPARTICLES REDUCE OXIDATIVE STRESS AFTER RAPID STRETCH INJURY IN VASCULAR SMOOTH MUSCLE CELLS
Yaping Zeng, BC, UTMB
Donald S Prough, MD, UTMB
Douglas S Dewitt, PhD, UTMB
Reactive oxygen species (ROS) have been implicated in trauma-induced cerebral vascular dysfunction. Carboxyfullerene nanoparticles are potent antioxidants that have not been studied in relation to traumatic brain injury. To identify carboxyfullerenes that might reduce cerebral vascular injury, the toxicity and antioxidant properties of several species were compared in vitro.
The toxicities of carboxyfullerenes, C3, DF-1, DF-1 mini and PW-75, were measured 24 hours after treatment using propidium iodide (PI) staining in vascular smooth muscle cells (A7r5, rat aorta). Antioxidant properties were measured in cells subjected to moderate rapid stretch injury (40 PSI) using 5′-(and 6′)-chloromethyl-2′, 7′-dichlorodihydroflurescein diacetate acetyl ester (CM-H2DCFDA). Cells were first treated with a 100 μM dose of one of the four carboxyfullerenes and ROS levels were quantified 24 hours after injury by measuring the fluorescence intensities of CM-H2DCFDA stained cells using confocal microscopy.
The lowest toxicity levels were observed in the vascular smooth muscle cells exposed to C3 and the highest were found in PW-75 treated cells. At a 500 μM dose, the percentage of propidium iodide stained cells for C3, DF-1 and DF-1 mini were 0%±0%, 6.2%±3.69%, and 7.9±5.64%, respectively. PW-75 had a much higher toxicity (25%±5.56%) at only a 100 μM dose. When measuring ROS levels, the fluorescence intensities of DF-1 mini and PW-75 exceeded the upper quantifiable limits, indicating extremely high levels of ROS. C3 and DF-1 were both able to reduce the fluorescence intensity of CM-H2DCFDA stained cells. C3 was the more effective antioxidant, reducing ROS fluorescence intensity levels below those of uninjured controls (C3=9.4±13.4 relative fluorescence units (RFU); uninjured, untreated controls=19.4±11.3 RFU). ROS levels in the DF-1-treated cells (187.8±78.2 RFU) were significantly reduced compared to the injured, untreated controls (355.6±39.7 RFU).
These results show promise for the use of carboxyfullerenes, C3 or DF-1, as a new therapeutic strategy for reducing traumatic cerebral vascular injury in vivo. C3's results were particularly promising, but somewhat surprising, as chemical structures and previous studies in different injury models had suggest that DF-1 and DF-1 mini would be more effective antioxidants with lower toxicity levels. However, toxicity and antioxidant properties have been found to vary among cell lines, which could account for these unexpected results as this is the first work of this nature to be done in vascular smooth muscle cells. Another possible explanation may be that because C3 is the most soluble of the four nanoparticles, it is therefore the least likely to form aggregates. Aggregation of nanoparticles is known to greatly alter their properties and may explain the difference in performance of DF-1 and DF-1 mini.
N/A
Nanoparticles, Vascular Injury, ROS
NEUROPROTEOMICS AND BIOLOGICAL PATHWAY ANALYSIS OF CALPAIN VS CATHEPSIN B/L MEDIATED POSTISCHEMIC BRAIN PROTEIN ALTERATIONS
Joy Guingab, Ph.D., Banyan Biomarkers
Firas Kobeissy, Ph.D., Banyan Biomarkers
Jackson Streeter, M.D., Banyan Biomarkers
Ronald Hayes, Ph.D., Banyan Biomarkers
The novel calpain inhibitor B27-HYD and cathepsins B/L inhibitor CP-1 preserve brain tissue and reduce neurological deficits after stroke. Here, we demonstrate the utility of selective protease inhibition, 2D-DIGE and mass spectrometry for neuroproteomic and biological pathway analysis of calpain vs cathepsin B/L-mediated ischemic brain injury.
Brain proteins extracted from sham, untreated stroke and inhibitor-treated stroke rats were labeled with 3 different CyDyes respectively. The labeled proteins were then mixed together and resolved by two-dimensional difference in gel electrophoresis (2D-DIGE). The gel images were scanned using Typhoon TRIO (Amersham Biosciences) and analyzed by Image Quant software (version 6.0, Amersham), followed by in-gel analysis using DeCyder software (version 6.0, Amersham). The fold change of the protein expression levels was obtained from in-gel DeCyder analysis and protein spots of interest were picked up by Ettan Spot Picker (Amersham). Tryptic digests from the gel spots were analyzed by MALDI-TOF MS and TOF/TOF tandem MS/MS on an ABI 4700 mass spectrometer or LTQ Velos ion-trap LC-MS/MS (Thermo), followed by database search for protein identification and biological pathway analysis.
Over 2000 proteins were found to undergo protein expression changes in the brain following transient middle cerebral artery occlusion (MCAO)-induced ischemic stroke. Our selective protease inhibition based neuroproteomic and systems biology approach identified differentially expressed proteins that were dually or differentially modified by calpain or cathepsin inhibition. Molecular pathway analysis using Pathway Studio (Ariadne V 9.0) revealed protein changes affected by both B27-HYD and CP-1 treatment to be associated with cell growth, survival, neurite outgrowth, and cell proliferation. Interestingly, CP-1 and B27-HYD differentially modulated specific proteins and biological pathways.
There is a significant body of evidence accumulated over the past 15 years that strongly implicates upregulation and hyperaction of cytosolic calpains 1 and 2 and lysosomal cathepsins B and L in the pathogenesis of ischemic and traumatic brain injury via multiple molecular and cellular pathways. Our selective protease inhibitor based neuroprotectant treatment strategy for acute brain injury has strong clinical relevance because, in addition to assessing tissue preservation and functional outcome, our biomarker-guided neuroproteomic and systems biology approach is an effective mechanism based therapy development strategy.
NIH
Calpain, cathepsins, ischemic, brain, proteomics
ENCAPSULATED MESENCHYMAL STROMAL CELLS ARE NEUROPROTECTIVE IN AN ORGANOTYPIC CULTURE MODEL OF CEREBRAL ISCHEMIA
Ijaz Ahmed, Ph.D., Rutgers University
Rene Schloss, Ph.D., Rutgers University
Martin L. Yarmush, M.D., Ph.D., Rutger University
David I. Shreiber, Ph.D., Rutgers University
We aim to circumvent the limitations of direct implantation of human mesenchymal stromal cells (hMSCs) for CNS injury treatment using cellular encapsulation. In this study, the therapeutic benefit of alginate-encapsulated hMSCs was evaluated in an organotypic culture model of cerebral ischemia.
Organotypic hippocampal slice cultures (OHSC) were harvested from postnatal day 8–10 rat pups. After 14 days in culture, ischemic injury was induced by oxygen and glucose deprivation (OGD) for 60 minutes, in a 1% oxygen environment. Following injury, the OHSC were co-cultured with monolayer hMSCs or alginate-encapsulated hMSCs (500μm capsules, 200 cells/capsule) and allowed to recover in a normoxic environment for 24 hours. Starting at 24 hours after injury, the OHSC were stained with propidium iodide (PI) and imaged to assess for ischemia-induced cell death. Cell death was quantified in the CA1, CA3, and dentate gyrus (DG) regions of the hippocampus. Injury was expressed as an area percentage above a pre-determined threshold.
hMSC treatment of OGD-injured cultures resulted in a marked decrease in PI labeling as compared to injured, untreated cultures. At 24 hours post injury, a reduction of PI staining was observed in the CA3 region of hippocampal slice cultures treated with monolayer hMSCs (p=0.0007) or encapsulated hMSCs (p=0.005). A similar reduction of PI staining in the CA3 region was also observed at 48 hours post-injury for monolayer hMSC (p=0.0377) and encapsulated hMSC (p=0.0054) treated cultures. No significant difference in PI labeling was observed between monolayer hMSC and encapsulated hMSC treatment groups at any timepoint.
Currently, hMSC treatment by direct implantation is limited by the inability to control long-term survival and localization. We hypothesize that alginate encapsulation of hMSCs will overcome these limitations, allowing for sustained therapeutic benefit. In this study, we have demonstrated that hMSCs cultured as a monolayer and within alginate capsules protect against ischemic cell death within 48 hours of OGD injury. These results are the basis of ongoing studies to determine the mechanism of protection and the long-term effects of encapsulated hMSC treatment, as well as design considerations for alginate encapsulation and subsequent implantation.
Funding by: NJ Commission for Brain Injury Research Grant 10-3215-BIR-E-0, NSF Stem Cell IGERT Fellowship 0801620, and NIH Biotechnology Training Fellowship T32 GM00008339-21
ischemia, stroke, cell transplantation
DETERMINATION OF CRITICAL GAP LENGTH IN A MOUSE MODEL OF SCIATIC NERVE INJURY
Andrew Callion, The Ohio State University
H. Francis Farhadi, MD, PhD, The Ohio State University Medical Center
Mouse models of peripheral nerve injury (PNI) allow for the use of homogeneous genetic backgrounds and of stem cell technology. As a first step in preparation for future studies, we set out to define the critical gap length in a mouse sciatic nerve transection model repaired with Neuragen™ tubes.
Seventeen male C57BL/6 mice bearing myelin basic protein (MBP) promoter-driven LacZ transgene expression were used in this study. While under anesthesia, the left hind limb was incised and the sciatic nerve identified. At a uniform distance from the distal trifurcation, either a 1, 2, 3, or 4 mm length of the nerve was cut resulting in a 3, 4, 5, or 6 mm tension free inter-stump gap, respectively. In six animals, the inter-stump gap was bridged with 1.5mm internal diameter Neuragen™ tubes filled with C57BL/6 fibroblasts. Sciatic functional index (SFI) was calculated weekly to assess for functional improvement. Mice were allowed to survive for 5 weeks post-injury. Regeneration was assessed by gross inspection, histochemical analysis of ß-galactosidase activity, and analysis of 1μm sections for myelinating Schwann cell profiles.
Animals with a nerve gap of either 3 or 4 mm showed gross evidence of nerve regeneration. This was confirmed by positive ß-galactosidase activity in the distal portions of the nerves. Animals with a 5 mm gap showed minimal regeneration, and those with a 6mm gap showed no regeneration. Neither 5 nor 6 mm gap animals showed improvement in SFI at 5 weeks post injury. In animals with either a 5 or 6 mm gap bridged with a NeuragenTM tube, there was evidence of minimal re-growth within the tube. However, analysis of 1μm sections distally demonstrated some newly myelinating Schwann cell profiles with the 5 mm gap, yet no new myelin formation with the 6 mm gap.
Mouse PNI models have numerous advantages that allow for the evaluation of repair strategies. We demonstrated in a mouse sciatic nerve transection model that small gaps of either 3 or 4 mm spontaneously regenerate without intervention, making these gap lengths inappropriate for the evaluation of therapeutic strategies. We defined 6 mm as the critical gap length in a NeuragenTM tube model of PNI repair. Knowledge of this critical gap will allow for the rational design of further experiments evaluating emerging therapies such as stem cell implantation.
We thank the Center for Brain and Spinal Cord Repair at The Ohio State University.
peripheral nerve injury, repair
INTEGRATED 16S NEXT-GENERATION SEQUENCING AND METAPROTEOMICS DIFFERENTIATE THE HEALTHY URINARY MICROBIOME FROM ASYMPTOMATIC BACTERIURIA IN NEUROPATHIC BLADDER ASSOCIATED WITH SPINAL CORD INJURY
Derrick Fouts, PhD, J. Craig Venter Institute
Sebastian Szpakowski, PhD, J Craig Venter Institute
Rembert Pieper, PhD, J Craig Venter Institute
Hans Pohl, MD, Children's Nat Med Ctr and George Washington University
Moo-Jin Suh, PhD, J Craig Venter Institute
Inger Ljungberg, MPH, National Rehabilitation Hospital
Bruce Sprague, BS, Children's Nat Med Ctr
Sarah Lucas, BS, J Craig Venter Institute
Mannolito Torralba, BS, J Craig Venter Institute
Karen Nelson, PhD, J Craig Venter Institute
Suzanne Groah, MD, MSPH, National Rehabilitation Hospital and Georgetown University Hospital
Urinary tract infection (UTI) is the most common cause of rehospitalization after spinal cord injury (SCI), but diagnosis is challenged by sensory impairment. Moreover, urinalysis is time-consuming, leaving health care providers to choose between delaying treatment, or empirically prescribing antibiotics that increase costs and contribute to microbial antibiotic resistance.
New next-generation sequencing (NGS) technologies are significantly reducing the time and expense of nucleic acid analysis, enabling sequencing approaches to potentially be considered as diagnostic tools. Therefore, we utilized a NGS strategy to characterize the urine microbiome in patients with SCI, and compare the results to standard urine culture tests. Urine was collected from 26 healthy controls without SCI, and from 27 individuals with asymptomatic (no-UTI) neuropathic bladder (NB) secondary to SCI. Among those with NB, 8 voided spontaneously, 8 utilized clean intermittent catheterization, and 11 used indwelling Foley catheterization. Aliquots were prepared for standard urinalysis/cultures and microbiome analysis. Purified microbial genomic DNA was used as a PCR template to amplify the V1-V3 region of 16S rRNA for subsequent NGS using an Illumina pipeline. Bacterial taxonomic profiles were generated based on operational taxonomic units (OTUs), as determined with the RDP classifier.
Microbial signatures from these samples revealed that the urinary microbiomes differed by gender, presence and severity of NB, and duration of exposure to urinary catheter. Urine from healthy controls was not sterile, instead it had a diverse bacterial profile. Probiotic Lactobacillus Sp. dominated in profiles from healthy control females, whereas Corynebacterium Sp. and other commensal gram-positive microbes were present in healthy control males. In subjects with NB, the microbial population became more homogeneous as time progressed after SCI (0–>40 months). Moreover, Enterococcus and Escherichia emerged as major components of NB profiles, and they predominated in urine from subjects with indwelling catheters. Klebsiella, Proteus, and Morganella also signifcantly contributed. All of these have previously been associated with UTI, suggesting that their presence in asymptomatic bacteriuria may be an indicator of increased UTI risk. In contrast, Lactobacillus Sp. decreased in females with NB, and longer exposure to urinary catheterization. A subset of samples was also evaluated with shotgun proteomics (LC-MS/MS). Although recent improvements in metaproteomic databases allowed more microbes to be identified with proteomics than with standard culture methods, 16S analysis identified more fastidious anaerobes and microaerophilic bacteria than either of those methods. However, metaproteomic data also revealed proteins associated with anti-microbial (peroxidases and hydrolytic enzymes) and pro-inflammatory (annexins, galectins) responses in subjects with NB, in the absence of diagnosed UTI.
The increased sensitivity of 16S NGS over standard culture analysis identified complex microbiomes in urine from healthy controls, and from subjects with NB. In contrast to the historical assumption, based on cultivation results, that urine must be sterile, the data revealed a gender-specific “probiotic” microbial community in healthy urine. Moreover, the microbiome of people with asymptomatic NB differed from healthy controls, and was affected by length of time after SCI, and exposure to urinary catheters. We suggest that NGS methods allow a more complete description of the microbial status of NB than current diagnostic tests, and that this provides a framework for development of targeted therapeutics.
Supported by NCMRR/NINDS R24 Medical Rehabiltation Network (CNMC), NCRR/NCATS UL1RR031975, and Georgetown University CTSI.
microbiome urine neuropathic
LIPIDOMICS OF SPINAL CORD INJURY: TARGETING MEMBRANE DYSFUNCTION AND PERMEABILITY
Crystal Simon, PhD, OmniGuide, Cambridge MA, USA
Sibali Bandyopadhyay, PhD, Georgia Tech
Al Merrill Jr., PhD, Georgia Institute of Technology
Michelle C. LaPlaca, Ph.D., Georgia Institute of Technology
The plasma membrane plays a fundamental role in maintaining cellular homeostasis; its integrity is crucial for normal cell function. Non-specific pores in the plasma membrane occur in the initial phase of traumatic CNS injury as a direct result of mechanical forces that have surpassed the structural threshold of the cell.
We investigated citicoline (an intermediate in the generation of phosphatidylcholine from choline) as a potential treatment aimed at facilitating membrane repair and improved post-SCI behavioral outcomes. Citicoline was administered to spinally contused rats at high doses and at frequent repeated delivery times (500 mg/kg given immediately and 3h after injury then daily thereafter). BBB gait analysis was carried out 1, 8, 15, 22, 29, 36 day post-injury, grid walk trials occurred 29 and 36 days post-injury and testing for hind limb thermal hyperalgesia was performed 8, 15, 22, 29, 36 days post-injury. Furthermore, we investigated, using tandem mass spectrometry (MS/MS) and liquid chromatography, tandem mass spectrometry (LC MS/MS), the acute effects of a severe contusion injury on the phosphatiylcholine (PCh) content and the sphingolipid content respectively, of spinal cord tissue at the site of injury in rats at 1, 5 and 10 minutes post-injury.
Our data showed little to no benefit when citicoline was administered to rats at high doses and at frequent repeated delivery times. Furthermore, our MS/MS analysis of injury-site tissue did not reveal any significant changes in the population of PCh species. These data back up similar analysis of injured brain tissue carried out in our lab and by others (Hankin et al, 2011). Considering the lack of evidence that CNS trauma affects PCh levels within injured tissue we generated, using LC MS/MS, data detailing the concentrations (pmol/mg protein) of sphingomyelin (SM), hexosyl ceramide species, ceramide species, sphingosine, sphinganine, sphingosine-1-phosphate (SoP) and sphinganine-1-phosphate (SaP). Sphinganine is an intermediate in the de novo synthesis of ceramide and ultimately SM, and sphingosine is an intermediate in the degradative pathway of SM. Initial data suggest that both SM and hexosyl ceramide (or glycosphingolipid) concentrations are reduced in a time-dependent manner. Analysis of the breakdown products of SM reveals some potentially important insights into the fate of neural sphingolipid species immediately following SCI.
Sphingolipids act both as structural components of cell membranes and as lipid mediators of signaling pathways that control ion transport, metabolism, proliferation, cell migration and cell survival, as well as death (via apoptotic and non-apoptotic pathways). The intermediates in the de novo synthesis and breakdown of sphingomyelin (SM), sphingosine-1-phosphate for example, are highly bioactive. Therefore, significant changes in the concentration of individual sphingolipids and their signaling metabolites within the “sphingolipidome” are likely to influence cell survival and the generation of secondary injury phenomena. Futher analysis of the “sphingolipidome” of the injured CNS may prove more fruitful with respect to membrane repair than focusing on their phospholipid cousins.
Funded by: NS061153, Institute of Bioengineering and Bioscience, Georgia Tech (Seed Grant)
Sphingolipids, Mass Specrometry, Membrane Permeability
(-)-EPIGALLOCATECHIN-3-GALLATE (EGCG) INTRAVENOUS INFUSION IMPROVES MOTOR AND SENSORY NEUROBEHAVIORAL DEFICITS AND PROTECTS SPINAL NEURONS IN CONTUSION SPINAL CORD INJURY RAT MODEL
Ghanim Al-Khaledi, M.D., Kuwait University - Faculty of Medicine - Department of Pharmacology
Alyaa Mousa, Ph.D., Kuwait University - Faculty of Medicine - Department of Anatomy
Prof. Habib Abul, Ph.D., University of Kuwait - Faculty of Medicine - Department of Pharmacology
Shaima Karam, Ph.D., Kuwait University - Faculty of Medicine - Department of Pharmacology
Sami Asfar, MBBS, M.D., Kuwait University - Faculty of Medicine
Spinal cord injury (SCI) causes long lasting neurobehavioral disabilities, chronic pain, and autonomic dysreflexia. EGCG has been shown to have neuroprotective role in many neurodegenerative disease models. This study was designed to elucidate the effects of EGCG on spinal neurons and the motor and sensory functions following SCI.
In a double blind study, 72 Female Sprague-Dawley rats were randomized and subjected to SCI using MASCIS Impactor or laminectomy as follows: (I) Sham: T9 -T10 laminectomy with no injury to the SC. (II) Acute SCI+EGCG (continuous intravenous (i.v.) infusion initiated within the first 4 hours post SCI for 36h). (III) Acute SCI+saline. (IV) Chronic SCI+EGCG (initiated at 1 year post SCI). (V) Chronic SCI+saline. Standard batteries of behavioral tests were performed weekly (1 test/day) for 6 weeks starting first week after infusion cessation and included BBB open field scale, the Louisville swim test, Von Frey, paw pressure, hot plate, and tail flick. Rats in all groups were anaesthetized, perfused with saline followed by 4% paraformaldehyde. Mid thoracic spinal cord was dissected and sections of spinal cord from rostral, and caudal to site of lesion were stained with cresyl violet, GAP-43 and GFAP immunostaining.
EGCG treatment of acute and chronic SCI animals showed significant (P<0.005) recovery in motor function as measured by the standard BBB test and Louisville forced swim test. Likewise, the sensory function tests showed significant recovery in the EGCG infused acute and chronic groups compared to saline treated SCI groups. Tactile allodynia, reflex and mechanical nociception thresholds from both hind limbs showed that the EGCG treated SCI animals significantly recovered the lost sensation. The spinally mediated nociception threshold measured by tail flick withdrawal reflex test improved significantly following EGCG i.v. infusion in both acute and chronic post SCI groups throughout the testing period. EGCG treatment significantly reduced the size of the lesion area and increased the number of neurons in the spinal cords of SCI animals compared to control. Immunostaining of GAP-43 and GFAP showed significant increase in the EGCG treated SCI group.
For the first time we show that EGCG systemic administration depicts substantial beneficial behavioral outcomes yielding significant functional improvements in terms of motor as well as sensory and nociception responses in both acute and chronic SCI animals compared to vehicle treated SCI rats. And the window of opportunity to array treatment regimen evolving systemic administration of EGCG is vast. In addition, treatment with EGCG reduces the lesion area and protects neurons from degeneration due to contusion injury to the spinal cord. These data suggest that EGCG treatment possibly would be an important therapeutic agent to mend the functional neurological deficits following SCI.
This work was supported by Kuwait University, Research Grant No. [MA01/08]. Many thanks to Ms. P George and Ms. A Mathur for their technical assistance.
green tea, nociception, regeneration, EGCG
NLRP1 INFLAMMASOME PROTEIN EXPRESSION IS INCREASED IN THE CEREBROSPINAL FLUID OF PATIENTS AFTER SPINAL CORD INJURY
George Lotocki, PhD, Miami Project to Cure Paralysis
Stephanie E. Adamczak, B.S., Miami Project to Cure Paralysis
Michael Y. Wang, M.D., Miami Project to Cure Paralysis
Michael D. Norenberg, M.D., Miami Project to Cure Paralysis
W. Dalton. Dietrich, PhD, Miami Project to Cure Paralysis
Robert W. Keane, PhD, Miami Project to Cure Paralysis
Inflammation is a response against damage and plays a role in the pathogenesis of spinal cord injury (SCI). The innate immune response is in part regulated by protein complexes termed inflammasomes that are involved in the activation of caspase-1 and the processing of the pro-inflammatory cytokines IL-1β and IL-18.
In this study we analyzed cerebrospinal fluid (CSF) samples of SCI patients (N=7) by immunoblotting of NOD-like-Receptor Protein-1 (NLRP1) inflammasome proteins. The American Spinal Cord Injury Association (ASIA) scale of these patients at admission to the emergency department ranged from AIS A to B. CSF from uninjured individuals (postmortem) was used as a control from 3 males and 2 females ranging from 67 to 91 years old obtained from the Brain Endowment Bank at The University of Miami. For immunohistochemical analysis of inflammasome proteins spinal cord sections were obtained from The Miami Project to Cure Paralysis' Human Spinal Cord Bank. In this study we analyzed inflammasome protein expression of diaminobenzidine (DAB) immunostained sections at the epicenter, penumbra and an area distal to the epicenter.
Here we show that the expression of components of the NLRP1 inflammasome NLRP1, apoptosis speck-like protein containing a caspase recruitment domain (ASC) and caspase-1 are significantly elevated in spinal cord motoneurons and oligodendrocytes of the human spinal cord after SCI. Moreover, ASC was also present in microglia/macrophages. In addition, increased levels of NLRP1, ASC and caspase-1 were also present in the cerebrospinal fluid (CSF) acutely after SCI. Interestingly, 3 patients who showed improved outcomes as determined by improved ASIA scores had lower levels of caspase-1 at day one after injury, whereas patients showing higher levels of caspase-1 in the CSF showed no improvement.
These results are consistent with our previous studies in rodents showing that neurons and other cells in the spinal cord express the NLRP1 inflammasome and that the inflammasome can be used as a therapeutic target to improve outcomes after central nervous system injury. In conclusion, we show that the NLRP1 inflammasome plays an important role in inflammatory responses after SCI, thus in the future these proteins may be used as sensitive biomarkers for the diagnosis of SCI severity in humans since individuals that present with low levels of caspase-1 in the CSF acutely after SCI seem to have a better prognosis than those individuals who show increased levels of these biomarkers early after injury.
This work was supported by NIH/NINDS NS059836 to RWK and The Miami Project to Cure Paralysis. We thank David Sequeira for technical assistance with immunohistochemistry.
CELLULAR AND TEMPORAL EXPRESSION OF NADPH OXIDASE ISOFORMS AFTER BRAIN OR SPINAL CORD INJURY
Kimberly Byrnes, Ph.D., Uniformed Services University
Traumatic brain and spinal cord injury result in up-regulation of the reactive oxygen species generating enzyme NADPH oxidase (NOX). The cellular and temporal expression profile of NOX isoforms, including NOX2 and NOX4, is currently unclear. The purpose of this study is to resolve this expression profile.
Briefly, adult male rats were subjected to either moderate controlled cortical impact injury (CCI) using a Leica Impactor (5m/s, 2mm deformation depth) or moderate spinal cord injury (SCI) using the Infinite Horizons Impactor (150 kdynes). Immunofluorescence was performed at 24 hours, 7 days and 28 days post-injury. Confirmation of cellular source was achieved by immunofluorescence of microglia and neuronal cells in vitro.
In uninjured brain, CD11b positive microglia were NOX2 and NOX4 negative. By 24 hours post-injury, CD11b positive cells were NOX2 and NOX4 positive. By 28 days post-injury, CD11b and NOX4 colabeling was lost, and the number of CD11b/NOX2 positive cells was reduced. NeuN positive neurons expressed both NOX2 and NOX4 in both injured and uninjured brain, with a slight decrease in NOX4 colabeling over time. Spinal cord injured tissue demonstrated a different pattern. In uninjured tissue, both neurons and microglia were positive for NOX2, but not NOX4. NOX2 labeling increased in CD11b positive cells by 24 hours after injury, at which time CD11b cells began to also show NOX4 expression. By 28 days post-injury, co-labeling of CD11b and NOX2 decreased, but was still increased compared to uninjured spinal cord, and NOX4 double-labeling was not observed. As in the brain, NOX2 expression in NeuN positive cells remained consistent between injured and uninjured tissue. Unstimulated microglial cells in vitro were positive for NOX2 and NOX4. NOX2, but not NOX4, labeling increased after exposure to the microglial stimulant lipopolysaccharide (LPS). Neuronal cells in vitro were also positive for NOX2 and NOX4; exposure to activated microglia led to a decrease in NOX4 labeling and an increase in NOX2 expression.
We now demonstrate that NOX isoforms have an expression pattern that is dependent on both cellular source and regional localization. Microglia in spinal cord, brain and in vitro all demonstrate different responses to injury/activation in terms of NOX responses. Neurons, on the other hand, show a more conserved NOX expression response across regions, although neurons in vitro demonstrate some alteration in this response. In conclusion, this study illustrates the regional and temporal influence on NOX isoform expression. This information will be useful in future studies of understanding reactive oxygen species production after injury and therapeutic approaches.
This work was funded by the Uniformed Services University Intramural Program and the NINDS/NIH (Grant number 1R01NS073667-01A1).
NOX, TBI, SCI, microglia, inflammation
FENBENDAZOLE IMPROVES LOCOMOTOR RECOVERY AFTER SPINAL CORD INJURY IN MICE
Carolyn Crowdus, Ph.D. candidate, University of Kentucky College of Medicine, Spinal Cord and Brain Injury Research Center
Kashif Raza, University of Kentucky College of Medicine, Spinal Cord and Brain Injury Research Center
James Geddes, Ph.D., University of kentucky College of Medicine, Spinal Cord and Brain Injury Research Center, Dept of Anatomy and Neurobiology
Pathogenic antibody is involved in locomotor disability which is major complication of spinal cord injury (SCI). This study evaluated the ability of oral administration of fenbendazole (FBZ), a FDA-approved anthelmintic, to reduce pathological antibody production at the lesion site, spare spinal tissue, and improve locomotor function following SCI.
The female, C57BL/6 mice, weighing 20–25 g, were treated with medicated feed with 150 ppm FBZ (ensuring 8 mg/kg body weight/day) for 4 weeks prior to contusive SCI. Control mice received the normal diet without FBZ. SCI was produced in the mice using the Infinite Horizon impactor at the 50-kdyn force setting. Open-field locomotion was evaluated until 6 weeks post-injury using the Basso Mouse Scale. Histological assessment of tissue sparing was performed at 6 weeks after SCI. Immunohistochemistry was used to measure IgG levels at the lesion epicenter.
The results demonstrated that FBZ pretreatment significantly reduced IgG immunoreactivity at the lesion epicenter and improved locomotor function and tissue sparing 6 weeks after contusive SCI in mice.
Reducing spinal production of autoantibody by fenbendazole improves locomotor recovery after spinal cord injury in mice. Thus, this FBZ represents a novel promising candidate for the treatment of SCI.
This research was supported by KSCHIRT grants #7-6A and 11-19A.
Pathogenic antibody responses; Paralysis; Fenbendazole
EFFECTS OF LOW INTENSITY VIBRATION ON BONE AND MUSCLE IN RATS WITH ACUTE SPINAL CORD INJURY
W. Dalton Dietrich, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Alex Marcillo, MD, University of Miami Miller School of Medicine
Lana Mawhinney, PhD, University of Miami Miller School of Medicine
Ofelia Furones-Alonso, BS, University of Miami Miller School of Medicine
Amade Bregy, MD, PhD, University of Miami Miller School of Medicine
William A. Bauman, MD, James J. Peters VA Medical Center
Christopher Cardozo, MD, James J. Peters VA Medical Center
Weiping Qin, MD, PhD, James J. Peters VA Medical Center
Following spinal cord injury (SCI), sublesional bone is rapidly and extensively lost. Low intensity vibration (LIV) has been suggested to reduce bone loss. The purpose of this study was to perform an initial characterization of the effects of LIV on bone and bone cells in an animal model of SCI.
Female rats with contusion injuries of the spinal cord after a 12.5 mm weight drop at T10, resulting in motor-incomplete SCI, were studied. Beginning 28 days post-SCI, LIV was administered for 15 minutes twice daily for 35 days.
LIV appeared to be tolerated well by the animals. LIV did not attenuate the significant 5% decline in bone mineral density (BMD) at the distal femur and proximal tibia, determined by dual energy x-ray absorptiometry. LIV did not reduce the 37% elevation of levels of a plasma bone resorption marker, type I collagen cross-linked C-telopeptide. However, LIV did reduce by 68% the 2-fold increase in osteoclast numbers observed in ex-vivo osteoclastogenesis assays performed using marrow hematopoietic stem cells obtained from the tibia and femur. Gene expression analysis after osteoblastic differentiation of marrow stromal cells from the femur and tibia were also performed. LIV completely reversed the 2-fold elevation in mRNA levels for SOST in osteoblasts, and it more than reversed the 40% reduction in mRNA levels for Runx2, an osteoblast differentiation factor. Following SCI, gastrocnemius muscle weights were reduced by 13% and were not significantly increased by LIV. SCI-mediated reduction of the mRNA levels in gastrocnemius were mitigated by LIV for PGC-1a, an activity sensitive gene, and for MCIP1.4, a transcript regulated by NFAT and sensitive to calcineurin signaling, and, consequently, to neuromuscular activity, although these changes did not reach statistical significance.
In this rodent model of neurologically motor-incomplete SCI, LIV did not reduce bone loss but did significantly reduce osteoclastogenesis in ex-vivo cultures and expression of SOST, a Wnt inhibitor that plays key roles in initiating and continuing bone resorption programs after bone unloading. These preliminary results are quite encouraging concerning the effect of LIV on parameters of sublesional bone resorption and formation. Perhaps, a longer duration of LIV after SCI, or a more severe model of injury, would have resulted in noticeable and favorable changes to BMD, as strongly suggested by the cellular and biochemical changes.
Supported by the Veterans Health Administration (B4162C, B4055X, and B3347K), the Department of Defense (SC090504) and The Miami Project to Cure Paralysis
SCI, vibration, bone, muscle, Wnt
POTENTIAL OF METHYLENE BLUE AND POLYETHELENE GLYCOL IN AN ANIMAL MODEL OF ACUTE SPINAL CORD INJURY
Kiran Kumar Rokkappanavar, MD, Wayne State University School of Medicine
Jay M. Meythaler, MD, JD, Wayne State University School of Medicine
Aleksej Zuzek, BA, BS, University of Texas at Austin
George Bittner, Ph.D., University of Texas at Austin
Recent studies by the Bittner lab found rapid functional repair of cut or crushed nerves using a series of solutions that includes methylene blue (MB) and polyethylene glycol (PEG). The potential of this treatment strategy was investigated in acute spinal cord injury (SCI).
Male Lewis adult rats received a mild contusive SCI using the MASCIS device. Twelve rats were randomly assigned to receive a series of 4 solutions that included MB and PEG or the same series of solutions without PEG. Immediately after injury, the dura was opened at the injury site and solutions were applied consecutively to the surface of the spinal cord. Bladders were expressed twice/day until return of function. In the 5 weeks after injury, the rats were blindly evaluated weekly using the BBB locomotor test, von Frey hairs to pads of the hindpaws and plantar thermal testing of the hindpaws using an automated Hargreaves device.
In the first week after injury, the mean BBB score of the vehicle treated group (3.17±1.25 SEM) was higher than the PEG treated group (2.83±0.91 SEM). However at 2 to 5 weeks after injury, the mean scores of PEG treated group was 3-4 points higher than the vehicle-treated group and were significantly different at 3 weeks after injury (p<0.05). Using a two-way ANOVA with Bonferroni post-test, the curves were statistically different (p<0.0001). Results from von Frey and thermal plantar testing appeared to be quite variable between the 2 sides and across time in a given animal even when care was taken to target the exact middle of the pads of the hindpaws.
The significant functional improvement in BBB scores as a result of using a series of solutions that contain PEG and methylene blue is encouraging. Overcoming certain technical difficulties such as quick removal of the dura after injury may improve results even more. Clinical translation may be eventually feasible using methods developed to extend the time to axonal fusion. Clinical feasibility is also enhanced because all of the solutions developed for maximal axonal fusion are already FDA approved.
This study was supported by the Lone Star Paralysis Foundation.
fusion, PEG, axons, SCI, BBB
PERSISTENT VEGETATIVE STATE FOLLOWING SEVERE TRAUMATIC BRAIN INJURY IS ASSOCIATED WITH BRAIN-REACTIVE ANTIBODIES IN SERUM
Boxuan Yang, PhD, Banyan Biomarkers
Susie Zoltewicz, PhD, Banyan Biomarkers
Andrea Gabrielli, MD, University of Florida
Sheila Catani, MD, Ospedale di Riabilitazione Fondazione Santa Lucia
Jixiang Mo-Seany, Banyan Biomarkers
Kara Schmid, PhD, Walter Reed Army Institute of Research
Frank C Tortella, PhD, Walter Reed Army Institute of Research
Kevin K.W. Wang, PhD, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
Rita Formisano, MD, PhD, Ospedale di Riabilitazione Fondazione Santa Lucia
Brain-reactive autoantibodies have been found in human TBI patients. The objective of this study was to evaluate the prevalence of brain-reactive autoantibodies in sera of patients suffering from vegetative state (VS) following severe TBI, and the relationship between autoantibody levels and initial severity of injury.
This retrospective case study enrolled 28 patients aged 17–59 years in persistent VS with mean duration of 105 days following severe TBI, and with Glasgow Coma Scale (GCS) scores of ≤8. Forty-one healthy subjects were used as controls. Serum samples were collected from each patient at several time points post-injury and Western blots were used to measure serum levels of brain-reactive autoantibodies.
The prevalence of brain-reactive autoantibodies in sera was 75% among patients in VS, compared with 17% among normal controls. Brain-reactive serum autoantibodies were statistically higher in patients with persistent VS compared to healthy controls (p<0.0001, Mann Whitney U test). Serum levels of autoantibodies were not affected by age, whereas they were higher in patients with lower GCS (median, 1 [Densitometric Units] for GCS 3–5 versus 0.5 [Densitometric Units] for GCS 6–8) or higher Marshall scores, but the differences were not significant. Serial evaluation of brain-reactive autoantibody levels in the same subjects showed persistence over time.
These studies suggest that severe TBI in humans is associated with alterations in BBB integrity that allows immunocompetent cells to access central nervous system antigens, which promotes emergence of autoantibodies against brain antigens. High prevalence and persistence of brain-reactive antibodies in human sera of patients suffering from VS following severe TBI leads us to propose that these autoantibodies may contribute to the initiation and/or evolution of enduring traumatic VS.
We thank the participants and their families for their invaluable contributions. We are very grateful to medical and nursing staff for their valuable cooperation.
Biomarkers, TBI, Persistent Vegetative State
PATTERNS AND DURATION OF SYMPTOMS IN MILD TRAUMATIC BRAIN INJURY
Stephen McCauley, PhD, Baylor College of Medicine
Harvey Levin, PhD, Baylor College of Medicine
Brent Masel, MD, Transitional Learning Center at Galveston
James McCarthy, MD, UT Health
Eli Mizrahi, MD, Baylor College of Medicine
Ponnada Narayana, PhD, UT Health
Andrew Papanicolaou, PhD, LeBonheur Children's Hospital
Claudia Robertson, MD, Baylor College of Medicine
Paul Swank, PhD, UT Health
Conventional wisdom regarding mild traumatic brain injury (mTBI) holds that there are minimal sequelae. We examined the frequency, pattern, and duration of physical, emotional, cognitive, and sleep symptoms in 61 mTBI subjects and 38 orthopedic-injury (OI) controls and found a high percentage of subjects with persistent symptoms.
For this prospective observational study, subjects age 18–50, right-handed, fluent in English or Spanish, with no injuries > 3 on the Abbreviated Injury Scale were recruited by experienced research nurses soon after discharge from the ED. Pre-existing medical conditions, including previous head injury, psychiatric conditions, and substance abuse were exclusion criteria. All mTBI subjects had a documented TBI with LOC<30 minutes, PTA<24 hours, GCS 13–15, and no abnormalities on head CT. Baseline evaluation occurred within 24 hours of injury, including symptom assessment, cognitive/behavioral testing, EEG, MRI, and blood samples for biomarker/genetic analysis. Four follow-up assessments were done through 6 months post-injury. Symptom data was collected using the Symptom Check List of the Acute Concussion Evaluation (ACE), which includes 10 physical symptoms, 4 cognitive symptoms, 4 emotional symptoms, and 4 sleep symptoms. The SF12v2 was used to record functional health status.
Groups were similar for age (mTBI 30±9; OI 28±9), gender (mTBI 70% male; OI 71% male), and education (some college/college grad: mTBI 48%, OI 53%; high school/GED: mTBI 26%, OI 26%). Mechanism of injury (mTBI: vehicle accidents 45%, falls 27%, assault 13%; OI: lacerations 45%, falls 27%, crush injury 13%) and Injury Severity Scores (mTBI 4.0±2.1; OI 1.7±1.3; p<0.0000) differed. Baseline ACE Scores (0–22) were 5.8±4.5 mTBI and 2.0±2.9 OI (p<0.0000). The most frequent mTBI symptoms were headache (67%), dizziness (50%), fatigue (48%), and feeling “foggy” (45%). In the OI group, headache (21%), numbness/tingling (21%), irritability (16%), and fatigue (16%) were most frequent. The ACE Score increased at 1 Week (mTBI 7.5±5.5; OI 2.3±3.4; p<0.0000) to the highest level for both groups. Fewer mTBI subjects reported headache (61%), dizziness (43%), or feeling “foggy” (36%); however, difficulty concentrating (52%) and irritability (38%) increased. mTBI subjects reported sleeping more (44%), drowsiness (41%), and problems falling asleep (36%). In the OI group, one symptom, fatigue, was reported more frequently (28%); all others decreased. There was also a significant difference in the SF12 Mental Health Scores (mTBI 48±12; OI 56±8; p<0.0000).
This difference persisted at 1 Month for both ACE (mTBI 4.6±4.9; OI 1.56±2.3; p<0.0002) and SF12 MHS (mTBI 48±12; OI 54±8; p<0.002), and a difference in the Physical Health Score appeared (mTBI 47±10; OI 51±6; p<0.005).
By 6 Months, the mTBI ACE Score was still significantly higher than the OI (mTBI 3.42±4.1; OI 0.3±0.5; p<0.0004). OI subjects reported no occurrence for 19 of 22 ACE symptoms. Fatigue persisted for 1 subject (8%), and two reported sleep problems. In the mTBI group, however, all symptoms were still being reported by mTBI subjects, with headache (29%), sleeping less (25%), trouble falling asleep (21%), dizziness (18%), and irritability (18%) most frequent.
The ACE Scores fall in the lower range (0–22), which is to be expected with these mild injuries. However, the mTBI scores are consistently higher than those seen with the OI control group. Despite a diagnosis of “mild” TBI, these patients endure a wide range of unpleasant and function-limiting symptoms that persist over time. Controversy about the diagnosis of post-concussive syndrome (PCS) focuses on the observation that other trauma patients also show PCS symptoms. In this report, both groups reported headache, fatigue, irritability, sadness, and sleep problems; however, the percentages were notably higher in the mTBI group, and the duration of the symptoms over time differed too. These differences were also reflected in the SF12 Scores at 1 Week and 1 Month. Understanding the types, patterns, and duration of symptoms is a crucial component in treating any condition. In mTBI, this information is only beginning to emerge.
This work is supported by the Department of Defense through the Congressionally-Directed Medical Research Program, Award # W81XWH-08-2-0131.
mild TBI, post-concussion syndrome, headache
VISUALIZATION OF TRAUMATIC MENINGEAL INJURY ON POST-CONTRAST 3D-FLAIR MRI IN PATIENTS WITH SUSPECTED ACUTE TBI
Jessica L. DeStefano, BS, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Katherine J. Williams, ScB, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Marcus Dean, BA, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Joanna Lund-Shay, RN, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Anita Moses, BSN, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
Traumatic Meningeal Injury (TMI) is a potentially novel, conspicuous, imaging marker that appears as hyperintensity of the dura and meninges on post contrast FLAIR MRI following acute head injury. Here we compare 2D to 3D FLAIR sequences for the detection and characterization of TMI.
Patients presenting to a Level-1 trauma center within 48 hours of head injury were enrolled into IRB approved THINC study and imaged on a 3T Philips Achieva MRI using a standardized exam. Following administration of Gd-DTPA and a 3 minute T1 sequence; a 2D FLAIR sequence with a duration of 1:21 min, 0.47×0.47×3.5 mm voxels, then a 3D “VISTA” FLAIR of 3:25 min, of 1×1×1 mm voxels were acquired. Acquired pre-contrast 2D FLAIR images using the same parameter were used for comparison.Images were de-identified and pre- and post-contrast scans were paired. Two raters independently scored first 2D- then the 3D-FLAIR in separate sessions for presence, conspicuity, size (used for grading TMI), and location of enhancing regions.Visualization using multiplanar reformatting and volume rendering were permitted. Cohen's kappa was used to test concordance. A third rater formed consensus used to compare between techniques. 3D Visualization used for description of TMI.
During the period of January 15 to March 19, 2012, 17 consecutive subjects, imaged with both post-contrast 2D and 3D FLAIR, were included. Based on consensus ratings, 9 (.53%) were positive for TMI on 2D imaging compared with 10 (59%) on 3D, k=0.88. The additional subject identified by 3D had enhancement of the frontal lobe, better visualized in the sagittal and coronal planes on 3D FLAIR. Agreement on conspicuity between the two techniques was perfect, with TMI on 4 of 9 (44%) subjects positive on both. Agreement between techniques on location of enhancement ranked from good to very good: tentorium k=.679 and temporal k=.85. For the independent reads, inter-rater agreement ranged from good (kappa>.60) to perfect. Inter-rater agreement for grading showed higher agreement for 3D (.765) compared to 2D (.647). Overall, good agreement was found between the techniques in terms sensitivity and intra-rater agreement.
Visualization of structure and morphology of the TMI was facilitated by 3D FLAIR. Using volume reconstruction, enhancement previously thought to be homogeneous and continuous along both along the convexity of the brain, and in the falx, could be seen to have structure. In the four subjects noted to have highly conspicuous TMI, regions of enhancement were observed to be discontiguous with irregular boundaries. In the most severely injured subject with SAH, SDH, IVH and a shunt placed to manage ICP, TMI appeared much more homogeneous and confluent.
It was anticipated that true 3D FLAIR imaging would provide a more sensitive method of detecting meningeal enhancement with improved conspicuity. Results indicate 3D showed a higher agreement for TMI grading measurements. Furthermore, significant improvement in visualization of the morphology of the injury was realized by 3D volume rendering. Scalp and skull appear hypointense compared with normal parenchyma, allowing both to be easily rendered transparent without further image editing. This permits volume rendering of the meninges enhanced by the contrast agent. This 3D visual representation may assist in defining trauma related enhancement, appearing more disconnected against the more uniform enhancement seen in meningitis or following intercranial hypotension caused by a CSF leak. While the two methods appear to be equivalent for the purpose of detecting TMI, 3D FLAIR in combination with volume reconstruction may provide further insight into the pathology and evolution of this novel imaging marker.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
THINC TBI, TMI, 3D FLAIR
AN EXPLORATON OF INSULIN-LIKE GROWTH FACTOR-1 REFERENCE RANGES AND GROWTH HORMONE DEFICIENCY: WHO ARE WE MISSING?
Sarah E. Johnson, M.A., Centre for Neuro Skills
Jessica Ashley, Ph.D., Centre for Neuro Skills
Mark Ashley, Sc.D., CCC-SLP, CCM, CBIST, Centre for Neuro Skills
Growth hormone deficiency (GHD), is an often undetected consequence of brain injury (BI). The incidence of GHD in patients with brain injury (BI) is 15–37%. Symptoms of GHD include fatigue, increased abdominal adiposity, memory and attention deficits, anxiety & depression. These symptoms overlap considerably with deficits observed in patients with BI.
Growth hormone (GH) levels are assayed through stimulation testing; i.e. glucagon stimulation test (GST). Medical history, clinical findings and biochemical evidence are all considered in the decision to pursue GST since the testing is expensive and time consuming. Insulin-like growth factor −1 (IGF-1) is considered the best marker of GH activity currently available. Low levels of IGF-1 increase the likelihood that GH levels are also deficient. However, IGF-1 levels in isolation do not reliably predict GH status in patients who have sustained BI. Approximately, 50% of adults with GHD have IGF-1 levels within the normal reference range (Lissett et al 2003). Ninety-four patients with a documented history of BI were assessed for GHD using IGF-1 levels and GST results. IGF-1 levels were compared to reference ranges and actual GH status to determine if the reference range captured the majority of patients tested.
Preliminary results indicate that by relying on IGF-1 reference ranges in isolation, approximately one-third of patients with IGF-1 levels within the “normal” reference range are actually GHD as determined by GST resuts. However, by using a cutoff of 200 ug/l and clinical symptoms of neuroendocrine dysfunction, only 12 of the patients tested who were actually GHD were missed by this criteria.
IGF-1 levels alone are not a good biomarker of growth hormone deficiency and “normal” reference ranges can miss many patients who are actually growth hormone deficient. It is important to pursue stimulation testing (i.e. glucagon stimulation test) and monitor clinical symptoms of growth hormone deficiency, especially in patients with brain injury.
N/A
growth hormone deficiency, reference ranges,brain injury
POST-TRAUMATIC FOCAL CORTICAL ENCEPHALOMALACIA PREDICTED BY FOCAL CORTICAL DIFFUSION ABNORMALITIES ON ACUTE MRI
Elizabeth Magrath, BS, Henry Jackson Foundation/CNRM Center for Neuroscience and Regenerative Medicine
Dzung Pham, PhD, Henry Jackson Foundation/CNRM Center for Neuroscience and Regenerative Medicine
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
To characterize the pathogenesis of post traumatic focal cortical encephalomalacia (FCE), we identified sites of FCE on MRI 1–3 m following head injury with sites of focal cortical diffusion abnormalities (FCDA) identified on acute MRI obtained within 48 hours of the TBI.
Patients (n=31) with head injury presenting to the Trauma/ED were enrolled in the Center for Neuroscience and Regenerative Medicine (CNRM) THINC (Traumatic Head Injury Neurioimaging Classification) study and received MRI within 48 h of injury and at 1–3 month follow-up. Evaluable MRIs included, a 3D T1 weighted IR SPGR (∼1mm cubic voxels), and isotropic trace diffusion weighted images (isoDWI), computed from 15 gradient direction b=1000 DTI obtained at 3.5 mm slice thickness. First, the follow-up (1 m or 3 m) 3D T1 images were reviewed to identify sites of focal cortical encephalomalacia (FCE). Next, the acute isoDWI were evaluated for sites of hyperintensity involving the cortical grey matter. Finally, the cases of discordance between the two interpretations were reviewed with all imaging data available to understand the basis for the discordance.
A total of 18 sites of FCE were identified on the follow-up 3D T1 in 5 patients. FCE was characterized by complete loss of the ribbon of cortical gray matter such that white matter was “exposed” to CSF at 17/18 sites. One site demonstrated cortical thinning. In the 5 patients with FCE, the acute MRI demonstrated corresponding sites of FCDA. In 2 patients, FCDR was present acutely, but no FCE was identified on the follow-up MRI, even in retrospect. No patients with FCE on the follow-up scan had a normal acute isoDWI. In 24 patients, no FCE or FCDA was identified. Thus, FCE was highly associated with FCDA P=0.00012, Fisher's exact test).
Much attention has been directed towards the white matter injuries that result from TBI as well as to global/quantitative measures of atrophy. Here we identify focal cortical gray matter encephalomalacia as a relatively common consequence of mild to moderate TBI. In most cases, the presence of FCDA developed into sites of FCE, whereas no FCE developed when FCDA was initially absent. Thus, in a manner analogous to the evolution of ischemic stroke, FCDA represents cytotoxic edema (demonstrated as diffusion restriction) which can be readily identified by acute diffusion weighted MRI, and, as in stroke, evolves to encephalomalacia. Thus, FCDA may represent gray matter “at risk” and thus represent a salvageable target for therapeutic intervention, motivating the use of MRI in the acute TBI setting.
This work was supported by CNRM and the NIH Intramural Program. The authors gratefully acknowledge the investigators of the THINC study.
contusion, DWI, cortex, mTBI
MICRODIALYSIS LACTATE/PYRUVATE RATIO CORRELATES WITH MAGNETIC RESONANCE SPECTROSCOPY N-ACETYLASPARTATE IN ACUTE SEVERE TRAUMATIC BRAIN INJURY
Mr. Nathan R. Stein, BS, Department of Neurosurgery, UCLA
Jeffry Alger, PhD, Department of Neurology, UCLA
David L. McArthur, PhD, MPH, Department of Neurosurgery, UCLA
Paul Vespa, MD, FAAN, FACN, University of California, Los Angeles
Traumatic Brain Injury (TBI) results in hemorrhagic injury, regional edema and metabolic dysfunction. The metabolic dysfunction can be measured with regional, minimally invasive, continuous monitoring, like microdialysis, and global, noninvasive imaging techniques like Magnetic Resonance Spectroscopy Imaging (MRSI).
We hypothesize that the elevated Lactate/Pyruvate ratio (LPR) in acute severe TBI patients will correlate with the metabolites measured with MRSI: N-acetylaspartate (NAA), Choline (Cho) and Creatine (Cr). We prospectively studied six severe TBI patients the first four days post injury with microdialysis monitoring, multivoxel MRSI and apparent diffusion coefficient (ADC). The microdialysis catheter was inserted in normal appearing frontal lobe white matter and hourly samples were taken. We averaged the results of LPR and Glucose from the five samples closest to the scan time (+/− 12 hours). Multivoxel MRSI was done at the microdialysis catheter level. We used two voxels closest to the microdialysis catheter tip and we collected NAA, Cho and Cr, and calculated the NAA/Cho and NAA/Cr ratios. ADC was co-registered with MRSI.
We studied 12 voxels from the MRSI and 30 samples from the microdialysis. The ADC values were above 600 in all voxels, indicating no ischemia. The LPR was elevated and glucose and NAA were reduced. Cr was not stable throughout the voxels. LPR correlates negatively with NAA (r=-0.70, p=0.037) and Cr (r=-0.94, p<0.001) and glucose correlates positively with Cho (r=0.62, p=0.007) and NAA/Cho (r=0.44, p=0.028).
Elevated LPR and reduced NAA indicate metabolic dysfunction and occur in the absence of ischemia. LPR has significant correlation with NAA and Cr. Glucose has strong correlation with Cho. The results suggest that we should focus on reversal of metabolic dysfunction rather than on reversal of ischemia, and that microdialysis and/or MRSI can be used to determine the extent of metabolic distress.
We would like to acknowledge the faculty and staff of UCLA Brain Injury Research Center (BIRC).
TBI, microdialysis, spectroscopy, LPR, NAA
PRE-HOSPITAL CHALLENGES IN NEUROSURGICAL EMERGENCIES IN DEVELOPING COUNTRIES
Rashim Kataria, MCH Neurosurgery, SMS Medical College, Jaipur, INDIA
India is the largest country in South Asian region representing all the problems faced by rapidly developing nations. . It has 1% of the total vehicles in the world yet accounts for 6% of total road accidents globally. Traumatic brain injury comprises major cause of death and disabilities.
Our study is observation and evidence based with data collected from the Emergency Medical Service including ambulance services, emergency department and department of neurosurgery at SMS Medical College, Jaipur (India). The multifactorial prehospital delay in transportation and the trauma-surgery interval is analyzed and compared with the experience of other cities in India including Delhi and Bangalore and other developing countries.
In our study, we found that prehospital care of traumatic brain injuries is grossly deficient in India and other developing countries. The errors during trauma reception and resuscitations have contributed to as high as 55–91% of preventable deaths. Triaging, planning and execution are lacking even in fairly well equipped metropolitan cities Early transportation and decrease in trauma- operation interval has shown to reduce the mortality in EDH from 19.0% to 11.2%, 7.8% to 4.7% in EDH (pediatric group) and from 51.2% to 37.9% in Acute SDH.
It is observed in the study that while transfer, they are often taken to the nearest hospital, regardless of the hospital's capabilities for dealing with trauma. Furthermore, there is often no pre-hospital notification about the arrival of trauma patients. Attendance to a trauma patient is often delayed and triage may be ineffectual or performed by non-clinical staff. In the majority of Indian hospitals there is no trauma team or callout system in place, and hence no interdisciplinary approach to trauma reception. It is also seen in the study that Intubation and securing airways by under-trained paramedical staff leads to increase incidence of airway injuries. Likewise over administration of intravenous fluid can lead to cerebral edema.
Trauma is one of the leading causes of mortality and morbidity in India There are approximately 400,000 road traffic accidents in India each year, resulting in 100,000 deaths and 1.2 million individuals who are seriously injured. We must begin efforts to improve prehospital care, as it is the most critical phase of head injury. In Developing countries the pre-hospital care can be improved by training laypersons involved in pre-hospital transport and care. Management of hypotension, ensuring airway, rapid and efficient transport to a neurosurgical care center are the pillars of prehospital care of traumatic brain injured patients. Research in prehospital care is needed and should include inputs from the paramedics. Prehospital care is a major determinant of long-term outcome for patients with traumatic brain injury.
We are thankful to Mr. Anurag Kishore Bhatnagar, Regional head 108 Ambulance, for providing the data.
Prehospital care, Head injury
IMPACT OF IMMEDIATE AIRLINE TRAVEL ON MTBI IN NFL AND NHL PLAYERS
Dave Milzman, MD, Georgetown University School of Medicine
Jeremy Altman, BS, Georgetown U
Zach Hatoum, BS, MBA, U Maryland
Jordy Sax, BS, Johns Oskins U
Air travel may be associated with unmeasured neurophysiological changes in an injured brain that may impact post-concussion recovery. No study has compared impact of commercial air travel on acute concussion injuries despite rather obvious decreased oxygen tension and increased dehydration impact on acute mTBI. Objectives: To determine if air travel within 4–6 hours of concussion is associated with increased recoverytime in professional hockey (NHL) and football (NFL) players.
Prospective cohort study of all active-roster NHL and NFL players during the 2010–2011 seasons. Internet website review of League sites for injury identification of concussive injury and when player returned to play solely for mTBI. Team schedules and flight times included only players who flew immediately following game (within 4–6 hr.) Players withmultiple injuries were excluded.
In the 2010–2011 seasons, 223 players experienced a concussion: NFL-122(7.7) and NHL- 101(13.0) (percent of total players). Of these, 68 NFL (57) and 39 NHL (39) concussed players flew within 6 hours of the incident injury. Mean distance flown was shorter for NFL: 850 miles (SD 576) VS. NHL 1060 (SD 579). Mean games missed for NFL and NHL players who traveled by air immediately after concussion was increased by 29 and 24 (respectively) than those who did not travel by air: NFL- Fly 3.8 games missed (SD 2.2) VS. No Fly 2.6 (SD 1.8) and NHL- Fly 16.2 games missed (SD 22.0) VS.No Fly 12.4 (SD 18.6); p<0.03.
This initial report of an increased rate of recovery in terms of more games missed, for professional athletes flying commercial airlines post mTBI. The obvious changes of decreased oxygen tension with altitude equivalent of 7,500 feet, decreased humidity with increased dehydration and duress of travel accompanying pressurized airline cabins alllikely increase the concussion penumbra in acute mTBI. Early air travel postconcussion should be further evaluated and likely postponed 48–72 hr. untilinitial symptoms subside.
MedStar Sports Georgetown University School of Medicine
concussion professional sports post-concussive injury head injury
TRAUMATIC MENINGEAL INJURY (TMI): A NOVEL MRI MARKER OF ACUTE HEAD INJURY
Kara L. Kuntz-Melcavage, PhD, Johns Hopkins HealthCare LLC
John M. Hallenbeck, MD, Stroke Diagnostics and Therapeutics Section of the National Institute of Neurological Disorders and Stroke
Steven Warach, MD, PhD, Seton/UT Southwestern Clinical Research Institute of Austin
Lawrence L. Latour, PhD, Stroke Diagnostics and Therapeutics Section of the National Institute of Neurological Disorders and Stroke
Following acute head injury, we have observed focal, regionally specific enhancement of the meninges on post-contrast FLAIR imaging. The pathophysiology of this imaging abnormality is unknown. We present the characteristics and temporal evolution of TMI.
Patients presenting to Suburban Hospital (Bethesda, MD) and Washington Hospital Center (Washington, DC) with head injury were enrolled within 48 hours of injury in the THINC Study and received a 1.5T or 3.0T MRI with single-dose gad contrast. Following their initial MRI, subjects returned for a 1-week time point (TP) and for optional visits at 30 and 90 days post-injury. Prospective interpretation of MR scans for intracranial pathology was completed following each TP. Subsequent evaluation of post-contrast FLAIR was performed to identify location of enhancement, change over time, and to assign an ordinal TMI Score: 0=none, 1=present, 2=present and highly conspicuous, 3=present, highly conspicuous, and thick (>2.5 mm). Clinical CT scans and additional MR sequences sensitive to extra-axial collections were used to distinguish meningeal injury from other pathologies. The association of TMI score with pathology on CT and MRI was tested (Fisher's Exact, p-value<.05).
From October 2010 to March 2012, 168 subjects were enrolled, 142 received contrast, and 71 (50%) were found to be positive for TMI. Of the 71 studied, age=45 (34–62), 19 (27%) female, 48(68%) had LOC, 42 (59%) had PTA, 54 (76%) with GCS of 15, 25 (35%) had positive CT scans, and 64 (90%) were admitted to the hospital.
TMI was most frequently localized to the falx cerebri in 60 (85%), followed by frontal 33 (47%), temporal 27 (38%), parietal 22 (31%), brainstem 12 (17%), tentorium 10 (14%), and occipital in 10 (14%). The 71 subjects were assigned a TMI Score of: 1: 26 (37%). 2: 21 (30%)or 3: 24 (34%).
50 subjects had first follow-up MRI at 6.2 (3.9–7.4) days in which TMI was noted to increase in 7 (14%), decrease in 16 (32%), disappear in 22 (44%), and have no change in 5 (10%). The subjects had TMI Scores of 0: 22 (44%), 1: 10 (20%), 2: 12 (24%), or 3: 6 (12%). Of the 36 subjects returning for a 30 or 90 day TP, TMI increased in 1 (3%), decreased in 6 (17%), resolved in 28 (78%), and had no change in 1 (3%). The TMI Scores at the 30 and 90 day TP were 0: 28 (78%), 1: 5 (14%), 2: 1 (3%), or 3: 2 (6%).
A significant association was found between the TMI Score and other abnormalities seen on acute imaging. Subarachnoid hemorrhage was identified on CT in 17 (24%) subjects and MRI in 14 (20%) subjects (CT: p=0.008, MRI, p=0.026). Other significant pathologies on MRI included punctate hyperintensities on DWI in 25/71 (35%) subjects (p=0.01), extra-axial collection in 22/71 (31%) subjects (p=0.009), and intraparenchymal hematoma >1cm in 10 (14%) of subjects (p=0.028).
We have identified a marker specific to head trauma that is highly conspicuous, prevalent in half the population studied, heterogeneous in appearance and presumed severity, and associated with other traumatic injury to vasculature and the brain. Injury to the vasculature caused by trauma can result in a separation of the dura from the arachnoid membrane where blood and fluid may collect. FLAIR MRI is a null technique that is highly sensitive to small concentrations of contrast agent in fluid. One explanation for TMI includes the extravasation of contrast from injured vessels in the meningeal layers into the fluid filled, subdural space allowing highly conspicuous visualization on post-contrast FLAIR imaging. Alternative explanations include that TMI is reflective of a CSF leak or the result of an acute inflammatory process. The etiology remains unclear, however the appearance and time course suggest TMI may be related to post-concussion syndrome or secondary injury.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
TMI, THINC Study, MRI, meninges
ACUTE AND 1 YEAR MRS/DTI FINDINGS AFTER PEDIATRIC TBI: COMBINED DATA MAY BE MORE SENSITIVE IN PREDICTING CHRONIC INJURY
Nirmalya Ghosh, PhD, Loma Linda University Medical Center
Richard Sun, PhD, Loma Linda University Medical Center
Karen A. Tong, MD, Loma Linda University Medical Center
Jamie Pivonka-Jones, PhD, Loma Linda University Medical Center
Melissa Rundquist, LVN, Loma Linda University Medical Center
Stephen Ashwal, MD, Loma Linda University Medical Center
We present preliminary MR spectroscopic (MRS) and diffusion tensor imaging (DTI) findings on a subset of pediatric traumatic brain injury (TBI) patients who were enrolled as part of an ongoing NIH sponsored study and have completed their 1 year imaging and neuropsychologic evaluations.
Children with TBI were enrolled in this prospective IRB approved study if the GCS<13 or if intracranial hemorrhage was present on their admitting CT study. MRI/MRS were acquired at 3T early after injury (7–15 days) and at 1 year. MRI included 3D-T1W, 3D-T2W, SWI and DTI (TR/TE=3700/102 ms, 1.2×1.2×3 mm, b=0,1000 s/mm2, 30 directions). 3D proton MR spectroscopic imaging (MRSI) was acquired with TR/TE=1700/144 msec in four 10 mm thick slabs covering from the corpus callosum through the superior brain stem. DTI skeleton data were overlaid onto the MRSI grid at each level to obtain DTI parameters for each voxel for comparison to metabolite ratios (N-acetylaspartate (NAA)/creatine (Cr), NAA/choline (Cho) and Cho/Cr). Each voxel was assigned to an anatomic brain region. Neurological and neuropsychologic evaluations were done at 3 months and 1 year after injury.
Findings from five TBI patients (age:12.4±2.6 yr) and five age-matched controls (12.5±2.8 yr) were analyzed. GCS scores ranged from 4 to a complicated 15 (mean:11±5). MRI was acquired from 7–9 days after initial CT (mean: 7.6±1 days). On the initial study, MRSI showed significant decreases of NAA/Cr in the frontal white (p=0.03), basal ganglia (p=0.01), frontal gray (p=0.04), parietal white (p=0.04) and temporal gray matter (p=0.03) in TBI patients compared to controls. Fractional anisotropy (FA) mean values taken from the same anatomical areas as the MRSI were significantly decreased in the frontal white matter (p=0.008), temporal white matter (p=0.007) and corpus callosum (p=0.003), in TBI patients compared to controls. Mean FA values did not correlate with metabolite ratios. When data were pooled as cortical (gray and white matter in the frontal, parietal, occipital and temporal regions), subcortical (basal ganglia, thalami and corpus callosum) and posterior fossa (superior brain stem and cerebellum), significant decreases in the NAA/Cr in the cortical and subcortical regions and the mean FA in the cortical region were found initially. At 1 year, NAA/Cr remained decreased only in frontal white matter (p=0.009) and mean FA remained decreased only in the pooled cortical region (p=0.035) in TBI patients compared to controls. The frontal white matter was the only region that showed significant decreases of NAA/Cr and mean FA values initially and was the only region that did not show normalization of NAA/Cr at 1 year. This suggests that injury occurred as a result of mechanical structural forces as well as metabolic dysfunction which acted synergistically to prevent recovery. These findings will be correlated with neurological and neuropsychologic evaluations.
Preliminary data indicate that the primary metabolite changes seen after moderate/severe pediatric TBI involves NAA reduction that is usually attributed to neuronal loss or dysfunction. The lack of normalization of NAA ratios in frontal white matter suggests chronic injury, whereas normalization in the other brain regions suggests neuronal recovery. The most salient observation of our study is that the only brain region (e.g. frontal white matter) that showed chronic impaired neuronal metabolite function at 1 year was the only brain region that showed reduced NAA ratios and FA values soon after injury. Such findings suggest that using combinations of MRI data that reflect different aspects of the metabolic and biomechanical injury process may provide more sensitive and accurate methods to predict late recovery vs. permanent injury.
This project was supported by grants from the National Institute of Neurological Disorders and Stroke from the National Institutes of Health.
Trauma, Brain, Spectroscopy, Diffusion Tensor
A LONGITUDINAL PERSPECTIVE ON DIFFUSION ABNORMALITIES FOLLOWING PEDIATRIC MILD TRAUMATIC BRAIN INJURY
Josef Ling, BA, Mind Research network
Amanda Pena, BA, Mind Research network
Zhen Yang, Ph.D., Mind Research network
Ron Yeo, Ph.D., UNM
Stefan Klimaj, BA, Mind Research network
Pediatric mild traumatic brain injury (pmTBI) is the most prevalent neurological insult in children and can be associated with both acute and chronic neurobehavioral sequelae. However, little is known about the pathophysiology underlying the injury and how these injuries change as a function of recovery.
Although diffusion tensor imaging (DTI) holds promise for the in vivo characterization of white matter pathology, both the direction and magnitude of anisotropic water diffusion abnormalities in axonal tracts are actively debated during the semi-acute stage of pmTBI. The current study therefore examined fractional anisotropy (FA), axial diffusivity (AD) and radial diffusivity (RD) in 16 (fourteen males; 13.50+/− 2.13 years old) pmTBI patients and 16 (twelve males; 13.19+/− 1.97 years old) well-matched controls. New analytic strategies were applied to capture spatially heterogeneous white matter injuries, minimizing implicit assumptions of uniform injury across diverse initial biomechanical forces. An extensive clinical and neuropsychological battery was also administered to characterize neurobehavioral functioning. Finally, a prospective component was included to measure recovery of function during which both patients (N=11) and controls (N=12) returned for a follow-up study (both MRI and clinical) approximately 4 months after their first visit.
All pmTBI patients were evaluated with both neuropsychological testing (15.33±4.58 days) and imaging (15.87±4.93 days) within 21 days of injury. No evidence of focal pathology was observed on T1- and T2-weighted images. Cognitive dysfunction was present in the domains of attention (p=0.02; d=−0.92) and processing speed (p=0.05; d=−0.73) semi-acutely, with moderate effect sizes also present in terms of memory functioning (d=−0.59). Region of interest analyses indicated that the multivariate effect of group was not significant for the corpus callosum or right hemisphere white matter tracts (p>0.10), although a significant univariate was present in the right anterior corona radiata (p=0.03; d=0.82). The multivariate effect of group was significant for left hemisphere tracts (p=0.04), with univariate results indicating significant or trend effects within the left anterior (p=0.01; Cohen's d=0.96) and superior (p=0.052; d=0.74) corona radiata, as well as left cerebral peduncle (p=0.03; d=0.83). Across all structures, FA was elevated for the pmTBI group in conjunction with reduced RD. Metrics designed to reduce the necessity for spatial overlap between different clinical injuries indicated that both the number (t28=6.97; p<0.00001; d=2.6) and volume (t28=6.97; p<0.00001, d=2.5) of clusters with high anisotropy were greater in pmTBI, with no differences for clusters with low anisotropy (p>0.10). Metrics of increased anisotropy were able to objectively classify pmTBI from HC at 90% accuracy, but were not associated with neuropsychological deficits. Finally, there was no evidence of recovery of function following a four month interval for any of the measures of increased anisotropy.
In summary, the current pattern of diffusion abnormalities (i.e., increased FA with reduced RD) suggests that the ratio of intracellular and extracellular water may be altered following pmTBI, resulting in cytotoxic edema. A novel analytic method indicated that a spatially heterogeneous pattern of increased anisotropic diffusion exists in various other white matter tracts. The magnitude of these effects were much greater relative to previous results from an adult population, suggesting that developing white matter may be more susceptible to initial mechanical injury forces. This may be secondary to physical differences in myelin development or from under-development of the musculoskeletal system for mitigating initial biomechanical forces. Little evidence of recovery in white matter abnormalities was observed over a four month interval, indicating that physiological recovery may be lag behind neurobehavioral recovery. Future studies are needed to replicate current findings in a larger sample of patients.
This research was supported by grants from The Mind Research Network [DOE Grant No. DE-FG02-99ER62764] to A.M.
pediatric; DTI; axonal injury; cognitive
THINC MRI PILOT: RAPID MR IMAGING EVALUATION AND VALIDATION
Jessica L. DeStefano, BS, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Kara Kuntz-Melcavage, PhD, Johns Hopkins HealthCare LLC
Dzung Pham, PhD, Henry Jackson Foundation/CNRM Center for Neural and Regenerative Medicine
Dima Hammoud, MD, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
John Butman, MD, PhD, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
Steven Warach, MD, PhD, Seton/UT Southwestern Clinical Research Institute of Austin
A rapid MRI protocol for the Traumatic Head Injury Neuroimaging Classification (THINC) study was developed and tested for detecting intracranial pathology in patients with minor head injury. Patients presenting with symptoms of acute stroke were used as a control population for blinded image evaluation.
Head injured patients presenting to a Level 2 trauma center who received a clinical CT were enrolled and imaged with the THINC MRI protocol within 48 hrs of head injury and compared with a cohort presenting as “rule out stroke” by the NINDS Stroke Team. THINC MRI, as with the acute stroke MRI, included isotropic DWI and ADC derived from 15 direction b=1000DTI, T2*-weighted, FLAIR, 3D T1, DSC-PWI, T1-post, and FLAIR-post contrast with a nominal imaging time of 24 minutes. Images were blinded to clinical diagnosis and evaluated and each case was classified as trauma, stroke, or undetermined. In a second read, DWI, FLAIR, and T2* imaging were evaluated for specific acute MRI abnormalities independently by two neuroradiologists. A third rater resolved discrepancies. Intra-rater agreement was assessed and consensus ratings were compared to classification rating. Median (IQR) and Cohen's kappa are reported.
During 9 months, Sep 2009 to Jun 2010, 27 subjects with head injury, were enrolled; age=48(30–56), 37% female, 25 with GCS=15, 15 with positive LOC or PTA, injury to triage 55 (42–82) min. The clinical CT was 80 (48 to 114) min from injury and negative in 21 (78%) while the THINC MRI was initiated 5.1 (3.9–18.9) hrs from injury. A total of 61 subjects presenting with symptoms of acute stroke were included with an admission diagnosis of ischemic stroke in 33 (54%), TIA in 12 (20%), and “other” (stroke mimic) in 16 (26%).
Blinded to diagnosis, MRI abnormalities were seen on MRI in 48 (55%). Cases were correctly classified in 85% of stroke and 14 (48%) of trauma subjects based on pathology on MRI. MRI was negative in 13 (52%) of trauma subjects and no stroke, TIA, or mimic subjects were incorrectly classified as trauma (i.e. no false positives).
Abnormalities on MRI consistent with trauma was seen most frequently on FLAIR-post (44%), T2* (41%), FLAIR (30%), and DWI (15%) in the head injured subjects. Of the subjects with negative CT, 10 (48%) had findings on MRI. Enhancement of the meninges on FLAIR-post contrast (p<0.01) and linear hypointensities in the white matter on T2* (p=0.03) were associated with trauma. “Spherical” microbleeds” were not specific to trauma (p=1.0). Inter-rater agreement varied from fair (k=0.40) for SAH on T2* imaging to very good (k=0.85) for enhancement on FLAIR-post.
MRI is superior to CT for detecting pathology in acute stroke. Using a rapid MR protocol designed for acute stroke, it was feasible to image patients with acute head injury in the ED and to detect abnormalities consistent with trauma-induced injury to the brain parenchyma, vasculature, and meninges. In particular, two novel pathological entities were identified; i) linear hemorrhages in the white matter on T2* and ii) enhancement of the meninges on post contrast FLAIR MRI. Both were significantly associated with head injured population and thus may be a specific marker of traumatic injury. MRI provides objective evidence, complementary to CT, for classifying head injury patients as “positive” or “negative” for injury to the brain and vasculature caused by trauma. Observations made in this pilot will be independently verified in a larger population of patients in the multi-site THINC study.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
THINC, MRI, mTBI, Concussion, Diagnosis
MODIFIED DECOMPRESSIVE CRANIECTOMY: A NEW TECHNIQUE FOR DESTINATION AND PRESERVATION OF THE BONE FLAP
Bradley Roth, MD, Providence Holy Cross Trauma Center
David Hanpeter, MD, Providence Holy Cross Trauma Center
In the treatment of increasing intracranial pressure (ICP) decompressive craniectomy has become a commonly used surgical technique. Debate has developed on the proper placement of the bone flap. We have developed a modification of the Tucci flap to address this issue.
A new technique for placement and preservation of the bone flap with a single incision was developed. The technique preserves the bone flap in the same site as the craniectomy. This is accomplished with the use of a newly designed Quinonez-Bone Flap Elevator (Q-BFE) to secure the flap above the skull defect. The bone flap is prepared for replacement; three or four Q-Bone Flap Elevators (Q-BFE) are attached to the flap with titanium screws. The bone flap is fixed to stay 1 or 1.5 cm above the skull level. This results in an increase of intracranial volume change from 70 to 100 cm3 depending on the area of the flap and the adjustment of the elevator plate. Once the patient has recovered from his injury and cerebral edema has resolved the patient returns to the operating room and the bone flap is replaced.
Three trauma patients underwent this modified decompressive craniectomy. All three patients had an immediate decrease in their ICP. All patients recovered from their trauma and returned to operating room for replacement of their bone flap. All patients had favorable outcomes and could function semi-independently.
In this small series modified decompressive craniectomy using the Q-BFE resulted in an immediate decrease in ICP. It also demonstrated that this surgical approach can be safely performed. Further procedures will help define when it is the most appropriate surgical procedure for a patient with increasing ICP secondary to a traumatic brain injury.
NONE
Traumatic brain injury, decompressive craniectomy
IDENTIFICATION OF CHANGES OF WHITE MATTER TRACTS IN DIFFUSE AXONAL INJURY BY DT MRI
Hongfa Yang, M.D., The First Hospital of Jilin University
Zheng Jin, M.D., The First Hospital of Jilin University
Jingduo Zhou, M.D., The First Hospital of Jilin University
Diffusion-tensor MRI (DTI) allows us to perform qualitative and quantitative evaluation of damage of white matter tracts. Dynamics of structural changes in corpus callosum (CC) and corticospinal tracts (CST) at various periods after severe diffuse axonal injury (DAI) is not studied enough.
Dynamic 1.5T MRI studies were primarily performed in 33 patients (pts) aged 13–60 y (average 32±5.7 y) with severe DAI (GCS≤8) within 2 – 19 days after injury, and was repeated in pts (check every 6 days). DTI was repeated 2 times within 3 months and 6 months.
In pts primarily examined 10−18 days after injury the values of fractional anisotropy (FA) in the CC and CST significantly correlated with pts outcome (GOS score). A significant decrease of FA and thinning of CST fibers were observed in pts with unfavourable outcome and motor deficit. Moreover, the data analysis of the DTI studies in 33 pts in 3 months and 6 months after injury showed different degrees of CC degeneration in 27 of them. Only 6 pts showed some signs of partial recovery of CC fibers and favourable outcome in 6 months.
DTI in pts with DAI identifies the changes of the white matter tracts and is correlate with pts outcome.
The authors declare that they have no competing interests.
DAI, DTI, CTS, CC
IMPACT OF DIRECT TRANSPORT VERSUS SECONDARY TRANSFER ON 6-MONTH OUTCOME IN ADULT PATIENTS WITH TBI
Daniel Nishijima, MD, UC Davis Medical Center
Garth Utter, MD, UC Davis Medical Center
Gene Gurkoff, PhD, UC Davis Medical Center
Marike Zwienenberg-Lee, MD, UC Davis Medical Center
Professor Paul Muizelaar, MD, PhD, UC Davis Medical Center
Kiarash Shahlaie, MD, PhD, UC Davis Medical Center
Adult patients with traumatic brain injury (TBI) often present to outside hospitals prior to being transported to a regional trauma center. It remains unknown if secondary transfer to a trauma center, rather than direct transport from the scene of injury, affects neurological outcome at 6 months.
Data was prospectively collected from 868 adult (>16yrs) patients with TBI treated at the UC Davis Medical Center between October 2008 and February 2011. Forty-four demographic, clinical, laboratory and radiographic data points were collected, including age, gender, race,education, injury type, admission laboratory values, and Glasgow Coma Scale (GCS) and computed tomography (CT) scores on admission. An experienced investigator collectedGlasgow Outcome Scale (GOS) scores at 3 and 6 months after TBI. Multivariate logistic regression analysis was used to identify independent predictors of GOS at 6 months afterinjury.
Patients that are transferred from an outside hospital to a tertiary care center appear to havethe same functional outcome as patients that are admitted directly, even inpatients with severe injuries or injury patterns on CT scan that often promptacute surgical intervention. This study supports other reports that stabilization and subsequent transfer of TBI patients from smaller hospitals to experienced neurotrauma centers may be a viable model for delivering adult TBI care. The study population had an average age of 47.1 years and was comprised of 70.0 males. Injuries were classified as mild (GCS 14–15, 67), moderate (GCS 9–13, 19) or severe (GCS 3–8, 14). Of the 868 TBI patients, 617 were transported directly to UCDMC and 251 patients were secondarily transferred from an outside hospital. Thesetwo groups differed only with respect to the percentage of severe injuries(19.1 of direct transports versus 13.5 of secondary transfers, p=0.05). Independent predictors of functional outcome included age and GCS score. Transfer from an outside hospital was not an independent predictor of functional outcome in the entire population, or in the following subpopulations: (1) patients with severe TBI, (2) patients with epiduralhematoma, (3) patients with >5mm midline brain shift.
Patients that are transferred from an outside hospital to a tertiary care center appear to havethe same functional outcome as patients that are admitted directly, even inpatients with severe injuries or injury patterns on CT scan that often promptacute surgical intervention. This study supports other reports that stabilization and subsequent transfer of TBI patients from smaller hospitals to experienced neurotrauma centers may be a viable model for delivering adult TBI care.
We would like to acknowledge the assistance of Karen Smith, RN, and Nancy Rudisill, RN for their assistance in compiling our database.
TBI, outcome, transfer
THE COMPARATIVE ANALYSIS OF POSTOPERATIVE IMAGING FEATURES BETWEEN LATTICE DURAPLASTY AND ROUTINE METHOD IN DECOMPRESSIVE CRANIECTOMY
To compare the postoperative imaging features of patients operated using lattice duraplasty with routine method in decompressive craniectomy for severe head injury.
Sixty patients suffered from severe head injury with brain swelling treated by decompressive craniectomy were randomly divided into two groups, among them 30 cases operated with the routine method, and the new lattice duraplasty technique was applied for the other 30 patients. The correlated imaging features were compared postoperatively between these two groups.
One week after operation the CT scan demonstrated that the occurrence of brain hernia at the edge of bone window in lattice duraplasty group was statistically lower than that in routine method group(p<0.05); However, there was no statistic difference between two groups for the midline shift and ambient cistern compression on postoperative CT imaging(p>0.05).Three weeks after decompressive craniectomy the CT scan revealed a lower occurrence of brain hernia or encephalomalacia beyond bone window in lattice duraplasty group compared with that in routine method group(p<0.05). Moreover, the perfusion CT scan after 3 weeks showed that, compared with the un-herniated brain tissue, the herniated brain tissue beyond bone window had significant lower cerebral perfusion(p<0.01).
Decompressive craniectomy with lattice duraplasty could significantly reduce the postoperative occurrence of brain hernia and encephalomalacia at the site of decompression.
We thank all the staff members in the Department of Neurosurgery, The affiliated Jiangyin Hospital southeast university school of Medicine, China, for their outstanding contributions.
duroplasty; decompressive craniectomy; head injury
EARLY CSF LEVELS OF UCH-L1 ARE ASSOCIATED WITH MEASURES OF INJURY SEVERITY IN PATIENTS WITH SEVERE TBI
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
H. Julia Hannay, PhD, University of Houston
Shelley Heaton, PhD, University of Florida
Andrea Gabrielli, MD, University of Florida
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Ronald Hayes, PhD, Banyan Biomarkers
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Claudia Robertson, MD, Baylor College of Medicine
Steven Robicsek, MD, PhD, University of Florida
Ubiquitin C-terminal hydrolase (UCH-L1), also called neuronal-specific protein gene product (PGP 9.3), is highly abundant in neurons. This study assessed whether early levels of UCH-L1 in CSF were associated with acute and long-termmeasures of injury severity in severe TBI patients.
This study was designed as prospective observational study that enrolled adults with severe TBI presenting to 2 Level I Trauma Centers. Patients were included if they had a blunt head injury with a GCS score of ≤8 and required intracranial pressure monitoring. Ventricular CSF was sampled from each patient within 24 hours of TBI and analyzed for UCH-L1 (ng/ml) using ELISA. Measures of injury severity included the post-resuscitation GCS score, pupillary reactivity, pre-hospital hypoxia and hypotension, Rotterdam score and the 6-month dichotomized GOS score.
There were 131 patients enrolled in the study, patients were an average of 38 years old and 78 were male. At 6 months 104 patients had follow-up data and, of these, 84 had CSF collected for biomarker analysis within 24 hours of injury. Fifty five (66) patients had poor outcome per GOS at 6 months. Early levels of UCH-L1 were significantly higher inpatients with a post-resuscitation GCS score 3–5 compared to those with a GCSscore 6–8 (P=0.014). Patients with at least one unreactive pupil had significantly higher initial UCH-L1 levels than those with bilaterally reactive pupils (P=0.010). UCH-L1 was markedly elevated in those with pre-hospital hypoxia (P=0.043) but not with hypotension (P=0.170). Moreover, UCH-L1 was strongly associated with the Rotterdam score (P=0.005). When these factors were assessed in combination, the most predictive factors of pooroutcome at 6 months were age OR=1.06 [95CI 1.01–1.12], Rotterdam Score OR=2.97[95CI 1.58–5.56] and early levels of UCH-L1 OR=1.01 [95CI 1.00–1.02].
Early levels of UCH-L1 are associated with both early and late measures of TBI severity.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury”.
Biomarker, TBI, UCH-L1, prediction model
TIMING OF SECONDARY INSULTS FOLLOWING SEVERE TRAUMATIC BRAIN INJURY
Megan Brenner, MD, MS, University of Maryland School of Medicine
Peter Hu, MS, University of Maryland School of Medicine
XianShu Zhu, MS, PhD Candidate, University of Maryland School of Medicine
Lynn Stansbury, MD, University of Maryland School of Medicine
Bizhan Aarabi, MD, University of Maryland School of Medicine
Thomas Scalea, MD, University of Maryland School of Medicine
Intracranial hypertension (ICH) and cerebral hypoperfusion (CH) worsen outcome after severe traumatic brain injury (sTBI). There is little objective data that describe when these occur. We objectively evaluated the timing of ICH and CH to see if there were temporal differences in patients with good vs. poor outcome.
Patients with head AIS≥3, age >14 years, and need for intracranial pressure (ICP) monitoring were prospectively enrolled over 2 years. Continuous, automated, digital data was collected every 6 seconds and 5 minute means were calculated for the duration of monitoring. ICP and Cerebral Perfusion Pressure (CPP) were captured over 12-hour time periods from admission through hospital day 7. The Brain Trauma Index (BTI=CPP/ICP) was calculated for these time periods. Outcome was measured by Extended Glasgow Outcome Scale (GOSE) or dichotomized functional evaluations at least 3 months after injury.
207 patients were enrolled. Mean age was 39 years (range 16-84), mean admission GCS 6.8±3.6, mean head AIS 4.3±0.7, and mean Marshall score 2.6±0.8. The in-hospital mortality was 21.7%, with 29 (14%) dying from sTBI. Of the 162 survivors, 123 had functional evaluations. 95 had good functional outcome (77%). Lowest mean BTI occurred hours 12-24 (9.0) and hours 132-144 (8.9) after admission. When stratified by functional outcome, survivors demonstrated dramatically different temporal patterns of ICP increases and CPP decreases as measured by BTI, with the first 72 hours being nearly indistinguishable between the 2 groups.
Early in the hospital course, secondary insults occur frequently. Patterns of ICP elevation and CPP decline, as measured by the BTI, are the same in the first 3 days but then differ significantly based on functional outcome. Understanding the temporal nature of secondary insults has significant implications into developing more evidenced-based management approaches.
Supported in part by W81XWH-07-2-0118
TBI, intracranial hypertension, cerebral hypoperfusion
INTRACRANIAL PRESSURE (PRX) AND CEREBROVASCULAR PRESSURE REACTIVITY (CVRX) AS MEASURES OF CEREBROVASCULAR REGULATION AFTER TRAUMATIC BRAIN INJURY
Denis E. Bragin, Ph.D., Department of Neurosurgery and Brain Imaging Center, University of New Mexico School of Medicine
Tracey Berlin, R.N., B.S.N., University of New Mexico
Suguna Pappu, M.D., Ph.D., Department of Neurosurgery, University of New Mexico School of Medicine
Howard Yonas, M.D., Department of Neurosurgery, University of New Mexico School of Medicine
The CNS system enables continuous monitoring of cerebral variables of tissue cerebral blood flow (lCBF) and intracranial pressure (ICP) and dynamic cerebrovascular regulation by fluctuations in cerebral perfusion pressure (CPP) for PRx and CVRx. CBF and ICP are focal and global measurements, respectively. We examined their correlation using varying time epochs in the assessment of cerebral autoregulation.
Brain parenchymal intracranial pressure (ICP) and local CBF (lCBF) were continuously monitored using a Hummingbird guide (Innerspace, Inc.) and with cerebral perfusion pressure (CPP). The data were continuously collected on the CNS monitor in binary and converted to ASCII in one minute intervals. The correlation between spontaneous changes in ICP and lCBF with respect to CPP were plotted over various time epochs and correlation coefficients determined. The correlations were made over different time intervals ranging from three to 24 hrs over a period of up to six days.
A plot of 18 three hour episodes between PRx and CVRx had a correlation coefficient of 0.221 indicating no significant correlation between these variables over a period 54 hrs. Plots of ICP vs CPP and lCBF vs CPP also showed no correlation between these variables over the 54 hr period although there were instances when there was a significant correlation between CBF and CPP with an R value of 0.8473 as opposed to 0.0700 between ICP and CPP. The correlation between ICP and CBF was 0.1526 in this interval. Daily correlations between ICP and CBF versus CPP and between ICP and CBF showed significant correlations between these variables.
Our results indicate that the difference between a global measurement of ICP and the local white matter tissue measurement of lCBF with respect to CPP are not correlated in their reflection of the presence or absence of cerebrovascular regulation. However, the lack of correlation between PRx and CVRx may also be due to the long time intervals (3 and 24 hrs) over which the data were analyzed which could be remedied by correlations made at shorter time intervals.
Supported by the Department of Neurosurgery
Cerebrovascular ICP reactivity cerebrovascular reactivity
EXTENDED GLASGOW OUTCOME SCALE: VALIDITY AND SENSITIVITY TO CHANGE AT 6-, 12- AND 24-MONTHS AFTER SEVERE TRAUMATIC BRAIN INJURY
Ava M. Puccio, RN, PhD, University of Pittsburgh
Yue-Fang Chang, PhD, University of Pittsburgh
Allison J. Hricik, MS, University of Pittsburgh
David O. Okonkwo, MD/PhD, University of Pittsburgh
The extended Glasgow Outcome Scale (GOSE) has been recommended as the core measure of global outcome following traumatic brain injury (TBI) in clinical trials by the Common Data Elements TBI Outcomes Workgroup. This study goal was to examine the validity and sensitivity of GOS-E at 6-, 12- and 24-months.
Upon admission to our Level 1 hospital, the legal authorized representative (LAR) of all severe TBI patients were approached for consent of an IRB-approved protocol for data collection and neurological outcome assessment (Neurobehavioral Rating Scale-R (NRS), Disability Rating Scale (DRS), GOS, and GOSE) at 6-, 12- and 24-months post-injury from the years 2004–2010. Regression models evaluated the linear association between the GOS or GOSE and NRS and DRS at each outcome time point. Sensitivity to change was assessed by comparing the distributions of the GOS vs. GOSE categories at 6–12 months and 12–24 month timepoints.
Surviving adults who sustained a severe TBI (N=121; mean age±SD=33.25±14.26; 78.5% male; ISS=34.6; median GCS=6) were analyzed. Comparison of R2 values indicated that the GOSE was a comparable predictor of outcome to the GOS for functional measures at 6, 12, and 24-months postinjury. For the 6 month vs. 12 month comparison (n=121), 61 (50.4%) had a different category on the 12-month GOSE relative to the 6-mo GOSE. This contrasts with 43 subjects (35.5%) who had a change in category on the GOS (Test of symmetry; p=0.005). For the 12 mo vs. 24 mo comparison (n=79), 39 (49.4%) had a different category on the 24-month GOSE relative to the 12-mo GOSE. This contrasts with 25 subjects (31.6%) who had a change in category on the GOS (p<0.031).
The GOSE is a valid measure of outcome at 12 and 24 months postinjury and is more sensitive to change than the GOS. These findings support the GOSE's selection as a measure of disability in long term clinical trials that assess recovery.
NS30318; Brain Trauma Research Center: Department of Neurosurgery, University of Pittsburgh
Traumatic Brain Injury Glasgow Outcome Scale Outcome
BIOMARKERS AS PREDICTORS OF HEALTH-RELATED QUALITY OF LIFE FOLLOWING TBI IN CHILDREN
Vicki Anderson, PhD, Murdoch Childrens Research Institute
Anne-Marie Guerguerian, MD, PhD, University of Toronto
John Beca, MD, Starship Children's Hospital
P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital
In children with acute traumatic brain injury (TBI) clinicians are unable to accurately predict which children will have long-term cognitive and behavioural sequelae that impact their health-related quality of life (QOL). Serum biomarkers have the potential to help us accurately predict health related QOL outcomes.
Study design: A prospective observational cohort study of children, aged 2 to 17 years, hospitalized with mild to severe TBI. Biomarker measurements: We will measure 5 brain-specific and inflammatory proteins (e.g. glial fibrillary acidic protein and soluble endothelial selectin) and 5 lipids (e.g. eicosanoids and oxidized lipids) using ELISAs, multiplex immunoassays and mass spectrometry. Demographic and clinical factors, which are known to be associated with outcome in pediatric TBI, will be collected during the hospital admission: Glasgow coma score, intracranial pressure, hypoxia, hypotension, fever, features on neuro images and pupil reactivity. Pre-injury variables will also be collected while the patient is hospitalised (e.g. pre-injury adaptive ability and social and behavioural function). Primary outcome: Pediatric Quality of Life Outcome score at 6 and 12 months post-injury. Secondary outcomes: mortality, functional outcome, health status, adaptive abilities, behavioural function and social factors such as family burden.
We have recruited 54 children, between June of 2011 and February of 2012, at 3 children's hospitals, in Toronto, Ontario (N=17), Montreal, Quebec (N=8) and Melbourne, Australia (N=29). Research ethics board applications and contracts are in progress in Phoenix, Arizona and Aukland, New Zealand. The value of biomarker concentrations to explain the health-related QOL and secondary outcomes at 6 and 12 months will be explored with regression models. The best fitting models with and without biomarkers levels, using pre and post-injury clinical and risk variables, will be compared. We will present data on feasibility and methods of international collaboration, electronic data entry, biobanking and outcome measurements. We will also present our preliminary biomarker analyses and results of regression modeling.
Biomarkers and clinical data, collected in the acute setting in children hospitalized with TBI, have the potential to accurately predict long-term health-related QOL, neurocognitive and behavioral outcomes. This information will be used to implement rehabilitation and neuropsychological interventions at the best possible times to improve the QOL of these children and their families.
Ontario Neurotrauma Foundation and Victoria Neurotrauma Initiative
Pediatric, brain injury, biomarkers, outcomes
HEALTH RELATED QUALITY OF LIFE IN TRAUMATIC BRAIN INJURY: IS A PROXY REPORT NECESSARY?
Joan E. Machamer, MA, University of Washington
Sureyya Dikmen, PhD, University of Washington
Despite its importance, clinicians and researchers often discount patient-reported outcomes in favor of proxy reports in persons with TBI. The objective of this study is to determine the validity of patient-reported health-related quality of life by evaluating its relationship to injury severity and more objective indices of outcome.
Participants: Consecutive sample of 374 persons with TBl at least 14 years old and having post-resuscitation Glasgow Coma Scale score 12 or below, an acute seizure or a CT scan showing TBI-related findings. Main Outcome Measure: Life Satisfaction Survey at 6 months post injury
The greatest decrease in satisfaction was in ability to think and remember, work, income, and leisure activities. Dissatisfaction significantly relates to the functional limitation in that area as judged by them (p<.001) or someone who knows them well (p<.001). The most severely injured group reported the most dissatisfaction for 15 out of 17 areas assessed.
Patients with TBI, in general, do not need a proxy to report on their behalf regarding their limitations or health related quality of life. Degree of dissatisfaction relates to degree of functional limitations judged by them or their proxy.
Research Supported by DoD grant #W81XWH-08-2-159, NIDRR grant #H133A070032, and NINDS #R01 NS 19643
Quality of Life Awareness Proxy
DECREASING ADRENERGIC OR SYMPATHETIC HYPERACTIVITY AFTER SEVERE TRAUMATIC BRAIN INJURY USING PROPRANOLOL AND CLONIDINE (DASH AFTER TBI STUDY): FEASIBILITY OF A RANDOMIZED CONTROLLED TRIAL
Oscar D. Guillamondegui, MD, MPH, Division of Trauma & Surgical Critical Care, Vanderbilt University Medical Center
JoAnn M. Alvarez, MA, Department of Biostatistics, Vanderbilt University School of Medicine
Judith M. Jenkins, RN, MSN, Division of Trauma & Surgical Critical Care, Vanderbilt University Medical Center
Pratik P. Pandharipande, MD, MSCI, Division of Critical Care, Department of Anesthesia, Vanderbilt University Medical Center
Sympathetic hyperactivity after TBI manifests as catecholamine excess, abnormal heart rate variability, agitation, and is associated with poor neuropsychological outcomes. Propranolol and clonidine are centrally acting drugs that may decrease sympathetic outflow, brain edema, agitation, and storming. However, there is no prospective randomized evidence available for adrenergic blockade after TBI.
The DASH after TBI study is an actively accruing, single-center, randomized, double-blinded, placebo-controlled, two-arm trial, where one group receives centrally acting sympatholytic drugs, propranolol (1 mg intravenously every 6 hours for 7 days) and clonidine (0.1 mg per tube every 12 hours for 7 days), and the other group, double placebo, within 48 hours of severe TBI. The study uses a weighted adaptive minimization randomization with categories of age and Marshall head CT classes. The current primary endpoint is plasma norepinephrine level reduction of 450pg/mL on day 8. Secondary endpoints include comprehensive plasma and urine catecholamine levels, heart rate variability, arrhythmia occurrence, infections, agitation measures using the Richmond Agitation-Sedation Scale and Agitated Behavior Scale, medication use (anti-hypertensive, sedative, analgesic, antipsychotic), coma-free days, ventilator-free days, length of stay, and mortality. Neuropsychological outcomes are measured at hospital discharge and at 6 and 12 months.
Of the 85 patients that have been screened for eligibility since August 2011, 69 are ineligible. Of those ineligible, 65 subjects met a priori exclusion criteria (pre-existing condition, brain related, physiologic, demographic), three subjects were enrolled in another RCT, and one subject had no surrogate available for consent. Of the remaining 16 eligible subjects, 100% were enrolled using surrogate consent and randomized. The median age of enrollees is 26 (IQR: 22–36) and 81.3% are male. The Marshall CT Classification distribution is as follows: 68.8% Class II, 25% Class III, and 6.2% Class VI. While on treatment, no subject has experienced a cardiac complication (dysrhythmia, myocardial infarction, or cardiac arrest). The clinical team has not required blinding to be broken or needed the use of beta-blockers or alpha-2-agonists during the treatment period. Of subjects completing the entire 7-day treatment, the cumulative treatment compliance rate is 99.4% (501 of 504 doses). Non-compliance has been related to either protocol deviation by the intensive care unit team or related to protocol hemodynamic parameters. Of the enrollees who did not complete treatment, one was discharged early, and three died from their injuries.
The DASH After TBI Study is the first randomized, double-blinded, placebo-controlled trial powered to investigate adrenergic blockade in patients with severe TBI. Preliminary results indicate once eligible patients are identified, enrollment using surrogate consent and treatment compliance is very high. If the DASH After TBI Study results in positive trends, this could provide pilot evidence for an entire class of neuroprotective agents and open doors for a larger multicenter RCT. If there is no effect of therapy, this trial would still provide a robust prospective description of sympathetic hyperactivity after TBI.
MBP is supported by the Vanderbilt Institute for Clinical and Translational Research award (CTSA grant 5UL1RR024975-05) and AHRQ Health Services 5T32HS013833-08.
TBI, sympathetic hyperactivity, adrenergic blockade
IMPACT OF PEDIATRIC NEUROCRITICAL CARE ON TRAUMATIC BRAIN INJURY OUTCOMES
Jeffrey Leonard, MD, Washington University School of Medicine
Mazotas Ioanna, MD, Washington University School of Medicine
Keller Martin, MD, Washington University School of Medicine
Jeff Gill, PhD, Washington University
Allan Doctor, MD, Washington University School of Medicine
Traumatic brain injury is a leading cause of death and disability in children. The impact of neurocritical care on outcome is largely unknown. The objective of this study was to determine whether implementation of a pediatric neurocritical care team improved outcome for children with severe traumatic brain injury.
Retrospective cohort study. Two 6-year intervals before and after implementation of the neurocritical care team were analyzed. Participants included children age 3 to 219 months (0–18 years) consecutively admitted to the Pediatric Intensive Care Unit with severe traumatic brain injury (Glasgow Coma Scale Score≤8). Exclusion criteria:gunshot wound to the head, non-accidental trauma, cardiac arrest prior to ICU admission, and GCS=3 + fixed and dilated pupils on admission.
The primary outcome was disposition at hospital discharge. Secondary outcomes included the Glasgow Outcome Scale score at hospital discharge and length of ICU and hospital stay.
Implementation of the neurocritical care team improved outcome despite no changes in discharge practices per degree of disability. This improvement included a 72% reduction in mortality and improved disposition at hospital discharge for survivors. There was no difference in age, gender, mechanism of injury and severity of injury between the two groups. There was no difference in secondary outcomes.
Implementation of a neurocritical care team improved outcome in children with severe traumatic brain injury. Further studies are needed to identify best clinical practices and their impact on long term outcomes.
St. Louis Children's Hosptial Foundation
TBI, pediatric, neurocritical care
HIGH MOBILITY GROUP BOX-1 (HMGB1) EXPRESSION IN NEURONS OF RATS SUBJECTED TO LATERAL FLUID PERCUSSION INJURY (LFPI)
Enoch Wei, PhD, Anatomy and Neurobiology, Virginia Commonwealth University
John T. Povlishock, PhD, Anatomy & Neurobiology, Virginia Commonwealth University
HMGB1 originally identified as a 216 amino acid (29-kDa), is a ubiquitously expressed, abundant, nonhistone DNA-binding protein. Recently, HMGB1 has been reported to be an early mediator of inflammation after reperfusion in the pathophysiology of cerebral ischemia–reperfusion. Little information exists on HMGB1 expression in traumatic brain injury (TBI).
Male adult Sprague-Dawley rats were subjected to a 2.4 atmosphere LFPI. The animals were allowed to survive 10 min, 2 h, 6 h, and 24 h following injury. At the appropriate survival times, cerebrospinal fluid (CSF) and serum were sampled and the brain was fixed for immunohistochemical analysis. Brain sections were reacted with antibodies targeting HMGB1 and neurons (NeuN). HMGB1 labeled sections were also analyzed by electron microscopy (EM). HMGB1 levels in CSF and serum were analyzed by HMGB1 ELISA kit.
In shams, HMGB1 was expressed in the nucleus of neurons. In cortical contusion, a loss of HMGB1 expression was observed in both the nucleus and cytoplasm of neurons. In para-contusional zones, cytoplasmic expression of HMGB1 was routinely observed in neurons within 10 min of injury. The number of cytoplasmic HMGB1-containing neurons decreased with time. EM revealed necrotic change in those neurons containing cytoplasmic HMGB1. HMGB1 levels in CSF were elevated at 10 min post-injury compared to shams and then decreased at 2 and 6 hours post-injury. However, at 24 h post-injury, HMGB1 levels in CSF were again elevated. No elevation of HMGB1 level was observed in serum.
Collectively, our results suggest that HMGB1 translocates from the nucleus to the cytoplasm as early as 10 min post injury and that such neuronal cytoplasmic HMGB1 is associated with necrosis in the early phase of LFPI. This suggests that cytoplasmic HMGB1 may be a marker of neuronal necrosis in the acute phase of TBI.
We thank Susan Walker and Lynn Davis for their excellent technical assistance. This study is supported by NIH grants HD055813 and NS047463.
HMGB1, neuron, necrosis, TBI
VIRTUAL SCREENING ACCURACY OF AUTODOCK VINA IN DISCRIMINATING CALPAIN-1 INHIBITORS
Joy Guingab, Ph.D., Banyan Biomarkers, Inc
Anatoliy Vakulenko, Ph.D., Banyan Biomarkers, Inc
Ronald Hayes, Ph.D., Banyan Biomarkers, Inc
John Anagli, Ph.D., Banyan Biomarkers, Inc
The hyperactivation of calpain in traumatic brain injury makes the enzyme an attractive target for drug inhibition. Computational simulations are becoming an important part of drug discovery research. In this study, AutoDock Vina's accuracy was tested in discriminating biologically active compounds from decoys in inhibiting calpain-1.
The structures of known calpain inhibitors from literature were drawn using ChemBioDraw. Decoy structures were downloaded from DrugBank.ca, NCGC Pharmaceutical collection and the Directory of Useful Decoys (DUD). These files were converted to *.pdbqt format using Openbabel. All of the structures were visually inspected and minimized. To avoid artificial enrichment, the selected decoys' physical properties were matched with those of the true inhibitors. These properties include mass (MW<500) and rotatable bonds (<4). The core of calpain-1(2R9C) was downloaded from the Protein Data Bank. The water molecules bound to the enzyme were removed using the AutoDock tools suite. Non-polar hydrogens from the enzyme were also removed. The software Autodock Vina was used to dock the calpain inhibitors and the decoys. The compounds were ranked based on their predicted binding energies.
In a pool of decoys and known calpain inhibitors, Autodock Vina was able to select the active compounds significantly better than random (AUC=0.91, p<0.001). However, when the decoys from NCGC Pharmaceutical collection, DUD and Drugbank were removed and the known calpain inhibitors were grouped as active (IC50<300 nM) and decoys (IC50>1 μM to no activity), Vina's screening ability was almost random (AUC=0.48, p<0.001).
In the process of drug discovery, millions of compounds from publicly available databases can be virtually screened by Autodock Vina to generate hit compounds. These hits can then be tested in vitro for activity against calpain. If active, the structures can be optimized to increase potency in inhibiting calpain. However, the software should not be used in lead optimization process. In a database that contains closely related structures, Vina was not able to discriminate between biologically potent compounds and non-actives.
Banyan Biomarkers
Calpain, brain injury, virtual screening, AutoDock
NEUROPROTECTION WITH AN ERYTHROPOIETIN-MIMETIC PEPTIDE (PYROGLUTAMATE HELIX B SURFACE PEPTIDE) IN A RODENT MODEL OF SEVERE CONTROLLED CORTICAL IMPACT INJURY
Samson Gaddam, MD, Baylor College of Medicine
Leela Cherian, PhD, Baylor College of Medicine
Roberto Garcia, MD, Baylor College of Medicine
Raymond Grill, PhD, Univ TX Houston
Carla Hand, MD, PhD, UNC School of Public Health Dept of Epidemiology
Claudia Robertson, MD, Baylor College of Medicine
Treating severe traumatic brain injury (sTBI) with Erythropoietin (EPO) enhances neurological recovery. Pyroglutamate Helix B Surface Peptide (pHBSP), an EPO-mimetic peptide, retains EPO's neuroprotective actions while reducing risk of thrombotic complications seen with EPO. This study examines effects of pHBSP on cognitive recovery in a rodent model of sTBI.
Long Evans rats were randomly assigned to one of three groups (pHBSP, saline, sham injury). The animals were anesthetized with isoflurane in oxygen, intubated, and ventilated to maintain normal arterial blood gases. An 8 mm right parietal craniectomy was made and a controlled cortical impact was induced with a velocity and deformation of 5m/sec and 3 mm, respectively. Anesthesia was discontinued and animals were allowed to awaken. pHBSP (30 ug/kg) b.i.d. by intra-peritoneal route starting 1 hour after surgery for 3 days was administered. The control group received saline. On days 1, 2, 7, and 14 post-injury, animals were subjected to the Novel Object Recognition (NOR) task to analyze non-spatial memory function. On day 11–15 post-injury, the animals were tested on the Morris water maze task to analyze spatial memory function.
47 rats were included in the study (pHBSP-17, saline-21, sham-9). 8 animals died (pHBSP-1, saline-6, sham-1) during the experiment. The mortality rate of the pHBSP-treated group (1/17, 5.9%) was lower than the saline control group (6/21, 28.57%), but the difference was not statistically significant.
Morris Water Maze
Each animal performed a total of 22 trials on days 11–15 post-injury. For trials where the animal successfully found the platform, the length of time was used in the analysis. The average length of time required for all animals to find the platform decreased progressively over time. This decrease with repeated trials was largest on the first day of testing (day 11 post-injury). Animals treated with pHBSP took significantly less time to find the platform than control (p=0.036, by repeated measures ANOVA test). The day 11 values for the treated animals were significantly better than the control animals when adjusted for multiple comparisons with the Bonferroni test.
Novel Object Recognition
Substitution of a familiar object for a new object (non-spatial change) was done in the NOR trials separated by one hour on days 1, 2, 7 and 14 following sTBI. The time spent with a novel object was calculated as a percentage of the total time spent during exploration. There was a statistically significant difference (p=0.002) in the percentage of time spent with the novel object between the treatment and control animals. The pHBSP-treated animals spent more time with the novel object than control animals. On day 2, the percentage of time spent with the novel object was significantly different using Bonferroni test to adjust for multiple comparisons.
pHBSP significantly improved cognitive recovery in this rodent model of sTBI. Recovery of spatial memory (hippocampal function) and non-spatial memory (entorhinal cortex and subiculum function) was statistically significant in the pHBSP group when compared to the control group.
This study was supported by National Institutes of Health (NIH) grant #P01-NS38660.
Traumatic brain injury (TBI)
INHIBITION AND EXCITATION IN THE CLUSTERS OF HYPER/HYPOACTIVE NEURONS IN SENSORY BARREL CORTEX AFTER TRAUMATIC BRAIN INJURY
Fritz W. Lischka, Ph.D., Center for Neuroscience and Regenerative Medicine, Uniformed Services University of the Health Sciences
Zygmunt Galdzicki, Ph.D., Center for Neuroscience and Regenerative Medicine, Uniformed Services University of the Health Sciences
Disruption of balance between excitatory and inhibitory networks underlies causes of neurological disorders e.g. epilepsy and schizophrenia. We hypothesize that traumatic brain injury (TBI) using controlled cortical impact (CCI) produces a shift in this balance by strengthening inhibitory circuits, which in turn impairs the ability of the neocortex to recover.
In this study we focused on the assessment of network activity by two-photon calcium imaging in combination with whole-cell patch clamp recordings. This approach enabled us to monitor calcium changes and electrical-activity in astrocytes, excitatory and GABAergic neurons to identify alterations in network dynamics followed by craniotomy CCI injury. We used GAD67 male adult knock-in mouse that expressed GFP in all GABAergic neurons. In these mice, we recorded from three groups of cells: excitatory neurons (EGFP-negative and SR101-negative), GABAergic neurons (EGFP-positive and SR101-negative), and astrocytes (EGFP-negative and SR101-positive) with the goal to characterize cell-type specific TBI-induced dysfunction. In slices derived from TBI and sham-operated mice and in the proximity of the injury site (within 300–500 μm from injury center), we measured intrinsic passive and active characteristics, and synaptic activity in both excitatory/inhibitory neurons. We also monitored calcium dynamics using two-photon Ca2+ imaging in neurons and astrocytes.
We tested our hypothesis using intact 400 μm thick coronal cortical slices containing barrel cortex. Quantitative calcium measurements revealed an increase in number of silent neurons and decrease in neuronal activity by 2-fold in layer 2/3 cortical neurons in response to injury. Unexpectedly some of the neurons exhibited an increase in the frequency of spontaneous Ca2+-transients by 15-fold followed by CCI injury. The hyperactivity appeared to be due to a relative decrease in synaptic inhibition because of altered homeostasis balance between excitation/inhibition. Electrophysiological characterization of the cells in the vicinity of the injury in TBI mice shows altered spontaneous activity in comparison to the sham control. Spontaneous events in excitatory neurons in TBI mice appear to have a larger cumulative probability for amplitude of sEPSCs than in sham mice, whereas opposite trend occur in fast-spiking inhibitory neurons. Simultaneous calcium imaging and electrophysiological recording has revealed a substantial shift in action potential-induced intracellular calcium transients in both inhibitory and excitatory neurons due to TBI.
Our results suggest desynchronization of neocortical activity following TBI with an impaired inhibition underlying hyperactivity/hypoactivity of excitatory neurons possibly due to disruption of feedback inhibition. Although neurotoxicity and thereby focal pathology is observed in injured brain, we suggest that a redistribution of synaptic balance between excitation/inhibition of glutamatergic and GABAergic neurons provides a mechanism for the disturbed somatosensory cortical function in traumatic brain injury. Consequently, pharmacological restoration of cortical excitation/inhibition balance after TBI should facilitate functional improvement of cortical plasticity in injured cortex. Taken together the capacity to simultaneously image calcium dynamics in three different cell types with a high spatial and temporal resolution along with whole-cell recordings in individual neurons appears as a valuable tool for comprehensive and quantitative analysis of molecular mechanisms in barrel cortex and thalamocortical circuit dysfunction underlying the pathophysiology of TBI.
We thank Ms. Xiufen Xu for assistance. This work was supported by CNRM grant [G1703O] to ZG and postdoctoral career development grant [G1708X] to MKJ.
CCI, TBI, 2-Photon imaging, Inhibition/Excitation
IGF-1 N-TERMINUS TRIPEPTIDE, GPE, ATTENUATES MICROGLIAL AND CASPASE-3 ACTIVATION LEADING TO DECREASES IN NEURONAL AND AXONAL DEGENERATION FOLLOWING CLOSED HEAD INJURY IN THE IMMATURE 11-DAY-OLD RAT
Ramesh Raghupathi, Ph.D., Drexel University College of Medicine- Dept. of Neurobiology and Anatomy
The N-terminus tripeptide of insulin-like growth factor, glycine-proline-glutamate (GPE), is neuroprotective in adult and pediatric models of hypoxic-ischemic brain injuries, by attenuating microglial activation and activating anti-apoptotic pathways. We hypothesized that GPE will reduce neuropathologic events following closed head injury in the immature rat.
Anesthetized 11-day-old rats were subjected to an electronic controlled cortical impact to the intact skull over the left parietal cortex (3 mm displacement; 5 m/sec). Immediately following injury, rats were implanted with Alzet mini-pumps designed to deliver GPE at the rate of 6mg/Kg/day for 3 or 7 days (N=4/time period). As controls, brain-injured rats (N=4) were also implanted with Alzet mini-pumps filled with phosphate-buffered saline (PBS, 100μL); sham-injured rats received either GPE or PBS (N=2–3/condition). Animals were euthanized at either 3 or 7 days and brains were analyzed for activation of caspase-3 (apoptosis), neuronal and axonal degeneration (fluoro-Jade B) and reactive microgliosis (Iba1 immunoreactivity).
Sham-injured animals receiving either vehicle or GPE did not demonstrate caspase-3 activation, neurodegeneration or reactive microgliosis. Brain trauma in vehicle-treated, immature rats resulted in caspase activation in the injured cortex and thalamus at 3 and 7 days and in the corpus callosum at 3 days post-injury. Infusion of GPE for 3 days failed to reduce the extent of caspase-3 activation but when extended to 7 days was able to reduce caspase-3 activation in the cortex and thalamus. At 3 and 7 days post-injury, microglial reactivity in the injured cortex, subiculum, thalamus and corpus callosum of vehicle-treated rats was characterized by increased density of Iba1(+) cells which exhibited swollen cell bodies and shorter processes. Treatment with GPE for 7 days, but not 3 days, reduced the extent of Iba1 immunoreactivity in these regions which was accompanied by a reduction in the extent of fluoro-Jade B reactivity.
Collectively, these data demonstrate that prolonged infusion of GPE may provide neuroprotection following trauma to the immature brain by affecting microglia-mediated inflammation and apoptotic cell death pathways.
NIH Grant K08NS053651, NIH Grant HD061963
TRAUMA, BRAIN, PEDIATRIC, MICROGLIA, CASPASE
REGION SPECIFIC NEURODEGENERATION AND ASTROGLIOSIS IN BLAST-INDUCED NEUROTRAUMA IN THE RAT
Mikulas Chavko, Ph.D., Naval Medical Research Center
Grant Hennes, Bs.C, DRDC Suffield
Saleena Adeeb, Ph.D., Naval Medical Research Center
Tracy Weiss, Bs.C, DRDC Suffield
Richard McCarron, Ph.D., Naval Medical Research Center
Neurodegeneration and active astrogliosis have been considered hallmarks of traumatic brain injury (TBI). However, the susceptibility of various brain regions differs depending on the model of TBI employed. This study examined the regional differences in neurodegeneration and astrogliosis in rat brains exposed to blast.
Animals were subjected to a 120 kPa blast overpressure with a side-on orientation in a pneumatic-pressure driven shock tube and allowed to recover for 3h, 24 h, 1 week and 3 weeks following blast exposure. At the end of each observation period, animals were euthanized and immediately perfused transcardially with 0.1 M ice-cold phosphate-buffer saline (PBS, pH 7.4) followed by 4% formaldehyde. Coronal sections (20 μM thickness) of the entire brain were prepared and stained with antibodies against the axonal marker Neurofilament H (NFH) and the astrocyte marker glial fibrillary acidic protein (GFAP). Confocal fluorescent images were acquired and processed using a stitching module so that the entire structure of a particular region (e.g., hippocampus) could be viewed and analysed. This technique ensures that the same areas of interest can be compared between control and blasted animals.
Results showed that the most susceptible regions to blast-induced injury were the hippocampus and the brain stem. Starting from 24 h after blast exposure, GFAP staining was significantly increased across all areas of both hippocampi while NFH staining were mainly decreased in the CA2 and CA3 regions of the hippocampi. These changes peaked at 1 week and persisted for at least 3 weeks after blast exposure. No significant changes in GFAP and NFH staining patterns were observed in either the frontal cortex or the cerebellum. Interestingly, while there was no significant gliosis in the brain stem, the staining pattern of NFH changed dramatically. In normal brain stem, many organized neurofilament bundles representing vertical neuronal axons connecting the brain and the spinal cord were observed. However, these bundles were replaced with disorganized neurofilaments starting from 1 week after blast exposure.
Future studies will examine other areas of the brain to help characterize brain injury patterns in blast-induced neurotrauma.
N/A
TBI; neurodegeneration; astrogliosis; hippocampus
EARLY INDUCED HYPOTHERMIA ATTENUATES REPERFUSIONAL NEURONAL AND SUBSEQUENT GLIAL DAMAGES IN A RAT SUBDURAL HEMATOMA MODEL : A MICRODIALYSIS STUDY
Shyam Gajavelli, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Kevin K.W. Wang, PhD, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
Helen M. Bramlett, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
W. Dalton. Dietrich, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
M. Ross. Bullock, MD, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
There is strong evidence for the use of hypothermia in treating patients with ischemic/reperfusional (I/R) injuries. I/R is present in acute subdural hematoma (ASDH), therefore we hypothesized that hypothermia may also be beneficial in this injury, by blunting reperfusional injury in the phase of post decompressive craniotomy.
28 S-D rats had induced ASDH and were allocated to one of four groups (7 rats each). Their temperatures were manipulated as follows: Group1 (Normothermia), maintained at 37°C throughout. Group 2 (Early-hypothermia), brain temperature (BT) reduced to 33°C at 30 min prior to craniotomy and ASDH removal, continued for three hours. Group 3 (Late-hypothermia), BT lowered to 33°C at 30 min after decompression and maintained 3 hours. Group 4 (Sham), normothermia without ASDH. To estimate neuronal and glial cell damage, we used microdialysis (100kD cut-off) techniques and compared the extracellular concentrations of the “BIOMARKERS” Ubiquitin Carboxyl-Terminal Hydrolase-L1 (UCH-L1) and Glial Fibrillary Acidic Protein (GFAP) sampled in ischemic and reperfusional phases among the four groups.
In the early phase of reperfusion (30min- 2.5 hrs after decompression), extracellular UCH-L1 in the early induced hypothermia group was significantly lower than in the normothermia and late hypothermia groups. Also, in the late phase of reperfusion (>2.5hrs after decompression), extracellular GFAP in the early hypothermia group was lower than in the normothermia and late hypothermia groups.
Only in the early-induced hypothermia group, both neuronal and glial biomarkers were seen to be attenuated. In contrast, late induced hypothermia did not attenuate neuronal damage but reduced subsequent glial damage. These data support further study of early induced hypothermia in basic laboratory research and clinical trials.
This work was supported by funds from NINDS RO1 NS 042133, the Miami Project to Cure Paralysis, and Nippon Medical School, Tokyo, Japan.
Hypothermia, Ischemia, Reperfusion
ANIMAL MODELS OF MILD REPEAT HEAD INJURY: SELECTIVE AXONAL INJURY, NEUROINFLAMMATORY RESPONSES, AND THE ROLE OF TAU MUTATIONS
Leyan Xu, Ph.D., Johns Hopkins University School of Medicine
Jiwon Ryu, Ph.D., Johns Hopkins University School of Medicine
Tong Li, Ph.D., Johns Hopkins University School of Medicine
Christina R. Marmarou, Ph.D., Virginia Commonwealth University, Department of Neurosurgery
Chronic traumatic encephalopathies (CTE) found in the brains of athletes include progressive tauopathies resulting from repeat concussions. The role of genetic vulnerability on Tau aggregation and neurodegeneration is unknown. Here we take the first steps in developing repeat concussive mouse models to study vulnerability-insult interactions relevant to CTE.
The rat impact accelerator of Marmarou was redesigned for the mouse cranium. Weights varied from 20–60 gr., with a goal to achieve a height-weight impact allowing for >90% survival with rare/no fractures. Several repeat-injury schedules were implemented, including a subacute regimen with 4 hits at days 0, 1, 4 and 7. With a 40g-1m impact, death rate was 8.3% and no fractures were observed. Subjects from this group were perfusion-fixed 7 days after TBI and brain tissues were processed for: general histology; TUNEL and Fluoro-Jade for dying neurons; Gallyas silver and APP ICC for traumatic axonal injury; and ICC for neuroinflammation markers, i.e. GFAP, CD68 and IBA1. To explore the predisposition of mutant tau to injury-induced degeneration, we applied our subacute model on five-week-old P301S tau transgenic mice. Mice were sacrificed at 4 months. Saggital brain sections or the retina were processed for phosphorylated tau ICC (AT8 and S422).
Our mild repeat TBI model produces no focal injury, anoxic/ischemic hippocampal or cerebellar Purkinje cell loss. There was traumatic axonal injury in the optic nerve and tract, cerebellar white matter/peduncles, lower corticospinal tract, lateral lemniscus, and corpus callosum, evident with both Gallyas silver and Fluoro-Jade. There was microglial activation with phagocytic transformation in some areas, e.g. the optic tract and select cerebellar sites (climbing fiber system). There was also evidence for greater density of GFAP (+) astrocytic profiles at these sites. There was a close anatomical association between traumatic axonal injury and microglial activation. There was also microglial activation near the cell bodies of origin of injured axons, e.g. the retinal ganglion cell (RGC) layer in retina. To explore weather mutations that increase Tau's propensity to aggregation, e.g. the FTD-associated mutation P301S, become hyperphosphorylated and/or precipitate within neuronal cell bodies after injury, we applied our mild repeat TBI model on P301S tau transgenic mice. Animals were exposed to TBI at five weeks of age and then sacrificed at 4 months. Initial analysis focused on the retina, as one of the prime sites of injury using our mild repeat TBI protocol. In several transgenic subjects exposed to injury, we found intense degeneration of RGC axons and phosphorylated Tau accumulations in RGC cell bodies reactive with both AT8 and S422 antibodies. We are presently counting S422 (+) and AT8 (+) profiles in the retinas of P301S mice exposed to injury or sham treatment. We have obtained similar results in the primary motor cortex.
The rat impact accelerator can be adjusted for use in mice and help generate mild repeat protocols of TBI that result in no fractures or focal contusions and produce no significant anoxic neuronal injury. These mild repeat protocols result in traumatic axonal injury in select CNS sites, e.g. the optic nerve and tract and components of the cerebellar circuitry, as well as microglial activation and phagocytic transformation at these sites. Our mild repeat model is presently used in P301S transgenic mice predisposed to tauopathy with the question of whether this concussion-like injury can accelerate tau pathology in these animals including hyperphosphorylation, aggregation and misfolding. Such models may provide critical proof of concept that mild repeat injury may play a key role in tau-related neurodegenerative events in genetically vulnerable subjects.
This work was supported by NIH and Maryland TEDCO funds to VEK.
concussion,tauopathy, traumatic axonal injury, neuroinflammation, neurodegenerative
HUMAN RESULTS FROM ANIMAL MODELS: SCALING LAWS FOR BLAST NEUROTRAUMA
Cameron R. Dale Bass, Ph.D., Duke University
A major uncertainty in blast neurotrauma research is relating an experimental exposure and response model to humans. Well-established pulmonary blast scaling laws imply that many small-animal neurotrauma models do not simulate real-world exposures. A better understanding of the biomechanics of blast exposure, response, and size is essential.
Finite element models of a head were developed to study the internal response of the brain during blast exposure. The spherical head model consisted of a skull (with diploë), a cerebrospinal fluid (CSF) layer, and a brain. Five scaled replica models were developed ranging in size from 10 to 160 mm in diameter, representing the anatomical range between mouse and human. All biological materials were modeled using viscoelastic theory and were consistent across all models. Each model was loaded by blasts considered relevant for real-world conditions, which ranged from 100 to 1000 kPa peak incident overpressure, and 1 to 8 ms positive-phase duration. Biomechanical response was examined using brain tissue biomechanical response (pressure, stress, and strain) and the frequency response of the skull. A power-law scaling function was used to determine the relationship between biomechanical response, blast exposure, and head size.
Peak pressure, stress, and strain patterns in the brain were consistent across model size for a given blast pressure and duration. At lower blast overpressures, the largest pressures developed in the brain were opposite the site of impact (reminiscent of a contre-coup brain injury). At higher blast overpressures, the largest pressures were located at the site of impact. Peak stresses and strains were mainly located at the CSF interface and in the region opposite the site of impact. Brain pressure, stress, and strain were higher in the smaller head models. The 50th percentile (by volume) peak brain pressure, stress, and strain increased by approximately 6%, 40%, and 50% respectively when decreasing the diameter of the head model by half. The resonant frequency response of each skull model was independent of blast loading and ranged from 2.3 kHz (160 mm diameter model) to 41 kHz (10 mm diameter model) and was inversely proportional to skull diameter. Similarly, the final velocity of the head model caused by the blast would double when reducing the diameter by half. The power-law scaling function for scaling brain tissue response was an excellent fit for all response variables (R2>0.96), and peak blast pressure, duration, and model diameter were all significant factors (p<0.001). The scaling relations for the 50th percentile brain pressure and brain strain together can be used to determine the scaled exposure for an equivalent biomechanical response in a brain of different a size. For instance, the equivalent scaled blast in a human would be 5% higher in peak pressure, but approximately 6 times longer in duration than a blast exposed to a mouse with the same biomechanical brain response.
The results of this study suggest that the biomechanical responses of the brain to blast exposures are dependent on brain size and imply that smaller animal models may be more susceptible to blast neurotrauma than a human at the same exposure. The larger kinematic response and higher frequency deformation of the skull likely exacerbates the biomechanical loading of the brain tissue in smaller head models. Size effects must be recognized when developing blast neurotrauma models and the applied loads should be scalable for real-world conditions. This study also provides a scaling relation for blast neurotrauma in models of various size and exposure. The effect of pressure scaling was relatively small in comparison to duration scaling, which is similar to the scaling laws developed for pulmonary blast injuries and blunt impacts. Investigation into the non-biomechanical aspects of blast neurotrauma scaling is required for a more robust scaling law.
This work is supported by the Multidisciplinary Research Initiative (MURI) program (W911MF-10-1-0526) through the Army Research Office (ARO).
blast, neurotrauma, scaling, biomechanics, model
KALLIKREIN-RELATED PEPTIDASE 6 AS A PROMISING BIOMARKER OF TRAUMATIC BRAIN INJURY
Jacob VanLandingham, PhD, Neuroscience, Florida State University College of Medicine
KLK6 is most abundant in neurons and oligodendrocytes and is only expressed in astrocytes and activated microglia in response to injury and/or disease. As KLK6 levels increase, extracellular matrix proteins degrade and inflammatory cells are thought to migrate to injury sites leading to increased neuronal cell death and blood-brain barrier breakdown. Progesterone (PROG) and other neurosteroids are known to decrease inflammation caused by traumatic brain injury (TBI) and are involved in kallikrein regulation. PROG also increases the suspected inhibitor of (RAT)Klk6 - neuroserpin. Therefore, this study examined the hypothesis that (RAT)Klk6 increases with increased severity of TBI and neurosteroid treatment reduces the concentration of (RAT)Klk6.
Male Sprague Dawley rats were placed in one of 7 groups; 6h injured (with and without PROG), 24h injured (with and without PROG), 72h injured, 96h injured, and uninjured. Each was anesthetized and received a controlled cortical impact (CCI) to the medial frontal cortex at a depth of 3mm and velocity of 2.5m/sec. PROG (16mg/kg) and vehicle (22.5% 2-hydroxypropyl-β-cyclodextrin) was delivered intraperitoneally at 1h and subcutaneously at 6h post-CCI. Protein analysis was performed on all biopsies and serum. Homogenized tissue was centrifuged at 4 degrees C for 30min at 14,800xg. The supernatant was placed in sodium dodecyl sulfate and β-mercaptoethanol buffer and run on a 4–20% polyacrylamide gel (BioRad Mini PROTEAN TGX). Samples were then transferred to nitrocellulose. Following transfer, the membrane was blocked for 1h and incubated overnight with polyclonal antibodies generated in rabbit before visualization with LiCor Odyssey Imaging. Serum and standards were run in duplicate with a USCN Life Science, Inc ELISA Kit.
Relative abundance of (RAT)Klk6 in brain homogenate from the contusion core and the peri-contusion area increased over 7-fold within 6h (p<0.001) and remained approximately 2-fold higher than shams out to 96h (p>0.05). (RAT)Klk6 protein in serum increased to 0.1548 ng/mL greater than shams by 6h Approximately 34% increase) and 0.6177 ng/mL in serum by 24h (over 2-fold increase). Levels reduced to uninjured levels by 24h in brain homogenate and to 0.3803 ng/mL by 72 h in serum. No significance in (RAT)Klk6 abundance was noted between injured and uninjured at 72h and 96h. PROG treatment significantly increased (RAT)Klk6 protein levels above vehicle-treated levels by 6h (p<0.001) but this elevation fell by 24h.
CCI of Sprague Dawley rats clearly results in increased expression of (RAT)Klk6 in the brain. Increases are also observed in the blood although to a lesser degree. These levels remain slightly elevated out to 96h. Preliminary data in moderate and mild models of TBI in rodents show similary, graded responses in (RAT)Klk6 concentration. This data suggests that (RAT)Klk6 may be a valuable addition to the TBI biomarker panel in the immediate diagnosis of TBI. However, a detailed quantification of the changes in (HUMAN)KLK6 in mild, moderate, and severe TBI is still needed. The KLK response element is known to be responsive to steroids; therefore, it is not surprising that PROG treatment increases (RAT)Klk6. KLK6 contains a classical progesterone receptor regulatory element in the upstream untranslated region of (RAT)Klk6 but not human according to Christophi et al. This paired with the finding that some neurosteroid treatments increase the suspected endogenous inhibitor of (RAT)Klk6 (neuroserpin), suggest a potential mechanism through (RAT)Klk6.
We thank Dr. Blaber's Laboratory for the anti-rat K6 polyclonal antibody, and Tallahassee Memorial Hospital and the Biomedical Science Department for funding.
TBI; progesterone; kallikrein; biomarker; rodent
INTERPROFESSIONAL EDUCATION TO COMMUNITY OUTPATIENT REHABILITATION HEALTH PROFESSIONALS FOR TREATMENT TO U.S. WOUNDED WARRIORS WITH BRAIN INJURIES
Jacobus Donders, PhD, Mary Free Bed Rehabilitation Hospital
Lorraine Pearl-Kraus, PhD, FNP-BC, Sole Proprietor
Jeff Trytko, MS, Grand Valley State University
Lawrence Baer, PhD, Sole Proprietor
Theresa Bacon-Bauley, PhD, RN, Grand Valley State University
Susan Jensen, PhD, RN, CCM, Grand Valley State University
Jared Skillings, PhD, Pine Rest Christian Mental Health Services
This research evaluated the knowledge gain of community-based rehabilitation health providers who received traumatic brain injury (TBI) education with military relevance. These health professionals will treat U.S. OIF/OEF Wounded Warriors returning home with TBI. This abstract presents findings on the education component of this research.
Grand Valley State University (GVSU) provided Mary Free Bed Rehabilitation Hospital (MFBRH) staff educational modules from July to September 2011. Modules included Pathophysiology/Symptomatology, Behavioral/Mental Health, and Case Management/Community Reintegration. Pedagogy included blended learning utilizing classroom instruction, web-based content, and live simulations. Fifteen MFBRH health professionals signed informed consent and participated consisting of a medical physician (1), neuropsychologist (1), physical therapists (3), occupational therapists (3), speech language pathologists (2), social worker (1), rehabilitation driving instructor (1), project coordinators (2) and a secretary (1). GVSU administered pre-and post-knowledge tests and evaluation measures. Fourteen of fifteen MFBRH health professionals completed the education.
Health professionals demonstrated improved pre- and post-test knowledge change on all three modules regardless of education and clinical role (Table 1): clinicians with terminal degrees (2),63.3–79.3%; clinicians with bachelors and masters degrees (9), 53.5–75.2%, p<0.01; non-clinical administrative (3), 48.9–71.1%. Statistical inference was not calculated for staff categories with low numbers. Health professionals demonstrated improved pre- to post-test knowledge on all three modules (Table 2): Pathophysiology/Symptomatology (p<0.001); Behavioral/Mental Health (p<0.001); and Case Management/Community Reintegration (p<0.001); total (p<0.001). Program evaluations were positive overall concerning content, while requesting more access to information from Department of Defense/Veterans Affairs professional staff.
There was improved knowledge on military relevant TBI rehabilitation care across all modules. The MFBRH health professionals gave overall positive evaluations, providing areas for improvement including more military culture information and increased interaction with Wounded Warriors in simulations.
Sponsor: the U.S. Army Medical Research & Materiel Command and the Telemedicine & Advanced Technology Research Center under Contract Number: W81XWH-10-1-0607.
TBI Education Wounded Warrior Rehabilitation
DOES BILATERAL MODERATE FLUID PERCUSSION BRAIN INJURY ELICIT LONG-TERM NEUROBEHAVIORAL AND COGNITIVE DEFICITS IN THE ADULT RAT?
Amor Belmeguenai, PhD, Lyon Neuroscience Research Center - Tiger Team; INSERM U1028; CNRS UMR5292; University Claude Bernard Lyon 1; University of Lyon, France
Béatrice Georges, University of Lyon, France
Noëmie Mermet, University of Lyon, France
Raafat Fares, PhD, Lyon Neuroscience Research Center - Tiger Team; INSERM U1028; CNRS UMR5292; University Claude Bernard Lyon 1; University of Lyon, France
Thomas Lieutaud, Lyon Neuroscience Research Center - Tiger Team; INSERM U1028; CNRS UMR5292; University Claude Bernard Lyon 1; University of Lyon, France
Gen. Lionel Bourdon, MD, PhD, Institut de Recherche Biomédicale des Armées - Département Soutien Médico-Chirurgical des Forces, France
Jean-Jacques Risso, PhD, Institut de Recherche Biomédicale des Armées - Département Soutien Médico-Chirurgical des Forces, France
Emilie Carré, PhD, Institut de Recherche Biomédicale des Armées - Département Soutien Médico-Chirurgical des Forces, France
Laurent Bezin, PhD, Lyon Neuroscience Research Center - Tiger Team; INSERM U1028; CNRS UMR5292; University Claude Bernard Lyon 1; University of Lyon, France
In most studies, unilateral mild-to-moderate fluid percussion brain injury (LFPu) fails to yield robust long-lasting neurobehavioral and cognitive deficits in experimental rodent models. Here, we asked whether moderate fluid percussion brain injury elicited bilaterally (LFPb) can result in short- and long-term deficits in the adult rat.
We exposed adult male Sprague Dawley rats (270–300g) to Sham surgery, LFPu or LFPb (mean fluid pressure: 2.3Atm; diameter: 5mm; center: 3.3mm posterior and 3mm lateral to Bregma) under isoflurane anesthesia. We measured the expression level of the inflammatory markers IL-1β, TNFalpha, IL-6, MIP-1alpha, MCP-1 and CD-14 using RT-qPCR in the cortex, hippocampus and dorsal thalamus of LFPu, LFPb and Sham rats at 8 hours post-trauma. We measured anxiety-like behaviors at 7, 14 and 43 days post-trauma and spatial memory at 1 month post-trauma in another subset of LFPb and Sham rats using the Water Exploration Test (WET) and the Morris Water Maze (MWM), respectively. In these rats, we also measured hippocampal long-term potentiation (LTP), an index of the activity of the neural substrates of learning and memory, at 2 months post-trauma using in vitro slice electrophysiology.
LFPu and LFPb elicited an “explosive” inflammatory response in injured brain hemispheres at 8 hours post-trauma. There was also a slight, but significant, inflammation in the “uninjured,” contralateral, hemisphere in LFPu rats. Induction of inflammatory markers was of similar amplitude in the left and right hemispheres in LFPb rats. Although LFPb rats exhibited prominent brain inflammation, we only detected transient hyperactive traits in the short-term and no signs of anxiety in these rats using the WET test. At 1 month post-trauma, LFPb rats needed more trials to find the platform in the MWM learning test than Sham rats; however, by the end of the protocol, the latency to find the platform was similar in LFPb and Sham rats. Consistent with MWM test results, we did not detect any robust differences in LTP in CA1 pyramidal cells of the hippocampus between LFPb and Sham rats at 2 months post-trauma.
In our experimental conditions, an early and sharp inflammation was elicited in both brain hemispheres in LFPb rats. Bilateral inflammation in cortical, hippocampal and thalamic regions of LFPb rats was associated with behavioral deficits that were only transient in the short-term, without detectable cognitive dysfunction. However, it is possible that the isoflurane anesthesia that we used in our experimental set-up, and that is used by many, may have exerted significant neuroprotective effects in LFPb rats. Thus, our objective is to modify the anesthesia protocol in future experiments.
This work was supported by a DGA Grant (French Ministry of Defense).
TBI, Inflammation, Anxiety, Learning, LTP
ASSESSING CORTEX AND SUBCORTICAL WHITE MATTER PATHOLOGIES FOLLOWING TRAUMATIC BRAIN INJURY: COMPARISON OF DIFFUSION BASIS SPECTRUM IMAGING (DBSI) TO DIFFUSION TENSOR IMAGING (DTI)
Peng Sun, PhD, Washington University School of Medicine
Yong Wang, PhD, Washington University School of Medicine
Sheng-Kwei Song, PhD, Washington University
The widespread use of diffusion tensor imaging (DTI) to assess white matter injury (axonal/myelin) in human and animal traumatic brain injury has revealed the progressive nature of injury in this disease. We have recently proposed a novel diffusion basis spectrum imaging (DBSI) that can quantify neuroinflammation and axonal/myelin injury.
We developed a data-driven diffusion basis spectrum imaging (DBSI) treating the diffusion weighted MR signal as a linear combination of anisotropic diffusion tensors plus a spectrum of isotropic diffusion components to simultaneously assess axonal injury, demyelination, and inflammation. The cortex and corpus callosum (CC) were examined from controlled cortical impact TBI rats. Animals underwent a rapid cortical deformation (250ms) using a 3-mm diameter impactor (depth of 2 mm). Three days after injury, intra-cardiac perfusion fixation was performed on animals followed by ex vivo diffusion MRI examinations. Diffusion MRI data were analyzed using both conventional DTI and the newly developed DBSI models. Diffusion parameters including fractional anisotropy (FA), axial (AD) and radial diffusivity (RD) derived by DTI and DBSI, as well as the cellularity, and signal ratios of extracellular water and white matter tracts derived by DBSI, were compared between ipsilateral and contralateral CC and cortex at the epicenter.
At 3 days after TBI, DTI derived FA decreased by 35% and 42% for the ipsilateral cortex and CC respectively when compared with that of the contralateral tissues. Interestingly, the DTI derived AD in the ipsilateral CC was not affected while RD increased by 35% compared with that of the contralateral CC. The apparent diffusion coefficient (ADC) was not affected in the ipsilateral cortex. Similar to the DTI results, DBSI derived AD was not changed in the ipsilateral CC while the RD increased by 22% when comparing with that of the contralateral CC.
Changes of the DBSI specific parameters were observed in both cortex and CC. Increased cellularity was seen in the ipsilateral cortex and CC 3 days after TBI with a 22% and 55% increase, respectively. The extracellular water content in the ipsilateral cortex increased by 53% while a 127% increase was seen in the CC. The fiber tract ratio of the ipsilateral CC decreased by 13% compared to the contralateral CC.
Taken together, our results suggest that DBSI can provide extensive new information about both gray and white matter pathologies in TBI.
In TBI, the increased cellularity findings from our DBSI results are consistent with our immunohistochemical results of increased microglial invasion at the injury site. Further, the 35% and 22% increase in RD derived by DTI and DBSI, respectively, is strongly suggestive of ongoing myelin injury. However, no changes were observed in AD derived by DTI and DBSI suggesting that myelin injury in this model may precede axonal injury. This finding is at odds with previous studies that have concluded axonal injury precedes myelin degradation. The increased extracellular water ratio in cortex (53%) and CC (127%) strongly suggests the presence of vasogenic edema 3 days after TBI. Although the mechanism is still unclear, altered blood-brain-barrier permeability and inflammatory cell infiltration may play a role.
Thus, DBSI would appear to be a more sensitive measure of white matter injury, while providing additional physiological information on the role of neuroinflammation following TBI.
Authors thank M. Hamer for generating the TBI animals. Funding supports from NIH R01 NS 047592, P01 059560, and NMSS RG 4549A4/1.
Neuroinflammation, CCI, magnetic resonance imaging
FACILITATED ASSESSMENT OF TISSUE LOSS FOLLOWING EXPERIMENTAL TRAUMATIC BRAIN INJURY
Johanna Hedin, MS, Uppsala University
Fredrik Clausen, PhD, Uppsala University
Niklas Marklund, MD, PhD, Department of Neurosurgery, Uppsala University Hospital
All experimental models of traumatic brain injury (TBI) result in a progressive loss of brain tissue. This can be measured to evaluate the injury severity and the potential neuroprotective effect of experimental treatments. This work aims at reducing the time required to make these measurements.
Quantitation of tissue volumes is commonly performed using evenly spaced brain sections stained using routine histochemical methods and then digitally captured. The brain tissue areas are then measured and the corresponding volumes are calculated using the distance between the sections. Measurements of areas are usually performed using a general purpose image analysis software and the results are then transferred to another program for volume calculations. To facilitate the measurement of brain tissue loss, we developed novel algorithms that automatically separate the areas of brain tissue from the surrounding image background and identify the ventricles. We implemented these new algorithms by creating a new computer program (SectionToVolume) which also has functions for image organization, image adjustments and volume calculations.
We analyzed brain sections from mice subjected to severe focal TBI using both SectionToVolume and ImageJ, a commonly used image analysis program. The volume measurements made by the two programs were highly correlated (R2≥0.90) and analysis using SectionToVolume required considerably less time. The inter-rater reliability was also evaluated and found to be high (Intraclass correlation >0.9). If the images from each brain are located in separate folders, images from an entire project can beimported into SectionToVolume in a single step. The area of brain tissue is calculated automatically in each image and if bregma levels are entered, the corresponding volumes are also calculated. The threshold for tissue detection can be adjusted and areas of tissue can be added or removed manually as needed. Apart from the automated measurements, brain regions can also be measured manually if needed.
Given the extensive use of brain tissue loss measurements in experimental TBI research, SectionToVolume will likely be a useful tool for TBI investigators. SectionToVolume is free to use and both the source code and the program is provided on request.
This work was funded by the Swedish Brain Foundation, the Swedish Research Council, and Uppsala University.
TBI Tissue loss Image analysis
S-NITROSYLATION BASED MECHANISMS PROTECT THE NEUROVASCULAR UNIT AND AID IN FUNCTIONAL RECOVERY FOLLOWING TRAUMATIC BRAIN INJURY IN RATS
Tajinder S. Dhammu, PhD, Medical University of South Carolina
Manjeet Kaur, PhD, Medical University of South Carolina
Anandakumar Shunmugavel, MS, Medical University of South Carolina
Inderjit Singh, PhD, Medical University of South Carolina
Avtar K. Singh, MD, Medical University of South Carolina
Traumatic brain injury (TBI) is associated with neurovascular damage caused by endothelial dysfunction and neurodegeneration, leading to impairment of neurobehavioral function. S-nitrosylation mechanisms are shown to protect BBB integrity and aid functional recovery. Therefore, we investigated whether the S-nitrosylating agent, S-nitrosoglutathione (GSNO), ameliorates TBI and aids functional recovery.
TBI was induced by controlled cortical impact (CCI) in adult male rats. GSNO (50 μg/kg body weight) was administered orally at two hours after CCI. The same dose was repeated daily until endpoints. GSNO-treated (GSNO group) animals were compared with injured (TBI group) and sham animals in terms of neurovascular injury, neuroprotection, neurorepair process, and recovery of motor and neurological functions. Expression of GSNO-induced neurorepair mediators was also investigated in brain endothelial cells and astrocytes.
GSNO treatment of TBI at 2 h after injury and every 24 h thereafter for 1–3 days reduced BBB leakage/edema, protected myelin, and decreased axonal damage. The treatment with GSNO for 14 days increased the expression of vascular endothelial growth factor (VEGF), brain-derived neurotrophic factor (BDNF), and its receptor tyrosine kinase B (TrkB). The treatment also improved locomotor and neurological functions. Stimulation of the neurorepair mediators by GSNO in rat brain was supported by GSNO-mediated enhanced mRNA expression of VEGF and BDNF in a brain endothelial cell culture model. The treatment also enhanced the expression of S-nitrosylated proteins. Furthermore, GSNO treatment of astrocytes enhanced the expression of ciliary neurotrophic factor and glial fibrillary acidic protein.
These observations indicate that GSNO is an efficient neurovascular protective agent which also stimulates neurorepair mediators later in the chronic phase of the injury. Mechanistically, it invokes its beneficial effects mainly via S-nitrosylation. GSNO is already in use in human and its exogenous administration has not shown any evident toxicity or side effects. The study shows clinical relevance because it confers neuroprotection in the acute phase of TBI and stimulates the neurorepair process in TBI's chronic phase. Moreover, investigating S-nitrosylation as a mechanism for TBI therapy is a novel approach.
The study was supported by grant from the NIH NS-72511, C06 RR018823 and No C06 RR015455.
TBI, S-nitrosylation, S-nitrosoglutathione, Neuroprotection, Neurorepair
GENE EXPRESSION PATTERNS FOLLOWING UNILATERAL TRAUMATIC BRAIN INJURY REVEALS A LOCAL PRO-INFLAMMATORY AND REMOTE ANTI-INFLAMMATORY RESPONSE
Gregory D. Ford, Ph.D., Morehouse College
Monique C. Surles-Zeigler, Morehouse School of Medicine
Alicia S. Gates, B.S., Morehouse School of Medicine
Michelle C. LaPlaca, Ph.D., Georgia Institute of Technology
Byron D. Ford, Ph.D., Morehouse School of Medicine
Traumatic brain injury (TBI) results in irreversible damage at the epicenter and initiates cellular and molecular processes that lead to secondary neural injury in the surrounding tissue. We used gene microarray analysis to determine which genes, pathways and networks were significantly altered using a rat model of TBI.
Adult rats received a unilateral controlled cortical impact (CCI) and were sacrificed 24h post-injury. The ipsilateral hemi-brain tissue at the site of the injury, the corresponding contralateral hemi-brain tissue, and naïve (control) brain tissue were used for microarray analysis. Ingenuity Pathway Analysis (IPA) software was used to identify pathways and gene networks that were associated with the altered gene expression in brain tissues following TBI.
Several of the top fifteen biological functions associated with TBI in the ipsilateral tissues were related to inflammation. The analysis also indicated that inflammatory genes were altered on the contralateral side but many of the genes were inversely expressed compared to the ipsilateral side. Several of the genes that were altered on the contralateral side were associated with anti-inflammatory processes. We created a network of the inversely expressed common (i.e., same gene changed on both sides of the brain) IR genes and those IR genes included in pathways and networks identified by IPA that changed on only one side and ranked the genes by the number of direct interactions each had in the network, creating a gene hierarchy. Two well characterized signaling pathways, toll-like receptor/NF-kappaB signaling and JAK/STAT signaling, stood out in our gene hierarchy.
These findings could provide a context for evaluating the potential of therapeutic agents currently in development and identifying new therapeutic targets for TBI. This method of analyzing gene expression data is a critical contribution as, in this case, the development of an effective anti-inflammatory treatment would prevent the inflammation-induced delayed neural cell death observed following TBI. Similar genomic analyses could also be carried out for other critical biological functions using the model developed in this study.
This work was partially supported by NIH grants U01 NS057993, C06 RR-07571, G12-RR03034; DoD Contract W81XWH-10-2-0055 and the W.M. Keck Foundation.
brain injury inflammation gene microarray
A NOVEL PROGRAMMABLE FLUID PERCUSSION INJURY DEVICE REVEALS DISTINCTIVE PATHOLOGICAL CHANGES FOLLOWING HIGH-RATE AND STANDARD CONCUSSIVE BRAIN INJURY
Eric J. Neuberger, BS, University of Medicine and Dentistry New Jersey
Fatima S. Elgammal, BS, UMDNJ-NJMS
Vijayalakshmi Santhakumar, PhD, University of Medicine and Dentistry of New Jersey
Bryan J. Pfister, PhD, New Jersey Institute of Technology
Our aim is to design a TBI device capable of controlling the rate and severity of injury independently to examine whether differences in concussive wave parameters alter the neurological outcome. Our uniquely programmable device offers independent control of key variables defining the waveform and can generate high-rate blast-like injuries.
We developed an original FPI apparatus that utilizes a voice coil actuator to generate a precise temporal forcing function. The system was closed loop controlled with a proportional–integral–derivative (PID) motion controller and linear encoder with a 1 micrometer resolution. The voice coil was coupled to a syringe filled with water that delivers the fluid percussion injury. In this study, the device was programmed to alter the rates of pressure rise, keeping the pressure peak consistent. We generated lateral fluid percussion injuries on 25 day-old rats at moderate peak pressure (25 psi) with a standard FPI system waveform (10ms rise-time) or a faster, high-rate waveform (3ms risetime). The effect of the rate of rise in injury pressure on immediate behavioral outcome, and early neuronal degeneration was examined. Electrophysiological studies were performed on hippocampal slices from rats one week after injury to evaluate changes in dentate network excitability.
Immediately following trauma, fast-rate injuries resulted in behavioral seizures and fatalities in significantly fewer rats compared to standard injuries. However, there was no difference in the duration of post-injury apnea between rats injured using fast and standard waveforms. Unlike sham-operated control rats, hippocampal sections from rats injured with fast- and standard waveforms demonstrated neurons labeled with FluoroJade C within 4-6 hours after injury, indicating the presence of degenerating neurons. Stereological counts showed significantly fewer degenerating FluoroJade labeled dentate hilar neurons following fast-rate injuries compared to standard injuries. In field recordings of afferent-evoked granule cell responses obtained in slices from rats one week after trauma, both fast- and standard-injuries resulted in a significant enhancement of population spike amplitude compared to sham-controls. Despite dissimilarities in early neuronal degeneration, the afferent-evoked dentate population response between standard and fast injuries was not different. In recordings performed in SR95531, a GABAA receptor antagonist, the dentate population response evoke by perforant path stimulation was significantly enhanced after standard injury. However, the evoked population spike amplitude in slices from rats injured using fast-rate waveforms was not different form sham controls. These data indicate that the rate of injury may have differential impact on both excitatory and inhibitory networks in the hippocampus.
Pressure waveforms delivered by our original, voice-coil system can match the injury waveforms of commercially available systems and deliver injuries with faster-rising waveforms. The PID control and voice-coil force generation allows user controlled simulation of various primary injury waveforms with our compact and portable device. Our study shows that despite improved immediate physiological outcome, including fewer behavioral seizures, lower mortality and lower neuronal degeneration, in rats subject to fast, blast-like waveforms; the increase in dentate network excitability after fast-rate injury was similar to that following the standard injury. Our data show that a better immediate neurological outcome following fast injury may mask the severity of long-term neuropathology. Our findings suggest the strategies for clinical neurological assessment and management of patients sustaining fast-rate blast-like injuries may have to be different from those employed following slower concussive injuries.
This work has been funded by the New Jersey Commission on Brain Injury Research.
Fluid Percussion Injury, Neuronal degeneration
DESPITE ANATOMICAL CIRCUIT REORGANIZATION AFTER DIFFUSE BRAIN INJURY IN THE RAT, MOLECULAR RESPONSIVENESS TO CIRCUIT ACTIVATION REMAINS INTACT
Theresa Currier Thomas, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Ellen Magee, DPT, PT, University of Kentucky, College of Health Sciences San Juan Regional Medical Center
P. David Adelson, MD, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's Hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Following brain injury, secondary cascades can lead to cellular damage rather than death. Damaged tissue though provides a substrate for circuit reorganization responsible for late-onset gain-of-function sensory sensitivity that develops over 28 days post-injury. Our goal is to determine whether rehabilitation can alter the development of stable, reorganized circuits.
Because the activity-regulated cytoskeleton-associated (ARC) protein is a master regulator of synaptic plasticity and tightly coupled to behavioral encoding of information in neuronal circuits, the molecular response for circuit activation in the somatosensory cortex (S1BF) in adult male rats was quantified for reorganized whisker circuits post trauma day (PTD) 28 after midline fluid percussion injury (FPI) or native circuits after sham injury by measuring ARC protein gene expression. Additional groups of animals were “rehabilitated” by whisker stimulation or deprivation during the first or second week post-injury. At PTD 28 we quantified whisker stimulation-induced ARC gene expression as a function of (1) time between whisker stimulation and tissue biopsy (15 min vs. 24h) and (2) therapeutic stimulation or deprivation of whiskers during the 1st or 2nd week post-injury. We hypothesize that molecular responses of the whisker circuit to stimulation will depend on the type and timing of prior exposure to rehabilitation.
Whisker stimulation elicited comparable ARC gene expression in the S1BF of FPI and sham; no changes were evident 15 min after stimulation and significant elimination of ARC gene expression was evident at 24h after stimulation. Therapeutic stimulation during the 1st week of injury correlated with a 5-fold increase in ARC gene expression compared to control animals. The potential benefit of early rehabilitation was lost when whisker stimulation was given at 2 weeks. In absence of stimulation, there was comparable ARC gene expression in both FPI and sham. Rehabilitative sensory deprivation (whisker shaving) had no effect on ARC gene expression regardless of the week during which it was provided. In the sensory thalamus, initial results indicate reduction in ARC gene expression in brain-injured, but not uninjured animals.
Based on activity driven gene expression of ARC, the reorganized circuits that underlie behavioral morbidity after diffuse brain injury respond comparably to uninjured circuits. On a molecular level, reorganized circuits are functional, despite the adverse behavioral responses they mediate. Stimulation therapy provided during the acute reorganization phase produced long-term alterations in the molecular response of stimulated circuits. Better understanding of rehabilitation induced synaptic plasticity may guide timing and modality of rehabilitative treatment in the clinic.
Support: KSCHIRT #7-11 and NIH R01 NS-065052. This research was conducted at the University of Kentucky.
TBI, rehabilitation, activity-regulated-cytoskeleton-associated-protein(ARC), circuit reorganization
THERAPEUTIC SCREENING OF NEUROPROTECTIVE AGENTS IN COMBINED TRAUMATIC BRAIN INJURY PLUS HEMORRHAGIC SHOCK IN MICE: A PRELIMINARY REPORT
C. Edward Dixon, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Hülya Bayır, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Robert S.B. Clark, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Larry W. Jenkins, PhD, University of Pittsburgh/School of Medicine / Neurosurgery
Vincent A. Vagni, BA, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Mu Xu, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Keri Janesko-Feldman, BS, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Traumatic brain injury (TBI) is commonly accompanied by hemorrhagic shock (HS). Neuroprotective strategies addressing TBI+HS have rarely been explored. We developed a mouse model of TBI+HS are screening drugs. Mechanisms exacerbating secondary injury are mitochondrial failure, neuroinflammation and axonal injury. Minocycline targets mitochondrial failure and neuroinflammation; tacrolimus attenuates axonal injury.
We hypothesized that either minocycline or tacrolimus will improve behavioral and neuropathological outcome after TBI+HS in mice. C57/BL6 male mice (n=50) were anesthetized with isoflurane and underwent controlled cortical impact (CCI) followed by severe HS to MAP of 25–27 mm Hg for 35 min. Mice (n=10/group) were randomly given either 3 mg/kg tacrolimus; 20 mg/kg minocycline; tacrolimus vehicle; or minocycline vehicle followed by resuscitation to goal MAP >70 with Lactated Ringers for 90 min mimicking pre-hospital care. The shed blood was then re-infused and mice were recovered. Ten mice underwent sham surgery without CCI or HS and received no drug or vehicle. A battery of outcome tests were used including Morris water maze (MWM) on d 14–20 and volumetric analysis of tissue loss in the injured hemisphere on d 21.
Surprisingly, mortality was greatest in the tacrolimus treated mice (60%, p=0.044, Chi Square). Final latency to find the hidden platform in the MWM was lowest in minocycline group 20.36±3.44 sec but highest in the tacrolimus group 45.88±7.71 sec (ANOVA p=0.046). There was a trend towards reduced hemispheric volume loss (non-injured minus injured hemisphere) in the minocycline group vs. all other injury groups.
We conclude that TBI+HS represents a unique insult, vs. TBI alone, which requires individualized therapeutic testing. Tacrolimus, at a dose shown to be beneficial in models of TBI, was deleterious in TBI+HS. Given the effects of minocycline, therapies targeting mitochondrial failure and neuroinflammation merit additional study in TBI+HS.
DARPA N660001-10-C2124. The views and opinions and/or findings contained in this abstract are those of the author and should not be interpreted as representing the official views or policies, either expressed or implied of DARPA or the DOD.
neuroprotection, secondary insult, critical care
ELONGATION OF MICROGLIAL MORPHOLOGY WITHIN THE PRIMARY SOMATOSENSORY BARREL FIELDS AFTER EXPERIMENTAL DIFFUSE BRAIN INJURY
Tuoxin Cao, BSc, Spinal Cord & Brain Injury Research Center/University of Kentucky/KY, USA
Jenna M. Ziebell, PhD, Spinal Cord & Brain Injury Research Center, Department of Anatomy & Neurobiology, Department of Physical Medicine & Rehabilitation/University of Kentucky College of Medicine, Lexington, KY
Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Neurotrauma results in microglial activation, whereby the cell bodies typically enlarge and processes retract and thicken. An overlooked morphological response is the elongation of the microglial cell body into a rod-shape, which was first described by Franz Nissl in the late 1800s.
Adult male rats were subjected to a single moderate severity midline fluid percussion injury (1.9atm; 6–10 minutes righting reflex time), and perfused at 1, 2, 7 and 28 days post brain or sham injury. Neuroscience Associates Inc. were contracted for tissue sectioning and staining for the Iba-1 microglia marker. Morphology and alignment were compared between brain-injured and sham operated animals. The morphological characteristics of rod microglia and classical microglia within the cortex were investigated by microscopy (200x magnification). Images were then analyzed by Fast Fourier Transformation (FFT), which represents the pixel intensity of the original photomicrograph as a frequency domain. When adjacent pixels of the original photomicrograph describe a straight line, the FFT plots a straight line though the origin along the orthogonal angle. The summation of pixel intensities in radial coordinates around the origin is used to generate a value of alignment.
Microglia in S1BF of brain-injured rats showed swollen cell bodies and reduced microglial processes indicating classical microglial activation. Additionally, microglia in the S1BF of brain-injured animals showed radical alignment perpendicular to the dural surface. Elongation was evident as early as 1 day post-FPI. When the cell body elongates, the processes begin to polarize to the apical and basal poles of the cell. By 2 days post-injury, adjacent rod microglia couple together to form ‘trains’ that can extend through multiple cortical layers (up to 15 cells long). Trains circumnavigate blood vessels. Furthermore, clusters of activated microglia (typically between 3–6) were routinely found at the terminal poles of trains (referred to as ‘bumping posts’). Rod microglia alignment using FFT analysis was significantly greater at 1, 2, 7, and 28 days post-FPI compared to the sham, where 7 days post-FPI showed significantly greater alignment compared to 1 day post-FPI (p<0.05, One-way ANOVA).
After experimental diffuse brain injury, rod microglia elongate and align perpendicular to the dural surface. Franz Nissl first described these rod microglia, and Del Rio Hortega illustrated rod microglia in close association to nearby neurons. This association has been re-discovered by our laboratory. Maximum alignment at 7 days post-injury suggested that the morphological change into rod microglia is more likely a secondary response, possibly associated with synaptic stripping that precedes circuit reorganization. The presence of a ‘bumping post' appears to terminate the rod microglial ‘train’, preventing further coupling between rod microglia, and therefore limiting the extent of synaptic stripping and circuit damage. Further characterization of rod microglia, and their role in neuronal plasticity post-injury, will expand our current understanding and allow for a directed pharmacological approach towards diffuse brain-injured patients.
We are grateful to Amanda Lisembee for surgical expertise support: NINDS R01 065052.
Rod Microglia, Fast Fourier Transformation
BEHAVIORAL TASKS FOR ASSESSING COGNITIVE FUNCTIONING AFTER INJURY
Dr. Michael R. Hoane, Ph.D., Southern Illinois University-Carbondale
Cognitive impairment is the most frequent cause of disability in humans following brain injury, yet the behavioral tasks used to assess cognition in rodent models of brain injury are underrepresented in the field. Additionally, few of these tasks have been used to assess frontal lesions across degrees of injury severity.
The goal of the present study was to compare four behavioral tasks commonly used in the brain injury field in frontally-injured rats with both mild and moderate-to-severe injuries. Rats were assigned to two of the following behavioral tasks: the retrograde amnesia and working memory paradigms of the Morris water maze (MWM), novel object recognition (NOR), passive avoidance (PA), and the Dig scent discrimination task (Dig). Four days prior to injury, Dig rats were trained to dig in unscented sand and MWM rats were trained on the retrograde amnesia paradigm. Bilateral cortical contusion injuries were then induced (mild frontal, moderate-to-severe frontal, or sham). Following a seven day recovery period, each rat was tested throughout a 17 day testing schedule. Analyses were then performed on each task and across tasks.
The results indicated that, while all four tasks were effective at assessing injury, some of the behavioral tasks were more effective at differentiating between injury severities than others. Specifically, the MWM and PA were effective at differentiating between rats with no injuries, mild injuries, and moderate-to-severe injuries.
More work is needed assessing cognitive functioning in rodent models of brain injury.
Funded by ARRA funds provided by the NINDS, NS045647.
animal modeling, behavioral testing, cognition
A CELL-BASED ASSAY FOR SCREENING COMPOUNDS THAT AFFECT BRAIN ENDOTHELIAL TIGHT JUNCTION PROTEIN STABILITY; IMPLICATIONS FOR BLOOD-BRAIN BARRIER REPAIR IN TBI
Shongshan Fan, PhD, Temple University School of Medicine
Holly Dykstra, BS, Temple University School of Medicine
As the gate keeper of the CNS, the Blood-Brain Barrier (BBB) is an important physiological structure in regulating the microenvironment needed for neuronal communication. It is known that neuroinflammation resulting from TBI alters BBB function. Therefore, identification of molecular targets that restores barrier fidelity is of clinical importance.
We have developed a cell-based assay for evaluating the protein turn-over kinetics of two essential tight junction proteins found at the BBB. Screening of compounds (for example GSK3β inhibitors) with the assay coupled with the CCI-TBI model allows for selection of compounds that may minimize BBB-permeability. The system offers real-time monitoring, single cell evaluation and high-throughput capability. The vector system employs reporter constructs which express bicistronic mRNA under the control of an inducible promoter (a Tet-Off system). The construct contains two expression cassettes from one mRNA transcript. The first expresses human claudin5 or occludin fused with AcGFP. The second expresses the mCherry protein after an IRES transcriptional element. When in the induced state, both mCherry and AcGFP-claudin5 or -occludin are produced at a constant ratio. Addition of doxycycline halts transactivation and allows protein turnover to be monitored as function of time.
We have previous reported on the anti-inflammatory effects of glycogen synthase kinase 3β (GSK3β) inhibitors on brain endothelial cells. Here we show that GSK3β inhibitors also promote barrier tightness by affecting tight junction protein stability. Results using primary human brain microvascular endothelial cells and a novel vector based assay for the evaluation of occludin and claudin-5 protein stability showed a significant increase in the half-life of these proteins when GSK3β was inactivated. Of the inhibitors tested, GSK3β inhibitor SB216763 (5μM) and LiCl (5 mM) showed the best efficacy by preventing occludin and claudin-5 degradation by approximately 32% and 43% respectively. The effect on the tight junction proteins was further validated using transendothelial electrical resistance (TEER). Inhibition of GSK3β produced a gradual and sustained increase in TEER (as high as 22% over baseline). This phenomenon was attributed to the effect on turn-over and not as result of transcriptional regulation since mRNA levels of occludin, claudin-5 were unchanged. These analyses led us to test whether GSK3β inhibitors identified to stabilized tight junction proteins (such as SB216763) would offer protection of the BBB in our experimental CCI-TBI model. The results of BBB permeability by fibrinogen leakage, showed an approximately 70% decrease in barrier permeability at 24hrs when the selected inhibitor was present following moderate CCI-TBI.
We have developed a cellular assay using primary human brain endothelial cells to test the effect of experimental compounds on tight junction protein stability. Vector constructs which are introduced to the endothelial cell, are bicistronic for either claudin-5 or occludin fused to AcGFP that also contains an internal ribosomal entry site for transcription of the reference gene mCherry. The vectors also feature the Tet-off system in which doxycycline halts transactivation and allows the ratio of the fused tight junction protein and mCherry proteins to provide a measure for protein half-life. Using this assay in conjunction with CCI-TBI, we have identified specific GSK3β inhibitors that are effective in aiding barrier integrity via reduction of claudin-5 and occludin protein turn-over.
Temple University development grant (to SHR)
BBB, neuroinflammation, cerebrovascular, permeability, GSK3β
SCALING AXONAL INJURY AND UNCONSCIOUSNESS THRESHOLDS FROM INFANT TO TODDLER TO PRE-ADOLESCENT CHILDREN
Sarah Sullivan, BS, MS, University of Pennsylvania
Colin Smith, BSc, MBChB, MD, FRCPath, University of Edinburgh
Susan Margulies, BSE, MSE, PhD, University of Pennsylvania
Currently, pediatric TBI rotational velocity and acceleration tolerances are scaled from the adult using age-dependent brain mass and modulus values. Recent data from infant, toddler and pre-adolescent piglets can be used to determine if mass and modulus are sufficient to determine load tolerances in the immature brain.
Diffuse TBI was produced via rapid axial head rotation in female pre-adolescent pigs (N=10, 2 months, 19 to 26 kg), toddler piglets (N=18, 4 weeks, 7–10 kg), and infant piglets (N=20, 3–5 days, 2–4 kg). Uninjured shams (N=4–5) were studied in each age group. All subjects were anesthetized with isoflurane (1–4%), and vitals were recorded. Prior to TBI, isoflurane anesthesia was discontinued, until pinch reflex was detected post-TBI, then anesthesia was resumed until sacrifice 6 hours post-TBI. Brains were perfusion-fixed, sectioned coronally, and 6 μm slices stained with β-APP to assess brain volume fraction experiencing axonal injury. Linear regression was used to evaluate the correlation between fractional axonal injury volume and either rotational velocity or acceleration. All protocols were approved by the IACUC of the University of Pennsylvania.
Regardless of age, axonal injury volume increased linearly with velocity and acceleration. Rotational velocity and acceleration of the infant and toddler piglets were scaled by brain mass and modulus to the pre-adolescent. Scaling loads revealed that brain mass and stiffness alone did not account for the enhanced axonal injury volume in the infant. Specifically, to produce the same axonal injury volume in the infant as the older piglets, median scaled velocity in the infant would have to be 1.4x to 1.9x lower than scaling by mass and modulus alone would predict, and median scaled angular acceleration 2.3x to 4.0x lower. Furthermore, return of the pinch reflex after TBI was 7.5x longer than shams in the infant piglets (p=0.0075), but was indistinguishable from age-matched anesthetized shams in the toddler (p=0.1457) and pre-adolescent groups (p=0.2288), indicating more severe neurologic status of the infant group immediately following injury. Thus, both injury outcome metrics indicate an enhanced vulnerability in the infant brain, despite its smaller brain and stiffer modulus. When comparing only toddler and pre-adolescent age groups to each other (wherein brain modulus is similar), scaling velocity and acceleration by brain mass was sufficient to represent the age-dependent differences in axonal injury volume.
Mass and tissue modulus scaling laws are accurate between the toddler and pre-adolescent, but not between the either toddler or pre-adolescent and the infant. We find that the infant is 1.4x to 4.0x more vulnerable to axonal injury assessed at 6 hours, and this range is dependent upon whether acceleration- or velocity–based scaling is used. Also, we also find prolonged unconsciousness immediately following injury in the infant, that is not present in the toddler or pre-adolescent.
This research was supported by the US DOT NHTSA award DTNH22-07-00088, and the NIH award R01 NS039679.
Traumatic Brain Injury, Pediatric, Biomechanics
COMPUTATIONAL ANALYSIS OF INJURED TISSUE FOLLOWING REPETITIVE MILD TRAUMATIC BRAIN INJURY
Anthony Bianchi, B.S., Center for Research in Intelligent Systems, University of California, Riverside
Richard Hartman, Ph.D., Department of Psychology, Loma Linda University
Bir Bhanu, Ph.D., Center for Research in Intelligent Systems, University of California, Riverside
Monica J. Carson, Ph.D., Center for Glial Neuronal Interactions, University of California, Riverside
Andre Obenaus, Ph.D., Department of Pediatrics, Loma Linda University
Mild traumatic brain injury (mTBI) has become an increasing public health concern as subsequent injuries can exacerbate existing neuropathology and result in neurological deficits. Experimental models of repetitive mTBI have explored injuries induced to the same location; however our study focuses on tissue level alterations following two contralateral mTBIs.
Controlled cortical impact (CCI; 0.5 mm deformation for 250 ms) was used to induce mTBI in adult male rats. An initial mTBI was induced to the right cortex of all animals, followed by a second injury delivered to rmTBI animals 3 (rmTBI 3d) or 7 (rmTBI 7d) days later. Animals underwent T2 weighted magnetic resonance imaging (MRI) 1, 4, 8, 14, 21, and 60 days after the initial injury. MR images were analyzed using a novel automated computational method that analyzed lesion composition (i.e. blood and edema) and volume. Animals were then perfused and histology for blood (Prussian Blue) and immunohistochemistry for neuroinflammation (IBA1: microglia; GFAP: astrocytes) were undertaken. Computational MR data were then compared to histological findings.
Lesion volume was dramatically increased in rmTBI 7d animals compared to those that received a single or rmTBI 3d apart. Analysis of the first and second mTBI lesion volume revealed an increase in lesion volume at the site of the second injury within rmTBI 7d animals, in contrast to the rmTBI 3d group which exhibited a similar lesion volume to that of single mTBI controls. However, lesion volume increases in the rmTBI 7d animals was transient, as lesion volumes were similar to that of single animals by 14d post first injury. We then computationally evaluated the composition of the lesion for pixels containing blood, edema and normal tissue values based on quantitative T2 mapping. Comparison of lesion composition at 1d post last injury revealed increased edema within rmTBI 3d animals, while increased blood volume was observed in the rmTBI 7d group at the site of the second injury. By 14d post first injury, the rmTBI 3d animals exhibited similar lesion composition to those in the single group, while the rmTBI 7d animals continued to demonstrate increased blood deposition. Prussian Blue staining revealed increased extravascular blood at the injury site in rmTBI 7d animals compared to rmTBI 3d animals, consistent with our computational analysis. Immunolabeling for neuroinflammation revealed increased microglial activation and astrocyte reactivity within the rmTBI groups at the injury sites, primarily at the site of the second mTBI.
Repetitive mTBI is thought to result in cumulative injury leading to neuropathological and neurological abnormalities. We report that there appears to be a window of tissue vulnerability where a second mTBI 7d (but not 3d apart) after an initial injury results in increased tissue abnormalities (blood). The presence of an activated inflammatory response following rmTBI may play a role in defining this time window. Our findings are consistent with the concept of hemorrhagic transformation that has been reported in the clinical literature. Future work elucidating the underlying mechanisms responsible for the exacerbated injury and defining the temporal window in which injury can be worsened by subsequent injuries are needed. More importantly, use of automated computational analysis of MR images demonstrates the ability to extract tissue relevant quantitative information that may improve not only diagnosis but also allow for assessment of future pharmacological treatments.
Funding provided by DCMRP # DR080470 (AO) and NSF IGERT: Video Bioinformatics Grant DGE 0903667 (VMD).
edema, blood, T2
EFFECTS OF GLUCOSE ON OXIDATIVE STRESS AFTER CORTICAL CONTUSION INJURY IN RATS
Sima Ghavim, BS, UCLA
Nobuhiro Moro, MD, PhD, Nihon University
David A. Hovda, PhD, UCLA
Neil G. Harris, PhD, UCLA
Richard L. Sutton, PhD, UCLA
High glucose (Glc) levels after traumatic brain injury (TBI) are often thought to be detrimental, but we have previously found acute Glc treatments to be neuroprotective 24 h post-TBI. The current study was conducted to assess effects of acute Glc treatments on protein markers of oxidative stress.
Adult male Sprague-Dawley rats received Sham injury (n=12) or left cortical contusion injury (CCI; n=12) and injection of 50% Glc (2g/kg) or no treatment (NoTx) at 0, 1, 3 and 6 h post-surgery. At 24 h post-surgery rats were euthanized and brain tissue from left contused cortex, left peri-contusion cortex and left hippocampus was collected and stored at −80°C until homogenized. Total protein in samples was determined using RC-DC kits. Proteins analyzed using Western blots included glyceraldehyde-3-phosphate dehydrogenase (GAPDH), nitrotyrosine (NT) as an index of production of reactive nitrogen species (nitration of tyrosine residues), and 4-hydroxynonenal (4-HNE) as an index of polyunsaturated fatty acids exposure to peroxides and reactive oxygen species. Protein bands on enhanced chemiluminescent gel images (Fluoromax system, BioRad) were quantified by integrated optical density measurement using Quantity 1 software.
GAPDH, NT and 4-HNE protein levels did not differ significantly between Sham-NoTx or Sham-Glc groups, so the levels in each brain region for each CCI group were expressed as a percent of the similarly treated Sham group. In all regions ipsilateral to injury, there were no effects of CCI or Glc treatments on the levels of GAPDH. The NT protein levels were significantly increased in the left contused cortex (p=0.001), peri-contusion cortex (p=0.001) and hippocampus (p=0.011) of the CCI-NoTx group compared to Sham-NoTx controls. In the CCI-Glc group the NT levels in left contused cortex and peri-contusion cortex appeared reduced compared to CCI-NoTx, but these NT levels were significantly increased in the contused cortex (p=0.010) and peri-contusion cortex (p=0.041) as well as in the left hippocampus (p=0.025) compared to Sham-Glc controls. The left peri-contusion cortex was the only region where reductions in NT approached significance in the CCI-Glc group compared to CCI-NoTx controls (p=0.052). 4-HNE protein levels in the left contused cortex (p<0.001), peri-contusion cortex (p=0.001) and hippocampus (p=0.002) were significantly increased in the CCI-NoTx group compared to Sham-NoTx. The 4-HNE levels in the CCI-Glc group were significantly increased in the left contused cortex (p=0.004), peri-contusion cortex (p=0.008) and hippocampus (p=0.024) compared to Sham-Glc controls. While these protein levels were slightly lower in all ipsilateral brain regions of the CCI-Glc compared to the CCI-NoTx group, levels of 4-HNE in the CCI groups did not differ significantly.
Our current results indicate that multiple Glc treatments within 6 h of unilateral CCI in rats can mildly (but not significantly) attenuate measures of oxidative stress (NT, 4-HNE protein levels) at 24 h post-CCI. Importantly, Glc treatments did not increase levels of NT or 4-HNE in any brain region studied at this time point, as would be expected if increased levels of Glc were detrimental after TBI. In light of our other recent studies in this model of CCI showing that provision of exogenous Glc acutely after injury can significantly reduce neuropathology, enhance cerebral glucose utilization, and does not adversely affect behavioral recovery, we conclude that moderate hyperglycemia after TBI may not be detrimental.
This work was supported by grant NINDS P01NS058489 and the UCLA Brain Injury Research Center.
Contusion; 4-hydroxynonenal; Glucose; Nitrotyrosine; Western
THE USE OF IMAGING MASS-SPECTROMETRY TO IDENTIFY POTENTIAL BIOMARKERS FOR TBI AND TO STUDY TIME-DEPENDENT LIPID CHANGES IN BRAIN TRAUMA FOLLOWING CONTROLLED CORTICAL IMPACT INJURY IN RATS
Jeremy Post, B.S., National Instutute on Drug Abuse
Kathrine Baldwin, B.S., National Institute on Drug Abuse-IRP
Gregory P. Bull, B.S., CNRM
Shawn Gouty, M.S., Department of Pharmacology, Uniformed Services University
Amina S.. Woods, Ph.D., National Institute on Drug Abuse
The effect of traumatic brain injury on brain lipids has received little study.We have used imaging mass spectrometry (IMS) to characterize changes in the expression of selected lipid species in sections from the brains of rats subjected to CCI injury.
Sprague Dawley rats were subjected to one CCI strike to the left cerebral hemisphere after craniotomy. The brains of the rats were removed after euthanasia at one of three time points; 24 hours, 3 days, or 7 days post injury. IMS analysis was performed on coronal brain sections to characterize lipid biomarkers indicative of injury location and severity, using 2′,6′-dihydroxy-acetophenone as a liquid ionization matrix. The brains of other rats subject to the same CCI injury were visualized by T2-weighted MRI before and after injury. The IMS technique permits the differentiation of members of each lipid species [e.g., sphingomyelins (SM), phosphatidylcholines (PC), phosphatidyl-ethanoloamines (PE)] by size of acyl chain substituents. A MALDI-Orbitrap mass spectrometer (Thermo-Fisher) scanning at 25 μm steps (X and Y) was used for image acquisition.
Accurate mass measurement coupled with precise localization shows the incredible diversity and distribution of lipid species in brain. The generated images show anatomical variations in the distribution of selected lipids, demonstrating time-related changes in the lipid profiles of the injured areas after CCI injury. IMS results implicate several lipid species in the dynamics of membrane disruption and signaling following injury; variations in the relative abundances and location of numerous lipid species are readily observed. Brain areas showing changes in lipid expression correlated with brain areas showing increased levels of water after MRI imaging. Reductions in some SM, PC and PE species were observed following CCI, while increases were noted for other species of PCs and SMs. Other lipid species including certain sulfatides and phosphatidylinositols showed no variation following injury, based on comparisons of lipids in the injured cerebral hemisphere relative to both the contralateral (non-injured) hemisphere from the same brain, and to control brains.
IMS analysis is revealing spatially distinct time-dependent changes in the relative abundance of selected lipid species in brain following an experimental TBI. These studies may shed light on the mechanisms of injury and repair. Correlations of lipid changes in brain sections with changes in the levels of lipids in peripheral blood are underway. These studies may lead to the identification of lipid biomarkers in blood suitable for evaluation of the severity of traumatic brain injuries and monitoring of recovery after treatment.
Supported by a grant from the CNRM. We thank Drs. Reed Selwyn and Haiying Tang of the CNRM Translational Imaging Facility for MRI imaging.
CCI injury, imaging MS (IMS), brain lipids
ACTIVATION OF SONIC HEDGEHOG RESPONSIVE ADULT NEURAL STEM CELLS AFTER MILD TRAUMATIC BRAIN INJURY IN MICE
Genevieve Sullivan, BS, USUHS
Sohyun Ahn, PhD, NICHD (National Institute of Child Health and Human Development)
Regina C. Armstrong, PhD, USUHS
The subventricular zone (SVZ) is the largest germinal site in the adult mammalian brain and Shh maintains this neural stem cell niche. Shh activation of endogenous repair from neural stem cells following TBI was studied in Gli1CreERT2;R26-YFP mice in which Shh responding cells are heritably labeled by YFP reporter expression.
Two distinct models of mild TBI were produced using an Impact OneTM Stereotaxic Impactor device on 8 week old male Gli1CreERT2;R26-YFP mice. This reporter line heritably labels cells that are responding to Shh signaling at the time of tamoxifen administration. Controlled cortical impact (CCI) involved a craniotomy and direct impact to the dura to produce damage primarily to sensorimotor cortex with extension into corpus callosum. Concussive brain injury (CBI) used direct impact onto the skull resulting in acceleration-deceleration injury with axon damage primarily in the corpus callosum without overt cortical injury. Sham and naïve animals were included as controls and all animals received tamoxifen on 2 and 3 days post-injury. Mice were sacrificed at 3 days, 2 weeks, or 6 weeks post-injury.YFP and cell type markers were detected by immunohistochemistry and quantified using unbiased stereology of cells in the SVZ or manual counts with binning to assess cell localization.
Shh expression, evaluated by immunohistochemistry, is increased at 3d post-injury in astrocytes of the SVZ and corpus callosum in comparison with naïve mice. Tamoxifen administration at 2 and 3 days post-injury resulted in YFP expression among neural stem cells in non-injured mice and following both injury models. This heritable YFP labeling of Shh responsive cells at 2–3 days post-injury identified differences associated with SVZ cells. After CCI, YFP+ cells are increased in the SVZ ipsilaterally at 2wks post-injury and bilaterally at 6wks post-injury, compared to sham and naïve animals. After CBI, YFP cells in the SVZ increase bilaterally between 2 to 6 wks post-injury. In both models, at 2wks and 6wks post-injury, there is an increase in YFP+ cells extending laterally from the SVZ inferior to the corpus callosum. Many of the YFP+ cells colocalize with either a marker of transit amplifying cells (EGFR) or astrocytes and neural stem cells (GFAP).
After CCI and CBI, Shh colocalized with GFAP+ astrocytes in the SVZ and corpus callosum indicating that astrogliosis may account for the increased Shh expression seen at 3 days post-injury. Gli1CreERT2;R26-YFP mice showed heritable labeling of neural stem cells with tamoxifen administration at 2 and 3 days post-injury when Shh expression was robust. Shh-responsive cells increased in the SVZ and also extended from the dorsal lateral SVZ under the corpus callosum. YFP in cells associated with the SVZ were double labeled with GFAP suggesting that they are neural stems cells or with EGFR suggesting they are in an active state as necessary for migration and repair. Thus, the upregulation of Shh and Shh signaling in the SVZ following TBI may indicate a potential role for Shh in neural stem cells neuroregenerative responses that can be prolonged even after a mild injury in the adult CNS.
This work was funded by the U.S. Department of Defense in the Center for Neuroscience and Regenerative Medicine.
traumatic brain injury sonic hedgehog demyelination regeneration stem cells
BLAST BRAIN INJURY ELEVATES CATECHOLAMINE BIOSYNTHESIS IN THE NUCLEUS TRACTUS SOLITARIES AND OXIDATIVE STRESS IN THE HYPOTHALAMUS IN RATS
Dr. Stanislav Svetlov, Ph.D., M.D., Banyan Biomarkers, Inc.
Nataliya Kirichenko, M.S., University of Florida/ VA Med Center
Melissa Whidden, Ph.D, West Chester University
Benedek Erdos, M.D., Ph.D., Physiology, University of Florida
Dr. Victor Prima, Ph.D., Banyan Biomarkers, Inc.
Alexandra Sherman, DVM, Banyan Biomarkers, Inc
Dr. Firas H. Kobeissy, Ph.D. in Neuroscience, Banyan Biomarkers
Robert Yezierski, Ph.D., Orthodontics, University of Florida
Charles Vierck, Ph.D., Neuroscience, University of Florida
Dr. Kevin Wang, Ph.D., Department of Psychiatry, University of Florida
Traumatic Brain Injury (TBI) produces major health problems impacting the lives of both military and civilian personnel. TBI disrupts autonomic function but the nature of this disruption is unknown. Following blast brain injury, we assessed selective biochemical markers for autonomic function in adult male Sprague Dawley rats.
Rats were subjected to head-directed overpressure blast injury (OBI) of 358 kPa magnitude at the target. At the same time for sham controls, rats were anesthetized as the previous group but instead of OBI were exposed just to noise being placed at ∼2 m distance from the shock tube nozzle. Sympathetic nervous system activation of nucleus tractus solitaries and in the hypothalamus was evaluated at 6 hours following blast injury by assessing the expression of catecholamine biosynthesizing enzyme, tyrosine hydroxylase (TH) in the nucleus tractus solitaries and NADPH oxidase activity, a marker of oxidative stress,in the hypothalamus.
Following OBI there was a significant elevation in TH protein expression by 49% compared with control (P<0.05). In addition, NADPH oxidase activity was significantly increased by 36% following OBI (P<0.05).
Collectively, the increased catecholamine biosynthesis in nucleus tractus solitaries and oxidative stress in the hypotalamus suggest that OBI results in increased sympathoexcitation in the rat brain. Such effects may be one important factor contributing to autonomic dysfunction following OBI.
Supported by Department of Veteran Affairs; Rehabilitation R&D, GRECC, Medical Research Services, Banyan Biomarkers Inc, University of Florida Brain Institute, NIA, and AHA.
blast injury, biomarkers, autonomic dysfunction
EXPRESSION OF HYPERPHOSHORYLATED TAU PROTEIN FOLLOWING CONTROLLED CORTICAL IMPACT IN IMMATURE (POST NATAL DAY 17) RATS
Roxanne Perez, B.S., St. Joseph's Hospital and Medical Center
Lucy Treiman, PhD., PsyD., St. Joseph's Hospital and Medical Center
Dr. P. David Adelson, M.D., Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ
Traumatic Brain Injury (TBI) remains the leading cause of death and disability in children. As well, upwards of 30% of patients with Alzheimer's Disease (AD) have a history of a single incident TBI. Amyloid beta (Aβ) plaques and neurofibrillary tangles of tau protein (TP) remain as the hallmark postmortem findings.
To study whether precursors to AD may be precipitated by TBI sustained earlier in life, immature post natal day (PND) 17 rats were injured using CCI (4 mm tip, 3.5 m/sec, deflection=2.0 mm, respectively) and compared to sham (craniotomy only, n=15). We evaluated the brains of 15 rats following experimental TBI using controlled cortical impact (CCI) and then sacrificed at post injury day (PID) 1,3, or 7 using perfusion fixation. Paraffin embedded coronal sections (7um) were histologically (H&E) and immunohistochemically (Tau-5 anti tau antibody, AT8 anti-hyperphosphorylated tau, and 6E10 anti Aβ antibody) stained and analyzed using a light/confocal microscope.
There was no evidence of hyperphosphorylated TP in either the dorsal, frontoparietal cortex surrounding the impact zone or the hippocampi through PTD 7. However, 93% of the cortically impacted animals showed expression of hyperphosphorylated TP along the ipsilateral ventral cortex (opposite injury site), and 87% along the contralateral ventral cortex. Hyperphosphorylated TP staining was present mainly in the neuronal cytoplasm and to a lesser extent within the axons. Hyperphosphorylated TP staining was not present in any of the sham animals at any of the time points. Additionally, there was no evidence of Aβ staining at any of the time points in either the CCI or the sham group.
While there were no changes in the direct peri-injury region, our data indicates that in the developing (PND 17) rat brain, experimental TBI using CCI can induce expression of hyperphosphorylated tau as early as PTD 1, and persist to PTD 7 in the ventral cortex both ipsilateral and contralateral to the impact. CCI though did not induce Aβ expression in the acute setting up to one week post injury. Further, long term studies are needed to assess whether TBI sustained early in life can induce the pathological process seen in AD.
Supported in part by the American Association of Neurological Surgeons Medical Student Research Fellowship.
TBI, Tau, Amyloid, CCI, Rat
DETECTING THE ONSET OF INDUCED SEIZURES AFTER LATERAL FLUID-PERCUSSION INJURY IN RAT USING FUNCTIONAL MRI
Antti Airaksinen, MSc, University of Eastern Finland
Joanna Huttunen, MSc, University of Eastern Finland
Artem Shatillo, MSc, University of Eastern Finland
Juha-Pekka Niskanen, MSc, University of Eastern Finland
Olli Grohn, PhD, University of Eastern Finland
Epilepsy is a major co-morbidity after TBI, but mechanisms of post-traumatic epileptogenesis are unknown. Studies aiming at understanding epileptogenic mechanisms would benefit of a method that could identify the most epileptogenic region in injured brain. We hypothesized that fMRI can detect the area with increased seizure susceptibility after TBI.
Lateral fluid-percussion injury was induced for 14 Spraque-Dawley rats (7 controls). Animals were imaged with fMRI at 2 months post-TBI when animals were undergoing epileptogenesis. For simultaneous local field potential (LFP) and fMRI measurements, a tungsten wire was inserted in the frontal cortex. MRI was done using a 9.4 T horizontal scanner interfaced with a Varian DirectDrive console. fMRI data were acquired using a single shot SE-EPI sequence (TR 4 s, TE 40 ms, slice thickness 1.5 mm, 15 slices, image matrix of 64×64, and FOV of 2.5×2.5 cm). Simultaneous LFP and fMRI measurements were performed consisting of 1000 images of baseline. PTZ was injected (30 mg/kg, i.p.) and image acquisition was continued for 1000 images. Anatomical images were acquired with fast spin echo sequence. The fMRI data were analyzed using SPM8 along with in-house made Matlab code.
PTZ caused detectable responses in BOLD and LFP signals in all rats. In the sham group, positive BOLD response had a cortical onset that appeared simultaneously in both hemispheres. After initiation, it spread rostrocaudally along the cortex, and subsequently, triggered a positive response bilaterally in the thalamus (6/6) and a negative response in the contralateral hippocampus (5/6). In 46 (6/13) of rats with TBI, BOLD activation started bilaterally and spread similar to that in the sham group. However, in 54 of injure animals, the BOLD activation started unilaterally in the perilesional cortex, and then, spread to the contralateral cortex within 20–100 s. Thereafter, contralateral thalamus was activated in 6/7 animals and the ipsilateral thalamus in 2/7 rats. Positive contralateral hippocampal activation was detected in 2/7 rats.
Decreased threshold for PTZ induced seizures has been reported after lateral FPI induced TBI in rat. In the present study, we demonstrate that PTZ induced seizures originate in the perilesional cortex in 54 of animals at 2 months post-TBI. This suggests that perilesional area is undergoing epileptogenic circuitry reorganization at that time point. In line with our previous studies showing inter-animal variability in the rate of epileptogenesis, enhanced perilesional BOLD response was seen in a subpopulation of animals. Our results show that a combination of fMRI and pharmacological stimulation is a potential tool to study post-injury hyperexcitability.
Academy of Finland, The Sigrid Juselius Foundation
Epilepsy, Epileptogenesis, fMRI
BLAST-INDUCED BRAIN INJURY: INFLUENCE OF SHOCKWAVE COMPONENTS ON INJURY TO THE RAT BRAIN
Richard Hisel, GLR Enterprises, LLC.
Dr. Sarbani Ghoshal, Ph.D., Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Ms. Julie Corkins, University of Kentucky
Mr. Joshua Hoffman, University of Kentucky
Dr. Richard Kryscio, Ph.D., University of Kentucky
Dr. Braden Lusk, Ph.D., P.E., University of Kentucky
Dr. James Geddes, Ph.D., Department of Anatomy and Neurobiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Blast-induced traumatic brain injury (bTBI) has been referred to as the defining injury of Operations Enduring and Iraqi Freedom (OEF/OIF). Additionally, civilians are increasingly at risk from terrorist attacks and industrial accidents. Little is known about which components of the blast shock wave contribute to injury.
Rats were exposed to shockwaves from chemical explosives and compressed air-driven membrane rupture using a shock tube. Head movement was recorded by high speed camera and analyzed for angle and velocity of rotation. Rat brains were analyzed via immunohistochemistry and Western blot for evidence of injury by probing for blood-brain barrier breakdown, reactive astrocytosis, and inflammation. Brain injury was compared between shockwave production mode (chemical explosive vs. air) and with respect to peak pressure, impulse, and head rotation experienced by each animal during injury.
Compressed air-driven membrane rupture produced shockwaves that differed substantially from those produced by chemical explosives (oxyhydrogen) with respect to the pressure-time profiles of the shockwave in that compressed air-driven shockwaves exhibited longer positive phase durations and greater impulses than chemical explosive-driven shockwaves at similar peak pressure, as previously reported (Reneer et al., J Neurotrauma, 2011). The maximal angle of head rotation along with the average velocity of head rotation did not differ between shockwave sources. Gross pathology including hematoma size and the extent of petechial hemorrhage was found to be present in rats exposed to shockwaves from either source, however the size of hematoma and the extent of petechial hemorrhage was greater after compressed air-driven shockwave exposure. Immunostaining of brain sections from injured animals revealed IgG extravasation indicative of blood-brain barrier compromise that was more extensive in compressed air-driven shockwave-exposed rats compared to those exposed to chemical blast. More extensive astrocytosis, localized to the same area as the increased IgG immunoreactivty, was observed in animals exposed to compressed air-driven shockwaves as compared to chemical explosive -exposed rats. Additionally, microglia in the hippocampus of rats exposed to compressed air-driven blasts exhibited increased Iba1 expression and morphology consistent with a reactive phenotype when compared to chemical explosive blast-exposed animals or sham animals. Western blot analysis of locations within the brain showing differential response to shockwave source by immunohistochemical studies failed to reveal quantifiable differences between shockwave source in IgG, GFAP, or Iba1 protein levels.
These data suggest that compressed air-driven membrane rupture produces fundamentally different shockwaves than those produced by chemical explosives and that these differences in shockwave components may contribute to subtle differences in pathology. Additionally, shockwaves alone can result in significant head rotation, which does not appear to be a factor in the differences between the injuries due to different shockwave sources demonstrated here, thus providing a foundation for the use of the MBD in bTBI research.
DVR supported by NIH F31 NS074678, NIH T32 DA022738. Additional supported provided by NIH P30 NS051220 and GLR Enterprises, LLC.
Blast-Injury, mTBI, Military, Model, Shockwave
IMPORTANCE OF THE TRAUMATIC PENUMBRA IN THE NEUROPROTECTIVE EFFECTS OF THE LIPID PEROXYL SCAVENGER U-83836E IN THE MOUSE CONTROLLED CORTICAL IMPACT TBI MODEL
Mona Bains, Ph.D., University of Kentucky College of Medicine
John Cebak, B.S., University of Kentucky College of Medicine
Ayman Mustafa, Ph.D., University of Kentucky College of Medicine
Amy Wang, M.D., University of Kentucky College of Medicine
Lesley Gilmer, Ph.D., University of Kentucky College of Medicine
Indrapal N. Singh, Ph.D., University of Kentucky College of Medicine
The lipid peroxidation inhibitor U-83836E has been shown to improve mitochondrial respiration, Ca2+ buffering and decrease calpain-mediated cytoskeletal damage following controlled cortical impact traumatic brain injury (CCI-TBI) in mice. Presently, we examined U-83836E's effects on motor function, cortical contusion volume and neurodegeneration in injured brain.
Groups of male mice (N=8/group) were anesthetized with isoflurane and subjected to severe (1.0 mm) CCI-TBI followed by either vehicle (0.0% saline) or U-83836E (30 mg/kg i.v. at 15 min. followed by 30 mg/kg i.p.maintenance dosing at 1, 3, 6, 12 and 24 hr). At 24 and 48 hrs, we assessed motor functional recovery using the Composite Neuroscore which tests both forelimb and hindlimb function. Following the 48hr functional testing, the mice were perfused and the brain sections stained with the de Olmos aminocupric silver method which selectively identifies degenerating axons. Cortical contusion volume and the density of neurodegeneration in the ipsilateral hemisphere surrounding the contusion site were analyzed separately in a blinded fashion as to whether the mice had received vehicle or U-83836E treatment.
Administration of a repeated 24 hr dosing paradigm of 30 mg/kg U-83836E produced a small, but not significant improvement in the overall Neuroscore or measures of forelimb function. However a significant improvement in hindlimb function was seen at both 24 and 48 hrs post-injury. Histological analysis failed to show a decrease in cortical contusion volume in U-83836E treated mice. However, the drug did reduce the area of hemispheric neurodegeneration at 48 hr post-injury by approximately 37 to 50% depending upon the section through the contusion site.
The protective effects of U-83836E in the CCI-TBI model shown in our past and presently reported studies continue to demonstrate the importance of post-TBI free radical induced lipid peroxidation in acute secondary injury. Although these effects do not seem to translate into a reduction in the traditionally measured cortical contusion volume, the drug does appear to limit neurodegeneration in the surrounding hemispheric tissue (i.e. traumatic penumbra). The data collectively indicate that the neuroprotective effect of U-83836E during the first 48 hrs after injury is secondary to mitochondrial and cytoskeletal antioxidant protective actions in the brain tissue surrounding the injury epicenter, and that this is adequate to produce an improvement in the recovery of some motor function within the first 48 hrs.
Funding for these studies was provided by P01 NS058484, P30 NS051220 and the Kentucky Spinal Cord and Head Injury Research Trust.
U-83836E lipid peroxidation mitochondria calpain cytoskeleton
ASSESSMENT OF NICOTINAMIDE ON OUTCOME AFTER FLUID PERCUSSION BRAIN INJURY: AN OPERATION BRAIN TRAUMA THERAPY STUDY
Dr. Helen Bramlett, PhD, University of Miami Miller School of Medicine
Ofelia Furones-Alonso, BS, University of Miami Miller School of Medicine
Juliana Sanchez-Molano, MD, University of Miami Miller School of Medicine
David Sequeira, BS, University of Miami Miller School of Medicine
William Moreno, MD, University of Miami Miller School of Medicine
Operation Brain Trauma Therapy (OBTT) is a consortium designed to test therapies in different models of traumatic brain injury (TBI). The University of Miami site utilizes the fluid percussion model of TBI. This model produces focal and diffuse brain trauma resulting in clinically relevant histopathological damage and behavioral deficits.
Following FP brain injury, rats were administered nicotinamide at 50 mg/kg (TBI-50 mg/kg, n=10) or 500 mg/kg (TBI-500 mg/kg, n=10) or vehicle (TBI-Veh, n=10) at 15 min and 24 hrs post-injury. Sham rats (Sham, n=10) underwent all procedures except for drug administration. Rats were tested on day 7 post-injury for sensorimotor function using the rotarod, gridwalk and cylinder task. On days 13–21, rats were assessed for cognitive function utilizing the simple place task (hidden platform), probe trial and working memory task. Following behavioral testing, rats were perfused and brain tissue was processed for histological assessment.
One-way ANOVA was not significant for any of the sensorimotor tasks. However, for the hidden platform task, two-way repeated measures ANOVA for latency was significant for days (p<0.001) but was not significant for group (p=0.089) or group × days (p=0.064). Sham rats showed decreased latencies over the four day testing period. All three TBI groups had higher latencies than sham with the two treated TBI groups exhibiting in general more cognitive deficits on this task. There was no significant difference between groups for the probe trial. However, a two-way repeated measures ANOVA for latency in the working memory task was significant for group × trial (p<0.005). Sham rats were significantly (p<0.05) different from TBI vehicle treated and TBI treated with 50 mg/kg. While sham rats as expected showed the greatest improvement in the delay match-to-place task, Nicotinamide treatment at 500 mg/kg did demonstrate improvement on this task as there was no difference between this group and sham rats. Histopathological analysis for lesion volume was not significant between groups. An assessment of volumetric cortical loss was significant for group (p<0.05). However, there was no effect of nicotinamide in salvaging tissue following TBI.
Although there was a modest improvement in working memory with nicotinamide treatment, no other improvements were observed using this model of TBI. Future studies will compare the efficacy of other drugs to this experimental treatment.
Supported by W81XWH-10-1-0623.
TBI, nicotinamide, behavior, histopathology
A SELECTIVE AND POTENT CDK INHIBITOR PROVIDES NEUROPROTECTION AFTER FLUID PERCUSSION-INDUCED TRAUMATIC BRAIN INJURY
Dr. Bogdan A. Stoica, MD, University of Maryland, School of Medicine/Center for Shock, Trauma and Anesthesiology
Dr. David J. Loane, PhD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Dr. Alan I. Faden, MD, University of Maryland, School of Medicine/ Center for Shock, Trauma and Anesthesiology
Cell cycle activation (CCA) is key secondary injury mechanism induced by traumatic brain injury (TBI) and leads to neuronal death and neurobehavioral dysfuction. In the present study, we evaluated the neuroprotective potential of a selective and highly potent cyclin-dependent kinase (CDK) inhibitor, CR8 following lateral fluid percussion (LFP)-induced TBI.
Male Sprague Dawley rats (300–325 grams) were subjected to moderate LFP injury (2.1–2.4 atms), followed by a single systemic administration of CR8 (5 mg/Kg, i.p.) or vehicle at 3 hours. TBI-induced neurobehavioral dysfunction was evaluated using various behavioral tests: sensorimotor function (composite neurological scoring and forelimb placing), cognition (Morris water maze, novel object recognition, passive avoidance), exploratory and anxiety-based behavior (light-dark open field), olfactory function (buried pellet), and depression (Porsault's forced swim) at various time points. Lesion volume was quantified based on the Cavalieri method of unbiased stereology using the Stereologer 2000 program software. Neurodegeneration in various brain regions was estimated using the optical fractionator method of unbiased stereology.
LFP caused significant impairment of sensorimotor function in terms of lateral pulsion, contraflexion and inclined plane scores, and forelimb placing responses. In addition, TBI resulted in a significant decline in spatial (acquisition trials of Morris water maze), reference (probe trial of Morris water maze), retention or intact (novel object recognition), and fear-based retention (passive avoidance) memories. LFP-induced impairment of exploratory and anxiety-based behavior (light/dark open field), olfactory function (buried pellet), and depression were also observed. A single systemic administration of CR8 at 3 hours post-injury significantly attenuated TBI-induced sensorimotor, cognitive and other functional deficits. In addition to the improvement of neurobehavioral recovery, a single delayed administration of CR8 ameliorated histological outcomes.
These findings provide further evidence supporting the superior neuroprotective potential of cell cycle inhibitors following experimental TBI. We recently reported that delayed administration of a selective and highly potent second-generation CDK inhibitor, CR8 improved neuronal survival in the hippocampus, cortex and thalamus, decreased lesion volume and attenuated sensorimotor and cognitive deficits in a mouse model of controlled cortical impact-induced TBI. Given the increased efficacy and potency of CR8 in multiple models of TBI as compared to earlier purine CDK inhibitors, this drug should be a strong candidate for future clinical trials.
This work was supported by a grant from the National Institutes of Health, R01 NS052568 to Dr. Faden.
TBI, CR8, cell cycle, behavior
HISTOPATHOLOGICAL OUTCOMES FOLLOWING A POLYTRAUMA INSULT MODEL
Juliana Sanchez-Molano, MD, University of Miami Miller School of Medicine
Emily Packard, BS, University of Miami Miller School of Medicine
David Sequeira, BS, University of Miami Miller School of Medicine
Traumatic brain injury is usually accompanied by other insults to the body besides localized brain injury. Examples of these multiple insults can be respiratory distress and multi-organ trauma. Our laboratory has modeled ‘polytrauma’ by combining fluid percussion (FP) brain injury with secondary hypoxia and a systemic injection of IL-1beta.
Rats received parasagittal FP injury followed by 30 minutes of either normoxic (TBI-NO+IL-1β, n=8) or hypoxic (TBI-HY+IL-1β, n=8) gas levels followed by an IP injection of IL1-β or vehicle. Sham operated animals underwent all surgical procedures plus normoxia and vehicle injection (Sham-NO+Veh, n=6). Animals were sacrificed at 3 days after TBI and tissue sections processed for assessment of contusion volume, neuronal damage and white matter pathology.
Cortical contusion volume was significant for group (p<0.05). Posthoc analysis was significant (p<0.05) for both TBI-hypoxic groups compared to the TBI-1β treated animals. Neuronal cell counts within the ipsilateral cerebral cortex and dentate hilar region were significant (p<0.05) for group. Posthoc analysis for cortical cell counts was significant (p<0.05) for TBI-NO+IL-1b, TBI-HY and TBI-HY+IL-1β versus Sham and TBI-HY+IL-1β versus TBI-NO+Veh. For the dentate hilus, post hoc analysis was significant for TBI-NO+Veh, TBI-HY and TBI-HY+IL-1β versus Sham (p<0.05) and TBI-HY+Veh and TBI-HY+IL-1β versus TBI-Normoxia+IL-1β. Polytrauma exacerbates contusion volume after TBI with the most damage observed in those animals undergoing a secondary hypoxic insult. Neuronal cell loss within the ipsilateral cerebral cortex and dentate hilar region of the hippocampus is exacerbated by the double insult of hypoxia and systemic inflammation.
These findings indicate that polytrauma impacts on neuronal populations that are involved in cognitive as well as sensorimotor function. Therapies that can show efficacy in reducing this histopathological damage as well as improve functional outcome in this polytrauma model will be assessed in the future.
Supported by W81WXH-08-1-0146.
traumatic brain injury, polytrauma, neuropathology
PENETRATING BALLISTIC BRAIN INJURY REDUCES FOCAL AND GLOBAL BRAIN GLUCOSE UTILIZATION: A STUDY OF C-14 2DG AUTORADIOGRAPHY IN A RAT MODEL
Clayton Jackson, B.S., ; Markus Spurlock, B.S., ; Daniel Diaz, B.S., ; Dr. Shoji Yokobori, M.D., Ph.D., United States;
Dr. Shyam Gajavelli, Ph.D., ; Alexandra Wick, ; Helen M. Bramlett, Ph.D., ; Dr. Lai Yee Leung, Ph.D., United States;
Dr. Frank Tortella, Ph.D., ; Professor Ross Bullock, M.D., Ph.D., United States
Penetrating brain injuries due to gunshot wounds are a major cause of death among deployed soldiers and young civilian adults in the United States. The effects of the injury on brain metabolism remain unknown. This study aimed at assessing brain glucose metabolism in a penetrating ballistic brain injury (PBBI) model.
Male Sprague-Dawley rats (280–350 g) were allocated to sham (n = 5) or injury groups (n = 10). A burr hole was made in the right frontal skull of all animals, and an inflatable probe was inserted in the injury group to simulate the temporary cavity in a gunshot wound (Tortella PBBI mod-el). All animals were immobilized using a cast and allowed to recover from anesthesia under minimal visual and auditory stimuli. Radioactive 14C-2-deoxy-D-glucose (14C-2DG; 50 μCi) was administered two hours post-injury and serial blood samples were taken within 45 minutes after 14C-2DG administration. Brains were harvested and flash-frozen immedi-ately after acquiring the last blood sample. All brains were subsequently sectioned, exposed to X-ray films, and digitized using the software MCID Elite 6.0 Rev. 1.0. Based on the Sokoloff Method, densitometric measures and scintillation counts of blood samples were used to calculate the lev-els of 14C-2DG in nine serial sections per animal.
Whole brain analyses showed global suppression of glucose metabolism in injured animals (MEAN ± SEM; 6.0 ± 0.7 μmol/100g/min) compared to sham animals (15.6 ± 1.5 μmol/100g/min; p < 0.0001). Glu-cose metabolism was significantly reduced in the hemisphere contrala-teral to the injury (sham vs. injured; 22.5 ± 1.2 vs. 8.7 ± 0.8 μmol/100g/min; p < 0.0001) as well as the ipsilateral hemisphere (sham vs. injured; 19.7 ± 1.2 vs. 3.2 ± 0.7 μmol/100g/min; p < 0.0001). The in-jured hemisphere showed a steeper decrease in glucose metabolism be-tween Bregma levels 2 and −6 (2.2 ± 0.5 μmol/100g/min) compared to the more rostral and caudal sections (6.5 ± 1.4 μmol/100g/min; p < 0.001). The latter regions of decreased metabolism were consistent with the in-jury core identified in previous studies.
This is the first report describing both global and focal reduction of glu-cose metabolism following a penetrating ballistic brain injury. This occurs in contrast to the hypermetabolism seen in other focal and dif-fuse injuries. The mechanism underlying decreased glucose consumption is not clear, but future studies using perfluorocarbons in this model may distinguish whether hypometabolism occurs due to hypoxia or deficiency of glycolysis. In addition, the relationship between the severity of the in-jury and the degree of glucose metabolism needs to be evaluated.
We thank Jessie Truettner for her guidance in autoradiography. This study was supported by the DOD Grant PTO74521-W81XWH-08-1-0419.
INTRACRANIAL PRESSURES IN MICE SUBJECT TO SHOCKWAVES
David Gullotti, BS, University of Pennsylvania
David F. Meaney, PhD, University of Pennsylvania
The current war has reintroduced a need to study blast traumatic brain injury. While shock wave phenomenon has been thoroughly researched, there is little knowledge on the transmission of shockwaves into the brain. Determining the pressures in the brain during a blast event will further research into bTBI.
Freshly euthanized mice were used in this study.The muscles and the skin were cut away to reveal a posterior portion of the skull. The exposed portion of the skull was thinned and tourmaline pressure gages were inserted.Two different methods of generating a shockwave were used.The first method was the use of a shock tube. Four different peak pressures (20psi, 30psi, 40psi, and 50psi) were tested using the shock tube.Each mouse was used twice at the same pressure level.The second method used the explosive, pentolite.The mice were positioned at three different standoffs that yielded approximate free field peak pressures of 16psi, 33psi, and 38psi.
With the shock tube, the mice were oriented such that the pressure waves traveled dorsal to ventral direction. Peak pressures and impulse into the skull and brain for all pressure levels were higher than respective free field pressures. In this orientation, a double peak is sometimes apparent suggesting that shockwave transmission into the skull and brain is a complex phenomenon. In the free field blast events, mice were placed in two orientations. In the first orientation mice were placed “face forward.” In this orientation, the shockwave travels from anterior to posterior. In the second orientation, the mice were position such that the oncoming shockwave would travel in a similar direction as the shock tube pressure wave, dorsal to ventral direction. The pressures in the brain from the two different orientation revealed that the second orientation had similar pressure phenomenon as that see in the shock tube.In these mice, the pressures in the brain were always higher than free field. The pressures in the brain from mice placed in the first orientation were generally not that much different than free field peak pressures and had similar duration. In some mice, the peak pressures were much higher than free field. The closer the standoff between mouse and charge, the more variability in the peak pressures seen in the head. At the furthest standoff, peak pressures ranged from 11psi to 15psi for orientation one; for orientation two, the peak pressure reached 58psi. At the 28″ standoff for orientation one, peak pressures ranged from 40 psi to over 100 psi; for orientation two, peak pressures reached over 100psi. At the third standoff of 30″, the peak pressures for mice in orientation one ranged from 40psi to 106psi; in orientation two peak pressure of 165psi was reached.
These studies show that pressure transmission through the skull and into the brain is affected by the orientation of the mouse with respect to the direction of the shock wave. In the dorsal to ventral shock wave direction, the peak pressures in the brain were always higher than free field.In the anterior to posterior shock wave direction, pressures in the brain better matched free field.
The authors would like to thank NSWCCD for facilitating these experiments. This work was also funded by the Department of the Army (W911NF-10-1-0526).
Explosives, shock tube, ICP
MICROARRAY BASED GENE EXPRESSION ANALYSIS FOLLOWING CONTROLLED CORTICAL IMPACT INJURY AND POST INJURY FLUMAZENIL TREATMENT IN IMMATURE (POST NATAL DAY 17) RATS
Pawel G. Ochalski, M.D., University of Pittsburgh Medical Center
John Caltagarone, Ph.D., University of Pittsburgh
YueFang Chang, Ph.D., University of Pittsburgh
Wendy Fellows-Mayle, Ph.D., University of Pittsburgh
David O. Okonkwo, M.D./Ph.D., University of Pittsburgh
Phillip Stafford, Ph.D., Arizona State University
C. Edward Dixon, Ph.D., Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Lucy Treiman, Ph.D., Barrow Neurological Institute at Phoenix Childrens Hospital
P. David Adelson, M.D., Barrow Neurological Institute at Phoenix Children's Hospital
Current treatments for pediatric populations are based upon data from scientific inquiries into adult traumatic brain injury (TBI). Flumazenil, a benzodiazepine antagonist, has efficacy in immature TBI. Microarray gene expression analysis allows a better understanding of unique injuries within the developing brain and can elucidate current and potential therapeutic targets.
Post natal day (PND) 17 rats (total n=16; n=4/group) underwent a lateral frontoparietal controlled cortical impact (CCI) injury (velocity: 6 m/s, depth: 2 mm) or sham surgery. Of rats who received a CCI (n=8); four were treated with 10 mg/kg i.p. flumazenil for 2 weeks prior to sacrifice, the other four received vehicle i.p. injections initiated on post-injury day 1. At 2 weeks post injury, the hippocampus was extracted fresh for analysis. The Illumina RatRef-12 expression BeadChip was used for whole-genome expression analysis of the hippocampal tissue. Genes with no known function and repeated genes in the microarray were excluded. Efficiency analysis was used to identify the optimal methods for normalization and for identifying differentially expressed genes. We implemented a pathway-level Impact Analysis previously described (Draghici et al., 2007) within OntoTools. The validity of the inference of differential expression implied by the microarray analysis was studied by RT-PCR.
Data quality for comparisons had high between group comparisons (r>0.995), low all-gene coefficients of variation (CqV<0.03), and low confounding indices (CI<1.0). Data analysis identified differential expression of 373 genes in injured versus sham, and 22 genes in injured group treated with flumazenil versus injured treated with vehicle control. Pathway-level impact analysis identified 19 pathways significantly impacted (p<0.05) by identified gene changes in CCI versus sham. These included cell communication, long-term potentiation, and calcium signalling pathways. While no significant pathways were identified in the flumazenil treated CCI versus vehicle treated CCI animals comparison, the 22 gene differences identified initially included genes important for signal transduction, cell differentiation, mitochondrial function, and inflammation. Real-time quantitative RT-PCR analysis verified the direction and magnitude of change in six genes that included sclerostin domain containing 1, transthyretin, solute carrier organic anion transporter, transmembrane protein 27, glycosylation dependent cell adhesion molecule 1, glucosamine-6-phosphate deaminase 2.
Microarray analysis has led to an increased understanding of the mechanisms following experimental TBI in the immature rat by providing a preliminary characterization and identification of a number of potentially important genomic pathways that may lead to the future development of novel therapeutic interventions in TBI in children.
Walter L. Copeland Fund of The Pittsburgh Foundation and a grant from the Neurosurgery Research and Education Foundation of the American Association of Neurological Surgeons
immature, microarray, flumazenil, CCI
OLIGODENDROCYTE 2′, 3′-CYCLIC NUCLEOTIDE 3′-PHOSPHODIESTERASE PARTICIPATES IN LOCALIZED ADENOSINE PRODUCTION: POSSIBLE ROLE IN TRAUMATIC BRAIN INJURY
Travis Jackson, Ph.D., University of Pittsburgh
Rashmi Bansal, Ph.D., University of Connecticut
Patrick M. Kochanek, M.D., Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Edwin Jackson, Ph.D., University of Pittsburgh
Adenosine is an endogenous neuroprotectant whose extracellular concentration is greatly elevated post-CNS injury. Recently, we identified extracellular 2′,3′-cAMP as a source of localized adenosine production post-TBI. In these studies, we examine the contribution of oligodendrocytes to the proposed neuroprotective extracellular 2′,3′-cAMP-adenosine pathway.
Controlled cortical impact: TBI experiments were conducted in (CNPase) +/+ and CNPase −/− mice using the CCI method with a 3-mm flat tip impounder at a velocity of 4.0 m/s and a depth of 1 mm. Oligodendrocyte cell culture: Primary oligodendrocytes were cultured from E17 mouse embryo brains (CNPase +/+ and CNPase −/−). Purine metabolism studies: Oligodendrocytes were washed twice and treated PBS in the presence and absence of various substrates (3′,5′-, 2′,3′-cAMP, 5′-, 3′-, or 2′-AMP) for 1 hour. In some experiments, enzyme inhibitors were included in the incubation. Analytical methods: To quantitatively measure the levels of the indicated purines, the samples were resolved by reverse-phase liquid chromatography and were then quantified using a triple quadrupole mass spectrometer. Statistical analysis: Data were analyzed by ANOVA followed by post hoc analysis (p<0.05). All values are means±SEM.
CNPase +/+ and CNPase −/− mice were subjected to CCI and were monitored for outcome using hematoxylin and eosin stain. The CNPase −/− mice appeared to have augmented cell death and damage post-CCI when compared to their CNPase +/+ littermates. Because we previously showed that CNPase −/− mice have reduced adenosine formation post-CCI and because CNPase is most abundantly expressed by oligodendrocytes, we next examined the role of these cells in the extracellular 2′,3′-cAMP-adenosine pathway.
Primary oligodendrocytes were evaluated for their ability to convert extracellular 2′,3′-cAMP and 3′,5′-cAMP to their respective AMP metabolites. Oligodendrocytes robustly convert 2′,3′-cAMP to 2′-AMP and to a lesser extent (over 10-fold) 2′,3′-cAMP to 3′-AMP. The formation of 2′-AMP was modestly reduced by the phosphodiesterase inhibitor DPSPX, but 3′-AMP formation was unaffected. The cells were least efficient at metabolizing 3′,5′-cAMP to 5′-AMP (>100-fold less than 2′-AMP), a process completely abolished by DPSPX. Oligodendrocytes from CNPase −/− animals showed a drastic reduction (∼75%) in their ability to form 2′-AMP from 2′,3′-cAMP, although the other AMPs showed no genotype dependent differences.
We next examined the ability of oligodendrocytes to form adenosine from the individual AMPs. Oligodendrocytes efficiently converted 5′-AMP to adenosine, a process blocked by AMPCP (a CD73 inhibitor). Oligodendrocytes also converted, albeit at a lower rate than 5′-AMP, both 2′-AMP and 3′-AMP to adenosine, yet AMPCP had no effect on this metabolism. Finally, we examined both CNPase +/+ and CNPase -/- cells for their ability to convert either extracellular 2′,3′-cAMP or 3′,5′-cAMP to adenosine. The CNPase +/+ oligodendrocytes converted both 2′,3′-cAMP and 3′,5′-cAMP to adenosine at similar rates, while the CNPase-/- oligodendrocytes were less efficient at metabolizing both cAMPs.
The data present here reveal an unappreciated neuroprotective role for oligodendrocytes in traumatic brain injury. We have previously established that CNPase does metabolize 2′,3′-cAMP to 2′-AMP in mouse brain and that extracellular 2′-AMP is an important source of adenosine post-CCI. These studies suggest that oligodendrocytes are the predominate cell of the CNS responsible for this reaction. We show that oligodendrocyte-derived CNPase is the major enzyme responsible for 2′-AMP formation. These data further elucidate the characterization of the extracellular 2′,3′-cAMP-adenosine pathway as a neuroprotective process post-injury. The inhibitor studies indicate the enzymes responsible for the extracellular 2′,3′-cAMP-adenosine pathway are distinct from those responsible for 5′-AMP to adenosine formation. Future studies will examine the source of 2′,3′-cAMP in the brain, which specific enzymes are responsible for the conversion to adenosine, and how to manipulate this pathway for enhanced neuroprotection.
We would like to thank the National Institutes of Health (NIDDK, NINDS) for funding support for this project and Delbert Gillespie for technical assistance.
TBI, cAMP, CNPase, Adenosine, Oligodendrocytes
MECHANISM LINKING TRAUMATIC BRAIN INJURY TO AMYOTROPHIC LATERAL SCLEROSIS PATHOLOGY
Carlos Jaramillo, M.D. Ph.D., University of Texas Health and Science Center
Holly Van Remmen, Ph.D., University of Texas Health and Science Center
The goal of this study is to elucidate the mechanism linking TBI to peripheral nervous system (PNS) pathology in animal models of ALS, by testing the hypothesis that TBI will accelerate ALS disease progression.
The well-characterized mouse models of familial ALS (G93A SOD1) and sporadic ALS (TDP43) were used to study the effect of single and double TBI on ALS progression. Mice were subjected to a closed head injury (CHI) at 45 days using the TBI 0310 Impact Device (Percision Systems LLC). To characterize structural central nervous system (CNS) pathology following TBI we utilized magnetic resonance imaging (MRI) to confirm extent of injury after 24 hours. Histological techniques were used to assess cell loss and edema in brain and spinal cord at 24 hours, 50 days (Pre-Onset), 90 and 110 days (Onset) and 130 days (Symptomatic) of age in transgenic mice. PNS abnormalities were measured using electromyography (EMG). To assess ALS progression, we measured body weight and disease scores once prior to TBI (45 days) and twice weekly thereafter. Gait analysis was performed using the TreadScan (CleverSys, Inc.) system.
Our preliminary results show neuron loss, edema and astrogliosis at 24 hours post-TBI in wildtype (WT) and transgenic (TG) mice. We have also used MRI methods for structural analysis to show that our CHI does not cause excessive neurological trauma or hemorrhaging. Gait analysis did not indicate any clear abnormalities 5 days after single or double injuries. However, changes in gait presented earlier in TG mice with two injuries compared with TG mice with no injuries. Furthermore, body weight and disease score changes also demonstrated early onset of pathology in TG mice with single and double TBI compared with TG mice with no injury. In our preliminary studies, we have used EMG to test for muscle membrane instability, and evaluate the integrity of the final common pathway of the PNS. Bilateral hind limb muscle groups (gastrocnemius, tibialis anterior) were examined using standard monopolar needle electrodes. Muscles were assessed for the presence or absence of abnormal spontaneous activity (ASA) and given a score based on the severity of ASA. We found electrophysiologic evidence as early as 48 hours post-TBI of significant ASA in the form of denervation potentials (positive sharp waves and fibrillations) and fasiculations in both WT and TG animals. No ASA was seen in our controls. In general, WT/TBI animals had equal prevalence of ASA after a single TBI as TG animals. After a second TBI, ASA was significantly increased in WT mice. However, a single TBI potentiates ASA in TG animals that was significantly increased compared to WT/TBI mice. This was further potentiated by a second injury resulting in the most severe phenotype of all groups studied. This approach allows us to study longitudinally, in a minimally invasive manner and in a controlled environment, this poorly understood electrophysiologic finding first identified in humans with TBI.
Our data strongly suggests that muscle denervation occurs in WT and TG mouse models secondary to mild CHI. This is the first demonstration of muscle denervation after central injury in WT mice. In addition, our preliminary results are the first to show that single and double TBI, in an animal model of motor neuron disease (ALS), results in significantly increased muscle denervation and potentiates disease onset and progression. This study is the first to show that mild CNS injury may result in secondary muscle denervation and significant downstream PNS effects. Following CHI to TG mouse models of motor neuron disease, TBI significantly increased motor neuron pathology in a dose dependant manner further supporting that CNS injury results in PNS pathology. While, both TBI and ALS transgenic mice result in lower motor neuron dysfunction, their additive effects suggest that each are affecting different components of the peripheral motor pathway.
This work was supported by an institutional training grant NIA 5 T32 AG021890 as well as the Barshop Institute for Longevity and Aging.
TBI ALS Electromyography Motor Neuron
MORPHOLOGY AND DISTRIBUTION OF FOCAL HEMORRHAGE FOLLOWING PRIMARY BLAST INJURY
Jordan Walker, University of Utah
Dave Bell, University of Utah
Kenneth Monson, PhD, University of Utah
Understanding injury from blast overpressure is a major challenge facing modern military and civilian medicine. Mechanisms behind blast-induced brain injury are not yet fully understood. We have developed a model of primary blast injury that results in focal blood-brain-barrier disruption.
Using a .308 bolt action rifle and a modified steel casing fitted for a .50 caliber primer, we have developed a benchtop shock tube capable of producing overpressures of up to 440 kPa. Male Sprague Dawley rats were anesthetized with isoflurane and immobilized in a Kevlar hammock offset from the muzzle to eliminate loading by the exiting gas jet and burning propellant. Injury severity from lateral exposure was tested at three levels of blast overpressure: 150, 280, and 440 kPa, determined by distance from the shock tube. Animals were then transcardially perfused with PBS and 4% paraformaldehyde. Brains were cut into 50 micron coronal sections and examined immunohistochemically. Primary markers were Biotinylated GtxRt IgG1 and RbxMs anti-laminin IgG1 to detect displaced blood and to visualize vasculature, respectively. Sections were imaged with an Olympus E600 upright microscope and then analyzed with NI Vision Assistant to measure lesion size and location.
Pressure curves were characterized at varying distances to find appropriate levels of blast overpressure outside of the gas jet. In animals exposed to blast, small focal regions of high fluorescence were found in the parenchymal brain tissue, indicating the presence of rat IgG. Lesions were too small for detection via high resolution MRI. Lesion size and frequency increased with blast exposure. Initial study at the 280 kPa blast level (N=2) resulted in lesion densities of 0.25 and 0.14 injuries/mm3 of brain tissue. A total of 137 separate injuries were found in the brain of one animal in which every section was examined. Many of these lesions persisted through multiple sections. Just 79 injuries were identified in a second animal where every fifth section was sampled. Lower levels of blast injury were also examined (50, 100, 125 kPa), but no significant injury was found in these cases. Further study is ongoing. Lesion distribution was measured from the midline of each section. In one animal the center of the distribution was found to shift towards the hemisphere ipsilateral to the site of blast exposure, experiencing a 5.4% (0.9 mm) shift from the center of the brain. Injuries were most commonly found in the deep dorsal ipsilateral cortex, at the boundary of the cortex and the hypothalamus on both sides, and at the contralateral ventral surface of the cortex. Laminin staining revealed vascular structure in the brain and surrounding injury sites. Knowing vessel structure helped filter suspected areas of injury, as only sites with clearly associated vasculature were considered lesions. A common feature that many vessels shared was a branch point close to the injury site.
Despite widespread lesion distribution, general trends for injury location have emerged. The boundary between the cortex and deeper brain structures seems to be particularly vulnerable, possibly due to mechanical impedance mismatches between gray and white matter. Another common injury zone was the contralateral ventral surface of the cortex, suggesting possible coup-contrecoup injury resulting from the blast wave. Laminin staining helped identify lesion sites and revealed information about microscale injury factors. In standard engineering applications, corners and joints are sites of stress concentration. Blood vessels appear to follow the same rules, as branch points were subject to tension during tissue deformation. Blast brain injury remains a topic of intense research. Understanding mechanisms behind its mechanical and physiological effects on both neural and vascular components of the brain will aid in developing therapies to treat this important problem.
This work was funded by a grant from the Department of Defense (W81XWH-08-1-0295).
blast, traumatic-brain-injury, vasculature, blood-brain-barrier
FK506 INDUCED CHANGES IN AXONAL, GLIAL AND MATRIX METALLOPROTEINASE RESPONSES IN THE INTERNAL CAPSULE FOLLOWING TRAUMATIC BRAIN INJURY
Thomas M. Reeves, PhD, Virginia Commonwealth University
Linda L. Phillips, PhD, Virginia Commonwealth University
The internal capsule (IC) is a highly myelinated fiber tract vulnerable to cytoskeletal breakdown, glial cell responses, and changes in extracellular matrix, specifically matrix metalloproteinases, following diffuse traumatic brain injury. This study examined the effects of an established neuroprotective agent, FK506, on the IC following central fluid percussion injury.
Traumatic brain injury (TBI) responses in internal capsule were most prevalent at 3d post-TBI thus FK506 effects were assessed at the 3d time point. Adult rats were placed into 3 groups: sham-control, TBI-alone, and TBI+FK506. TBI+FK506 animals received 3.0 mg/kg of FK506 i.v. 30 min before central fluid percussion injury. Following injury, the IC was bilaterally dissected from fresh brain tissue (n=5 sham, n=5 TBI+FK506, n=4 TBI-alone). Western blot analysis of αII-spectrin proteolysis was performed as an indicator of cytoskeletal breakdown. Enzymatic activity of matrix metalloproteinases (MMPs) 2 and 9 was quantified using gelatin zymography. A second set of animals (n=3/group) was transcardially perfused using 4% paraformaldehyde at 3d post-TBI. Glial cell response was examined using confocal immunohistochemistry (IHC) with antibodies specific for astrocytes (GFAP) and microglia (IBA-1). To determine glial cell contribution to extracellular matrix changes co-localization of MMP2 and MMP9 was examined.
FK506 inhibits the activity of calcineurin, an enzyme that modulates axonal structure and function. In order to establish changes in axonal damage in the IC due to FK506 treatment, the proteolysis of αII-spectrin was analyzed. Calpain proteolysis of αII-spectrin generates 145 and 150kD fragments, which exhibited the most significant IC changes. The 145kD fragment was significantly increased following TBI (1133% of sham, p=0.0001). Pre-treatment with FK506 significantly attenuated this response (596% of sham) compared with TBI-alone (p=0.05) although remaining significantly above sham-controls (p=0.015). The 150kD fragment showed a similar, yet reduced effect, trending toward lower levels with FK506 (p=0.06), however, it remained 5.3 fold higher than controls (p=0.0005). Caspase-3 production of a proteolytic fragment at 120kD was significantly increased following TBI (142% of sham, p=0.005), although unaffected by FK506 (135%, p=0.03 vs. sham, 0.6 vs. TBI-alone). FK506 immunosuppression also affects microglial activation and attenuates astrocytic hypertrophy following multiple types of brain insult. IHC showed that microglia were reactive with elongated somas and short, thick processes although no qualitative differences were observed between TBI-alone and TBI+FK506 animals. In contrast, there was an apparent decrease in GFAP+astrocytic processes for animals treated with FK506. Both glial types exhibited co-localization with MMPs 2 and 9. FK506 induces MMP9 production in the brain and other organs, but has yet to be studied following TBI. MMP2 activity in the IC was significantly increased following TBI (247% of sham, p=0.0008). FK506 treatment decreased this activity relative to TBI-alone animals (168% of sham, p=0.05), but it also remained significantly greater than sham-controls (p=0.03). In contrast, MMP9 showed no post-injury IC change, but was significantly increased after FK506 treatment when compared with both sham (144% of sham, p=0.02) and TBI-alone animals (p=0.03).
These data support FK506 effects on axonal damage, glial cell responses, and MMPs in IC following diffuse TBI. These effects appear selective to calpain derived proteolysis of the axonal cytoskeleton and the hypertrophic astrocytic response, both were attenuated. The decrease in astrocyte activation could play a role in the differential MMP changes. Previous studies from our laboratory examined the effect of FK506 on a mixed fiber population in corpus callosum (CC), showing significant recovery from injury-induced deficits in compound action potential and attenuation of MMP responses 3d post-injury. MMP activity changes were exacerbated in IC when compared to CC although the trends remained similar. Because TBI affects multiple fiber tracts throughout the brain, it's important to establish similarities and differences between injury response in pathways with different axon composition. These differences may be critical to development of therapeutic interventions to enhance overall axonal recovery.
Supported by: NIH-NS56247, NS057758
TBI, Internal capsule, MMP, glia
CHANGES IN THE DISTRIBUTION AND EXPRESSION OF VOLTAGE GATED SODIUM CHANNELS (NAV1.6) AND CONTACTIN-ASSOCIATED PROTEIN (CASPR) WITHIN THE MOLECULAR DOMAINS OF TRAUMATICALLY INJURED AXONS (TAI) FOLLOWING TRAUMATIC BRAIN INJURY
Xiuyin Liang, MS, Virginia Commonwealth University, Department of Neurosurgery
Daniel J. Gardiner, MS, Virginia Commonwealth University
Shanaz Parveen, MD, Virginia Commonwealth University, Department of Neurosurgery
Naqeeb Abidi, MD, Virginia Commonwealth University, Department of Neurosurgery
Jeffrey L. Dupree, PhD, Virginia Commonwealth University, Department of Anatomy and Neurobiology
TAI contributes to mortality and morbidity following TBI. We posit that injury-induced changes in Nav1.6 density at the nodes of Ranvier in TAI are related to degradation of contactin-associated protein (CASPR), within paranodal domains. Consequently, irregular distribution/expression of these protein complexes may contribute to the pathobiology of TAI.
A rodent model of impact acceleration, closed head, diffuse-TBI was used to examine alterations in the distribution of nodal and paranodal domains of traumatically injured myelinated axons within the corticospinal (CSPT) and medial lemniscus (ML) tracts in sham-injured and TBI rats at 30 min, 3 h and 24 h post-injury. To examine changes in the distribution and expression of nodal Nav1.6 and paranodal CASPR, and explore their relationship with amyloid-precursor protein (APP), an accepted marker of TAI, we employed multiple label immunofluorescence techniques, ultrastructural cytochemistry, and biochemical analysis using western blot. Blood gasses and mean arterial pressure were routinely recorded 15 min and 30 min pre/post-TBI for sham and TBI animals. Analysis of the variance (ANOVA) and Tukey post-hoc tests were used, data is expressed as mean±SE and p<0.05 was considered significant.
Ultrastructural analysis of traumatically injured axons showed lengthening, bulging and occlusion within the nodal domain consistent with cytoskeletal compaction and bulb formation that are indicative of impaired axonal transport and TAI. Measurement of the nodal domains in CSPT and ML demonstrated a significant increase in nodal length over a 24-h period. Multiple label fluorescent microscopy showed a marked redistribution of Nav1.6, normally located at the node of Ranvier, into the paranodal area that are demarcated by the bilateral localization of CASPR. Furthermore, TBI induced an increase in the distribution/density of Nav1.6. Conversely, CASPR domains were dispersed and often truncated. These changes were collectively related to APP positive swellings, the historic marker of TAI. Biochemical analysis using western blot densitometry showed injury-induced up-regulation of Nav1.6 expression over 24 h; Nav1.6, normalized by comparison to β-actin expression (control protein), increased from 0.25±0.06 in sham operated animals to 0.38±0.07 at 30 min, 0.47±0.10 at 3 h, and 0.52±0.10 at 24 h post-trauma (p≤0.05). Conversely, quantitative analysis showed an injury-induced down regulation of CASPR; CASPR expression, normalized by comparison to cyclophilin-A (control protein), was 1.13±0.11 in sham animals and was significantly decreased to 0.83±0.09 at 30 min, 0.68±0.09 at 3 h, and 0.59±0.08 at 24 h post-TBI (p≤0.05). These changes were not due to TBI-induced depression of systemic function as no significant differences in pH, pO2, pCO2, or mean arterial pressure were found between the sham and TBI groups.
The results support our hypothesis that alterations in the molecular domains of injured axons contribute to cellular mechanism of TAI. We showed that TBI resulted in profound abnormalities in Nav1.6 and CASPR sequestration within the node and paranode domains that co-localized with classic markers of TAI. Up-regulation of Nav1.6 and down-regulation of CASPR expression began within 30 min and developed over a 24 h time-course post-TBI. Taken in concert, these studies suggest that abnormal Nav1.6 and CASPR sequestration within the molecular domains of injured axons plays a pivotal role in the progression of TAI. Future studies, will seek to relate the changes in Nav1.6 distribution to functional deficits; providing the rationale for reduction in conduction velocity and saltatory conduction that occurs in TAI. Identification of these novel structural and molecular changes in axons following TBI adds important insight into the complex pathobiology and provides novel questions for future TBI research.
Funding: RFP 07-302 from the Commonwealth Neurotrauma Initiative Trust Fund, Richmond, VA.
axonal, injury, sodium, CASPR, Nav1.6
COMPARISON OF INTRAVENOUS AND INTRACEREBROVENTRICULAR ADMINISTRATION OF AMNION-DERIVED CELLULAR CYTOKINE SOLUTION IN A RAT MODEL OF PENETRATING BALLISTIC-LIKE BRAIN INJURY
Christopher van der Merwe, B.S., Walter Reed Army Institute of Research
Francis Bustos, B.S., Walter Reed Army Institute of Research
Weihong Yang, Walter Reed Army Institute of Research
Zhiyong Chen, Ph.D., Walter Reed Army Institute of Research
Deborah A Shear, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Previous work indicates that the neuroprotective effects of Amnion-derived Multipotent Progenitor (AMP) cells may be mediated through the sustained secretion of AMP cell-secreted factors (Amnion-derived Cellular Cytokine Solution; ACCS). This study compared the neuroprotective efficacy of different routes of ACCS administration following penetrating ballistic-like brain injury (PBBI).
Isoflurane-anesthetized rats received either sham or PBBI surgery. ACCS was delivered directly into the CSF (intracerebroventricular; i.c.v.) or intravenously (i.v.) within 15 m post-PBBI. Chronic i.c.v. infusion of ACCS was delivered via Alzet osmotic pumps using a continuous flow rate of 1μl/hr (∼100 μl/kg daily) for 1 week. For i.v. administration, PBBI animals were given bolus infusions of ACCS (2 ml/kg) at 15 m and 6h post-injury and twice daily thereafter (∼4 ml/kg daily) for 5 consecutive days. Motor and cognitive outcome were assessed on the rotarod (7 and 14 days post-injury) and Morris water maze (MWM; 14–18 days post-injury) tasks respectively.
Results showed that the continuous i.c.v. infusion of ACCS into the CSF significantly improved motor outcome on the rotarod task as evidenced by an increased latency to fall (p<0.05). In contrast, i.v. administration of ACCS failed to produce significant improvement in motor outcome. Under these test conditions, neither the i.c.v. or i.v. route of ACCS administration showed efficacy on improving cognitive performance on the MWM task.
Overall the results of this study indicate that chronic i.c.v. ACCS delivery improves motor but not cognitive outcome in the PBBI model. However, the beneficial effects of ACCS were not apparent following i.v. delivery indicating that blood-brain-barrier (BBB) permeability may be a mitigating factor. Further work evaluating (1) intrathecal administration as an alternate CSF delivery route and (2) the ability of ACCS to penetrate both the intact and compromised BBB is ongoing.
This research was funded by the Army Combat Casualty Care Research Program and CDMRP (W81XWH-08-2-0127).
Amnion-derived Cellular Cytokine Solution
SIMILARITIES AND DIFFERENCES OF ACUTE EPILEPTIC ACTIVITIES FOLLOWING PENETRATING BALLISTIC-LIKE BRAIN INJURY AND FOCAL BRAIN ISCHEMIA IN RATS
Andrea Mountney, Ph.D., Walter Reed Army Institute of Research
Zhiyong Chen, Ph.D., Walter Reed Army Institute of Research
Guo Wei, Ph.D., Walter Reed Army Institute of Research
Ying Cao, B.S., Walter Reed Army Institute of Research
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Vivek Khatri, Ph.D., Walter Reed Army Institute of Research
Tracy Cunningham, M.D., Walter Reed Army Institute of Resarch
Jesse Harris, B.S., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
In this study, we characterized the similarities and differences of nonconvulsive seizures (NCS) and other epileptic events, such as periodic epileptic discharges (PED) and intermittent rhythmic delta activities (IRDA) in rat models of penetrating ballistic-like brain injury (PBBI) and permanent middle cerebral artery occlusion (pMCAO).
Rats received either unilateral frontal PBBI (n=43) or pMCAO (n=28). Brain activity was monitored continuously by EEG recordings for 72h (PBBI) or 24h (pMCAO). Epileptic activities examined included: NCS, PED, and IRDA. The NCS profile was characterized based on EEG waveform patterns, seizure incidence frequency, duration, and time distribution. The PED and IRDA activities were also characterized based on EEG waveform patterns, but the severity of these activities was rated based on the abundance of their occurrence and their associations with NCS.
NCS occurred in 70% PBBI rats and 79% pMCAO rats, manifested in three EEG waveforms: Type I–rhythmic high frequency (>2Hz) spikes, Type II–rhythmic low frequency (∼1Hz) sharp waves, and Type III–arrhythmic (0.5–2 Hz) spikes. Type I seizures resembled clinical EEG patterns of generalized/partial complex seizures and were predominant following both injuries. Types II/III seizures appeared to reflect inter-ictal activities. Overall, NCS occurred more acutely and intensely after pMCAO (latency=0.6h, frequency=25 episodes/rat) compared to post-PBBI NCS (latency=26h, frequency=10 episodes/rat). The most salient feature that differentiated post-traumatic and post-ischemic NCS was their time distribution. After pMCAO, >50% seizures occurred within the first 3h of injury, whereas after PBBI NCS occurred sporadically (0–5% per hour) throughout the 72h recording period. In both models, severity of PEDs was positively correlated to the severity of Type I seizures. But IRDA appeared to be an independent index, not predictive of other epileptic events.
This study provided comprehensive comparisons of post-traumatic and post-ischemic epileptic profiles. The identification of the similarities and differences across a broad spectrum of epileptic events may lead to differential strategies for post-traumatic/stroke seizure interventions, e.g. treating post-ischemic seizures more aggressively during the acute phase of injury, and treating post-traumatic seizures more selectively when time allows for referencing other epileptic activities.
This research was funded by the Army Combat Casualty Care Research Program and DHP grant (D10_I_AR_J6_414).
PBBI, Nonconvulsive seizure, EEG, Rat
FUNCTIONAL SIGNIFICANCE OF SYSTEMIC FACTORS IN THE PATHOPHYSIOLOGY OF BLAST-INDUCED NEUROTRAUMA
Yonas Alamneh, MSc, Walter Reed Army Institute of Research
Ying Wang, MD, Walter Reed Army Institute of Research
Peethambaran Arun, PhD, Walter Reed Army Institute of Research
Samuel Oguntaya, Walter Reed Army Institute of Research
Yanling Wei, MD, Walter Reed Army Institute of Research
Joseph Long, PhD, Walter Reed Army Institute of Research
Madhusoodana Nambiar, PhD, Walter Reed Army Institute of Research
Blast-induced traumatic brain injury is a complex type of brain injury that involves systemic factors and direct effects of shockwave on brain. The significance of activation of platelets and polymorphonuclear leukocytes as systemic factors in the development and progression of brain injury after blast exposure is still on debate.
We evaluated the functional role of systemic activation of platelets and polymorphonuclear leukocytes in the exacerbation of brain injury after blast exposure. Recently established repeated blast exposure mice model using shock tube (21 psi, 3 times with 1–30 min intervals) which showed significant neuropathology, neurobehavioral changes and brain regional specific alterations in various biomolecules was used for the study (Wang et al., 2011; Valiyaveettil et al., 2012a, b; Arun et al., 2012). The activation of platelets was analyzed by surface of expression of alpha IIb beta 3 by flow cytometry as well as by estimation of serotonin content by enzyme linked immunosorbent assay. Polymorphonuclear leukocyte activation was analyzed by myeloperoxidase activity and expression in the blood and plasma of repeated blast exposed mice. Histopathology with hematoxylin and eosin staining was used for the evaluation of cerebral vasospasm.
An acute increase in the activation of platelets at 4 h after repeated blast exposures was observed, indicating changes in platelet phenotype after blast exposures. Platelet serotonin concentration decreased at 4 h post-blast with concurrent increase in the plasma serotonin content confirming the early onset of platelet activation and release of inflammatory mediators after repeated blast exposures. The concentration of serotonin in the frontal cortex of blast exposed mice showed significant reduction, which may be an indicator of aggressive/suicidal behavior after blast neurotrauma. Increased activation and expression of myeloperoxidase enzyme was observed in the blood and plasma of repeated blast exposed mice at 4 h post-blast. Brain histopathology showed an acute constriction of blood vessels in animals exposed to repeated blasts compared to the respective sham controls, a classical phenomenon of reported cerebral vasospasm in blast affected victims.
In summary, repeated blast exposures in mice showed acute activation of platelets and leukocytes which can play a major role in driving the pathological effects of brain injury. Release of various biomolecules from platelets or leukocytes can exacerbate the brain injury after blast neurotrauma leading to chronic neurobehavioral/neuropathological changes. Platelet/leukocyte targeted therapies can be potential acute treatment strategies to protect against blast-induced neurotrauma.
Collaborative help from COL Paul Bliese, Center for Military Psychiatry and Neuroscience and members of blast-induced neurotrauma branch are greatly acknowledged.
TBI; blast exposure; platelets; leukocytes
CORRELATING BIOMECHANICAL INSULT WITH BIOMARKER LEVELS: A NEW MODEL FOR BIOMARKER EVALUATION IN TBI
Yan Li, M.S., Wayne State University
Srinivasu Kallakuri, Ph.D., Wayne State University
John Cavanaugh, M.D., Wayne State University
Thus far a growing number of biomarkers have been studied for assessing brain injury severity. However, a lack of unified models and knowledge of injury mechanisms has contributed to existing biomarker controversies. A modified Marmarou impact acceleration model was used to help screen reliable biomarkers to assess injury severity.
Fifteen anesthetized male Sprague-Dawley rats (375–425 grams) were subjected to a closed head injury by a modified Marmarou impact device from 2.25 m (n=8) and 1.25 m heights (n=7). Four rats were used as control. Linear and angular responses of the head were measured in vivo with an accelerometer and angular rate sensor affixed to the skull. Twenty-four hrs after impact cerebrospinal fluid (CSF) was collected from cistern magna and then 1 mL of blood was collected from the heart just before perfusion with 4% paraformaldehyde. CSF and serum levels of Amyloid beta (Aβ) 1–42, neurofilament H (NF-H) and interleukin (IL)-6 were assessed by ELISA. Biomechanical measurements of the head and biomarker levels between impact groups, or between impact group and control group were assessed using t-tests. Bivariate correlation analysis was used to evaluate the correlation between biomechanical parameters and biomarker levels by SPSS.
The peak linear acceleration, average linear acceleration, peak angular velocity and average angular velocity in the 2.25 m group were 730±36g, 301±14g, 154±13rad/s, and 108±9rad/s, respectively. These were significantly higher than in the 1.25 m group (445±48g, 184±21g, 104±13rad/s and 81±11rad/s). Significantly higher CSF (6.12±0.90ng/ml) and serum (0.70±0.18ng/ml) NF-H levels were observed in 2.25 m impacted group compared to 1.25 m (2.41±0.57ng/ml and 0.12±0.02ng/ml) and control samples (0.79±0.11ng/ml and 0.12±0.01). Compared to control, CSF levels in 1.25 m impacted group were also higher. Increased IL-6 levels were observed in 3 out of 8 rats in 2.25 m impacted group. Overall, the difference between 2.25 m (429±142pg/ml) and 1.25 m (146±4pg/ml), as well as 1.25 m and control (121±5pg/ml) was not statistically significant (p>0.05) in CSF and serum IL-6 levels. Levels of Aβ were not significantly different between groups. Correlation analysis showed NF-H level in CSF had positive correlation with average linear acceleration (r=0.683, p<0.01), which is a good predictor for traumatic axonal injury according to histologic assessment in our previous study. NF-H level in serum also showed good correlation with average linear acceleration (r=0.687, p<0.01).
In this study rat head kinematics were characterized for various TBI severities, and the level of potential biomarkers was measured in both CSF and serum. The results showed both CSF and serum NF-H could be a reliable predictor for severe TBI (2.25 m impact), whereas NF-H in CSF is also a good indicator for mild TBI. Furthermore, the level of NF-H had a positive correlation with average linear acceleration, suggesting it is directly related to the injury mechanism. Taken together, NF-H is a potential candidate biomarker for TBI. In addition the model used in this study showed promise in elucidating the relationship between biomarker and severity of the mechanical trauma to the brain, which cannot be determined in clinical trials.
The work was supported by NIH R01 EB006508.
Biomarker head kinematics traumatic brain injury rat
APOE4 IS NOT A RISK FACTOR FOR POOR COGNITIVE OUTCOME IN A MOUSE MODEL OF REPEATED MILD CONCUSSIVE BRAIN INJURY
William Meehan, MD, Harvard Medical School
Jimmy Zhang, BA, Department of Pediatrics, Massachusetts General Hospital
Jacqueline Berglass, BA, Department of Medicine, Children's Hospital Boston
Tory Gray, BA, Department of Pediatrics, Massachusetts General Hospital
Christopher Lee, BA, Department of Pediatrics, Massachusetts General Hospital
Michael J. Whalen, MD, Harvard Medical School
Repetitive concussions may impair cognitive function in susceptible individuals playing contact sports or military personnel suffering blast injuries. The apolipoprotein E4 (APOE 4) allele has been associated with worse severity of chronic neurologic deficits in boxers, but whether or not it is a risk factor for outcome after multiple,
Anesthetized adult wild type (WT) and transgenic APOE4 mice were subjected to multiple, mild concussive TBIs using a weight drop model (28 in drop height, 53 g weight) or to sham injury (n=10-16/group). Cognitive function was assessed using the Morris water maze (MWM) 3 days, 3 mos, and 6 mos after the final injury. Soluble Aβ40, total tau, and phosphorylated tau were assessed at 6 mos (ELISA). Brain were stained for Aβ plaques at 6 mos after injury and compared to Alzheimer Disease (AD) control mice. MWM data were analyzed by repeated measures analysis of variance. Mean quantities of Aβ40, total tau, and phosphorylated tau were compared using Student's t-test.
Three days after sustaining multiple concussions, injured APOE4 and WT mice performed worse on hidden trials of the MWM than sham injured mice (P<0.001, 8–10 mice per group), but there were no differences between injured APOE4 and WT. In both APOE4 and WT mice, MWM deficits persisted for 6 mos after injury (p<0.001 for injured vs. sham-injured), but injured APOE4 mice did not differ from injured WT at 6 mos after injury. Six mos after injury, there were no differences between any of the injured or sham-injured groups in soluble Aβ40, soluble total tau or soluble phosphorylated tau. Neither injured nor sham-injured APOE4 or WT mice demonstrated Aβ plaque formation 6 mos after injury compared to the AD control mice.
Using a weight-drop model of repetitive concussive brain injuries in mice, we found no difference in cognitive outcome or accumulation of Aβ plaques, soluble Aβ40, soluble total tau or soluble phosphorylated tau between APOE4 and WT mice. These data call into question whether APOE4 is a risk factor for worse outcome after multiple, mild concussive TBI. These data have implications for people at high risk for repetitive head injury such as athletes and military personnel.
This work was supported by grants from the NICHD T32 HD040128-06 (WM), National Football League Charities (WM, RM, and MJW), and NINDS 5RO1NS047447 (MJW).
traumatic brain injury, concussion, APOE4
EFFECT OF SELECTIVE BRAIN COOLING ON THE FUNCTIONAL RECOVERY FOLLOWING PENETRATING BALLISTIC-LIKE BRAIN INJURY IN RATS
Xiaofang Yang, M.S., Walter Reed Army Institute of Research
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Andrea Mountney, Ph.D., Walter Reed Army Institute of Research
Deborah A. Shear, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Previous work has shown that 2h exposure to selective brain cooling (SBC) protects against acute (24h) neuronal damage but does not improve cognitive outcome following penetrating ballistic-like brain injury (PBBI) in rats. In this study, we evaluated the effects of extended (8h) SBC exposure on PBBI-induced motor and cognitive impairment.
Unilateral frontal PBBI was produced in the right hemisphere of isoflurane anesthetized rats (10% injury severity level). SBC (34° C) was induced via extraluminal cooling of the bilateral common carotid arteries immediately following PBBI and continuously maintained under anesthesia for 8h post-PBBI. The control rats (PBBI alone) were exposed to identical procedures as SBC (including being maintained under anaesthesia for 8h) without the cooling of the brain. Neuroprotective efficacy was measured on (1) the fixed-speed rotarod task to measure motor coordination and balance at 7 and 10 days post-injury and (2) the Morris water maze (MWM) to measure spatial learning performance from 12–16 days post-PBBI.
Results showed that the extended (8h) SBC duration significantly improved motor outcome on the rotarod task with the most profound effect being evident at 10 days post-PBBI. Mean rotarod latencies were approximately 2-fold higher in the SBC-treated PBBI animals relative to non-treated PBBI controls (Sham=52±2s; PBBI=14±3s; PBBI+SBC=30±5s; means±SEM). However, no significantly improvements were detected in cognitive performance on the MWM task measured by latency to locate the hidden platform (sham=22±4s; PBBI=58±5s; PBBI+SBC=57±7s; means±SEM). Critically, neither prolonged anesthesia nor the 8h SBC protocol had an adverse effect on the neurofunction (i.e. motor or cognitive) parameters studied.
Overall, these results demonstrate that prolonged exposure to SBC translates into improved outcome on motor (but not cognitive) abilities following PBBI. Future work should focus on determining the potential long-term cognitive benefits of SBC by exploring the optimal SBC parameters or using SBC in combination with neuroprotective drugs.
This research was funded by the Army Combat Casualty Care Research Program.
PBBI, Hypothermia; Functional recovery
CO-CULTURE OF MESENCHYMAL STEM CELLS AND AN IN VITRO MODEL OF TRAUMATIC BRAIN INJURY
It has been shown that mesenchymal stem cells (MSC) have potential to evoke recovery after traumatic brain injury (TBI). We are developing a co-culture system of MSC and in vitromodel of TBI to investigate these neuroprotective effects of MSC and underlying mechanisms under a more control environment.
The hippocampi obtained from neonatal Sprague Dawley (SD) rats (P7–10) were cut and 400 μm-thick transverse slices were cultured using the membrane interface method. At 7 days in culture, organotypic hippocampal slices were subject to a focal mechanical injury by use of NYU weight-drop impactor. The impactor rod was positioned above the CA1 region of the hippocampal slices and dropped from a 6.25-mm height at the impact velocity of 0.35 m/s. Resultant cell death was identified 2 and 24h later with fluorescent dye propidium iodide (PI). To investigate whether MSC can decrease the cell death after the impact, immediately after the injury some of the damaged slices received MSC isolated from the 8-week old SD epididymal adipose tissue (ADMSC). Prior to transplantation, ADMSC were labeled with the green fluorescent dye (CMFDA) and 10000 cells were deposited directly on the surface of mechanically damaged slices.
The drop of the impactor rod caused hippocampal tissue destruction below the point of impact. The PI fluorescent – dead cells in the border zone around the destructed tissue were identified at 2h after injury. At 24h post-injury, PI staining of the dead cells spread out from the site of the primary trauma. Preliminary results of PI uptake measurements suggest that ADMSC have the ability to decrease cell death in mechanically injured slices.
In future studies, we will further explore ADMSC protective effects and mechanisms through which these cells decrease the neuronal death after mechanical impact. These data could be the groundwork for the development of autologous or allogeneic stem cell therapy for TBI.
Supported by the Department of Neurosurgery, Medical College of Wisconsin.
brain injury, stem cells, neuroprotection
INJURY INCREASES EXTRASYNAPTIC NR2B-CONTAINING NMDA RECEPTOR-MEDIATED CURRENTS IN CORTICAL NEURONS
Paulette B. Goforth, Ph.D., University of Michigan
Leslie S. Satin, Ph.D., University of Michigan
We previously found that in vitro TBI increases NMDA receptor activity immediately and 4 hours after injury. Our objective here was to determine which population(s) of NMDA receptors (NR2A vs. NR2B containing) mediate NMDA receptor up-regulation and whether extrasynaptic vs. synaptic receptors are involved.
Co-cultures of rat cortical neurons and astrocytes were plated on flexible membranes and a pulse of pressurized air was used to stretch the membrane and cells. Stretch injury occurred 15–18 days in vitro. Synaptic function was examined using whole cell patch clamp recordings immediately after injury.
Moderate stretch resulted in minimal cell death 30 min to 4 hours after injury (<18%). Recordings from neurons immediately after injury showed that their whole cell NMDA-elicited currents were increased, while the contribution of NMDA receptors to spontaneous miniature excitatory synaptic currents was unaltered. Increased NMDA receptor activity was predominantly due to NR2B-containing NMDA receptors as it was blocked by the selective NR2B antagonist Ro25-6981 (1μM). This appears to be due to an increase in extrasynaptic NMDA receptor function.
Moderate stretch injury produced minimal cell death but altered NMDA receptor function in surviving cortical neurons. Injury appears to potentiate neuronal NMDA receptors containing NR2B-subunits. The data are consistent with reports of increased NMDA receptor activity and a shift towards NR2B-containing receptors after in vivo TBI.Up-regulation of these receptors may alter neuronal plasticity and neuronal death after TBI.
This work was supported by NS069629 (C.R. Ferrario) and 5R01NS049519-05 (L.S. Satin).
NMDAR, glutamate, cortex, synaptic transmission
PYROPTOSIS IS A NOVEL CELL DEATH MECHANISM INDUCED BY HOST NUCLEIC ACIDS RELEASED AFTER TRAUMATIC BRAIN INJURY
Juan Pablo de Rivero Vaccari, PhD, Dept. of Neurological Surgery, The Miami Project to Cure Paralysis, University of Miami Miller School of Medicine
Gordon Dale, B.S., University of Miami
Doris Nonner, B.S., University of Miami Miller School of Medicine
W. Dalton Dietrich, PhD, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Robert W. Keane, PhD, Department of Physiology and Biophysics, University of Miami Miller School of Medicine
Host nucleic acids are released from necrotic cells and act as damage-associated molecular patterns (DAMPs) that activate the absent in melanoma-2 (AIM2) inflammasome, which cleaves caspase-1 and interleukin-1beta. The AIM2 inflammasome also induces pyroptosis, a caspase-1 dependent inflammatory mode of cell death.
dsDNA was purified from 200μl of cerebrospinal fluid (CSF) from TBI patients and non-trauma controls and quantified using a PicoGreen assay. Cortical neuron cultures were prepared by dissecting cerebral cortices from E18-E19 rat embryos. Neurons were incubated in 50% media-50% CSF from a TBI or non-trauma control patient or stimulated with dsDNA. Lysates and supernatants were immunoblotted for AIM2 inflammasome proteins and interleukin-1beta. Neuronal lysates were immunoprecipitated with anti-AIM2, IgG, or preimmune serum. DNA from the immune complex was subjected to PCR to determine if the fragment used to stimulate the neurons was bound to AIM2. Biochemical assay of the pyroptosome was performed followed by a cell viability assay. Small and large membrane impermeable dyes were used to determine pore formation. To establish caspase-1 dependency, neurons in which Ac-YVAD-CMK was added prior to simulation with dsDNA were run in parallel.
Cell-free nucleic acids are elevated in the CSF of TBI patients compared to non-trauma controls. Incubating neurons with CSF from a TBI patient induces an innate immune response characterized by caspase-1 activation. dsDNA stimulates an inflammasome in cortical neurons, as evidenced by caspase-1 activation in lysates and interleukin-1beta cleavage and release in the supernatants. Coimmunoprecipitation shows AIM2 forms an inflammasome in cortical neurons that activates caspase-1 upon binding to dsDNA. A modified chromatin immunoprecipitation assay shows AIM2 binds a cell-free, dsDNA fragment used to stimulate an innate immune response in cortical neurons. In addition, activation of the AIM2 inflammasome by dsDNA induces oligomerization of ASC, indicating assembly of the pyroptosome, a supramolecular complex of ASC oligomers that serves as a platform for caspase-1 activation in pyroptosis. There is also a significant reduction in neuronal cell viability after stimulation with dsDNA. There is a significant increase in the uptake of YoPro, a small 629 Da dye, while a larger 1293 Da dye, ethidium homodimer-2, is excluded in neurons stimulated with dsDNA, consistent with the formation of small, 1–2 nm pores characteristic of pyroptosis Pretreatment with YVAD blocked cell death and significantly decreased the uptake of YoPro, indicating pore formation and cell death in response to activation of the AIM2 inflammasome by dsDNA depends on caspase-1.
dsDNA is a known damage-associated molecular pattern that initiates an innate immune response by activating the AIM2 inflammasome. Here we show dsDNA is released from necrotic cells following injury and is acutely elevated in the CSF of TBI patients. The AIM2 inflammasome is present in cortical neurons and mediates the inflammatory response to cell-free nucleic acids. Upon stimulation with dsDNA, cortical neurons undergo caspase-1-dependent cell death characterized by assembly of the pyroptosome, formation of discrete 1–2 nm plasma membrane pores, and release of inflammatory cytokines. These features are consistent with pyroptosis, an inflammatory cell death mechanism. Characterization of this process yields novel insight into innate immunity after injury controlled by neurons. The implications of this study are widespread, as identification of a DNA-sensing inflammasome yields new insight into other CNS pathologies in which DNA stimulates an immune response, including autoimmune disease, infection, and stroke.
NINDS F31 and Lois Pope Development Award
inflammation, cell death
AMNION DERIVED MULTIPOTENT CELLS INHIBIT LIPOPOLYSACCHARIDE-INDUCED MICROGLIA ACTIVATION IN VITRO VIA CELL-CELL CONTACT
Jitendra Dave, Ph.D., Walter Reed Army Institute of Research
Zhiling Liao, M.S., Walter Reed Army Institute of Research
X-C May Lu, Ph.D., Walter Reed Army Institute of Research Stemnion Inc, N/A, Stemnion Inc
Frank Tortella, Ph.D., Brain Trauma Neuroprotection & Neurorestoration Branch, Center of Excellence for Psychiatry & Neuroscience, WRAIR
Deborah A. Shear, Ph.D., Walter Reed Army Institute of Research
Traumatic brain injury triggers acute neuroinflammatary responses. Activation of microglia plays a key role in posttraumatic neuroinflammation. In this study we used a well-established microglia inflammation model to evaluate the potential anti-inflammatory effects of amnion-derived multipotent progenitor (AMP) cells and amnion-derived cellular cytokine suspension (ACCS).
AMP cells and ACCS were tested in a human microglial cell culture model of inflammation. Human micro glia (106/43.2 cm2) were thawed and cultured in 50:50 DMEM: F-12 supplemented with 5 FBS and 10 ng/ml of M-CSF for 4 days to reach 70–80 confluence. Fluororuby (10 μg/ml) was added to the medium on Day 3 to label the microglial cells. On Day 4, fluororuby-labeled microglia were washed once with DMEM and exposed to lipopolysaccharide (LPS) (0.2 μg/ml in DMEM; 24h) insult. At the same time, either AMP cells (500/1.8 cm2-200000/1.8 cm2), ACCS (1:50 1:400) or human fibroblast (HFF) cells (control condition; 500/1.8 cm2-200000/1.8 cm2) were added to culture system and incubated for 24h. After 24 hours, the cultures were fixed, imaged, and ramified microglial cells were quantified.
Under normal conditions, microglial cells display ramified appearance. When activated inresponse to injury or insult, they retract their branches and present thick amoeboid shape.In the present study we used a low (0.2 μg/ml) concentration of LPS to induce a slow rate of microglial activation in order to maximize the potential anti inflammatory effects of AMP cells and ACCS. Preliminary analysis showed that co-culture with AMP cells reduced LPS-mediated microglial cell activation in a dose-dependent manner (minimal 8000/1.8 cm2 required). By contrast, neither ACCS treatment nor HFF co-cultures affected LPS induced microglia activation.
Our previous studies have shown that either AMP cells or ACCS treatment significantly reduced axonal degeneration and/or functional deficits in a traumatic brain injury rodent model. The results from the present study suggest that inhibition of microglial activation may be mediated by a reaction taking place in the AMP cells in response to the inflammatory environment that does not occur when treating with ACCS alone. These findings are consistent with previous research demonstrating that AMP cells (but not ACCS) were effective in modulating antigen-induced mononuclear cell proliferation (Banas et al, 2008). Based on these results, we hypothesize that the in vivo neuroprotective effects of AMP cells and ACCS may be mediated by different mechanisms in the presence of microglial activation. Further work is ongoing to test this theory.
This research was funded by the Army Combat Casualty Care Research Program and CDMRP (W81XWH-08-2-0127).
Amnion Derived Multipotent Cells, microglia
USING BIOMARKERS TO DISTINGUISH BETWEEN ISCHEMIC AND HEMORRHAGIC STROKE IN RATS
Changhong Ren PhD, Capital Medical University
Joy Guingab, PhD, Banyan Biomarkers
Boxuan Yang, PhD, Banyan Biomarkers
Firas Kobeissy, PhD, Banyan Biomarkers
John Anagli, PhD, Banyan Biomarkers
Jixiang Mo-Seany, Banyan Biomarkers
Dancia Scharf, MS, Banyan Biomarkers
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers
Yuchuan Ding, MD, PhD, Wayne State University
Ronald Hayes, PhD, Banyan Biomarkers
Jackson Streeter, MD, Banyan Biomarkers
Xunming Ji, MD, PhD, Capital Medical University
There are two primary categories of stroke, ischemic and hemorrhagic, which have fundamentally different mechanisms and treatment options. These two types of stroke were modeled in rats, with the aim of identifying biomarkers that distinguish between them.
Ischemic stroke was induced by permanent middle cerebral artery occlusion (MCAO) without reperfusion, and hemorrhagic stroke (ie, intracerebral hemorrhage) by injecting 1 ml of saline containing 0.5 U collagenase VII-S. Brain hemispheres and biofluids were collected at two time points, 3 and 6 hours (3h, 6h) after stroke. Two approaches were used for biomarker analysis, targeted and proteomics. In the targeted approach, known biomarkers were tested on the rat samples via quantitative immunoblotting (injured brain, CSF) and Banyan's proprietary ELISA assays (CSF, serum). In the proteomics approach, mass spectrometry (LC-MS/MS) was used on the 3h injured brain lysates to identify brain proteins that were differentially expressed between sham and each type of stroke. Based on the differentially expressed proteins, systems biology was conducted to identify associated cellular processes.
In injured brain, quantitative analyses revealed that αII-spectrin breakdown products (SBDP150, SBDP145) were strongly increased after 6h ischemia. Levels of total GFAP, and the calpain-cleaved 38 kDa GFAP fragment, were significantly increased in 6h ischemic brains. In CSF, SBDP145, GFAP and UCHL1 were elevated after 6h ischemic stroke, as seen both by westerns and ELISA. In serum, GFAP and UCHL1 levels were increased after 3h and 6h of ischemia. After LC-MS/MS analysis of the 3h brain lysates, 38 proteins were identified to be differentially expressed between sham and ischemic, while 86 were expressed differently between sham and hemorrhagic. Globally, ischemic stroke proteins were related to cell death, ischemia, inflammation, oxidative stress, caspase activation, and apoptosis. Hemorrhagic stroke proteins were involved in autophagy, ischemia, necrosis, apoptosis, calpain activation, and cytokine production.
In brain and biofluids, SBDPs, GFAP and UCHL1 were elevated after ischemic stroke but not hemorrhagic. Thus these biomarkers behaved differently in the two stroke models and may be capable of differentiating between them. Based on proteomics and systems biology, the neuroproteomes of the two types of stroke had some overlap, but many distinct elements, offering promise for discovering new biomarkers to distinguish between ischemic and hemorrhagic stroke.
Thanks to the staff of Banyan Biomarkers for their support.
Biomarkers, stroke, ischemic, hemorrhagic
BRACHIAL PLEXUS INJURY IN CHILDREN: A STUDY OF 33 PATIENTS FROM LEVEL I APEX TRAUMA CENTRE OF INDIA
Dr. Sumit Sinha, MBBS MCH, AIIMS
Dr. Deepak Gupta, MBBS, MS, MCH, All India Institute of Medical Sciences, New Delhi, india
Dr. Kanwaljeet Garg, MBBS MCH, AIIMS
Brachial plexus injuries usually result in a severe permanent handicap. These injuries represent a surgical challenge because of the complex anatomy of the brachial plexus, which is usually complicated by trauma induced changes.
The purpose of our study was to study all the patients of <18 yr of age with brachial plexus injury operated between april 2008 to march 2012 at level I apex trauma centre in India. We retrospectively analyzed various factors including mode of injury, type of injury, surgery performed and recovery following surgery. Thirty three patients, aged≤18 years were operated at our centre for brachial plexus injury in 4 year period. Mean age at presentation was 15.1 (range 4–18 years). Boys constituted 79 (n=26) of our patient population and girls constituting 21 (n=7). High velocity injury was the commonest mode of injury, responsible for 82 of cases. Other modes of injury included crush injury, fall or gunshot injury. Associated head injury and long bone injury was noticed in 12 and 33 of patients, respectively. Panbrachial injury was the commonest, observed in 85 (n=28) of patients. Others included 2 patients of cord injury and 3 patients of trunk injury.
Mean duration between injury and surgical intervention was 6 months (range 2–13 months, SD±2.6 months). Two thirds of the total patients underwent neurotization and one third underwent neurolysis alone. Mean follow up was 30.9 months (range 2–48 months). Twenty one patients could be followed up, and 60 patients out of those who could be followed up showed improvement in muscle power, (MRC grade≥3/5).
High velocity trauma is the most common mode on injury. Neurotization was the most commonly performed surgery and around 40 of the patients of total series and 60 (12/60) who had followed up showed improvement at a mean follow up of 31 months.
None
Brachial plexus, pediatric, trauma
NEURONAL MYOSIN-X IS UPREGULATED AFTER PERIPHERAL NERVE INJURY AND MEDIATES LAMININ-INDUCED GROWTH OF NEURITES
Johan Zelano, MD PhD, Uppsala University
Professor Mårten Risling, MD PhD, Karoinska Institutet
Professor Staffan Cullheim, MD PhD, Karoinska Institutet
Interaction between the extracellular matrix and the neuronal growth cone (via e.g. integrins) is a prerequisite for successful axonal regeneration after injury. The integrin-binding protein Myosin-X has been shown to be important for extension of filopodia in endothelial cells, but its function in the nervous system is unknown.
To examine the neuronal expression of Myosin-X after injury, we used in situ hybridization and immunohistochemistry and examined injured neurons using several experimental lesion paradigms (peripheral nerve injury, dorsal root transection, rubrospinal tract injury and cortical stab wound). Further, cellcultures of DRG neurons and motoneurons were used to study neurite outgrowth. We cultured cells on different substrates (laminin and N-CAM) and, by use of RNAinterference, examined the role of Myosin-X in neurite growth.
Expression of myosin-X mRNA is upregulated in adult rat sensory neurons and spinal motoneurons after peripheral nerve injury, but not after a spinal cord injury injury (transection of the rubrospinal tract) or mechanical brain injury (stab wound). Myosin-X is localized to neuronal growthcones and knocking down the protein using siRNA reduces neurite growth on laminin, but not on N-CAM.
Our studies indicate that Myosin-X is upregulated by neurons after lesions where a regenerative response is generally observed (such as peripheral nerve injury), but not in “non-regenerating” injuries (spinal cord injury, brain injury) and is crucial for integrin-mediated axonal outgrowth. These results also suggest that forced (over-) expression of Myosin-X in CNS neurons could be a potential strategy to improve regeneration after TBI or SCI.
Grant support: The Swedish Research Council, Marianne and Marcus Wallenbergs stiftelse, Hjärnfonden and Karolinska Institutet.
PNS, Spinal cord, TBI, ECM
DIFFUSION TENSOR IMAGING METRICS IN SUBACUTE TRAUMATIC CERVICAL CORD INJURY
Michael B. Jirjis, BS, Marquette University
Brian D. Schmit, PhD, Marquette University
John L. Ulmer, MD, Medical College of Wisconsin
Marjorie C. Wang, MD, MPH, Medical College of Wisconsin
Shekar N. Kurpad, MD, PhD, Medical College of Wisconsin
Few clinical studies have documented the profile of DTI metrics beyond the acute phase of spinal cord injury (SCI). The purpose of this study was to evaluate the role of diffusion tensor imaging (DTI) in patients with traumatic cervical SCI 2–7 days post-injury.
With appropriate IRB approval, seven patients (5 men, 2 women) with traumatic blunt cervical cord injury underwent DTI at a median duration of 99 hours post-injury. DTI indices were calculated for the cervical cord (C1-T1), upper cord (C1-2), injury zone (the level of injury) and the lower cord (C7-T1). DTI metrics of the cervical cord in 14 neurologically intact controls were used for comparison. DTI data were also compared to that provided by T2-weighted magnetic resonance images (T2W) to see if DTI provided additional information as to the extent of SCI.
The fractional anisotropy (FA) throughout the cervical cord (C1-T1) was reduced, while the anisotropy index (AI) was increased in patients as compared to controls. Similar changes in the FA and AI were seen at the injury zone (p=0.001, 0,007) and in the lower cord (p=0.007, 0.01), but not in the upper cord. Decreased FA values were also noted at cervical levels below the lower margin of T2W hyperintensity, indicating that DTI detected structural changes in caudal regions of the cervical cord that appeared normal on T2W images.
FA and AI are sensitive parameters in the assessment of patients with cervical cord injury 2–7 days post SCI. DTI metrics are useful in defining the extent of cord injury and show changes beyond those seen on T2W MR images. DTI metrics may also be useful to determine spinal cord plasticity distal to the injury site as early as 2–7 days post SCI.
Sponsors: 1. VA Rehab R&D grant #1 I01 RX000113-01, 2. Bryon Riesch Paralysis Foundation Endowment
DTI; spinal cord injury
SPINAL CORD CONTUSION INJURY DECREASES GLUTAMATE UPTAKE MEASURED IN VIVO USING MICROELECTRODE ARRAYS
Juan Anaya, BS, University of New Mexico Biomedical Engineering Program
George Quintero, PhD, University of Kentucky, Dept Anat & Neurobiol, Center for Microelectrode Technology
Greg Gerhardt, PhD, University of Kentucky, Department of Anatomy and Neurobiology
Excitotoxic injury and oxidative damage are two contributors to secondary tissue damage following spinal impact injury. We hypothesize that induction of oxidative events after impact injury in a rat model inactivate glutamate transporters, which increases extracellular glutamate concentrations and exacerbates excitotoxic tissue damage.
Amperometric recording with enzyme-based microelectrode arrays was used to measure extracellular glutamate clearance parameters in urethane anesthetized female 225–250 g Sprague Dawley rats after 1) control laminectomy, 2) a 200 kdyn severe impact injury to the dorsal cord surface at the T-10 vertebral level delivered with the Infinite Horizons® device and 3) after a single i.p. injection of 300 mg/kg tempol, a spin trap agent that sequesters peroxynitrite radicals, injected within five minutes post injury. Measures with calibrated microelectrodes were taken beginning 5 minutes after cord contusion in the injury site and continued up to 120 minutes post injury from a depth of 400–750 microns in T11 cord dorsal grey matter.
Average uptake rates (k−1) at 105–120 minutes after injury were significantly reduced by 58% and 61% in injury and injury+tempol, groups respectively, compared to control laminectomy. Conversely, while decline of the signal to 80% of maximum amplitude (T80) was significantly prolonged 2.9 fold in injured compared to control laminectomy rats at 105–120 minutes after injury, tempol administration normalized T80 in injured rats to within 36% of control laminectomy values.
The spinal cord injury reduced uptake rates and prolonged T80, indicative of reduced glutamate clearance by glutamate transporters in the spinal cord. Tempol given within 5 minutes of injury nomalized the prolonged T80 without affecting uptake rates in the injury condition. The changes induced by tempol suggest that glutamate clearance, attributed primarily to excitatory amino acid transporter function, is significantly reduced due to an oxidative inactivation of the transporters.
Supported in part by Award Number R21NS050430 from the NINDS and the Dedicated Health Research Funds from the UNM School of Medicine, C-2315-RAC.
Glutamate transport, EAAT, MEA, amperometry
SLEEP DISRUPTION IN A RODENT MODEL OF SPINAL CORD INJURY
Olivia Lied, Department of Chemistry, Susquehanna University, Selinsgrove, PA
Emily Swartz, BS, Penn State University College of Medicine
Emily Qualls-Creekmore, BA, MA, Comparative Biomedical Sciences, Louisiana State University School of Veterinary Medicine
Stefany Primeaux, BS, MA, PhD, Louisiana State University College of Medicine
Spinal cord injury (SCI) produces dramatic, life-long, impairment of multiple basic physiological processes, including the regulation of sleep. Our preliminary observations suggest similar impairments after experimental SCI. The ventrolateral preoptic nucleus (VLPO) has been implicated in rapid sleep/wake state transitioning. We hypothesized that VLPO activity is diminished after SCI.
Experiment 1: Male Wistar rats (200–300 g) were injured at the 3rd Thoracic spinal level. Controls received a laminectomy only. Animals were monitored for sleep/wake behaviors 3 hours into the light cycle. After for a period of 60 minutes, animals were deeply anesthetized and transcardially perfused. Using a standard protocol, immunohistological expression of c-Fos was visualized in the VLPO. Only c-fos positive cells which were above the 25th percentile in optical density threshold were counted.
Experiment 2: Rats were implanted with EEG and EMG recording electrodes for collection of pre-operative sleep baselines. Sleep stages were recorded continuously over two consecutive days. A lack of EMG, recorded from neck musculature, was used to determine to confirm REM-associated EEG patterns. Two to three weeks later, SCI or control surgery was performed and sleep patterns monitored 3, 14 and 21 days post-op.
In contrast to control animals, SCI rats display fewer behavioral sleep patterns during the early light phase. Rather than exclusive somnolence or wakefulness observed in control animals, SCI rats display frequent transitions between behavioral sleep/wake states. In these same animals, the average number of c-fos positive VLPO neurons are significantly higher in the uninjured control rats. (p<0.05).
Our preliminary EEG analysis indicates that rats with SCI experience greater sleep fragmentation than uninjured controls. As is characteristic of awake animals, SCI rats have higher levels of high frequency, low amplitude EEG in the absence of atonia. This data provides electrophysiological confirmation of our observations in Experiment 1.
Pontine cholinergic and monoaminergic neuronal groups are considered to be essential in ascending arousal systems. The reciprocal connections between the VLPO and these pontine nuclei form a bistable neural switch (also referred to as a flip-flop switch) which rapidly gates between sleep and wake states. Our behavioral analysis of control and SCI rats supports our earlier observation that SCI rats display fewer periods of somnolence during the light cycle. We propose that a loss of ascending information from below the spinal lesion center (after SCI) may be one mechanism which leads to an imbalance of neurotransmitters operating in the flip-flop switch between the VLPO and the pontine arousal networks. The precise levels of the neurotransmitters (for example, norepinephrine, serotonin GABA and orexin) in SCI rats could lead to an increased inhibition of the VLPO.
Support: CDMRP SC090514
arousal sleep hypothalamus EEG
TAIL NERVE ELECTRICAL STIMULATION COMBINED WITH SCAR ABLATION AND NEURAL TRANSPLANTATION IMPROVES FUNCTIONAL RECOVERY IN RATS WITH CHRONICALLY CONTUSED SPINAL CORD
Fengfa Huang, Spinal Cord Society Research Center
Mary Gates, Spinal Cord Society Research Center
Eric Holmberg, PhD, Spinal Cord Society Research Center
Treatments that lead to functional recovery from chronic spinal cord injury (SCI) have shown little progresses to date. Here we report that combination treatments with tail nerve electrical stimulation (TANES) significantly improves locomotor function in rats with chronic SCI.
TANES is a noninvasive physical therapy, which apparently activates the central pattern generator (CPG), inducing active weight-supported stepping in chronically spinal cord injured rats. Female, adult Long-Evans rats were subject to contusion injury at T10 produced by using the NYU impactor device (25 mm height setting). Basic treatments include rose Bengal-produced scar ablation followed by transplantation of a mixture of lamina propria of olfactory mucosa and cultured olfactory ensheathing cells into the lesion cavity in two subject groups 6 wks after SCI. Rats both with and without basic treatments received TANES 1 wk after secondary surgery or 2 wks after SCI. Subjects were assessed with BBB open field rating scale, horizontal ladder test, electrophysiological tests, histology, immuohistochemistry, and neurotracing.
Sixteen weeks after secondary surgery or 22 wks after SCI, rats in two groups receiving TANES significantly improved their locomotor outcome compared with those without TANES. TANES may promote the activity-dependent plasticity, which has been considered a major contributor to functional recovery through activation of the CPG. Additionally, TANES may also promote axonal regeneration, including those from the brain and brain stem.
Our results demonstrate that TANES has considerable potential for aiding in functional recovery in the rat SCI model.
This study was supported by the Spinal Cord Society, USA.
contusion; scar; OECs; recovery
THE EFFECTS OF VENLAFAXINE ON FUNCTIONAL RECOVERY AFTER SPINAL CORD INJURY
Candace Floyd, PhD, University of Alabama at Birmingham
Spinal Cord Injury (SCI) can significantly alter a person's physical and psychological health. Depression in persons with SCI is often treated with antidepressants which alter serotonergic and adrenergic signaling, both critical regulators of plasticity. Yet the effect of antidepressants on plasticity and functional recovery after SCI has never been evaluated.
The aim of this study was to evaluate the effects of venlafaxine (VEN) administration on functional recovery using a SCI rodent model. Venlafaxine is a serotonin-norepinephrine reuptake inhibitor. Adult male Sprague-Dawley rats were subjectedto amoderate contusion SCI at the T10 level. Therapeutic intervention began at day 30 post-SCI and continued for 30 days, after which the animals were euthanized and spinal tissue was harvested for histological evaluation. The treatment groups consisted of daily VEN administration or vehicle (VH), and swimming rehabilitation (RHAB) such that groups were VEN, VH, VEN+RHAB, or VH+RHAB. The animals were evaluated weekly for functional recovery of hindlimb locomotion using Basso, Beattie, and Bresnahan open-field test (BBB), the CatWalk® kinematic analysis, and the Louisville swim scale (LSS) with evaluations staggered to avoid fatigue. Depression was evaluated with the Forced Swim test prior to and after the therapeutic interventions.
We observed interesting trends in the functional recovery of the animals on both LSS and BBB. No significant difference was noticed between the groups which did not receive rehabilitation (VEN vs. VH), suggesting no main effect of venlafaxine treatment. Animals that received RHAB+VH showed improved locomotor recovery compared to the non-RHAB groups (VEN or VH). However, the animals that received both RHAB and VEN showed decreased BBB and LSS scores as compared to any other group, suggesting that VEN treatment hinders the spinal plasticity and locomotor recovery gained from RHAB. Another interesting and clinically relevant finding seen in both VEN and VEN+RHAB treatment groups is a significant increase in the incidence of priapism when compared to the VH and VH+RHAB groups. We hypothesize this novel finding is due to the depletion of NE at sympathetic inputs to the penile tissues, a result of the reuptake inhibition of norepinephrine by VEN. Evaluation of the effects of VEN or RHAB on depression-like behaviors is ongoing. Similarly histological markers of plasticity are currently being evaluated in the spinal cords of all animals.
Taken together, these data suggest that RHAB is beneficial after SCI by improving locomotor function. However, administration of the antidepressant venlafaxine during rehabilitation may prevent the potential gains from the therapy. Also, administration of a NE reuptake inhibitor such as VEN may lead to priapism, an unwanted and possibly dangerous side effect.
Supported by DoD CDMRP W81XWH-10-1-0839, UAB CCTS 5UL 1RR02577, and TL1RR025775.
depression, plasticity, serotonin, norepinephrine, priapism
CONDITIONAL OVEREXPRESSION OF INSULIN-LIKE GROWTH FACTOR-1 IMPROVES WHITE MATTER SPARING AND FUNCTIONAL OUTCOME FOLLOWING SPINAL CORD INJURY
Benjamin Tuttle, B.S., University of Kentucky/SCoBIRC
Chen-Guang Yu, Ph.D., University of Kentucky/SCoBIRC
James Geddes, Ph.D., Department of Anatomy and Neurobiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Kathryn Saatman, Ph.D., Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky
Insulin-like growth factor-1 (IGF-1) promotes cell survival, neurogenesis, synaptogenesis and myelination. IGF-1 has been shown to stimulate motor neuron axon outgrowth in vitro and increase growth of supraspinal axons after spinal cord injury (SCI). We hypothesized that IGF-1 would promote functional recovery and tissue sparing following SCI.
Transgenic mice with conditional overexpression of IGF-1 (IGF-1Tg) driven through the GFAP promoter were used to enhance delivery of IGF-1 to damaged regions of the cord undergoing reactive astrocytosis. Moderate spinal cord contusion (50 kdyn force) was performed at a T-10 level using a spinal impactor (n=8 IGF-1Tg and 9 wildtype). Sham mice received a laminectomy at T-10 (3 shams/genotype). Behavior was tested using the Basso Mouse Scale (BMS) on multiple time points ranging from 0–42 days, after which spinal cords were harvested for histological analysis. Lesion area and white and gray matter sparing was measured in eriochrome cyanine stained 20μm sections taken every 200 μm up to 800μm on either side of injury epicentre. Immunostaining for GFAP was used to visualise astrocytosis.
While recovery of motor function observed with simple and coordinated motor tasks was equivalent over the first week after SCI when assessed using BMS, IGF-1Tg mice exhibited improved function between 2 weeks to 6 weeks postinjury (p<0.05 at 2, 5 and 6 week for simple motor tasks and p<0.05 at 6 week for co-ordinated motor function). The size and rostro-caudal distribution of the lesion and the area of gray matter sparing were equivalent for injured WT and IGF-1Tg mice. However, white matter sparing was significantly greater in IGF-1Tg mice compared to WT mice when measured in spinal cord sections up to 800 μm rostral or caudal to the epicentre. Maximum white matter sparing was observed in the rostral- and caudal-most sections. Astrocytic GFAP immunoreactivity was distributed mainly in the white matter in sham mice of either genotypes. Injured spinal cords exhibited increased GFAP staining with glial scar formation. Qualitative assessment suggested that GFAP-positive cell numbers were higher in injured IGF-1Tg spinal cords.
These studies demonstrate that IGF-1 overexpression limits white matter loss following SCI. However, IGF-1 overexpression had no significant effect on lesion volume or gray matter sparing at 6 weeks, suggesting that motor neuron survival was not enhanced by IGF-1 after injury. However, the method we used only measures gross tissue sparing. Stereological counting of spared/survived neurons is necessary to reach a definitive conclusion regarding neuronal survival. Increased GFAP staining in injured IGF-1Tg spinal cords may be due to autocrine actions of IGF-1 on astrocytes. Further experiments are necessary to quantify this enhanced astrocytic response. Behavioural improvements may be due to the preservation of spinal cord white matter in the IGF-1Tg mice. Positive effects of IGF-1 overexpression on white matter and behavioural outcomes suggest its therapeutic potential for use in SCI.
Supported by NIH grants NS058484, NS072302, and NS051220, and Kentucky Spinal Cord and Head Injury Research Trust grant 7–20.
IGF-1, astrocytosis, behaviour, locomotor function
ROLE OF NEURONAL NUCLEAR FACTOR KAPPA BETA IN MODERATE SPINAL CORD INJURY
Kate Lykke Lambertsen, Cand. Science, PhD, University of Southern Denmark, Department of Neuroscience
Ditte Ellman, Cand. Science, University of Southern Denmark, Department of Neuroscience
John Bethea, PhD, The Miami Project to Cure Paralysis
Henrik Schrøder, MD, PhD., Doctor of Medicine, University of Southern Denmark, Clinical Pathology
Louise Jørgensen, Cand. Science, PhD, University of Southern Denmark, Clinical Pathology
Miriam Mohamed, University of Southern Denmark, Department of Neuroscience
This ongoing study investigates the extent of secondary tissue damage, following spinal cord injury in transgenic, IKK2 deficient mice. Secondary tissue damage results in cell death, denervation and paralysis. Previously, nuclear factor-kappa B (NF-κB) inhibition in astrocytes has been shown to improve functional recovery and reduce secondary injury in mice.
In the present study, neuronal NF-κB deficiency is achieved by inhibiting IKK2, a kinase essential to NF-κB activation, using loxP flanked IKK2 and edition by the neuron specific recombinase, SynCre. Behavior, locomotor- and neuromuscular function, was assessed using rotarod, grip strength, Y-maze and open field with computerized SMART tracking system. Neuronal and striated muscle morphology are currently being evaluated using EM, while H&E stained sections provide morphology of liver, spleen, lung and small intestine. T9 moderate spinal contusion is achieved using the Infinite Horizon Device. Locomotor performance is scored using the Basso Mouse Scale (BMS) in the open field, performed day 1 to verify lesion quality and every week to evaluate functional recovery. Mice are euthanized days 1, 3 and 8 weeks. Western Blot evaluating expression of the proteins Cre, IKK2 and NF-κB is performed on all survival periods, while stereological lesion size approximation is performed on 8 weeks only.
So far, phenotypical characterization of our TG mice has shown that these display a normal breeding pattern and that peripheral organs display apparent normal histological structures. In addition, TG and WT littermate mice display similar activity and anxiety levels in the open field test and motor function seems to be unaffected by neuronal NF-κB deficiency. The muscular evaluation does show signs of chronic leukocyte infiltration, in both TG and WT animals. We are awaiting spinal cord EM, Functional recovery, lesion size and protein expression.
The role of NF-κB in neurons exposed to mechanical injury is yet to be explored. Significant findings could provide advances in the identification of a future therapeutic target. Such a target would be relevant for post trauma treatment, thus reducing cell death and improving overall outcome, of a still untreatable condition.
Carlsbergfondet
Secondary Injury, Transgenic, Inflammation, Rehabilitation
BRAINBOOK: A STATEWIDE PILOT FOR EDUCATING HIGH SCHOOL ATHLETES ABOUT CONCUSSION
Nicholas Theodore, MD, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center
Yashar Kalani, MD, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center
Angela Barrus, PhD, Arizona State University
Robert Christopherson, PhD, Arizona State University
Renee Pilbeam, BS, Arizona State University
Quincy Conley, MS, Arizona State University
Brian Nelson, PhD, Arizona State University
Annually, 3.8 million sports related concussions occur, the majority in youth and adolescents. Concussion is associated with a higher incidence of decline in cognitive skills, adaptive functioning and academic performance. Through legislative and administrative channels, we developed an interactive educational module and tested its efficacy on student-athletes in Arizona.
We developed a comprehensive, age-appropriate e-learning module called Brainbook. This educational program was deployed along with a 12-item questionnaire to determine prior attitudes/awareness about concussions. A post-attitude survey was administered to determine if participation in this module influenced attitudes and awareness.
During 2011–2012 academic year, a total of 80,250 participants (45,876 male (57%) and 34,374 female (43%) enrolled in the Brainbook e-module on concussion education. The participants were between the ages of 13–19. Out of 26 possible points, the average pre- and post-test scores were 16.9 and 18, respectively. Using a Wilk's lambda analysis, this was noted to be a significant result. Moreover, participants reported increased awareness and attitudes underlining the seriousness of concussion.
Despite a small effect size, the Brainbook e-learning module had a significant impact on teaching student-athletes in the state of Arizona about concussion. Our results suggest that the students have a fair baseline awareness of mTBI, but that this and similar learning modules can enhance the education and prevention of TBI in this high-risk group.
The authors would like to thank the Arizona Interscholastic Association and the Arizona Cardinals for their cooperation in the development and implentation of this project.
mTBI, concussion, education
NEUROMEDIATOR STRUCTURES DAMAGING IN CLINICAL EVALUATION OF SEVERE TRAUMATIC BRAIN INJURY
Nataliya Zakharova, Ph.D., Burdenko NSI
Oleg Zaitsev, Ph.D., Burdenko NSI
Valeriya Tenedieva, Ph.D., Burdenko NSI
Yuryi Vasilevich. Vorobiov, Ph.D., M.D., Burdenko NSI
Ekaterina Sokolova, M.D., Burdenko NSI
Alexander Sychev, Ph.D., Burdenko NSI
Andrey Oshorov, Ph.D., Burdenko NSI
Alexander Polupan, M.D., Burdenko NSI
Anton Gavrilov, Ph.D., Burdenko NSI
Prof. Alexander Alexandrovich Potapov, Academician of RAMS and RAS, Burdenko NSI
Close interactions between the neuromediator systems of brainstem, subcortical structures and cortex play a crucial role in awareness, behavior, cognitive and motor functions regulation. Therefore evaluation of different neuromediator systems damaging may provide a key to outcome prediction and development a new treatment strategy in severe traumatic brain injury (TBI).
This study included 31 patients with severe TBI (GCS<8) (17 male and 14 female) aged 18–53 (29,2±10,7 y.o.). Brain damage localization was verified by 1,5-3T MRI (T1, T2, FLAIR, T2 GRE/SWI, DWI) in all patients. We analyzed the damage to different neuromediator structures: basal ganglia (Globus Pallidus interna and externa - GPi, GPe (mainly GABA neurons), nucleus caudatus (NC) (mainly GABA and cholinergic neurons)), Thalamus (mainly glutamate neurons) and brainstem (norepinephrinergic - locus coeruleus area, dopaminergic - ventral tegmental area, substantia nigra and cholinergic system - laterodorsal tegmental area, pedunculopontine tegmental area). Plasma catecholamine (Norepinephrine-NE, Dopamine-DA, Epinephrine-E) levels were evaluated in dynamics by HPLC with electrochemical detector. Consciousness levels ranged based on the following criteria: eye opening only /– tracking (unconsciousness), obeying commands /- attempts to speak (minimal consciousness state), answering questions /- orientation (recovered consciousness). Neurological state included evaluation of movement dysfunction: paresis degree, muslular hypotony/hypertony, hypokinesia/hyperkinesia.
All patients were divided in four groups accordingly to neurological syndromes: 1) hypertony/hypokinesia (10 patients, 32,3), 2) hypertony/hyperkinesias (3 patients, 9,7), 3) hypotony/hyperkinesia (8 patients, 25,8), 4) hypotony/hypokinesia (10 patients, 32,3). 1. Damage of GPi GPe was associated with muscular hypotonia (83,3), while GPe NC or GPi Thalamus damage – with hypertony/hypokinesia syndrome (80). 2. Isolated basal ganglia (mainly GPi and/or GPe) damage (without brainstem neuromediator structures injury) associated with hyperkinetic syndromes (85,7). 3. Cholinergic brainstem area damage mainly leads to hypotony/hypokinesia syndrome (p=0,03), dopaminergic brainstem structure damaging more often was accompanied with hypertony/hypokinesia syndrome (p=0,2). 4. It tended to more unfavorable outcome (evaluated by Glasgo Outcome Scale - GOS) in patients with dopaminergic and cholinergic brainstem areas damage compared to other patients with brainstem injury (p=0,06). 5. Prolonged unconsciousness (coma, vegetative state and akinetic mutism) (two month and more) was mole likely associated with concomitant damage to three subcortical structures (GPe, GPi, NC or Thalamus) (p=0,0002), than injury of neuromediator brainstem structures. The longest unconsciousness(118,3±39,9 days) period was noticed in patients with concomitant damage of Thalamus and GP (p=0,01). 6. Two variants of plasma catecholamine (CA) dynamics were found earlier during consciousness recovery: 1) CA-dissociation, when NE and DA level changes were oppositely directed, was observed in unconsciousness; 2) CA-association - unidirectional NE and DA level changes followed by the unconsciousness regress. Duration of plasma CA-dissociation correlated with minimal Glasgo Coma Scale - GCS (r=-0,6, p=0,002) and GOS (r=-0,45, ð=0,03) evaluation, but was not related to cholinergic and/or dopaminergic brainstem structures damage.
It is well known that consciousness and neurological function recovery often depend on regular choice of neurometabolic therapy like dopaminergic and cholinergic agents. The association between brainstem neuromediator structure damage, plasma catecholamine levels and clinical features was analyzed in this study. It is interesting that concomitant inhibitor subcortical structures and cholinergic brainstem areas damage lead predominantly to hypotony/hypokinesia syndrome, isolated inhibitor subcortical structure damaging associated with hyperkinetic syndromes, while concomitant inhibitor and exciting subcortical structure damage mainly cause hypertony/hypokinesia syndrome. Prolonged unconsciousness state was associated with concomitant damage of inhibitor and exciting subcortical structures, especially for combined damage of GABAergic and glutamatergic structures. We find these data useful for choosing a specific pharmacotherapy which will provide consciousness and neurological function recovery.
grant RFBR 11-04-12166 2011
TBI, neuromediator, unconsciousness, movement dysfunction
CLINICAL AND PATHOLOGIC PHENOTYPE OF DEMENTIA AFTER TRAUMATIC BRAIN INJURY
Nasreen Sayed, MS, University of Texas Southwestern Medical Center
Carlee Culver, BA, Center for Neuroscience and Regenerative Medicine/Uniformed Services University
Flora Hammond, MD, Indiana University School of Medicine
Traumatic brain injury (TBI) is associated with increased risk of dementia. It is unclear whether TBI results in Alzheimer's disease (AD) like pathology, or in some other condition such as chronic traumatic encephalopathy (CTE). This study was undertaken to assess the clinical and pathologic features of dementia associated with TBI.
Data from the National Alzheimer's Coordinating Center (NACC) Uniform Data Set (UDS) was obtained encompassing participants from September, 2005 through December, 2010 using the Initial Packet Visit (IVP) and when available, Neuropathology Packet (NP). Categorical data was analyzed using Fisher's exact test. Continuous parametric data was analyzed using Student's t-test. Non-parametric data was analyzed using the Mann-Whitney test.
The NACC UDS database contained 877 demented individuals who had sustained TBI. Of these, 567 had brief loss of consciousness (LOC), 248 had extended LOC, and 62 had chronic deficit or dysfunction. Only TBI with chronic deficit or dysfunction was associated with increased risk of dementia. Demented patients with TBI were significantly more likely to experience depression, anxiety, irritability, and motor disorders than age and gender matched patients with probable AD. Autopsy data was available for 20 of the 62 TBI patients, and for 16 of the 122 probable AD subjects used as controls. Of the patients with TBI, 62% met NIA-Reagan “High Likelihood” criteria for AD, and 69% met CERAD criteria for Definite AD.
TBI with chronic deficit or dysfunction is associated with an increased odds ratio for dementia. Clinically, patients with dementia associated with TBI were more likely to have symptoms of depression, agitation, irritability, and motor dysfunction than patients with probable AD. The majority were diagnosed with Probable or Possible AD during life, although a substantial minority did not have a definite clinical diagnosis.
P30 AG12300 (Core B), R01 HD048179, and NIDRR H133A020526 (to RD-A) and U01 AG16976 (NACC).
Dementia; Alzheimer's; Chronic Trauumatic Encephalopathy
EFFECTIVENESS OF MEDICAL THERAPIES FOR INTRACRANIAL HYPERTENSION IN PEDIATRIC SEVERE TRAUMATIC BRAIN INJURY - PRELIMINARY RESULTS
Nikki Ferguson, MD, Critical Care Medicine, Children's Hospital of Pittsburgh of UPMC
Elizabeth Tyler-Kabara, MD, PhD, Dept. of Neurological Surgery and Bioengineering, Children's Hospital of Pittsburgh of UPMC
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Michael J. Bell, MD, Departments of Critical Care Medicine and Neurological Surgery, Children's Hospital of Pittsburgh of UPMC
Control of raised intracranial pressure (ICP) is a cornerstone of therapy for pediatric severe traumatic brain injury (TBI). Despite widespread utilization of opiates, barbiturates and hyperosmolar agents to treat intracranial hypertension, no level I evidence supports their effectiveness on managing ICP, cerebral perfusion pressure (CPP) or cerebral metabolism.
Children with severe TBI (Glasgow Coma Scale score≤8) were prospectively enrolled and had ICP measured using both an external ventricular drain and an intraparenchymal pressure monitor. Standard neurocritical care included sedation, neuromuscular blockade, mild hyperventilation, maintenance of age-appropriate CPP and continuous cerebrospinal fluid drainage. Intracranial hypertension crises (ICP≥20 mmHg for >5 min) were treated with fentanyl, 3% hypertonic saline (HTS) and pentobarbital per the bedside clinician's preferences. Infusion times were prospectively recorded. Mean values of ICP, CPP and invasive mean arterial pressure (electronically collected every 5 seconds) were calculated for epochs preceding (5 min) and following (30 min and 60 min) each individual medication administration. If multiple medications were administered for a single event, the final medication was considered effective. Medians (25–75 percentile) were compared with Kruskall-Wallis ANOVA with Dunn's test. Proportions were compared with chi square or Fischer's exact test.
Fifty-three medication administrations were available for analysis (fentanyl=26; HTS=7; pentobarbital=20). Compared to the 5 min prior to administration, mean ICP was significantly decreased in the 60 min following the administration of fentanyl (19 [15 – 21] vs. 21 [20 – 23], p<0.05) and pentobarbital (19 [17 – 19] vs. 20 [20 – 22], p<0.05). No medication significantly affected CPP or mean arterial pressure. The effectiveness of an individual dose of medication differed significantly between HTS (7/7 doses), fentanyl (10/26 doses) and pentobarbital (10/20 doses) (p=0.015). Cumulative doses of fentanyl of ≥2.5 mg/kg trended toward increased efficacy (6/8 vs. 3/11, p=0.07). Pentobarbital trended toward increased efficacy during the initial 48hrs of admission vs. after 48hrs (5/7 vs. 5/13 doses) while fentanyl trended toward increased efficacy after 48hrs of admission vs. initial 48hrs (9/18 vs. 1/8 doses).
In this prospective study, fentanyl and pentobarbital significantly lowered ICP without affecting CPP in children with severe TBI. When intracranial hypertension crises were treated with multiple agents, HTS was most consistently associated with termination of the crisis. These preliminary data suggest that efficacy and time from admission may be related for these commonly used agents. A larger study that includes correction for multiple comparisons of data from the same subject will be required to fully explore the utility of these agents for intracranial hypertension control in children.
Sponsored by T32 HD 040686
pediatric, TBI, intracranial hypertension, therapy
NEGATIVE PREDICTIVE VALUE OF MENINGEAL ENHANCEMENT ON FLAIR-POST CONTRAST MRI
Jessica L.. DeStefano, B.S., Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Martin R. Cota, B.A., Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Lawrence L. Latour, PhD, Stroke Branch, NINDS, NIH
In the THINC MRI study, the most frequent MRI abnormality seen was traumatic meningeal injury (TMI) visualized on post-contrast FLAIR. In the absence of TMI, other abnormalities were less likely to be seen. We investigated the negative predictive value of TMI for otherwise normal MRI and CT.
Patients presenting to Suburban Hospital (Bethesda, MD) and Washington Hospital Center (Washington, DC) following head trauma between October 2010 and March 2012 were enrolled and imaged with a 1.5T or 3.0T MRI scanner within 48 hours of injury. Isotropic diffusion-weighted images (DWI), T2* weighted images, fluid-attenuated inversion recovery images (FLAIR), and 3DT1 imaging modalities were obtained; further imaging was completed following an injection of gadolinium contrast. T1 post-contrast, FLAIR post-contrast, and perfusion-weighted images were obtained. CT was obtained for clinical purposes.
Images were prospectively evaluated for abnormalities common to head injury using a standardized report form. FLAIR-post contrast images were evaluated for TMI, defined as enhancement to the meninges on FLAIR sequences following gadolinium contrast. Associations between TMI and other findings on both MRI and CT were tested using Fisher's exact. Negative predictive value (NPV) of TMI for assessing other imaging abnormities was calculated.
Of the 142 subjects enrolled who received a MRI scan with a FLAIR-post contrast sequence, 71 subjects were positive for TMI (50.0%). The 71 subjects found negative for TMI were used for subsequent negative predictive value (NPV) calculations.
All subjects received a CT for clinical purposes and 124 (87%) were found to be negative for injury to brain. 65 were negative for TMI and on CT for parenchymal injury resulting in a NPV of 91.5%. There were 19 (13.4%) subjects determined to have subarachnoid hemorrhage by FLAIR or T2* MRI. 68 subjects were negative for both subarachnoid hemorrhage and TMI; the NPV of TMI for subarachnoid hemorrhage was 95.8%. For extra-axial hemorrhage found exclusively on T2* images, the NPV of TMI was also 95.8%.
There were 52 (36.6%) subjects found positive for at least one abnormality on either T2* images, DWI, or FLAIR; 31 (21.4%) subjects were positive for DWI (punctate and focal cortical hyperintensities). The NPV of TMI for DWI abnormalities was high: 67 were also negative for DWI abnormalities (NPV=94.4%). 32 subjects were positive for spherical appearing punctate hypointensities, or “microbleeds”, on T2* images, TMI had a NPV of 93.0%. 21 subjects were positive for linear appearing hypointensities on T2* images, TMI had a NPV of 94.8%. 33 subjects were positive for a focal lesion on T1 (23.2%), 29 (20.4%) were positive for TMI. 67 subjects were negative for both focal abnormalities on T1 images and TMI; thus, the NPV of TMI for focal abnormalities on T1 imaging was 94.4%.
In summary, the NPV of TMI for assessing other abnormalities on MRI ranged from 93.0% to 95.8%.
Brain injury following trauma is heterogeneous and diverse. While there is overlap in injury identified across imaging modalities and contrast mechanisms on MRI, seldom is injury seen if the FLAIR-post contrast is negative. Without the presence of TMI, there is a 93% or greater chance there will be no other evidence of injury on any other MRI sequences, and a 91.5% chance CT is negative. Therefore FLAIR-post contrast imaging provides the best predictive marker for identifying patients unlikely to have further injury. We found the conspicuity of this novel marker to be high. There is excellent intra-rater agreement in detecting TMI. The duration of FLAIR sequences on MRI scanners is in the order of 2–3 minutes. These characteristics, combined with the high NPV of TMI, makes FLAIR post-contrast MRI a quick and effective screening tool for ruling out other injury to the brain following trauma.
For the investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
traumatic meningeal injury, FLAIR, screen
MONITORING OF AMYLOID-BETA DYNAMICS AFTER HUMAN TRAUMATIC BRAIN INJURY
Nina Farrokhnia, MD, Department of Neurosurgery, Uppsala University Hospital
Anders Hånell, PhD, Anatomy and Neurobiology, Medical College of Virginia, Richmond, VA
Per Enblad, MD, PhD, Department of Neuroscience, Neurosurgery at Uppsala University
Henrik Zetterberg, MD, PhD, Sahlgrenska Academy at University of Gothenburg
Kaj Blennow, MD, PhD, Sahlgrenska Academy at University of Gothenburg
Lars Hillered, MD, PhD, Uppsala University
Accumulation of amyloid precursor protein (APP) occurs in injured axons after traumatic brain injury (TBI), and APP may be cleaved to amyloid-β (Aβ) peptides playing an important role in Alzheimer's disease (AD). AD. We used intracerebral microdialysis (MD) to measure Aβ peptides in patients with severe TBI.
We evaluated 10 mechanically ventilated patients (7 male, 3 female, age 18–76 years) with severe TBI (Glasgow Coma Scale score≤8) monitored with intracranial pressure monitoring and MD (using a 100-kDa nominal molecular weight cut-off membrane; M Dialysis AB, Solna, Sweden). Each MD sample was analyzed every hour for routine energy metabolic markers (MD-lactate, MD-pyruvate, MD-glucose and MD-lactate/pyruvate ratio) and MD-urea. The remaining MD sample volume were analyzed for Aβ40 and Aβ42 in 2-hour fractions up to 14 days post-injury using the Luminex xMAP technique allowing detection with high temporal resolution of the key Aβ peptides Aβ1-40 (Aβ40) and Aβ1-42 (Aβ42).
Interstitial Aβ40 and Aß42 levels were analyzed in a total of 1530 MD samples. Both Aβ40 and Aß42 were detected in all 10 patients, and the interstitial Aβ40 and Aβ42 levels were only slightly variable with minor fluctuations during the course of the disease. We hypothesized that the Aβ levels were related to injury type (diffuse TBI versus focal/mixed TBI), the energy metabolic situation, age and/or the level of consciousness assessed using the motor components of the Glasgow Coma Scale. Both Aβ40 and Aß42 were found to be consistently higher in patients with diffuse axonal injury compared to patients with focal TBI at day 1- 6 post- injury, Aß42 being significantly increased at 113–116 hours post-injury (p<0.05). However, the Aβ levels did not show a significant correlation with the energy metabolic situation, age or the level of consciousness.
Using a microsphere-based Luminex xMAP technique, we were able to analyse and monitor Aβ40 and Aβ42 levels in bi-hourly samples in the severely injured human brain. Our data suggest that MD is a useful tool for studying Aβ dynamics following TBI. We also confirm our previous hypothesis that these Aβ peptides occur at higher interstitial concentrations in patients with axonal injury compared to focal TBI. However, we could not find a correlation between Aβ levels and patient age, the focal energy metabolic situation or the level of consciousness. Future studies using an increased number of TBI patients are needed to confirm these observations. Additional markers such as tau and neurofilament (NF) and/or Aβ subspecies should also be used for elucidating the potential role of Aβ in the pathobiology of TBI as well as the complex relationship between TBI and Alzheimer's disease pathology.
The Swedish Medical Research Council, Swedish Brain Foundation, Selander Foundation and funds from the Uppsala University Hospital.
Microdialysis, TBI, amyloid-β, axonal injury
PLEASANT OLFACTORY STIMULATION AUGMENTS CARDIOVAGAL MODULATION IN PATIENTS WITH MILD TRAUMATIC BRAIN INJURY
Thomas Hummel, MD, Dept. of Otorhinolaryngology, University of Dresden Medical School, Dresden, Germany
Harald Marthol, MD, Dept. of Neurology, University of Erlangen-Nuremberg, Erlangen, Germany
Julia Koehn, MD, Dept. of Neurology, University of Erlangen-Nuremberg, Erlangen, Germany
Anja Rossmeissl, MD, Dept. of Neurology, University of Erlangen-Nuremberg, Erlangen, Germany
Steven Flanagan, MD, Dept. of Rehabilitation Medicine, New York University, New York, NY, USA
Philip DeFina, MD, International Brain Research Foundation
Stefan Schwab, MD, Dept. of Neurology, University of Erlangen-Nuremberg, Erlangen, Germany
Theresa Rameder, Depts. of Neurology, Medicine, Psychiatry, New York University, New York, NY, USA
After mild traumatic brain injury (mTBI), patients have sympathetic predominance (Hilz et al.,2011) which might contribute to increased long-term mortality rates (McMillan et al.,2011). In healthy persons, olfactory stimulation with jasmine augments cardiovagal modulation (Inoue et al.,2003). We determined autonomic responses to jasmine stimulation in post-mTBI patients.
In 18 healthy persons (6 women, 12 men; age 30±12 years) and 19 post-mTBI-patients (3 women, 16 men; age 37±14 years; 15–83 months post-injury; GCS scores 13–15 assessed 30 minutes post-injury), we recorded respiratory frequency, RR-intervals (RRI), systolic and diastolic blood pressure (BPsys, BPdia) before and during 30 seconds of pleasant olfactory stimulation with jasmine (1-second stimuli, 3 second intervals). We calculated spectral powers of mainly sympathetic low- (LF: 0.04–0.15 Hz) and parasympathetic high-frequency (HF: 0.15–0.5 Hz) RRI-oscillations, RRI-LF/HF-ratios reflecting sympatho-vagal balance, sympathetic LF-oscillations of BP and compared parameters of patients and controls before and during jasmine stimulation (RANOVA, post-hoc t-test; significance: p<0.05).
Without stimulation, patients and controls had similar RRIs (871.7±113.4 vs. 925.0±123.3 ms), BPsys (125.4±12.2 vs. 136.4±19.8 mmHg), BPdia (65.8±6.1 vs. 70.8±10.2 mmHg), RESP (13.0±4.5 vs. 12.7±3.5 mmHg) and RRI-LF-powers (1444.9±1208.0 vs. 1798.3±1957.3 ms2), but patients had lower RRI-HF-powers (763.6±962.0 vs. 1654.2±3471.2 ms2, p<0.05), higher BP-LF-powers (17.4±21.3 vs. 9.6±9.3 mmHg2) and RRI-LF/HF-ratios (7.1±11.8 vs. 2.9±2.8, p<0.05) than controls. During jasmine stimulation, RRI-HF-powers increased (763.6±962.0 vs. 1290.0±1561.2 ms2, p<0.05), and RRI-LF/HF-ratios decreased (7.1±11.8 vs. 2.8±3.5, p<0.05) in patients only, while RRIs, BPsys, BPdia, RESP, RRI-LF-powers, and BP-LF-powers remained unchanged in both group.
Increased parasympathetic RRI-HF-powers and decreased sympatho-vagal RRI-LF/HF-ratios during jasmine stimulation suggest that jasmine stabilizes the sympatho-vagal balance in mTBI patients and buffers the augmented sympathetic tone after mTBI. Thus, pleasant olfactory stimulation, using e.g. jasmine might non-invasively improve autonomic stability in post-mTBI patients. Long-term effects still need to be evaluated.
The study was funded by the International Brain Research Foundation, Flanders, NJ, USA.
traumatic brain injury, cardiovagal modulation
UNVEILING THE ROLE OF THE DORSAL ATTENTION NETWORK: DISCRIMINATION ABILITY AND PROGNOSTIC VALIDATION OF EVENT-RELATED POTENTIALS IN BRAIN INJURED PATIENTS
Lara Prisco, MD, University College Hospital London
Walter Calligaris, University Hospital "Cattinara" Trieste
Antonio Draisci, University Hospital "Cattinara" Trieste
Giuseppe Romano, University Hospital "Cattinara" Trieste
Mauro Semencic, University Hospital "Cattinara" Trieste
Laura Ganau, University of Cagliari
Daniele Pescador, University of Trieste
Fabrizio Monti, MD, University of Trieste
Event-Related Potentials (ERP) is a promising technique for the assessment of comatose patients following brain injury. To this regard, Mismatch Negativity (MMN) and other long-latency components, such as N100 and P300, are advocated as useful tools in the evaluation of unresponsive patients and in the neurological outcome prognostication.
Acute brain injured patients, admitted to ICU with severe coma [GCS<7], presenting a prolonged unresponsiveness after >72h of sedation withdrawal were preliminary selected for this study. A history of neurological diseases was considered as exclusion criteria. Sixteen consecutive patients were enrolled and underwent neurophysiological evaluation of auditory pathways [short-latency evoked potentials]: ERP were recorded during an auditory-oddball-paradigm (AOP) in resting conditions with eyes closed [standard tone: 750Hz, target tone: 1000Hz at 130dB, ratio: 15, recording time: 2 sec, bandpass: 1–50Hz in Fz, Cz and Pz, test repeated to obtain 30 sweeps]. An EEG was performed to rule out seizures during ERP recording. Amplitude and latency of identified ERP components [MMN, N100, P300] were matched with 10 healthy controls. Patients' outcome was assessed 6 months later according to Coma-Recovery-Scale-Revised and classified as Minimally Conscious State (MCS), Persistent Vegetative State (PVS) and deceased (DEC).
ERP were recorded on average 15 days after injury, and EEG ruled out any seizures during neurophysiological evaluation. At follow up [6 months] 6 out of 16 patients died, 5 remained in an unconscious PVS and 5 evolved to a MCS. MMN was found in 2 patients who died, in 1 PVS patient and in 3 MCS ones; its average amplitude was higher in MCS than SVP or deceased [MCS: 4.72 mV, SVP: 0.78 mV, DEC: 0.504 mV; versus controls: 4.89 mV; p<.001] whereas latency was lower [MCS: 260 ms, SVP: 318 ms, DEC: 190 ms; versus controls: 219 ms; p<.01]. No statistical differences were found among patients concerning the amplitude [MCS: 5.2 mV, SVP: 3.1 mV, DEC: 6.4 mV; versus controls: 12.1 mV] and latency [MCS: 147 ms, SVP: 155 ms, DEC: 140 ms; versus controls: 117 ms] of N100. P300, a late cognitive ERP component, denoted a somehow voluntary control in the integration processes with a striking involvement of the dorsal attention network (DAN). Differences in amplitude [MCS: 8.6 mV, SVP: 1.2 mV, DEC: 4.6 mV; versus controls 11.9 mV; p<.001] and latency [MCS: 607 ms, SVP: 730 ms, DEC: 626 ms; versus controls: 402 ms; p<.01] of this component showed the highest discrimination ability and prognostic validity at 6 months, suggesting a promising accuracy for widespread clinical use in the near future.
The DAN shares with the ventral attention network the role of sensory orienting systems in the human brain: it is involved in voluntary (top-down) orienting and shows activity increases after presentation of cues indicating where, when, or to what subjects should direct their attention. Noteworthy, in our cohort the DAN emerged also as a pivotal centre in integration processes and cognitive functions. The AOP and the P300 were indeed excellent and reliable in assessing and predicting the recovery of consciousness through the evaluation of this network also in unresponsive patients whose behavioural assessment resulted otherwise impossible in an ICU setting. For the first time ever the results herein presented provide a neurological validation at 6 months of an ERP examination performed acutely in brain injured patients; those breakthrough findings need to be confirmed with more sophisticated imaging techniques (EEG - fMRI) to further unveil the roles of the DAN.
The authors do not report any financial, personal or professional conflict of interest concerning the materials, methods or findings presented in this study.
Acute Brain Injury, Neurophysiology, Prognosis
THE INTERNATIONAL TBICARE STUDY - PRELIMINARY RESULTS
David K. Menon, University of Cambridge, UK
Ari Katila, Turku University Hospital, Finland
Jonathan Coles, Addenbrookes's Hospital, Cambridge, UK
Janek Frantzén, Turku University Hospital, Finland
Joanne Outtrim, University of Cambridge, UK
The vast heterogeneity of TBI is one of the greatest challenges in providing evidence-based recommendations for the diagnostics or treatment of TBIs. Due to this heterogeneity, randomized controlled trials have often been unsuitable to promote clinical TBI medicine. In the EU-funded international TBIcare study, we have tried to develop methods
In this study we try to combine data mining, system simulation modelling, and sophisticated statistical methodology in analysing large both retrospective and prospective clinical databases of subjects with TBI. The databases used include existing clinical datasets from the participating hospitals in Cambridge (UK) and Turku (Finland), the IMPACT database, and a prospective dataset collected during the study. In the prospective dataset, detailed clinical data collection, serial MR imaging with modern methods, and extensive blood biomarker collection are used in combination.
This three-year project will yield a basic software tool to support clinical decision-making in individual cases of TBI, both in assessing the accurate nature of the injury and in selecting the optimal treatments. These are based on extensive modelling from the databases, comparing the individual injury profile with the existing information of respective cases from the database. Thus, the more common type of injury, the more reliable recommendations can be given. Besides these clinical tools, the TBIcare study yields a huge amount of new scientific information from the relationships between various clinical parameters, imaging findings, and blood biomarkers.
In this presentation we will present up-to-date information from the project, concerning: - system simulation modelling of TBI care - imaging analytics - blood biomarkers - basic clinical variables and connections - challenges in developing individualized solutions for TBI
This project is partially funded under the 7th Framework Programme by the European Commission.
Biomarkers, model, diagnostics, treatment
ACUTE AND 1 YEAR SUSCEPTIBILITY-WEIGHTED MRI OF HEMORRHAGIC SHEARING INJURY AFTER PEDIATRIC TBI
Ruba Al-Ramadhani, MD, Loma Linda University Medical Center
Melissa Rundquist, Loma Linda University Medical Center
Jamie Pivonka-Jones, PhD, Loma Linda University Medical Center
Barbara Holshouser, PhD, Loma Linda University Medical Center
Stephen Ashwal, MD, Loma Linda University Medical Center
We present preliminary findings on 1 year follow-up of hemorrhagic shearing injury and neuropsychological outcomes in pediatric TBI patients, using MRI susceptibility-weighted imaging (SWI). SWI accentuates paramagnetic properties of blood products and has been shown to improve detection of shearing-related hemorrhages.
Pediatric patients, aged 4 to 18, were enrolled if they sustained a moderate/severe traumatic brain injury requiring admission to hospital, defined either as having a GCS score <13 or evidence of intracranial injury on initial computed tomography scan. Patients underwent MRI scanning in the acute period (7-17d post TBI), and at 1 year after injury. SWI images were analyzed using an off-line post-processing program, “SPIN”, to semi-automatically count and measure the volume of hemorrhagic brain lesions. The number and volume of hemorrhagic lesions between the initial and one year follow-up study were compared, and also evaluated with initial GCS and gender. Initial and follow-up SWI data were also compared with neuropsychological outcomes at 3 and 12 months, specifically measures of memory, attention, and global IQ.
The current group of follow-up patients consists of 17 children or young adolescents between 6 and 17 years of age (12 males and 5 females). Six were involved in motor vehicle crashes, 2 in motorcycle accidents, 2 in ATV accidents, 2 were pedestrians struck by automobiles, 1 was struck by a car while riding a bicycle, and 4 sustained a fall. The total number of initial hemorrhagic lesions in each patient ranged from 0 to 299 (the latter with a GCS of 4). The initial GCS did not correlate with the initial number/volume of hemorrhages on the acute MRI. In the course of one year, the number of lesions decreased by up to 81.3%, and the volume of hemorrhages decreased by up to 84.5%. However, approximately half of the patients (9 of the 17) retained more than 50% of the original volume of hemorrhages. We did not observe a significant gender difference in the resolution of the number or volume of hemorrhages. The number of hemorrhages in the initial MRI study showed a moderate inverse correlation with memory scores at 3 month follow-up.
We have previously shown that SWI depicts significantly more traumatic hemorrhages, particularly small ones, than conventional MRI. In a prior retrospective study, we reviewed pediatric and adult TBI patients with MRI studies acquired acutely and at varying follow-up times, and observed that many patients had hemorrhages that persist for more than one year. In this preliminary prospective analysis of pediatric patients, 50% still had hemorrhages at one year. There was no apparent association in degree of improvement, with gender or initial GCS, although the sample size is small. Comparison with neuropsychological outcomes showed that 3 month memory scores had a moderate inverse correlation with the number of hemorrhages on the initial MRI. Further analysis will determine if there are regional differences in the resolution of hemorrhages, correlation with other clinical and imaging variables, or associations with location of lesions with specific neurological and neuropsychological outcomes at one year.
National Institutes of Health
Pediatric, TBI, MRI, hemorrhage, shearing
CYTOKINES AND BRAIN INJURY MARKERS IN TBI PATIENTS: DIFFERENCES IN FOCAL AND DIFFUSE BRAIN DAMAGE, AND NORMOXIC OR HYPOXIC STATUS AND THEIR RELATION TO NEUROLOGICAL OUTCOME
Nicole Bye, PhD, National Trauma Research Institute, Alfred Hospital and Monash University
Phuong Nguyen, PhD, Department of Health, Victoria, Australia
Thomas Kossmann, MD, Epworth HealthCare, Melbourne
Jeffrey Rosenfeld, MD, Alfred Hospital and Monash University
Edwin Yan, PhD, Department of Physiology, Monash University
Traumatic brain injury (TBI) can be aggravated by a hypoxic insult, a recognised contributor to poor outcome. To elucidate potential differences in the pathophysiology of secondary brain damage in focal and diffuse TBI and how these can be affected by post-traumatic hypoxia, we examined changes in cerebral inflammation, injury markers.
Forty-two TBI patients with Glasgow Coma Scale ≤8 and an extraventricular drain were recruited. Cerebrospinal fluid (CSF) and serum were collected for 6 consecutive days. Patients were divided into focal (n=15), diffuse (n=23), hypoxic (Hx; n=22), or normoxic (Nx; n=20) groups. Eight cytokines were measured in CSF by multiplex ELISA, while albumin (in CSF/serum), S100, MBP and NSE were measured in serum by ELISA. CSF/serum albumin quotient was used for BBB function assessment.
GM-CSF, IFN and to a lesser extent TNF, increased in CSF of Hx, but not Nx patients. These differences were evident at days 4–5 post-TBI-Hx. IL-2, IL-4, IL-6, IL-10 increased similarly in both groups, while IL-8 was not significantly different among the cohorts. When comparing focal and diffuse TBI patients, IL-6 was higher in the focal, and GM-CSF in the diffuse group. All biomarkers were more elevated in focal TBI patients. The occurrence of Hx coincided with higher MBP and S100 levels. However, S100, MBP and NSE were significantly higher in those Hx patients with unfavourable outcome. Interestingly, prolonged elevation of all biomarkers in diffuse and Hx patients coincided with sustained BBB dysfunction.
We demonstrate for the first time that post-TBI hypoxia amplifies neuroinflammation and extends BBB permeability. These changes were associated with prolonged biomarker extravasation and adverse neurological outcome. Subtle differences in cytokine levels also appeared between focal and diffuse patients. All biomarkers were higher in the focal than the diffuse TBI group but this did not reflect differences in outcome scores. Altogether this study highlights distinct pathological changes in response to individual forms of brain damage and the combination with secondary hypoxia.
Victorian Neurotrauma Initiative/Traffic Accident Commission #D009 and Fellowships to C.M-K., E.B.Y. and N.B., National Health Medical Research Council Project Grant #436815
cytokine, biomarker, hypoxia, diffuse, focal
TRAUMATIC HEAD INJURY NEUROIMAGING CLASSIFICATION STUDY (THINC): ACUTE MRI FINDINGS FROM THE FIRST 18 MONTHS OF ENROLLMENT
Jessica L. DeStefano, BS, Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Martin R. Cota, B.A., Center for Neuroscience and Regenerative Medicine/Henry M. Jackson Foundation
Kara Kuntz-Melcavage, PhD, Johns Hopkins HealthCare LLC
John Butman, MD, PhD, Radiology and Imaging Sciences, National Institutes of Health/Clinical Center
Jose Merino, MD, Stroke Diagnostics and Therapeutics Section of the National Institute of Neurological Disorders and Stroke
Lisa Davis, RN, MSN, CCRC, Stroke Diagnostics and Therapeutics Section of the National Institute of Neurological Disorders and Stroke
Leighton Chan, MD, MPH, Rehabilitation Medicine Department of the NIH Clinical Center
The Traumatic Head Injury Neuroimaging Classification (THINC) multisite study explores the use of MRI for identifying intracranial pathology in civilian patients with head injury, compares the findings to that from the US military population, and provides a mechanism for referral to longitudinal observational and therapeutic studies.
Patients presenting to two Washington DC Metro trauma centers within 48 hours of head injury were enrolled. Study visits include initial, day 4, and optional follow up at 30 and 90 days. Study procedures include clinical interview, MRI, biomarker collection, three outcome scales, and CT scans obtained for clinical purposes. The MRI protocol included isotropic DWI and ADC derived from DTI, conventional and segmented-EPI T2*, FLAIR, 3DT1 SPGR, dynamic susceptibility contrast perfusion, T1-post, and FLAIR-post contrast with a nominal imaging time of 24 minutes. Abnormalities were classified both by sequence (i.e. lesion on DWI) and by presumed pathology (i.e. subarachnoid hemorrhage based on multiple series). Data presented are median (IQR).
From October 2010 to March 2012, 168 subjects were enrolled; age=45(29–60), 32% female, 119 (71%) white, 41(24.4%) black, 151(90%) GCS=14 or 15, 113 (67%) LOC+, 85(51%) PTA+, and injury to triage of 49 (30–130) min. Disposition from ED included 33 (20%) discharged home, 88(52%) admit to floor, and 38(23%) to the ICU; polytrauma influenced admission.
Median time from injury to CT was 86 (56–188) min. Of 162 (96.5%) Acute CT scans that were available, 39 (23.2%) were positive by radiology report: 20 (11.9%) injury to parenchyma, 27 (16.1%) skull fracture, 24 (14.3%) subarchanoid hemorrhage, 13 (7.7%) subdural hematoma, 2 (1.2%) epidural hematoma.
Median time from injury to baseline MRI was 18.6 (8.1–28) hrs. Of 142 subjects receiving contrast, 71 (50%) of had enhancement of the meninges, referred to as traumatic meningeal injury (TMI). Hypointensities were seen in parenchyma on T2* weighted in 50 (30%) of subjects, including “microbleeds” in 39(23%), linear appearing in 22 (13%), and intracranial hemorrhage>1cm in 17(10.1%). On DWI, lesions<1cm were seen in 34 (20%), and multifocal lesions were seen in 17(10%), >1cm in 13 (7.7%). Edema/contusion was seen on FLAIR in 34 (20%). Focal deficits were detected on perfusion imaging in 6 (3.6%) and blood-brain barrier disruption was seen in 4 (2.4%). Heterogeneity was documented by a “Lesion Score” based on the number of series (DWI, T2*, FLAIR) in which parenchymal lesions were detected; “1” 17.3%, “2” 10.1%, and “3” 9.5%.
Of 123 subjects with a negative CT scan, abnormalities were seen on MRI in 55 (45%): specific to parenchyma in 29 (24%), and on T2* weighted in 24 (19%).
Mild TBI is a heterogenious injury. The misconception that patients with minor injury have completely negative neuroimaging findings persists in the literature. Data from the THINC study suggests that may not be the case. Abnormal findings on MRI were present in nearly half the subjects. Additional, diffuse injury that may have gone undetected, such as traumatic axonal injury, could potentially be revealed during group analyses. By using MRI to identify specific focal injury, clinical heterogeneity inherent to the mTBI population may be accounted for.
For the Investigators of the CNRM THINC Study, supported by The Center For Neuroscience and Regenerative Medicine (CNRM) and NIH-NINDS.
THINC, MRI, mTBI, Concussion, Diagnosis
RECOVERY FROM FULLY DISABLING TRAUMATIC BRAIN INJURY, 15 YEAR FOLLOWUP
John J. Briggs, RPh, NPh, MSC, Quantitative Imaging Inc
45 year old male fell 9 feet to concrete floor February 1997. Ttransiently unconscious. Profound hypotension Rx splenectomy. Severe headaches, photophobia, visual distortions, impaired focusing, vertigo, dizziness accompanied by devastating memory loss, totally diabled. Second head injury 2009 17mph bike collision resulted in loss of interval adaptation.
Baseline neurocognative tests and PET imaging May 2011. Rx AA+hypertension meds. Patient initiated hyperbaric oxygen, meditation, and continued rigorous bike riding exercise. Frequent interval neurocognative assessments. PET repeated March 2012.
Eleven month interval change from May 2011 to March 2012.: Processing speed 5 53; Verbal memory 12 90; Composite memory 21 79; Reaction time 42 73; Visual memory 45 53; Neurocognitive index 63 90 Executive function 68 95; Cognitive flexibility 70 94; Complex attention 82 82; Psychomotor speed 86 99 SUMS 494 808 March 2012 FDG QUANTITATIVE PET demonstrated marked improvement in frontal lobe areas that had been hypometabolic at baseline. Both medial temporal lobes and both amygdala continue to be severely hypometabolic. Patient is currently studying to regain his commercial real estate license. He supports his family and functions well socially.
Return to full social and economic function after total disability documented in1997 has been observed. Recovery from the second head injuryin 2009 has been documented despite persistance of severe bilateral medial temporal lobe hypometabolism on PET scan.
Dr. Robert Carroll and Major John Briggs are consultants to Quantitative Imaging Inc. Website Is www.quantitative brain.com
plasticity regeneration imaging cognition rehabilitation
NATURAL HISTORY OF SYMPTOMS AND SYMPTOM SEVERITY FOLLOWING GERIATRIC TRAUMATIC BRAIN INJURY
Tessa Rue, MS, University of Washington
Frederick Rivara, MD, MPH, University of Washington, Department of Pediatrics, HIPRC
Older adults with traumatic brain injury (TBI) are a significant and growing public health problem. Little is known about the natural history of TBI in older adults and traditional outcome measures (e.g. mortality, Glasgow Outcome Scale) are often of limited value in guiding intervention.
The purpose was to compare the natural history of symptom experience and severity in the year following a mild-moderate TBI between older (65+ years; n=15) and younger adults (21–64 years of age; n=19). Subjects with a clinical diagnosis of TBI (GCS>8) were enrolled within 24 hours of TBI and symptom endorsement, severity and total number of symptoms were assessed via the Head Injury Symptom Checklist (HISC) at 1 week, and 1, 3, 6, 12 months post-injury. In addition, health-related quality of life (HRQOL) was assessed at the same time points using the SF-12. Descriptive statistics and correlations were used to describe the natural history of symptom burden over time in the different groups and determine associations between symptom burden and HRQOL. Linear mixed modeling was used to explore differences in symptom experience between older and younger persons with TBI over time.
Older adults report higher mean total numbers of symptoms experienced from 1 month to 12 months post-TBI than do younger adults. These differences are significant at 3 and 6 months post-injury. In addition, the types of symptoms experienced differ significantly. The top three symptoms initially endorsed by older adults at one week post-injury include balance disturbance (73%), fatigue (67%), and dizziness (60%) while younger adults endorse headache (58%), balance (58%), and irritability (53%). Older adults also frequently endorse difficulty with memory (47%) but this was not reported by younger adults, who reported new or worsening anxiety (42%). These significant differences persisted among the groups with younger adults continuing to report new or worsening anxiety to 1 month post-injury; while older adults reported persisting issues with fatigue and issues with balance and coordination to 6 months post-injury (p<0.05). In older adults, symptoms that correlated negatively with with the physical component subscore (PCS) of the SF-12 at one week included headache, blurry vision, and sexual difficulties. At three months, fatigue and blurry vision were corrlated to lower PCS scores and at 12 months blurry vision remained an important influence on HRQOL in older adults following TBI. Interestingly, in younger adults, more severe irritability, anger and anxiety were significantly correlated to lower PCS scores while physical symptoms like blurry vision, balance issues and trouble with sleep were correlated with lower MCS scores at one week post-injury. At 3 and 12 months post-injury in younger adults, experiencing blurry vision significantly influenced physical HRQOL while fatigue and trouble concentrating influenced mental health scores in a negative manner.
This study provides beginning knowledge about the natural history of symptom experience and HRQOL following mild-moderate TBI in older adults. Following a mild-moderate TBI, older adults report different symptom clusters than do younger adults. A better understanding of these patterns and the recovery trajectory will allow for development of specific age-appropriate interventions aimed at reducing or eliminating symptoms and improving HRQOL. Blurry vision is a symptom associated with poorer HRQOL amongst all age groups and further attention to alleviating this symptom is warranted.
Supported in part by: John A. Hartford Foundation BAGNC Fellowship 06–202, NIH 5 KL2RR025015 and P30NR004001.
fatigue, anxiety, balance, symptom burden
LACTATE FUELS THE BRAIN AND BODY FOLLOWING HUMAN TRAUMATIC BRAIN INJURY
George Brooks, Ph.D., UC Berkeley
Michael Horning, MS, UC Berkeley
Matthew Johnson, PhD, UC Berkeley
David L. McArthur, MPH, PhD, UCLA
Paul Vespa, MD, FAAN, FACN, University of California, Los Angeles
David A. Hovda, PhD, University of California, Los Angeles
Neil Martin, MD, Department of Neurosurgery UCLA
Thomas C. Glenn, PhD, UCLA, Brain Injury Research Center, Department of Neurosurgery
Our previous studies have found that lactate is actively consumed by the brain following TBI. Using a dual isotope venous tracer infusion alongside conventional cerebral arterial-jugular bulb blood sampling, we attempted to trace the systemic and cerebral metabolic fate of glucose and lactate after TBI.
Traumatic brain injury patients (n=18, admission GCS<9) were entered into the study following consent of patients' legal representatives. Written and informed consent was obtained from age and gender matched controls (n=6). ([6,6-2H2]glucose and ([3-13C]lactate, Cambridge Isotopes) tracers were infused intravenously alongside standard arterial blood gas analysis. Infusion was performed 5.7 +/− 2.2 days after injury, with a range of 2–10 days. Dual lactate and glucose tracers were used to discriminate between systemic whole body production and cerebral uptake of the respective metabolites. Blood glucose concentrations and IE's were determined by gas chromatography/mass spectrometry (GC/MS; GC model 6890 Series and MS model 5973N, Agilent Technologies) of the pentaacetate derivative using an Agilent DB-17 GC column. A dual compartmental model was subsequently used to interpret findings. Statistical analyses were conducted in R, version 2.14.2.
As reported herein and in previous studies, the cerebral metabolic rate for glucose was depressed following TBI (p<0.04). Gluconeogenesis from precursor lactate accounted for 65% of lactate disposal, compared to 15% in controls, reflecting averaged values determined between 90–150 minutes. However, lactate fractional extraction approximated 12% in both healthy control subjects and TBI patients. Further, neither the CMR for lactate (CMRlac, i.e., net lactate release), nor tracer-measured cerebral lactate uptake differed between healthy controls and TBI patients. Hence, Total Cerebral Lactate Production (=Tracer Measured Uptake – CMRlac) did not differ between TBI patients and controls. The percentages of lactate tracer taken up released as 13CO2 into the jugular bulb accounted for 68 and 100% for control, and TBI conditions, respectively, suggesting that most cerebral lactate uptake was oxidized following TBI. Comparisons of isotopic enrichments of lactate oxidation from infused [3-13C]lactate tracer and 13C-glucose produced during hepatic and renal gluconeogenesis (GNG) showed that 75–80% of 13CO2 released into the jugular bulb was from lactate and the remainder for the oxidation of secondarily labeled glucose. Hence, either directly as lactate uptake, or indirectly via GNG, peripheral lactate production accounted for≈70% of carbohydrate (glucose+lactate) consumed by the injured brain. Near normal arterial glucose and lactate concentration levels in patients studied 72–96 hr post-injury belied 100% greater systemic lactate production that was largely cleared by greater (hepatic+renal) glucose production.
These findings illustrate the central role of lactate in a systemic metabolic feedback loop following traumatic insult to the brain. Lactate flux through gluconeogenic pathways results in the hyperglycemic state typically observed following TBI, which in turn is supplemented by corporal glycogen reserves. In summary, our findings indicate that following TBI, the body exerts enormous effort to provide for the energy needs of the injured brain, in which lactate plays a critical role in mediation of these diverse processes either directly (cerebral oxidation) or indirectly (as a gluconeogenic precursor). Undiminished cerebral lactate fractional extraction and uptake suggest the arterial lactate supplementation may be used to compensate for decreased CMRgluc following TBI.
We would like to thank Maria Etchepare, RN and the UCLA NeuroIcu staff. Funding: The University of California Neurotrauma Initiative, and NINDS NS058489.
Metabolism, stable isotopes, gluconeogenesis
A MATTER OF PERSPECTIVE: DETECTION OF SPREADING DEPOLARIZATIONS BY CONTINUOUS EEG
Adam Wilson, PhD, University of Cincinnati
Sebastian Pollandt, MD, University of Cincinnati
Norberto Andaluz, MD, Department of Neurosurgery
David Ficker, MD, University of Cincinnati
Lori Shutter, MD, University of Cincinnati
Cortical spreading depolarizations are a primary mechanism of secondary injury after stroke and severe traumatic brain injury (TBI) and are associated with poor clinical outcomes. Conventional wisdom for over 60 years was that depolarization waves can not be detected by scalp electroencephalography (EEG), but recent evidence suggests otherwise.
Thirteen patients with severe TBI were enrolled in a study of spreading depolarizations, measured by electrocorticography (ECoG) from electrode strips placed on the surface of the brain adjacent to the injury focus. Only patients with a clinical need for surgery were enrolled. After surgery, ECoG was recorded during intensive care, as previously described, and scalp EEG was recorded as clinical standard-of-care following the local protocol for multi-modal monitoring of severe TBI. Sixteen scalp electrodes were placed bilaterally according to the international 10–20 standard and were acquired in a ‘double-banana’ bipolar montage with a Grass amplifier system (0.5 Hz high-pass). Depolarizations were first identified on ECoG by slow potential changes (0.01–0.1 Hz) and depressions of 0.5–50 Hz activity propagating between electrodes. ECoG/EEG signals were then merged for EEG analysis. Leaky power integrals (120 s decay time constant) were calculated for quantification of EEG changes. Research protocols were IRB-approved.
ECoG and EEG were recorded for an average 5.0 and 3.4 days, respectively, and the total duration of simultaneous recordings was 39.5 days. A total of 427 spreading depolarizations occurred in 11/13 patients on ECoG, but only 164 occurred during simultaneous ECoG/EEG in 8 patients. All 8 patients had moderate-to-severe generalized slowing on EEG and 6 had continuous polymorphic delta activity localized to the injured hemisphere. EEG correlates of ECoG depolarizations were assessed by visual inspection on a highly compressed time scale of 40–90 mm per hour. EEG amplitude depressions occurred in association with 61 (37%) of 164 ECoG depolarizations in up to 3 bipolar EEG channels closest to the ECoG electrode strip. The maximal depression of EEG power relative to baseline was 55% (median; inter-quartile range: 44–63), which developed progressively over a period of 12 min (IQR: 8-17). In most cases (53/60), EEG amplitude recovered prior to subsequent ECoG depolarizations. For these, the total time envelope from start of depression to recovery to a steady-state amplitude was 17 min (IQR:13-24). Of the remaining 103 ECoG depolarizations that did not induce EEG depression periods, 58 occurred during a period of continued maximal depression (n=15) or partial depression with incomplete recovery (n=43) after a prior depolarization. Thus, these instances exhibited a fusion of EEG depression periods induced by multiple repetitive depolarizations; the intervals between these depolarizations were only 36 min (IQR: 24-47). This is significantly shorter than intervals preceding depolarizations that were associated with EEG depressions (88 min; IQR: 47-111, p<0.001). Nineteen ECoG depolarizations without EEG depression could not be explained by persisting depression periods, but also occurred at short intervals (39 min; IQR: 31-46). Finally, 26 depolarizations occurred at the start of EEG recordings so that baseline amplitudes could not be determined. We found that 37% of spreading depolarizations detected by the gold standard of ECoG are evidenced on EEG by inducing amplitude depressions to 55% of baseline power. An additional 35% of spreading depolarizations were evidenced in continuing depression periods that, after these initial amplitude reductions, persisted through a series of depolarizations occurring at short intervals. Initial EEG amplitude depressions developed over a period of 5–30 min, in contrast to <1 min in ECoG. This is likely due to the broad spatial sampling (∼10 cm2) of EEG electrodes and the time required for a depolarization spreading through cortex at 1–5 mm/min to traverse and depress a sufficient portion of this area. Thus, detection of spreading depolarizations by EEG has been possible since their discovery in 1944 and is simply a matter of perspective: EEG must be viewed on a highly compressed time scale to appreciate these slow, creeping changes.
This work was funded by the U.S. Army CDMRP PH/TBI Research Program, Contract No. W81XWH-08-2-0016.
EEG; spreading depression; secondary injury
MARSHALL CLASSIFICATION VERSUS THE ROTTERDAM SCORE IN PREDICTING 6-MONTH MORTALITY IN SEVERE TBI
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
Andrea Gabrielli, MD, University of Florida
Shelley Heaton, PhD, University of Florida
H. Julia Hannay, PhD, University of Houston
Ronald Hayes, PhD, Banyan Biomarkers
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Gretchen M. Brophy, PharmD, Virginia Commonwealth University, Medical College of Virginia
Claudia Robertson, MD, Baylor College of Medicine
Steven Robicsek, MD, PhD, University of Florida
This study compared the Marshall Classification to the Rotterdam score in predicting 6-month mortality in patients with severe TBI.
This study was designed as prospective observational study that enrolled adults with severe TBI presenting to 2 Level I Trauma Centers. Patients were included if they had a blunt head injury with a GCS score of ≤8 and required intracranial pressure monitoring. Marshall Classification and Rotterdam Score were calculated for each patient using the initial CT scan on admission. The Core IMPACT score (age, pupillary reactivity and GCS motor score (www.tbi-impact.org) was calculated for each patient to determine their risk of mortality at 6 months. Area under the ROC curve (AUC) was calculated to determine if the Marshall and/or Rotterdam added prognostic value to the Core IMPACT score.
There were 131 patients enrolled in the study. Mean age of patients was 38 years and 78 were male. At 6 months 104 patients had follow-up data and 36 (35) did not survive to 6 months. Distribution of non-survivors in the Marshall Classification was: DI=0 (0), DII=7 (19), DIII=15 (42), DIV=2 (6), EM=5 (14), and NEM=7 (19). For the Rotterdam scorethe distribution was: R1=1 (3), R2=10 (28), R3=11 (31), R4=12 (33), R5=2(6), R6=2(6). The correlation between the 2 scales was 0.378 (p<0.001). Marshall Classification, as an ordinal scale, was not significantly correlated withmortality (p=0.07) but it was significant when dichotomized into DI, DII vsDIII, DIV, EM, NEM (p=0.01). Rotterdam was significantly associated with mortality(p<0.001). The area under the ROC curve (AUC) for the Core IMPACT forpredicting mortality was 0.80 (95CI 0.72–0.89). When the Rotterdam score was added to the IMPACT model the AUC was 0.82 (95CI 0.74–0.90). When the dichotomizedMarshall Class was added to the IMPACT model the AUC was 0.82 (95CI0.73–0.89). When both were added the AUC remained at 0.82 (95CI 0.74–0.90).
The Rotterdam score was more strongly associated with 6-month mortality than Marshall Class but each provided the same amount of prognostic information to the Core IMPACT score in this cohort of patients with severe TBI.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
TBI, CT analysis, predictive model
INCIDENCE OF ICU COMPLICATIONS AND THEIR IMPACT ON OUTCOME IN MODERATE TO SEVERE TRAUMATIC BRAIN INJURY
Susanne Muehlchlegel, MD, MPH, University of Massachusetts Medical School/UMASS Memorial Medical Center
Cynthia Ouillette, RN, University of Massachusetts Medical School
Wiley Hall, MD, University of Massachusetts Medical School/UMASS Memorial Medical Center
Fred Anderson, PhD, University of Massachusetts Medical School
Robert Goldberg, PhD, University of Massachusetts Medical School
Non-neurologic organ failure contributes to 2/3 of all deaths after traumatic brain injury (TBI). Admission characteristics are known predictors of outcome, but only explain 1/3 of outcome variability. The incidence rates of specific intensive care unit (ICU) complications and their impact on patient outcomes in TBI remain poorly defined.
This is a prospective cohort study of 170 consecutive moderate-severe TBI patients admitted to a single academic Level I trauma center with a closed neurotrauma unit 11/2009-2/2012. Pre-specified medical and neurological complications were collected and adjudicated weekly by three board certified neurointensivists. Functional outcome was collected at 3 months post TBI using the Glasgow Outcome Scale (GOS). We identified the ten most common medical and neurological complications after ICU admission. Data was analyzed using descriptive statistics, as well as multiple logistic regression analysis to identify predictors of poor 3-month outcome (GOS≤3).
The mean age of the study sample was 51 years, 72% were men, and the median Glasgow coma scale and injury severity scores were 4 and 29, respectively. Incidence rates of the ten most common medical complications in the ICU were: hyperglycemia (75%), fever (62%), systemic inflammatory response syndrome (38%), cardiac complications (36%), hypotension requiring vasopressors (35%), pneumonia (any type [34%]); sepsis (33%), anemia requiring transfusion (31%), other pulmonary complications (ARDS, pulmonary edema [26%]), and hyponatremia (sodium≤134mEq/L; [23%]). Using multiple logistic regression analysis, hypotension requiring vasopressors (HRV) was the strongest predictor of poor outcome [OR 2.8; 95% CI 1–7.5]) among medical complications. After combining medical with neurological ICU complications, brain herniation (OR 5.8; 95% CI 1.1–30.2) and intracranial rebleeding (OR 2.9; 95% CI 1–8.4) were the strongest predictors of poor outcome, while HRV approached significance (OR 2.4; 95% CI 0.9–6.4).
In our cohort of moderate-severe TBI, medical complications were very common. Hypotension requiring vasopressors was the strongest medical predictor of poor outcome, while brain herniation and intracranial bleeding were the strongest neurological predictors of poor outcome. We identified important potentially modifiable predictors of adverse outcomes after moderate-severe TBI. Confirmation of our findings in a larger cohort is warranted.
None
Critical Care, Outcomes, TBI, Complications
A CLINICAL INTERVENTION: VETERANS WITH TRAUMATIC BRAIN INJURY (TBI) AND POSTTRAUMATIC STRESS DISORDER (PTSD)
Eric Elbogen, Ph.D., ABPP, University of North Carolina (UNC) at Chapel Hill
Aysenil Belger, Ph.D., University of North Carolina (UNC) at Chapel Hill
Franc Donkers, Ph.D., University of North Carolina (UNC) at Chapel Hill
Robert Hamer, Ph.D., University of North Carolina (UNC) at Chapel Hill
Sally Johnson, MD, University of North Carolina (UNC) at Chapel Hill
Half of military service members with combat-related traumatic brain injury (TBI) meet criteria for posttraumatic stress disorder (PTSD). Deficits in attention, executive function, affective and cognitive control associated with TBI, are compounded in the presence of PTSD. Problems with anger and violence are also common and predictive of poor outcome treatments.
To better understand the neural circuitry and neurocognitive functions underlying these deficits and to allow us to better identify therapeutic targets, we propose a randomized clinical trial using cognitive rehabilitation intervention with the participation of N=100 (n=50 experimental, n=50 usual care) veterans diagnosed with both PTSD and TBI. The experimental group is involved in a 6-month cognitive rehabilitation intervention program targeted at improving neurocognitive function. At baseline and at 6 months we will evaluate cognitive/ affective control and neuroanatomical measures to determine any group differences. We will do so by assessing behavior and neurocognitive performance, evaluating changes in cortical function using fMRI, neural activity using EEG, and white matter connectivity using diffusion tensor imaging (DTI).
To date, our efforts have focused on validating the DTI analysis pipeline and establishing analysis methods for tract-based statistical analysis procedures.We will present data demonstrating that our new analysis models enable the tracing and quantification of white matter properties along fronto-temporal and fronto-limbic pathways with high reliability.
We hypothesize that cognitive and affective control and performance will be enhanced and significantly improved in veterans in the experimental group and that there will be greater changes in brain structure and function after intervention as compared to the control group.
United States Department of Defense (DOD), University of North Carolina (UNC) School of Medicine,Howard Hughes Medical Institude (HHMI), Durham VA Medical Center
Rehabilitation, TBI, PTSD, DTI, Biomarkers
S100B AS A PROGNOSTIC BIOMARKER IN OUTCOME PREDICTION FOR PATIENTS WITH SEVERE TRAUMATIC BRAIN INJURY
Christian Niyonkuru, MS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Michelle Carter, BS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Krutika Amin, BS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Anthony Fabio, PhD, MPH, University of Pittsburgh - Department of Epidemiology
Rachel Berger, MD, MPH, Children's Hospital of Pittsburgh
Amy Wagner, MD, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Studies suggest that extra-cerebral contributions reduce the sensitivity of serum S100b for outcome prediction after TBI. We assessed longitudinal serum and cerebrospinal fluid (CSF) of adults with severe TBI to characterize contributing extra-cerebral sources of S100b to serum and identify the most sensitive S100b estimate for outcome prognostication.
This study evaluates S100b as a prognostic biomarker in adults with severe TBI (GCS≤8) by comparing outcomes with S100b temporal profiles generated from both CSF (n=138 subjects; n=499 samples) and serum (n=80 subjects; n=224 samples) across the first 6d post-injury. S100b was measured using Nexus Dx S100b enzyme-linked immunosorbent assays. Group based trajectory analysis (TRAJ) was used to generate subpopulations having unique temporal biomarker profiles. Long-bone fracture, injury severity score (ISS) derived from Abbreviated Injury Scores (AIS) to anatomical regions, and isolated TBI status were used to assess extra-cerebral sources of serum S100b. Isolated TBI included head AIS≥3 and AIS<3 for other anatomic regions. Non-isolated TBI included head AIS≥3 and AIS score≥3 for >1 other anatomic region. Acute mortality and Glasgow Outcome Scale (GOS) and Disability Rating Scale (DRS) scores were assessed 6 months post-TBI. Serum and CSF from healthy controls was used to assess injury-induced changes in S100b levels.
CSF and serum S100b levels in TBI subjects were significantly increased over healthy control values across the first six days post-TBI of sampling (p≤0.005 and p≤0.031, respectively). There was strong correlation between CSF and serum levels across early time points post-TBI, but this association diminished over time. Bivariate analysis showed that CSF TRAJ group membership with higher S100b concentration was linked to unfavorable outcomes, including higher acute care mortality (p<0.001), worse 6-month GOS scores (p=0.002), and worse 6-month DRS scores (p=0.005). Mean and peak CSF levels were less predictive of outcome and TRAJ group membership. Temporal serum profiles were strong predictors of acute mortality (p=0.001) in bivariate analysis. Peak serum S100b levels showed further evidence toward discriminating morality status (p=0.001). Evidence of significant extra-cerebral sources of serum S100b was represented best by ISS scores compared to other measures (e.g. fracture, isolated head injury status), where higher scores were associated with TRAJ group membership having higher serum S100b levels (p=0.034). Membership in TRAJ groups having higher CSF S100b levels were observed for women (p=0.021) and older subjects (p=0.004). Multivariate logistic regression analysis controlling for demographic and injury covariates confirmed CSF S100b TRAJ as a strong predictor of GOS (p=0.011) and DRS (p=0.094) following severe TBI, while multivariate analysis confirmed serum S100b TRAJ as predictive of acute mortality (p=0.016).
We demonstrate that S100b in CSF is a strong predictor of outcomes like acute care mortality, GOS, and DRS. Temporal serum S100b profiles were also strong in their predictive ability, despite extra-cerebral sources that could potentially confound estimates of S100b derived from cerebral sources. However, CSF-serum S100b correlation were strongest early after injury, and peak serum levels which typically occur early after TBI, were predictive of outcome. Future studies may confirm the relationship between CSF and serum S100b levels and should also focus on the prognostic value of early (e.g. admission) serum values when cerebrally-derived S100b is highest.
UPMC Trauma Registry team, Brain Trauma Research Center. This work was supported by CDC Grant: #R49/CCR323155 and DOD Grant: #W81XWH-07-1-0701.
TBI, S100b, CSF, serum, outcome
QUANTIFICATION OF THE BURDEN OF INTRAOPERATIVE SECONDARY INSULTS DURING ORTHOPEDIC SURGERY IN PATIENTS WITH TRAUMATIC BRAIN INJURY
Nelson Algarra, MD, University of Washington
Sumidtra Prathep, MD, University of Washington
Michael Souter, MbCh, FRCA, University of Washington
Monica Vavilala, MD, University of Washington
There are currently no data on the prevalence of intraoperative secondary insults during extracranial surgery in patients with traumatic brain injury (TBI). The aim of this descriptive study is to report the prevalence of secondary insults during orthopedic surgery within 2 weeks of moderate-severe TBI.
After IRB approval, we retrospectively reviewed the medical records and electronic anesthesia records of patients older than 18 years, with moderate-severe TBI (Glasgow Coma Scale score<13) who underwent one orthopedic surgery (long bone or pelvis) within two weeks of TBI at our institution between 2007 and 2011. We defined secondary insults as follows: • Hypotension: Systolic blood pressure<90 mmHg or CPP<50 mmHg (in patients with ICP monitoring) • Hypercarbia: End-tidal CO2>45 mmHg • Hypocarbia: End-tidal CO2<30 mmHg in the absence of intracranial hypertension • Hyperglycemia: Blood glucose>200 mg/dL • Hypoglycemia: Blood glucose<60 mg/dL • Intracranial hypertension: Intracranial Pressure (ICP)>20 mmHg
Data are reported as number (percent), median (range), and mean (SD) as appropriate.
Seventy-one patients (68% males) met eligibility criteria and were included in the final analysis. Admission Head Abbreviated Injury Score was 4±1, admission Glasgow Coma Scale (GCS) score was 4±2, preoperative GCS score was 9±3. Sixty-eight percent patients were intubated preoperatively and the timing of surgery after TBI was 0–2 days (50%), 3–7 days (46%) and 7–14 days (4%). Median duration of anesthesia and surgery were 200 (44–456) and 140 (18–378) minutes, respectively. Prevalence of intraoperative secondary insults was as follows:
Hypotension 64%
Intracranial hypertension 48%
Low cerebral perfusion pressure 37%
Hypercarbia 32%
Hypocarbia 29%
Hyperthermia 11%
Hyperglycemia 7%
Hypoglycemia 0%
Three-fourth patients had 1–2 secondary insults. Other intraoperative data were as follows (data as medians and ranges): lowest systolic blood pressure was 88(62-120) mmHg for all patients and 84(62-88) for hypotensive patients. Number of episodes of hypotension was 2 (0-26). Increase in ICP was 11 (-1-34) mmHg and the lowest CPP in patients with ICP monitoring was 55(29-101) mmHg. Lowest EtCO2 in patients with hypocarbia was 27 (23-29) mmHg and highest EtCO2 in hypercarbic patients was 56(46-68) mmHg. Glucose was monitored intraoperatively in 29(40%) patients and the highest glucose was 132(83-293) mg/dL.
Secondary insults are common during orthopedic surgery in patients with moderate-severe TBI, with hypotension, intracranial hypertension and reduced cerebral perfusion being most common and most patients sustaining 1–2 secondary insults. Anesthesiologists providing intraoperative care to these patients should be vigilant to minimize secondary insults to avoid worsening of primary injury to the brain. Future studies should evaluate the impact of these secondary insults on outcomes of TBI.
None
TBI, Anesthesia, Extracranial surgery
ECHOCARDIOGRAM UTILIZATION AND FINDINGS IN ISOLATED TRAUMATIC BRAIN INJURY
Deepak Sharma, MD, University of Washington
Sumidtra Prathep, MD, University of Washington
Matthew Hallman, MD, University of Washington
Aaron Joffe, MD, University of Washington
Burkhard Mackenson, MD, University of Washington
Traumatic brain injury (TBI) may have adverse cardiac consequences and brain-heart interactions may impact outcome (1,2). The aim of this study was to examine echocardiogram utilization and findings after TBI.
After IRB approval, data from patient's age >18 years, with isolated TBI (ICD 9 codes 800–801.9, 803–804.9 or 850–854.1) who underwent echocardiograms between 2003–2010 were examined. Patients with preexisting cardiac disease and chest injuries were excluded. Recorded clinical characteristics and echocardiogram data abstracted were ejection fraction (EF) and regional wall motion abnormalities (RWMA). Data are presented as mean +/− SD (range) and p<0.05 reflects significance.
Of 3,325 patients with isolated TBI, 157 (4.7%) had echocardiogram evaluation. Patients were 55±20 years, had admission Glasgow Coma Scale score of 10±5 (3–15), had subdural hematoma (64%) and were mostly male (68%). The most common indication was history of fall (43.9%) and most patients had one exam on hospital day 8. Low EF (<50%) was documented in 8% (EF 42±7%; ), and 17% had abnormal RWMA (mild hypokinesis 9%, moderate hypokinesis 2%, severe hypokinesis 5%, akinesis 1%). 8 (5%) patients received the vasopressor drug on the day of echocardiogram. Of 3,325 patients with isolated TBI, 157 (4.7%) had echocardiogram evaluation. Patients were 55+20 years, had admission Glasgow Coma Scale score of 10+5 (3–15), had subdural hematoma (64%) and were mostly male (68%; Table 1). The most common indication was history of fall (43.9%) and most patients had one exam on hospital day 8. Low EF was documented in 8% (EF 42+7%), and 17% had abnormal RWMA. Only 8 (5%) patients received the vasopressor drug on the day of echocardiogram.
Echocardiograms are infrequently obtained in isolated TBI patients but when performed, the most common indication was for a history of fall. Despite lack of pre-existing cardiac disease, or chest injury, cardiac dysfunction after isolated TBI was common, suggesting adverse cardiac effects of TBI.
Funding: NIH /R01 NS072308-01
traumatic brain injury, echocardiogram
EXPOSURE CHARACTERISTICS OF UCH-L1 IN CSF ARE ASSOCIATED WITH 6-MONTH OUTCOME IN SEVERE TBI
Linda Papa, MD, Orlando Regional Medical Center
Andrea Gabrielli, MD, University of Florida
Ilona Schmalfuss, MD, NF/SG Veterans Administration and University of Florida
H. Julia Hannay, PhD, University of Houston
Shelley Heaton, PhD, University of Florida
Kevin K.W. Wang, PhD, University of Florida McKnight Brain Institute
Ronald Hayes, PhD, Banyan Biomarkers
Claudia Robertson, MD, Baylor College of Medicine
Steve Robicsek, MD, PhD, University of Florida
We examined whether the exposure and kinetic characteristics of UCH-L1 measured in CSF were associated with 6 month outcome in patients with severe TBI.
This is prospective observational study that enrolled adults with severe TBI from to 2 Level I Trauma Centers. Patients were included if they had blunt head injury with a GCS score of ≤8 and required ventricular CSF drainage or access. CSF was sampled from each patient at enrollment, 6, 12, 18, 24, 48, 72, 96, 120, 144, 168, 192, 216 and 240 hours following TBI and analyzed for UCH-L1 (ng/ml) using ELISA. Results are presented using medians and interquartile range. Injury severity was assessed by 6-month dichotomized GOS defined by the traditional dichotomy of the GOS score (death, vegetative state and severe disability). Exposure and kinetic parameters included AUC, Cmax, Tmax, MRT and half-life.
At 6 months, 93 patients had both follow-up and exposure data available for analysis, 63 (68) patients had poor outcome per GOS. Mean age of patients was 38 years and 78 were male. When exposure and kinetic characteristics were compared to the type and severity of injury assessed by the Rotterdam CT score, AUC and Cmax were significantly correlated with the Rotterdam score rho=0.42 (p<0.001) and rho=0.41 (p<0.001). The median AUC (n=93) in those with good and poor recovery respectively was 1540 [987–2085] vs 2913 [1455–6815] ng/ml*hrs (p=0.001), the median Cmax (n=93) was 71.2 [30.6–96.8] vs 132.2 [67.8–208.7] ng/ml (p=0.001), the median Tmax (n=93) was 15.7 [11.2–19.5] vs 18.0 [11.2–24.6] hrs (p=0.393), the median half-life (n=72) was 4.5 [2.3–6.7] vs 7.9 [3.9–10.5] hrs (p=0.010), and median MRT (n=87) was 38.0 [17.1–53.1] vs 32.3 [22.9–81.4] hrs (p=0.309).
Exposure and kinetic analysis of UCH-L1 in the CSF showed a significantly larger AUC, higher peak concentration and longer half-life in patients that had a poor recovery as compared to those with a good recovery. There was also a significant correlation between AUC, Cmax and Rotterdam scores. These differences support UCH-L1 as a potential biomarker for predicting outcomes in severe TBI.
This study was generously supported by NIH RO1 NS052831 “Biochemical Markers of Severe Traumatic Brain Injury.”
biomarker, TBI, UCH-L1, kinetics, outcome
CLINICAL EFFECTIVENESS OF FENTANYL AND MIDAZOLAM FOR INTRACRANIAL HYPERTENSION IN PEDIATRIC TRAUMATIC BRAIN INJURY
Michael Wallendorf, PhD, Washington University School of Medicine
Evan Kharasch, MD, PhD, Washington University School of Medicine
Allan Doctor, MD, Washington University School of Medicine
Timothy P. Welch, MD, MPH, Washington University School of Medicine
Traumatic brain injury (TBI) is often complicated by intracranial hypertension (ICH). Treatment for ICH includes pharmacologic reduction in intracranial pressure (ICP). Sedation and analgesia with benzodiazepines and opioids is considered first-tier therapy, despite of lack of evidence of effectiveness and potentially harm due to cerebral hypoperfusion (low CPP).
Pediatric TBI patients who received at least one dose of fentanyl or midazolam in response to a sustained elevation of ICP (>20 mmHg>5 minutes).
High-resolution data recording of ICP and CPP was obtained at 1.8 second intervals. This data was imported into LABpilot, allowing area-under-the-curve (AUC) calculations.For each dose administration of the study drugs, we confirmed the absence of confounding coadministration of additional ICP-directed therapies. We then examined the AUC for both ICP over 20 and CPP deficit (age threshold CPP – patient CPP) in the 15 minute epochs before and after drug administration. AUC were compared using a paired t-test. We also examined the effects of putative confounding variables, such as age, time after injury, drug dosage type, and injury mechanism and severity. Pearson correlation coefficients were computed to assess for potential covariate effects of these factors. If present, factors were entered into a mixed random effects ANCOVA model.
A total of 278 discrete date-time records were obtained from 37 patients. Sustained reduction in ICP below 20 mmHg was observed in only 10 records (3.6%). AUC-ICP increased by 13.67 mmHg*min (CI 6.9, 20.5; p=0.0002) after drug administration.When accounting for differences in drug dose and combination as a fixed effect, the mean change in AUC-ICP was 5.12 mmHg*min (CI −7.1,17.3; p=0.399). AUC-CPP deficit increased in 87 of 205 records (42%). Age and severity of injury by GCS were not significant covariates.There were no meaningful correlations between absolute magnitude of change between peak and nadir ICP and CPP, and AUC for each parameter.
Fentanyl and midazolam have a defined role in the management of pediatric ICU patients, and under certain conditions, may contribute to ICP control. However, our data does not support their routine usage as bolus therapy in treating intracranial hypertension.Furthermore, their use may contribute to both increases in ICP burden and reductions in cerebral perfusion.Ongoing research will help us determine other patient-specific factors which might predict responsiveness to these medications.
St. Louis Children's Hospital Foundation
Brain injury, sedation, pediatric
CHARACTERIZATION OF βII-SPECTRIN DEGRADATION UNDER BOTH EXCITOTOXIC CHALLENGE AND AFTER TRAUMATIC BRAIN INJURY
Joy Guingab, PhD, Banyan Biomarkers
John Anagli, PhD, Banyan Biomarkers
Kevin Wang, PhD, Department of Psychiatry, University of Florida
Ronald Hayes, PhD, Banyan Biomarkers
A major consequence of TBI is the rapid proteolytic degradation of structural cytoskeletal proteins. In this study, we investigate the integrity of βII-spectrin protein and its proteolytic profile both in primary rat cerebrocortical cultures under apoptotic, necrotic and excitotoxic challenge and in a rat model of experimental TBI.
A controlled cortical impact (CCI) device was used to model TBI in rats. Brain trauma was produced by impacting the ipsilateral cortex with a 5 mm diameter aluminum impactor tip at a velocity of 3.5 m/s with a 1.6 mm compression and 150 ms dwell-time. Cerebrocortical cells harvested from 1-day old Sprague-Dawley rat brains were plated for cell culture studies. the following conditions were used: NMDA (N-methyl-D-aspartate; 300 M) for 3 24 hrs as an excitotoxic effect; apoptotic inducers STS (staurosporine) that activates calpain and caspase-3 for 24 hrs; the Ca2 chelator EDTA (2 mM; for up to 24 hrs as a caspase-dominant challenge). For pharmacological intervention, cultures were pretreated 1 hr before the STS, EDTA or NMDA challenge with 30 M of the calpain inhibitor SNJ-1945 (Senju Pharmaceuticals, Kobe, Japan or with 20 M the caspase-3 inhibitor IDN-6556).
Results revealed that the intact 260 kDa βII-spectrin was degraded into major fragments (βII-spectrin breakdown products; βsBDPs) of 110 kDa, 108 kDa, 85 kDa and 80 kDa in rat brain (hippocampus and cortex) at 48 hrs post injury. These βsBDPs profile were further characterized and compared to an in vitro βII-spectrin fragmentation pattern of naïve rat cortex lysate digested by calpain-2 and caspase-3. Results revealed that βII-spectrin was degraded into major fragments of 110 kDa/85 kDa by calpain-2 activation and 108 kDa/80 kDa by caspase-3 activation.
These data strongly support the hypothesis that the in vivo overactivation of multiple protease system induces βII-spectrin proteolysis via a specific calpain and/or caspase-mediated pathway resulting in a signature, protease-specific βsBDPs that is dependent upon the type of neural injury mechanism.
We would like to thank Dr. M C Liu for his help.
Biomarkers, proteases, βII-spectrin, TBI
CSF PROTEIN BIOMARKER PANEL FOR ASSESSMENT OF NEUROTOXICITY
Andreas Jeromin, PhD, Banyan Biomarkers, Inc.
Danny Johnson, Banyan Biomarkers, Inc.
Juan Martinez, PhD, Banyan Biomarkers, Inc.
Ronald Hayes, PhD, Banyan Biomarkers, Inc.
Stefania Mondello, MD, MPH, PhD, Banyan Biomarkers, Inc.
The goal of this study was to investigate CSF concentrations of ubiquitin C-terminal hydrolase 1 (UCHL-1), glial fibrillary acidic protein (GFAP), αll-spectrin break down products (SBDP150, SBDP145 and SBDP120) and their relation to neuropathology and neurodegeneration in an animal model of kainic acid (KA) excitotoxicity.
Methods: Neurotoxicity in adult rats was induced by subcutaneous injections of KA (9 mg/kg). Brain tissues and CSF sample were collected at different time points after injection. The levels and cellular localization of biomarkers were examined using ELISA in CSF and immunohistochemical (IHC) analysis on paraffin-embedded 6 μm brain sections, respectively. Overall brain pathology was examined using H&E staining and DAPI fluorescence, and neuronal degeneration was examined using Fluoro-Jade stain. Occurrence of status epilepticus with duration for at least 2 h after KA injection was confirmed in all animals from treatment group used in analysis. Statistical analysis for quantitative data were performed using One-way ANOVA followed by Bonferroni's Multiple Comparison Test. Data were reported as mean μ standard error of mean. P value less that 0.05 was considered as significant.
Results: Triple fluorescent labeling and Fluoro-Jade C staining revealed a reactive gliosis and appearance of degenerating neurons in hippocampus of KA-treated rats. Gliosis but no signs of neurodegeneration were observed in the cerebral cortex. IHC showed upregulation of GFAP expression in hippocampus and in cortex starting 24 h post-KA injection with a peak at 48 h. At these time points concurrent with extensive neurodegeneration, SBDP150, SBDP145 and SBDP120 were observed throughout the brain. Expression of UCHL-1 in hippocampus of vehicle-treated animal was observed as appearance of immunopositive material surrounding neuronal nuclei. In experimental animals at 24 h post-KA injection was observed a loss of structural integrity in cellular distribution of UCHL1 that correlated with increase in immunopositive material in the extracellular matrix. CSF levels of UCHL-1, GFAP and SBDPs were significantly increased in KA-treated animals compared to controls. The temporal increase in CSF biomarkers correlated with brain tissue distribution and neurodegeneration.
Conclusion: This study demonstrated validity of use CSF levels of glial and neuronal proteins for assessment of neurotoxic damage in humans. The spatiotemporal profile of the proposed CSF biomarker panel including UCHL1, GFAP and αll-spectrin break down products (i.e. SBDP150, SBDP145 and SBDP120) is associated with progression of neurodegeneration in specific brain regions. The difference in the temporal profile of accumulation of these biomarkers in CSF might provide tools for better assessment of neurotoxicity during acute and chronic stages. The molecular nature and the differences in cellular and subcellular origins of these biomarkers can also provide critical information on the underlying mechanisms of neurotoxicity that might facilitate the identification of drug candidates and allow physicians to monitor drug safety.
None
Neurotoxicity, Biomarker
EFFECT OF SINGLE DOSE VS REPEATED DOSE OF HYDROPHILIC CARBON CLUSTERS IN MILD TRAUMATIC BRAIN INJURY MODEL SUPERIMPOSED WITH HEMORRHAGIC SHOCK IN RATS
Brittany Bitner, Ph.D, Baylor College of Medicine
Daniela Marcano, MS, Rice University
Roderic Fabian, MD, Ph.D, Baylor College of Medicine
Thomas Kent, MD, Baylor College of Medicine
James Tour, Ph.D, Rice University
Claudia Robertson, MD, Baylor College of Medicine
Traumatic brain injury (TBI) impairs cerebral blood flow (CBF) due to oxidative damage by increased superoxide and hydroxyl radicals. Polyethylene glycolated hydophilic carbon clusters (PEG-HCCs) have antioxidant properties in in vitro models of oxidative stress. This study determines if a second dose of PEG-HCCs can increase CBF in TBI.
Long Evans rats were anesthetized with isoflurane in oxygen, intubated and ventilated to maintain normal arterial blood gases. An impact injury of 3 m/sec and 2.5 mm deformation was induced at the right parietal cortex followed by blood withdrawl (2.0 mL/100 g body weight) to reduce mean arterial pressure (MAP) to 40 mmHg. This hypotensive phase was maintained for 50 min in room air. Next, a ‘prehospital care’ phase included infusion of lactated Ringer's solution in 1 mL boluses to achieve MAP of 50 mmHg and lasted 30 min. During the ‘definitive hospital care’ phase, the anesthesia was switched to isoflurane in 100% oxygen and blood was re-infused. The drugs were given intravenously prior to definitive hospital care phase and 2 h afterwards. The animals were monitored for 6 h post TBI. The CBF was measured with Periscan PIM3 imager and expressed as % from baseline.
Treatment with a single dose of PEG-HCCs significantly increased cerebral blood flow at the center of the impact as compared tophosphate buffered saline (PBS ) treated animals at 140 and 200 min post-TBI. (RM ANOVA: time effect p<.0001, treatment effect p=0.0059, Bonferroni post test). The relative CBF was maintained in the injured brain at the end of the monitoring period (6 h post TBI) with a second dose of PEG-HCCs (91.9±29.5of baseline CBF in 2 dose compared to 57.1±11.9 of baseline in single dose study). The relative CBF in the uninjured contralateral cortex was similar between 2 dose PEG-HCCs and vehicle-treated rats, 104.1±6.9 and 93.5±5.9, respectively. The mean arterial blood pressure was not significantly different between groups (p=0.507, RM ANOVA with Bonferroni post test).
These results suggest that treatment with a second dose of PEG-HCCs sustains the cerebral blood flow after MTBI complicated by hemorrhagic shock. This effect may be due to the antioxidant properties of the PEG-HCCs by decreasing superoxide anion and normalizing the vasodilator nitric oxide. The antioxidant's ability to increase CBF lasted for 2 −3 h, which correlates with the half life of the drug in the blood.
Supported by Department of Defense ( DOD ) W81XWH-08-2-0141
Traumatic brain injury
TREATMENT WITH CANDESARTAN, THE ANGIOTENSIN II TYPE I RECEPTOR ANTAGONIST, AFTER TRAUMATIC BRAIN INJURY IS NEUROPROTECTIVE IN MICE
Trevor T. Logan, Center for Neuroscience and Regenerative Medicine (CNRM)
Alexandra Yaszemski, Center for Neuroscience and Regenerative Medicine (CNRM)
Kwame Affram, MD, Center for Neuroscience and Regenerative Medicine (CNRM)
Juan M. Saavedra, MD, NIH
Aviva J. Symes, PhD, Center for Neuroscience and Regenerative Medicine (CNRM)
Angiotensin II is produced in the brain and Angiotensin II AT receptoroverstimulation is vasoconstrictive and pro-inflammatory. We wanted to examine the effects of Angiotensin II receptor blockers (ARBs) administering candesartan after injury, and to determine an effective dose.
We have previously shown that the ARB candesartan is neuroprotective when administered 5 hours before controlled cortical impact (CCI) in mice. We performed moderate controlled cortical impact (CCI) injury on 9-week-old male C57BL/6 mice (3.6 m/s, 1 mm depth, 100 msec dwell time). We then sought to determine if the ARB, candesartan, would mitigate the effects of the injury. Mice were injected i.p. once a day with either 0.1 or 1 mg/kg candesartan, or vehicle starting one hour after injury, and were sacrificed at 1 or 3 days post injury (dpi). Lesion volume and cell death were examined at 1 dpi. Rotarod and open field tests were performed at 1 and 3 dpi. Cerebral blood flow (CBF) was also measured at 1 and 3 dpi around the cortical impact area using a laser-Doppler flowmeter, and tail blood pressure was obtained.
We found that the lower dose of candesartan (0.1 mg/kg) was more effective, reducing the lesion volume at 1 dpi by approximately 37% compared with a 22% reduction of lesion volume with the higher dose of candesartan (1 mg/kg). Only the 0.1 mg/kg dose of candesartan significantly reduced the number of dead cells and oxidative stress damage in the pericontusional tissue. Also, both dose reduced the number of activated microglial cells, despite the low dose was more noticeable. However, both doses of candesartan raised the cerebral blood flow at one day and three days after CCI in comparison with injured mice treated only with vehicle. Interestingly, the higher dose of candesartan, but not the lower dose, reduced systemic blood pressure. Rotarod tests indicated that 0.1 mg/kg candesartan improved the motor function outcome of the mice at 3dpi.
Our data indicate that post-injury administration of low dose candesartan (0.1 mg/kg) can ameliorate deficits resulting from TBI, but further work is necessary to determine the time window after injury where candesartan treatment will remain beneficial.
This work was supported by a pilot grant and a post-doctoral fellowship from the Centre for Neuroscience and Regenerative Medicine.
neuroprotection, blood flow, inflammation, candesartan
DYNAMIC CHANGES IN ATP-BINDING CASSETTE/MULTI-DRUG RESISTANCE TRANSPORTERS AFTER TRAUMATIC BRAIN INJURY IN MICE
Philip E. Empey, PharmD, PhD, School of Pharmacy, University of Pittsburgh
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Christina Hosler, MS, Safar Center for Resuscitation Research
Yaming Chen, MD, Safar Center for Resuscitation Research
Robert S.B. Clark, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
ATP-binding cassette (Abc)/multi-drug resistance-associated protein (Mrp) transporters are key components of blood-brain and brain-CSF barriers, regulating the brain's biochemical milieu. Changes in expression after traumatic brain injury (TBI) are unknown. We hypothesized that Abcb1/p-glycoprotein, Abcc1/Mrp1 and Abcc2/Mrp2 expression are altered after TBI in mice.
Anesthetized adult male mice were subjected to moderate controlled cortical impact (CCI; 1.2 mm depth, 6 m/s velocity) then sacrificed at 6, 24 and 72 h (n=4/group) by perfusion fixation with 4% paraformaldehyde. Brains were removed, post-fixed, cryoprotected, cut in coronal sections, and processed for immunohistochemistry using antibodies against Abcb1, Abcc1 and Abcc2. Brain sections were dual-labeled using antibodies against NeuN, GFAP and Glut1 to identify neurons, astrocytes, and endothelial cells, respectively. Naive and sham injured mice served as controls. Immunoreactivity was qualitatively scored in coronal brain sections through the dorsal hippocampus using a 0–4 scale by an observer blinded to experimental group.
Baseline Abcb1 and Abcc2 immunoreactivity was seen in endothelial cells, choroid plexus and ependymal cells, while baseline Abcc1 immunoreactivity was seen primarily in ependymal cells. Abcb1 and Abcc2 expression in choroid plexus at the brain-CSF barrier was abluminal and did not appear to change after CCI versus controls. Abcb1 immunoreactivity appeared increased in endothelial cells (bilateral) and was now seen in neurons (ipsilateral) at 24 and 72h after CCI versus controls; although Abcb1 abundance scored qualitatively was similar (p=0.1; Kruskal Wallis). Neuronal immunoreactivity was primarily somal and was evident at 24 h after CCI. Abcc1 immunoreactivity was increased over time in ependymal and peri-ependymal cells, and neurons in the pericontusional area versus controls (p=0.02; Kruskal Wallis/Student Newman Keuls). Neuronal immunoreactivity was primarily in neurites and was most evident at 72 h after CCI. Abcc2 immunoreactivity was increased over time in endothelial cells and neurons in the pericontusional area versus controls (p=0.01; Kruskal Wallis/Student Newman Keuls). Neuronal immunoreactivity was primarily somal and was evident at 24 h after CCI. Expression of Abcb1, Abcc1 or Abcc2 in astrocytes was not detected after TBI or in controls.
TBI results in increased expression and/or redistribution of the Abc transporters Abcb1, Abcc1 and Abcc2, not only in cells lining the blood-brain and brain-CSF barriers, but also in neurons. These findings could have important implications for TBI related to transport of pharmacological and endogenous substrates, not only in and out of the brain, but in and out of neurons after injury as well. As an example, temporal changes in Abcb1, Abcc1, and Abcc2 may influence brain bioavailability and effectiveness of many therapeutics commonly used in the clinical management of TBI patients, and adjusted dosing requirements for agents known to be transporter substrates (e.g. narcotics, antiepileptics, and antibiotics) may be required. Further investigation and pharmacokinetic studies in humans after TBI appear warranted. Finally, agents that inhibit Abcb1, Abcc1 and/or Abcc2 transporters may represent novel pharmacologic adjuncts for future TBI therapies.
NIH R01 NS069247
blood-brain barrier, brain-CSF barrier, p-glycoprotein
REPETITIVE MILD BRAIN INJURY ELICITS CEREBRAL VASCULAR ABNORMALITIES WHICH CAN BE ATTENUATED BY THERAPEUTIC INTERVENTION
Motoki Fujita, PhD, Advanced Medical Emergency and Critical Care Center/Yamaguchi University/Japan
Enoch Wei, PhD, Anatomy and Neurobiology, Virginia Commonwealth University
John T. Povlishock, PhD, Anatomy & Neurobiology, Virginia Commonwealth University
While several studies suggest that repetitive mild injury exacerbates axonal and synaptic abnormalities, no evidence exists for comparable dysfunction of the cerebral vasculature which has been inferred in human second impact syndrome. In this study, we examined the effect of low intensity impacts on the cerebral microcirculation in rats.
Adult male Spague Dawley rats were subjected to a single impact from a height of 0.5 to 1 meter. Cerebral vascular reactivity to ACh was assessed at 3 and 4 h post injury. In a separate series of animals a second impact was delivered from the same height 3, 5 and 10 h later. Following the second impact and in response to the ensuing vascular abnormalities (see below) we employed two therapeutic interventions which included the use of hypothermia (33 °C), or FK506 initiated 1 h after the second impact.
A single impact from the height of 0.5 to 1 meter exerted no alteration of cerebral vascular function. The same was true when the weight was dropped a second time from a height of either 0.5 or 0.75 meters. In contrast, a second impact from 1 meter, initiated 3 h following the initial traumatic insult, rendered the cerebral arterioles fully non-responsive to ACh. This dramatic cerebral vascular dysfunction, however, was reduced or eliminated when the interval between the insults was elongated to 5 to 10 hours, respectively. Additionally, the delayed use of either hypothermia, or systemically administered FK506 provided significant vascular protection.
Collectively, these studies illustrate that repetitive traumatic injuries of specific injury severity and interval time span can evoke profound vascular dysfunction which can be attenuated by therapeutic targeting. While the overall implications of these vascular findings remain unclear, they most likely would place the repetitively injured brain at risk. Further, these vascular abnormalities may be consistent with the vasoparalysis described in some cases of human second impact syndrome.
This study is supported by HD055813.
cerebral vascular reactivity, hypothermia, FK506
DEMONSTRATION OF HIGHER ANXIETY LEVELS IN RATS WITH BLAST INJURIES COMPARED TO ROTATIONAL INJURIES OR CONTROLS
Holly Sikora, BS, Medical College of Wisconsin
Michael McCrea, PhD, Medical College of Wisconsin
Matthew Budde, PhD, Medical College of Wisconsin
Alok Shah, MS, Medical College of Wisconsin
Frank Pintar, PhD, Medical College of Wisconsin
Although biomechanical mechanisms for blast- and rotationally-induced mTBI are quite different, the two injuries are generally treated using similar clinical protocols. Whereas acute and chronic deficits for rotationally-induced mTBI are somewhat well defined, blast-related symptoms are more complex and are commonly associated with post-traumatic stress disorder (PTSD).
Sprague-Dawley rats were anesthetized using Isoflurine and exposed to rotationally- or blast-induced mTBI. Rotational injuries were induced using the MCW Rotational Injury model, with head angular acceleration characteristics sufficient to induce mild TBI. Blast injuries were induced using a shocktube apparatus generating shockwaves of 436 kPa peak overpressure with 0.2 ms duration. Control rats were subjected to rotational or blast protocols without exposure to head rotational acceleration or shockwave. Rats were exposed to behavioral testing protocols on post-injury days 2–4. Elevated Plus Maze (EPM) assessments were conducted on post-injury day 2 (5 min), Open Field testing (OFT) was conducted on post-injury day 3 (15 min), and Morris Water Maze Assessment (MWM) was conducted on post-injury days 2–4 (2 sets of 4 trials per day). These assessments were focused on determination of elevated levels of post-injury anxiety, aggressive behavior, and spatial learning deficits.
All experimental rats survived shockwave (n=4) or head rotational acceleration (n=4) without skull fracture or cervical spine injury. The visuo-spatial learning paradigm of the MWM did not identify spatial learning deficits for rats with rotational or shockwave injury as latencies were not different between controls (n=6) or injured groups. OFT identified a number of differences between experimental groups and controls. Both injury groups were less active at later time points in the OFT, demonstrating less total distance traveled and less distance traveled around the outer border of the maze. Whereas both injury groups entered the center of the maze less frequently than controls, rats receiving shockwave injury entered the center of the maze a considerably fewer number of times. During the total duration of OFT, control rats entered the center of the maze an average of seven times, the rotational group entered the center of the maze an average of five times, and the shockwave injury group entered the center of the maze only two times. Likewise, the amount of time in the center of the maze and total distance traveled in the center of the maze was considerably less in the shockwave injury group than other groups. Some differences were also identified in the EPM. Rats receiving shockwave injury were considerably more active early in the event (minutes 1–3) than controls or the rotational injury group. That early activity included more frequent arm changes and more time spent in the open arms for the shockwave injury group. Likewise, the shockwave injury group demonstrated considerably less activity late in the trial, with fewer arm changes and less time in the open arms than the other groups.
Although the number of rats in each group was somewhat limited in this preliminary study, post-injury behavioral deficits were evident between the injured groups and controls. Trends for the two injured groups were similar, although deficits in rats receiving shockwave injury were enhanced. For example, less late OFT activity was evident for both injured groups although rats receiving shockwave injury had considerably less activity than rats receiving rotational injury. The types of behaviors demonstrated by injured rats in this study (e.g., more EPM closed arm time and less OFT center time) can generally be attributed to higher anxiety levels, although other factors may also induce these behaviors. Results of this ongoing study will be used as a basis to continue to define post-injury differences between rotationally and shockwave induced injuries.
This research was supported by Department of Veterans Affairs, Veterans Health Administration, Rehabilitation Research and Development Service, and Advancing a Healthier Wisconsin.
blast, behavioral testing, anxiety, mTBI
ALTERATIONS IN HIPPOCAMPAL CIRCUITRY IN A SWINE MODEL OF DIFFUSE BRAIN INJURY
Brian Johnson, Ph.D., Children's Hospital of Philadelphia
Constance Mietus, B.A., University of Pennsylvania
Kevin Browne, B.A., University of Pennsylvania
Victoria Johnson, MBChB, Ph.D., The University of Pennsylvania
Douglas Smith, M.D., The University of Pennsylvania
M. Sean Grady, M.D., Hospital of the University of Pennsylvania
Akiva S.. Cohen, Ph.D., Children's Hospital of Philadelphia
D. Kacy Cullen, Ph.D., University of Pennsylvania
Progressive neurodegenerative changes induced by rapid head rotational acceleration have been extensively studied using a model of non-impact diffuse brain injury (DBI) in swine. The current investigation advances these studies by establishing whether this inertial DBI leads to changes in hippocampal circuitry and function.
We developed a slice paradigm for swine hippocampus and examined limbic function following DBI or sham conditions using extracellular recording techniques. Either eight hours (8hr) or seven days (7d) following rotational head injury, pigs were transcardially perfused with artificial cerebral spinal fluid (aCSF), one of the hemispheres was extracted, and the hippocampus isolated. The contralateral side was perfused with paraformaldehyde and prepared for histological examination. Transverse slices of the live hippocampal tissue (350μM) were prepared and extracellular recordings were performed in an interface chamber while slices were bathed in oxygenated aCSF. Input-output curves were generated and paired-pulse paradigms were utilized to examine changes in excitability and neurotransmitter release probabilities in slices from injured versus sham animals in both area CA1 and the dentate gyrus.
Swine undergoing rotational injury in the coronal plane regained balance and began feeding within a few hours. Swine undergoing sagittal injuries either had difficulty regaining consciousness and were sacrificed at 8 hours, or regained consciousness and were sacrificed at 7 days. Hippocampal slice recordings revealed overt changes in the waveforms of the responses to stimulation in traces from CA1 and dentate at 8hr and 7d. These alterations suggest an increased post-synaptic excitability in the cell layers receiving the EPSPs, or a shift in excitation-inhibition balance of the local circuitry. In addition, there was a loss of paired pulse potentiation in area CA1, potentially due to changes in neurotransmitter release probability. Paired pulse depression was evident in the medial perforant path in sham, and was significantly further decreased at 7d post injury. In addition, there were significant changes in initial fiber volley slope post injury, with the largest decrease in the sagittal plane at the 8hr time point. This was most significant in the CA1 and lateral perforant path of the dentate. At 7d, this had resolved in the sagittal plane. In the coronal plane, the changes in CA1 were significant, and also for a subset of currents in the lateral perforant path. There were no significant changes in the input/output curves. When plotted against the fiber volley slopes, the injured 7d responses were greater than sham. If the fiber volley was considered to be the “input” instead of the current, this would imply a greater response (“output”) from the CA1 EPSPs. Histopathological examination of the contralateral CA1 (i.e. Schaffer collaterals) and perforant path axons was performed. This analysis has thus far not revealed a degree of axonal degeneration consistent with the reduction in fiber volley amplitudes, suggesting axonal dysfunction or reduced efficacy.
These data indicate that non-impact inertial DBI in swine leads to dysfunction in various aspects of hippocampal circuitry, and suggests that this may be a useful pre-clinical model of human TBI. Further histopathological examination may indicate potential substrates for neuronal excitability changes and circuit disruption. In addition, we are currently utilizing in vivo electrophysiological techniques to validate these findings.
R01-NS050598
diffuse electrophysiology hippocampus excitability circuitry
ACOUSTIC RADIATION FORCE IMPULSE IMAGING IMPROVES ULTRASOUND RESOLUTION IN NEURAL TISSUE: EFFECTS OF TEMPERATURE AND CONFINEMENT ON BRAIN MATERIAL PROPERTY CHARACTERIZATION
Mark Palmeri, M.D., Ph.D., Duke University
Ned Rouze, Ph.D., Duke University
Nick Kloppenborg, M.S., Duke University
Cameron R. Dale Bass, Ph.D., Duke University
Brain injury remains a leading cause of mortality and morbidity in the United States. To model the biomechanics of injury in acute brain trauma, the mechanical behavior of the brain must be well-characterized. Unfortunately, in vitro viscoelastic brain stiffness and relaxation parameters in the literature provide little consensus.
This study provides in vivo and in vitro brain properties using the Acoustic Radiation Force Impulse (ARFI) imaging technique. Briefly, multiple powerful acoustic-radiation pulses were applied, causing local tissue strain propagating laterally from the focus. These displacements were tracked through time. For the in vivo study, a healthy pig was anesthetized, and bilateral holes were drilled tangentially through the temporal skull cap. An AcuNav ultrasound catheter (Siemens, CA) was inserted into one hole, B-mode ultrasound was used to visualize region before applying ARFI. For the in vitro part, specimens were removed from the skull and placed in a saline water bath at either 20°C or 37°C. Resulting displacements were analyzed using a computational algorithm that accounts for both instantaneous elastic and viscoelastic response using a novel level set algorithm to determine time-to-peak displacement and attenuation at laterally offset positions along every vector from the line of excitation.
Results from this study suggest that the new techniques for estimation of the ARFI shear wave speed, elastic shear modulus and viscoelasticity of the white matter are robust across experimental condition and variation in acoustic input. Experimental white matter testing using ARFI with in vitro brains at 20° C yielded a shear modulus of 3.03±0.87kPa while brains in a water bath at 37° C yielded a shear modulus of 1.06±0.37kPa. Linear regression was used to correlate shear modulus with temperature (R2=0.69) for values normalized to mean shear modulus at 20° C. These results suggests that brain temperature has a strong effect on local material properties. In vivo ARFI data produced shear moduli of 3.06±0.49kPa. Interestingly, this is statistically different from the in vitro modulus at 37° C (p<0.001) but not 20° C (p=0.91). This result suggests local physiological pressure effects or confinement effects on white matter brain material properties in vivo. Viscoelastic attenuation values in both in vitro and in vivo testing indicate that the white matter is strongly viscoelastic with power law frequency dependence. Results from this study overcome limitations of previous studies, particularly compromised anatomical and physiological conditions, and introduction of unrealistic boundary conditions and local geometric complexity.
The goal of this experimental series was to use AFRI and ultrasound imaging to evaluate the effects of temperature on shear modulus in brain tissue to compare in vivo and in vitro material response. White matter brain properties in this study show strong temperature and in vivo vs in vitro confinement changes. Since many of the experimental results used in numerical modeling are derived from room temperature testing, the results of this study may be used to produce more biofidelic brain models. In addition to establishing reasonable values for modelers, this technique may have use in vivo to assess pathological conditions such as elevated intracranial pressure or local pathological changes in brain tissue stiffness. Continued testing is needed to discriminate the detailed temperature dependence with respect to other environmental changes.
NAVAIR - Naval Air Systems Command
Ultrasound, brain material properties, viscoelasticity
NEUROSTEROIDS IMPROVE FUNCTIONAL OUTCOMES FOLLOWING REPETITIVE MILD TRAUMATIC BRAIN INJURY
Vedrana Marin, Ph.D., Florida State University College of Medicine
Gabriel Glaun, none, Florida State University College of Medicine
Shanese Doss, none, Florida State University College of Medicine
Helen Phipps, Ph.D. Candidate Biomedical Science, Florida State University College of Medicine
Jacob VanLandingham, Ph.D. Neuroscience, Florida State University College of Medicine
Repetitive mild traumatic brain injury (mTBI) can lead to long term disability. The purpose of this study was to determine if prophylactic or acute post-treatment with a neurosteroid improves motoric, emotional and cognitive function in an animal model.
Male Sprague Dawley rats were placed in one of four groups (n=10/group); injury-vehicle, injury-pre-treated progesterone (PROG, 16 mg/kg), injury-post-treated PROG, sham. All injured groups received 3 mTBI, 30 min apart. Pre- treatments were administered 15 min prior to the first impact and post-treatments at 15 min following the third impact. Briefly, the animals were anesthetized and stabilized in prone between the 6th and 7th cervical vertebrae. A scalp incision was made followed by placement of a 1mm thick rubber disk over the right frontal plate. A metal impactor was activated at a 15° angle, perpendicular to the disk, at a velocity of 6.6 m/s. The impact induced a downward rotation of the head to the left. Behavioral analysis using Morris Water Maze (MWM, spatial learning and memory), Elevated Plus Maze (EPM, anxiety-like behavior) and Rotarod (balance) testing were performed on allanimals in the study.
At 3 and 24h, animals treated with either pre-or post PROG had a significant reduction in latency to find the MWM platform on the first and second trial compared to the injury-vehicle group (p<0.05). These same findings were evident when comparing thigmotactic behavior. Significant (p<0.05) improvements in Rotarod performance were seen in the treated groups compared to vehicle at 3 and 24 hrs. Anxiety-like behaviors were also significantly (p<0.05) reduced with treatment.
Our findings show that both prophylactic and acute post-injury treatment with PROG improves cognitive, stress and motoric behaviors in an animal model of repetitive mild TBI.
Florida State University College of Medicine, Department of Biomedical Sciences
Repetitive brain injury, cognition, balance
CARDIAC ISCHEMIC CHANGES IN A SWINE MODEL OF BLAST TBI
Valerie Coppes, SF VA Hospital
S. Scott Panter, PhD, SF VA Hospital
Blast-induced traumatic brain injury (BI-TBI) is a significant cause of morbidity and behavioral dysfunction in service personnel. We have developed a short-term survival model of BI-TBI in swine that we believe will help us understand BI-TBI better and help develop new therapies to treat survivors.
This study was funded by the DOD, and carried out with IACUC and DoD ACURO approval. Two PVC piping airguns with compression chambers were constructed. The guns were calibrated using a digital ballistics chronometer over a series of firings with controlled pressures in the air chambers. Four female Yorkshire swine were used in the first set of experiments; two were injured with the smaller gun at 100 PSI, one with the larger gun at 80 PSI and one sham . In the second set of six animals, two pigs were injured with the larger gun at 80 PSI, three with the larger gun at 100 PSI, and one sham.
Two of the three animals that faced the 100 PSI blast overpressure developed ischemic changes in their EKG-Lead II within 5 minutes of the blast. The ST elevation remained sustained even at 7 days post-BI-TBI, suggesting cardiac dysfunction following neurological insult. Other physiological parameters remained unremarkable, including their neuro-cognitive scores, as well as their performance in the neuro-cognitive tests.
To our knowledge this is the first report of ischemic changes in a BI-TBI swine model. Its occurrence in further studies needs to be observed for long-term sequelae in the neuro-cardiovascular system. Our focal BI-TBI swine model targeting the skull and the brain is a robust platform for directing therapeutic and rehabilitative efforts towards our military personnel.
DOD Funding
Blast TBI, Swine, Ischemia
LONG TERM EFFICACY OF HUMAN BONE MARROW MESENCHYMAL STEM CELLS IN TRAUMATIZED MICE BRAIN IS NOT AFFECTED BY IMMUNOSUPPRESSIVE TREATMENT
Elisa R Zanier, MD, Mario Negri Institute for Pharmacological Research
Giovanna D'amico, PhD, M. Tettamanti Research Center
Federica Marchesi, Mario Negri Institute for Pharmacological Research
Andrea Biondi, MD, M. Tettamanti Research Center
Ettore Biagi, MD, M. Tettamanti Research Center
Giuseppe Citerio, MD, Neuroanesthesia and Neurointensive Care Unit, San Gerardo Hospital, Monza
Maria-Grazia De Simoni, PhD, Mario Negri Institute for Pharmacological Research
Most of cell types transplanted after traumatic brain injury(TBI) show poor survival. Immunosuppressants are used to enhance graft permanence. Studies addressing the dependence of mesenchymal stem cell (MSC) efficacy on immunosuppression in the injured brain are lacking. We explore the need of immunosuppression after MSC transplantation in TBI mice.
C57Bl/6 mice were subjected to severe TBI (by controlled cortical impact, 1 mm depth) or sham surgery. At 24h they received an intracerebroventricular infusion of PBS (control) or human bone marrow MSC (150,000/5μl). Mice receiving MSC were subjected to immunosuppressive treatment with cyclosporin A (CsA, 10 mg/kg ip, immunosuppressed: IS) or no treatment (immunocompetent: IC). Cortical mRNA expression (real time RT-PCR) was assessed at 72h to analyze immunosuppression (INFγ) and MSC rejection (MHCII, CD86). Mice were evaluated for sensorimotor (neuroscore) and cognitive (Morris water maze) dysfunctions weekly and at 4 weeks after surgery respectively. Mice were sacrificed at 5 weeks and immunohistochemistry was performed to analyze: cell distribution, pericontusional astroglial activation (glial fibrillary acidic protein, GFAP), pericontusional vessel density (CD31) and subventricular zone (svz) neurogenesis (doublecortin, DCX).
Immunosuppression was confirmed by a significant reduction of INFγ mRNA expression in IS compared to IC mice (-73%) and no signs of rejection were highlighted by MHCII or CD86 expression (MHCII: control=7.4, IC=7.7, IS=7.6; CD86: control=2.3, IC=2.5, IS=2.4 sham relative mRNA expression) 72h post-injury in pericontusional cortex. MSC improved sensorimotor (neuroscore (mean) at 4 weeks: control=4.6; IC=6.9; IS=6.8) and cognitive (mean latency to platform: control=46.7, IC=38.0, IS=39.3 sec) functions compared to controls. Post mortem analysis at 5 weeks revealed that MSC were present in the injured brain in IS and IC mice in a comparable amount. At this time point, MSC reduced GFAP positivity in the scar region (IC:-34%, IS: −23%), increased vessel density (CD31, IC: +19%, IS: +20%) and induced neurogenesis in the svz (DCX, IC: +79%, IS: +82%) compared to controls.
Long term MSC survival and efficacy in the injured brain are not dependent on immunosuppressive treatment. MSC induce comparable sensorimotor and cognitive improvements in IC and IS mice and produce structural protection and repair characterized by increased vessel density in the pericontusional tissue, reduction of the gliotic scar and increase of neurogenesis into the svz. These results have important clinical implications since immunosuppressive agents have toxic side effects for TBI patients, therefore a primary goal for translational research should be to avoid unnecessary suppression of a patient's immune system.
We thank the association “Esserci con Cate per i Bimbi” who supported this work.
“mesenchymal stem cells,” transplantation, neuroprotection, neurorepair, “stem cell transplantation”
DIFFUSION TENSOR IMAGING OF THE BRAIN IN RATS EXPOSED TO PRIMARY BLAST
Alok Shah, MS, Medical College of Wisconsin
Frank Pintar, PhD, Medical College of Wisconsin
Brian Stemper, PhD, Medical College of Wisconsin
Traumatic brain injury (TBI) due to blast exposure is a consequence among many military personnel serving in modern combat. A rat model of primary blast injury was employed to identify the role that diffusion tensor imaging (DTI) may have in detecting brain injury following exposure to blast.
A helium-driven shock tube with a 3.04 m driven length and 0.03 m driver length was used to generate shockwaves. Rats were anesthetized, placed in a holder to protect the body, and exposed to a blast shockwave with a peak overpressure and duration of 100 kPa and 370 μs(n=6), 450 kPa and 210 μs(n=7), or sham procedures(n=5). Animals were euthanized two days after injury, and the fixed brains were imaged on a 9.4T MRI. Twelve diffusion-weighted images(b=1200 s/m2) and two non-diffusion weighted images(b=0 s/m2) were acquired at a spatial resolution of 0.2x0.2x0.5 mm3 using a spin-echo sequence(TR=2000 ms, TE=21 ms). DTI maps of fractional anisotropy (FA) and mean diffusivity (MD) were computed and registered to a common atlas. Voxelwise and region of interest (ROI) analysis were performed using an unpaired Student's t-test to assess group differences and a paired t-test for within-group comparisons.
DTI revealed lateralized abnormalities related to the blast severity. Specifically, a blast wave directed to the left side of the head resulted in diffuse injury predominantly in the left hemisphere. Using a region of interest analysis, the Blast100 group was not significantly different than control animals. However, the Blast450 group had significantly decreased FA in the left cortex (p=0.008), left hippocampus (p=0.007) and portions of the left white matter (p=0.03) compared to the control group, whereas the right cortex, hippocampus, and white matter were unaffected. Although MD was not significantly different between control animals and those exposed to blast, MD had a significant effect of laterality in the Blast450 group, with the left hippocampus (p=0.05) and white matter (p=0.04) being significantly different than the right. Lateral effects were not present in the Blast100 or control animals.
The primary observations of the study are that 1) DTI is sensitive to injury in the brains of rats exposed to primary blast and 2) the effects are related to either the blast severity (i.e. peak overpressure) or the duration of shockwave. The DTI-detected injury occurs in the absence of gross imaging or pathological features such as edema or hemorrhage. The decrease in FA is likely associated with the pathological features of axonal damage and gliosis that have been implicated in other central nervous system injuries. The detection of injury related to primary blast is important as several aspects of blast mTBI remain unclear, including the mechanism of injury, presence of symptoms, and influence of shockwave characteristics on outcomes. Overall, the results demonstrate the utility of DTI as a marker of injury and support its application in preclinical studies of blast injury.
Funded through the Advancing a Healthier Wisconsin endowment(M.D.B.), the Wisconsin Injury Research Center Seed Project(B.D.S), and the Veterans Health Administration Rehabilitation Research and Development Service(F.A.P.).
diffusion tensor imaging, blast, magnetic resonance imaging
HISTOLOGICAL ASSESSMENT OF EFFECT OF PERFLUOROCARBON (PFC) ON NEURODEGENERATION IN RAT MODEL OF PENETRATING BALLISTIC BRAIN INJURY (PBBI)
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Ross Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Markus Spurlock, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Daniel Diaz, B.S., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Natalie Aguirre, High School, Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Clayton Jackson, B.S, University of Miami
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Posttraumatic hypoxia contributes to poor outcomes. Recently, hypoxia has been reported in a rat model of PBBI. PFCs have been shown to alleviate hypoxia in experimental brain injury models. However, the PFC mediated mitigation of neurodegeneration is poorly characterized. Here, we present histological assessment of neurodegeneration mitigated by PFC.
PBBI was induced in male Sprague-Dawley rats (280–350g) that were randomly allocated to following groups: (i) Group1 (no injury), (ii) Group2 (Probe+no inflation), (iii) Group3 (Probe+inflation to 10% brain volume+treatment with Vehicle), (iv) Group4 (same as group3 but treatment with PFC). Animals were sacrificed at 24h or 72h (n=10 /group) and perfusion-fixed for quantitative histopathological evaluation. Flurojade B (FJB) staining was used to estimate neuronal degeneration. Using the physical fractionator method (MicroBrightField Stereo Investigator software), FJB positive neurons in five sections spaced 400 μm apart within the region spanning lesion (∼10 mm) were estimated.
Consistent with previous publications FJB positive neurons were present in large numbers ipsilateral to the lesion, radiating throughout caudate putamen up ∼4mm from lesion center. Fewer FJB cells were seen in the regions remote from the lesion site. FJB positive cell distribution was similar in both vehicle treated and PFC treated animals.
Similar to previous work with other pharmacological agents the neurodegeneration within the core lesion is not mitigated by the presence of PFC. The estimation of FJB cells in the penumbra is ongoing and will be presented.
These studies were made possible by support from CDMRP award PTO74521 (W81XWH-08-1-0419).
PBBI, FJB, PFC, neuroprotection, neurodegeneration
EXPLORING SPATIAL VARIATIONS IN RESTING STATE DEFAULT-MODE NETWORK IN MILD TRAUMATIC BRAIN INJURY USING INDEPENDENT COMPONENT ANALYSIS
Ping-Hong Yeh, Ph.D., Traumatic Brain Injury Image Analysis Lab, USUHS, Bethesda, MD; Henry Jackson Foundation, Rockville, MD; National Capital NeuroImaging Consortium, Bethesda, MD
Rachel D. Wolfowitz, BS, Department of NeuroImaging, National Intrepid Center of Excellence, WRNMMC, Bethesda, MD; Henry M. Jackson Foundation, Bethesda, MD
Jamie Harper, MPH, Department of NeuroImaging, National Intrepid Center of Excellence, WRNMMC, Bethesda, MD
Wei Liu, Ph.D., National Intrepid Center of Excellence, WRNMMC, Bethesda, MD; Traumatic Brain Injury Image Analysis Lab, USUHS, Bethesda, MD; Henry Jackson Foundation, Bethesda, MD
John L. Graner, Ph.D., National Intrepid Center of Excellence, WRNMMC, Bethesda, MD; National Capital NeuroImaging Consortium, Bethesda, MD
Hai Pan, MS, Traumatic Brain Injury Image Analysis Lab, USUHS, Bethesda, MD; Henry Jackson Foundation, Bethesda, MD
Binquan Wang, Ph.D., Traumatic Brain Injury Image Analysis Lab, USUHS, Bethesda, MD; Henry Jackson Foundation, Rockville, MD; National Capital NeuroImaging Consortium, Bethesda, MD
Terrence R. Oakes, Ph.D., National Intrepid Center of Excellence, WRNMMC, Bethesda, MD; Traumatic Brain Injury Image Analysis Lab, USUHS, Bethesda, MD; National Capital NeuroImaging Consortium, Bethesda, MD
Gerard Riedy, M.D., Ph.D., Department of NeuroImaging, National Intrepid Center of Excellence, WRNMMC, Bethesda, MD; National Capital Neuroimaging Consortium, Bethesda, MD
Connectivity analysis is frequently used to understand resting state networks in the brain. We use a data driven approach to investigate resting state networks in functional MRI (fMRI) data relating to mTBI. Regions of activation detected via this approach were subsequently found to be correlated with several neuropsychiatric measures.
Functional MRI (fMRI) data were acquired on 12 healthy and 15 mTBI subjects. The fMRI data were transformed to the MNI template. Groups were combined and independent component analysis (ICA) was performed to decompose the data into 25 independent components. Further analysis was performed using dual regression ICA and voxel-wise t-tests. A resting state Default Mode Network (DMN) map was derived from the healthy group and subsequently used as a template. Individual subject DMN maps were threshold using a Z-statistic of 3.498 (p corrected<0.005). Spatial information from each mTBI and normal subject was extracted using the DMN template. Regions of interest (ROIs) were identified using the Harvard and probabilistic cerebellar atlases. Spatial information of subjects was analyzed using ANOVA, and independent samples t-test to identify group and region differences after correction for multiple comparisons. Pearson's correlation was calculated between mTBI subject ROI spatial information with neuropsychological assessments to identify potential neural correlates.
Analysis of DMN spatial information: ANOVA results suggest that significant group differences (p<0.05) were present in right cortical and subcortical regions. Analysis of individual ROI spatial information comparing mTBI and healthy subjects revealed significant differences (p<0.05) in regions of the left parietal lobule, left posterior cingulate gyrus, right precentral gyrus, right inferior temporal gyrus, right cuneal cortex, left cerebellar crus II, left hippocampus, left amgydala, and right thalamus. The findings from our study correlate with results from current literature specific to mTBI. However, the detection of abnormal activations in regions such as the cerebellum, precentral cortex, supplementary motor area and sub cortical regions are unique to our analysis method of using the ICA data driven approach, and previously not reported for mTBI.
Analysis of neuropsychological assessment correlations: Significant correlations (R2>0.6, p<0.05) were observed between neuropsychological measures and the DMN spatial information of mTBI subjects in the left amygdala, left posterior cingulate gyrus, left anterior supramarginal gyrus, right precentral cortex, right supplementary motor cortex and inferior temporal gyrus. The correlations were centered on neupsychiatric assessment scores of working memory, emotion, mood and behavior. These findings are consistent with the literature in which abnormal DMN activations have been associated with neurological pathologies or injury.
Functional connectivity analysis in mTBI frequently reports on group analyses using seed based correlations. Our results indicate that the data driven approach of ICA is able to identify DMN activation within specific brain regions and furthermore, this analysis method is sensitive in distinguishing variances of activation patterns seen in mTBI subjects. We also observe specific variances of activation patterns in DMN regions associated with cognitive and executive control networks, previously not reported in the literature. Furthermore, the correlations between mTBI spatial information with neuropsychological assessments provide potential as a biomarker for assessing the effects of mTBI on an individual subject level. Additional analysis comparing information from modalities such as diffusion tensor imaging (DTI) could compliment this study by providing insight into the structural morphology of the DMN regions. Our next step would be to replicate this work on a larger sample size. DISCLAIMER: The views expressed in this presentation are those of the author and do not reflect the official policy of the Department of Defence, or U.S. Government.
We thank Dr. Cara Olson, and Jacob Hershorin. Research supported by Center for Neuroregenerative Medicine (CNRM), HJF Award Number: 300606-07.01-60855 and CDMRP Award Number: W81XWH-08-2-0165.
mTBI, DMN, imaging, ICA, neural-correlates
A NOVEL SHOCK TUBE TO INDUCE A BLAST RELATED TRAUMATIC BRAIN INJURY IN RODENTS
Vijayalakshmi Santhakumar, Ph.D., University of Medicine and Dentistry of New Jersey
Bryan J. Pfister, Ph.D., New Jersey Institute of Technology
We have developed a novel pneumatic air pressure-driven shock tube to generate a range of mild-to-severe shock waves that can be employed to induce a blast-related TBI (bTBI) in small rodents. This compact and controllable device will safely facilitate bTBI research in the general lab setting.
The shock tube system is driven by compressed air that is quickly released creating a shock wave that propagates down an expansion tube. The shock driver section is comprised of two chambers: a Compressed Air Driver Chamber and a Shock wave Trigger Chamber. The Driver Chamber is filled with compressed air causing an internal piston to seal the chamber. The Trigger Chamber is pressurized to disengage the piston seal from the Driver Chamber and release the compressed air into the Expansion Tube. The rapid expansion of air creates a shock wave that propagates down the long tube. A piezoelectric pressure transducer was placed at 4 and 6 locations along the 12 and 24 inch Expansion Tube, respectively, to directly measure the blast pressure and calculate the shock loading characteristics including overpressure peak, rise time, rate of rise, positive phase duration, decay time constant and impulse.
The shock tube generated a range of pressure waveforms with peak pressures from 30 kPa to those greater than 100 kPa. Use of the 24 inch Expansion Tube resulted in the formation of the underpressure not seen in the 12 inch Expansion Tube. The pressure waveforms in the 12 inch Expansion Tube were more defined and the overpressure peak recordings were more consistent than in the 24 inch Expansion Tube. Two positions were then chosen inside the 12 inch Expansion Tube for further study. Pressure measurements were made when the pressure transducer was located 6 and 9 inches away from the explosive source inside the Expansion Tube. Our data indicate that shock loading characteristics such as the overpressure peak can be modified by manipulating the user defined compressed air pumped into the Compressed Air Driver Chamber.
The propagation of the shock wave in the 24 inch Expansion Tube resulted in the formation of the typical Friedlander curve comprised of the positive and negative phase. This implied that shock waves generated by this unique shock tube mimic pressure waveforms found in literature. Manipulation of the user defined compressed air and position inside the Expansion Tube resulted in different shock loading characteristics. Animal studies will be conducted to investigate the regional cell loss of the hippocampus as well as physiological and behavioral consequences of blast exposure. Replicating blast injury using this safe device eliminates the need for a specialized lab setting for bTBI research and will accelerate the use of animal models to study the unique pathology associated with blast injury to the brain and other organ systems.
The NJ Commission on Brain Injury Research Grant BIR11PJT003, Professor Robert Stoll & Dr. William R. Spillers
traumatic brain injury, shock tube
INFLAMMATORY PROFILE IN A MOUSE MODEL OF TBI AND EFFECTS OF POTENTIAL THERAPEUTICS
Benoit Mouzon, M.S, Roskamp Institute
Jon Reed, MS, Roskamp Institute
Myles Mullan, Roskamp Institute
Alex Bishop, BS, Roskamp Institute
Gogce Crynen, PhD, Roskamp Institute
Venkatarajan Mathura, PhD, Roskamp Institute
Michael Mullan, MD, PhD, Roskamp Institute
Fiona Crawford, PhD, Roskamp Institute
We have used proteomic technology to analyze the brains response to TBI in a mouse model and have identified target pathways for therapeutic intervention. Using drugs that modulate those pathways, including ARC031 and RCP006, we have explored the acute and chronic effects on outcome after TBI.
Given that the APOE4 allele predisposes an individual to a poorer outcome from TBI, we used APOE3 and APOE4 transgenic mice to elucidate potential targets for therapeutic intervention. Proteomic profiling of the reponse of human APOE3 and APOE4 transgenic mice revealed differences in APP and NF-kB related molecules up to 3 months after TBI. To evaluate the potential therapeutic benefit of targeting these pathways, C57BL/6J mice received either a CCI or sham surgery (craniectomy only) and then received either placebo treatment, treatment with ARC031 (known to inhibit inflammatory responses and b-amyloid production) or RCP006 (an NFkB inhibitor). Neurobehavioral testing consisted of Rotatarod on days 1, 3, 5, and 7 after surgery followed by Barnes maze acquisition testing on days 8–13 with a probe trial on day 14. IL-6 and MCP-1 were also examined by ELISA in a separate cohort to study the acute inflammatory response.
Treatment with ARC031 improved the motor and memory outcomes; ARC031 significantly improved fall latency and ARC031 treated mice showed no statistically significant differences from sham animals in travel time to the target hole during the probe trial of the Barnes maze. Placebo (DMSO) treated mice showed a significantly longer travel time and had a higher primary error during the probe trial. Both the ARC031 and DMSO treated TBI mice showed a significantly greater cumulative distance to the target hole during the acquisition trials, and this was not an effect of velocity or motor impairment. ELISA results showed a significant increase of IL-6 and MCP-1 after TBI, but this was not significantly decreased by the administration of ARC031. Thus ARC031's effects on outcome after TBI may be mediated through other mechanisms such as effects on APP processing and b-amyloid production. Generation of data from treatment with RCP006 and other potential therapeutics is ongoing at the time of submission.
Treatment with ARC031 significantly improves neurobehavioral outcome in a mouse model of TBI as measured by Rotarod and Barnes maze. IL-6 and MCP-1 increase in response to injury at acute timepoints, but are not extinguished by the administration of ARC031. Further exploration into the mechanism of action for ARC031's effects after TBI is suggested, and additional compounds targeting inflammation may prove useful individually or in combination with ARC031.
This research was funded by a Department of Defense award (W81XWH-07-1-0700) to Dr. Fiona Crawford and by the Roskamp Foundation.
Inflammation therapeutics TBI CCI CHI
POLYNITROXYLATED PEGYLATED HEMOGLOBIN ATTENUATES FLUID REQUIREMENTS AND BRAIN EDEMA IN COMBINED TRAUMATIC BRAIN INJURY PLUS HEMORRHAGIC SHOCK IN MICE
C. Edward Dixon, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Hülya Bayır, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Brian Blasiole, MD, PhD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Robert S.B. Clark, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Vincent A. Vagni, BA, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Professor Li Ma, PhD, Georgia Southern University
Carleton J.C. Hsia, PhD, SynZyme Technologies, LLC
Patrick M. Kochanek, MD, Safar Center for Resuscitation Research, University of Pittsburgh School of Medicine
Polynitroxylated pegylated hemoglobin (PNPH), a bovine hemoglobin decorated with niroxide and polyethylene glycol moieties, demonstrated neuroprotection in vitro and in vivo vs. lactated Ringer's (LR) in experimental TBI plus hemorrhagic shock (HS). Hypothesis: Resuscitation with PNPH will reduce intracranial pressure (ICP) and brain edema vs. LR in experimental TBI+HS.
C57/BL6 male mice (n=20) were anesthetized with isoflurane and underwent controlled cortical impact (CCI) followed by severe HS to mean arterial pressure (MAP) of 25–27mm Hg for 35 min. Mice (n=10/group) were then resuscitated with an initial bolus of 20ml/kg of either 4% PNPH or LR followed by 10ml/kg boluses every 5 min for MAP<70 mm Hg for 90min, mimicking pre-hospital care. The shed blood was then re-infused as part of the hospital phase, and at the completion of the model the mice were sacrificed. ICP was monitored for the duration of the study. The mice were sacrificed at the end of the study and percent brain water was measured by the wet/dry weight method.
Mice resuscitated with PNPH vs. LR required less fluid for resuscitation (26±0 vs. 167±10.7 ml, P<0.001) and had a higher MAP at the end of the pre-hospital phase (78±1.5 vs. 60.1±2.8 mm Hg, P<0.001). The PNPH mice required only the initial resuscitation bolus of 20ml/kg, while the LR mice required multiple boluses during resuscitation for MAP<70mmHg. PNPH-treated mice also had a lower peak ICP (14.5±0.97 vs. 19.6±1.1 mm Hg, P=0.003) and higher cerebral perfusion pressure (CPP) at the end of the pre-hospital phase (67.1 vs. 45mmHg, P<0.001). The PNPH mice also had a reduced % brain water in the hemisphere ipsilateral to injury vs. LR (80.3±0.12 vs. 80.9±0.12%, P=0.02). Hemoglobin and lactate levels were not different between the groups at any time point. Methemoglobin levels were mildly elevated in the PNPH group in the pre-hospital and hospital phase (P<0.001).
We conclude that resuscitation with PNPH dramatically lowers fluid requirements, improves ICP and CPP and reduces brain edema vs. LR resuscitation. Despite severe HS, PNPH-treated mice required only a single bolus of fluid and were adequately resuscitated while the LR mice were refractory to resuscitation. Our data support ongoing pre-clinical development of this novel resuscitation fluid for TBI resuscitation.
Supported by U44 NS070324
resuscitation, brain edema, blood substitute
EXPRESSION OF VOLTAGE-GATED SODIUM CHANNEL NAV1.3 IS ASSOCIATED WITH SEVERITY OF TRAUMATIC BRAIN INJURY IN ADULT RATS
Qin Mao, MD, PhD, Department of Neurosurgery, Shanghai Renji Hospital, Shanghai Jiaotong University School of Medicine
Yong Lin, MD, PhD, Department of Neurosurgery, Shanghai Renji Hospital, Shanghai Jiaotong University School of Medicine
Jun-feng Feng, MD, PhD, Department of Neurosurgery, Shanghai Renji Hospital, Shanghai Jiaotong University School of Medicine
Ji-yao Jiang, MD, PhD, Department of Neurosurgery, Shanghai Renji Hospital, Shanghai Jiaotong University School of Medicine
Secondary injury induced by traumatic brain injury (TBI), includes influx of Na+ and Ca2+ ions. Depolarization of neurons mediated by voltage-gated sodium channels (VGSCS) leads to cellular abnormalities and neurological dysfunction. Expressions of different α-subunits of VGSCs vary during the early stage following TBI.
Adult male Sprague-Dawley rats were randomly assigned to sham-TBI, mild-TBI (mTBI), or severe-TBI group (sTBI). TBI was induced using a lateral fluid percussion device at magnitudes of 1.5–1.6 ATM (mTBI) and 2.9–3.0 ATM (sTBI). The sham animals undergo the exact surgical preparations but with no injury. In study 1, expressions of mRNA and protein for Nav1.3 in the ipsilateral-injured cortex were examined at different timepoints following injury (2h, 12h, 24h, and 72h) by real-time quantitative polymerase chain reaction and Western blot. In study 2, brains from all groups were collected at different timepoints (24h, 72h, and 7d) for TUNEL staining and cell count analysis. In study 3, immunofluorescence was performed to localize the expression of Nav1.3 protein in the ipsilateral-injured cortex.
Expression of Nav1.3 mRNA and protein were significantly upregulated in the ipsilateral-injured cortex in mTBI and sTBI groups when compared with the sham-TBI group at both 2h and 12h post-TBI. However, when comparing at the 12h timepoint, Nav1.3 mRNA and protein level in the sTBI group were much higher than that in the mTBI group. There was no significant difference between groups at 24h and 72h post-TBI. The number of the TUNEL-positive cells was significantly higher in the sTBI group than that of the mTBI group at 24h, 72h, and 7d post-TBI. However, the numbers of apoptotic cells in the mTBI and sTBI group decreased significantly from 24 hrs to 72 hrs post-TBI. Rare TUNEL-positive cells were detected in the sham-TBI group. The expression of Nav1.3 occurred predominantly in neurons of the cortex.
Expression of Nav1.3 mRNA and protein were significantly upregulated in the rat cortex at the very early stage post-TBI and correlated to TBI severity.
We thank Dr. Bruce Lyeth and Dr. Ken Van, Center of Neurosurgical science, University of California, Davis, USA, for reviewing this paper.
Nav1.3, Sodium Channel, Traumatic brain injury, rat
A COMPARISON OF TWO ANTI-INFLAMMATORY DRUGS, EPOETIN ALPHA AND ANAKINRA FOR TREATMENT AFTER TRAUMATIC BRAIN INJURY
Todd C. Peterson, M.S., Southern Illinois University-Carbondale
Dylan Greeney, Southern Illinois University-Carbondale
Aaron Jannings, Southern Illinois University-Carbondale
Gail D. Anderson, Ph.D, University of Washington
Michael R. Hoane, Ph.D, Southern Illinois University-Carbondale
Recent studies have shown that anti-inflammatory drugs have strong potential for treatment of TBI. Out of these studies, two likely candidates have emerged: epoetin alpha (Procrit) and anakinra (Kineret). Our laboratory performed a direct comparison between the two to assess which drug leads to better recovery of function following TBI.
The dosing for the drugs was determined from a study that examined the pharmacokinetics of each. A unilateral CCI injury over the posterior sensorimotor cortex was induced. The drugs were then administered twice daily for three days with an i.p loading dose at 2 hours post-CCI and then s.c. doses at 12, 24, 36, 48, 60 and 72 hours post-injury. One group of injured rats was given epoetin (2500 units/kg). Another injured group received anakinra (100 mg/kg). The other injured group and sham group received vehicle (1.25 mL/kg). Groups of rats were euthanized at 24 hours, 72 hours and 7 days to examine gene expression via a gene-chip assay. A separate group of rats were tested on sensorimotor and cognitive behavioral performance. At day 25 post-injury, rats were euthanized and brains were sliced to assess lesion size.
No significant improvement from either drug was found on sensorimotor or cognitive behaviors. There was no reduction in lesion size in drug-treated rats. However both drugs were able to differentially change genetic expression following injury. This leads to several possible conclusions regarding these drugs.
One conclusion is that there may be additional factors that are of the utmost importance in evaluating these specific drugs (such as injury location, time of administration, dosing). Another potential conclusion may be that these drugs are not as effective as previously reported. Regardless, additional investigations need to be performed to identify the factors that influence positive versus negative outcome following administration of these two drugs as well as reconcile the differences in genetic expression changes versus actual behavioral function.
Funding provided by NIH grant #HD061944
Behavior, Treatment, Neuroprotection, Drug
THE EFFECT OF THERAPEUTIC HYPOTHERMIA ON THE CCL-2 AFTER TRAUMATIC BRAIN INJURY
Li Bie, M.D. Ph.D., The First Hospital of Jilin University
Yang Ju, M.D. Ph.D., The First Hospital of Jilin University
Jingduo Zhou, M.D., The First Hospital of Jilin University
Traumatic brain injury (TBI) elicits acute inflammation that in turn exacerbates primary brain damage. In previous reports, the chemokine CC ligand-2 (CCL-2) is implicated in macrophage recruitment into damaged parenchyma after TBI. In this study, we investigated the influence of therapeutic hypothermia on CCL-2 inflammasome signaling after TBI.
Adult male Sprague-Dawley rats (n=47, 235 - 385 g) received a moderate traumatic brain injury by lateral fluid percussion (fluid percussion pressure, 2.3 - 2.8 atm) were divided into two groups. Temperature manipulation (33.0°C) was initiated 30 min after TBI and maintained using cooled air and heating lamps for a period of 4h (n=24). The normothermia groups (n=23) were maintained at 37.0°C throughout the procedure. At the end of the temperature manipulation or at 24 h after TBI, rats were sacrificed and the samples of brain tissue were immediately excised from the traumatized cortex and prepared for Western blot analysis.
In the normothermic groups, CCL-2, macrophage inflammatory protein-2 (MIP-2), caspase-7 and caspase-11 tended to increase overtime. In contrast, the hypothermic groups demonstrated an increase at 4h after TBI, and then decreased at 24 h after trauma.
Therapeutic hypothermia has been reported to improve outcomes in several animal models of brain injury and has been successfully translated to specific patient populations. Taken together, the above results show that CCL-2 is functionally important in the modulation of injury and inflammation in TBI.
The authors declare that they have no competing interests. The authors also declare there was no external funding for this study.
TBI, CCL-2, hypothermia
ALTERED NOCICEPTION INDUCED BY CORTICAL TRAUMA IN MICE IS PARTIALLY INHIBITED VIA CALCITONIN GENE-RELATED PEPTIDE BLOCKADE
Christine Macolino, BS, Thomas Jefferson University
Mason Love, Thomas Jefferson University
Jack I. Jallo, MD, PhD, Thomas Jefferson University
Melanie B. Elliott, PhD, Thomas Jefferson University
Post-traumatic headache (PTH) can be a new headache resulting from head trauma or worsening of a pre-existing headache disorder. In many patients, it resolves in three months; in others, it persists for much longer. Previously, cortical injury in mice resulted in altered nociception that outlasts cortical glial responses.
In this study, male C57BL/6 mice received either controlled cortical injury (CCI) or craniotomy, and were administered repeated treatments with a CGRP antagonist, Sumatriptan, or saline at either 1 or 4 weeks post-injury. Incision only mice were included as controls because craniotomy induces transient neuroinflammatory responses and changes in sensory behavior. Baseline and weekly periorbital and paw von Frey (mechanical) sensory testing for the presence of allodynia was performed for up to 4 weeks post-injury. Allodynia is an abnormal response to a mechanical stimulus that is normally innocuous. Changes in Fos labeled neurons and calcitonin gene-related peptide (CGRP) within the trigeminal ganglia, brainstem, or somatosensory cortex were determined using ELISA or immunohistochemistry.
Results show CCI causes a significant reduction in periorbital and forepaw von Frey mechanical thresholds (allodynia) but not hindpaw thresholds, p<0.001. CCI mice showed a significant increase in CGRP levels in the caudal medullary trigeminal nucleus and the number of CGRP positive cells in the trigeminal ganglia compared to controls, p<0.001. Neuronal activation was increased in the somatosensory cortex and caudal trigeminal nucleus after injury compared to controls, p<0.001. Sumatriptan increased mechanical thresholds and reduced CGRP levels after injury compared to saline, p<0.05. Treatment with the CGRP antagonist increased mechanical thresholds in a dose-dependent manner and reduced neuronal activation in the trigeminal nucleus compared to saline, p<0.05. There was no difference in the number of Fos-labeled neurons in the somatosensory cortex between Sumatriptan, CGRP antagonist, or saline treated mice.
In conclusion, mechanical injury to the somatosensory cortex results in altered nociception as evidenced by reductions in periorbital and forepaw von Frey thresholds (allodynia), and increases in CGRP and neuronal activation. Treatment with Sumatriptan and a CGRP antagonist partially diminish mechanical allodynia and abnormal activation of trigeminovascular neurons. Findings indicate the involvement of additional mechanisms in mediating allodynia that may be amenable to treatment. Studies to test the optimal treatment strategy and to compare additional pharmacological agents are warranted.
This work was supported, in part, by Investigator Initiated Studies Program, Merck Sharp & Dohme Corp.
Post-traumatic, headache, allodynia, post-concussive, CGRP
REPEAT MILD TRAUMATIC BRAIN INJURY IS ASSOCIATED WITH ACUTE WHITE MATTER ABNORMALITIES IN THE JUVENILE RAT
Kristi Clark, Ph.D., University of California
David A. Hovda, Ph.D., University of California
Mayumi Prins, Ph.D., University of California
Diffusion tensor imaging (DTI) has been used to characterize white matter abnormalities following TBI, but acute and chronic effects of repeat mild TBI (mTBI) have yet to be described. We hypothesized that TBI-induced axonal injury would acutely decrease anisotropy in a dose-dependent manner, with recovery to sham levels by 6 months post injury.
Postnatal day 35 (PND35) rats received sham, single, double (RTBI2), or quadruple (RTBI4) TBI using a closed-head injury model, with repeat injuries occurring at 24 hour intervals. One week and 6 months after the last injury, animals were scanned in vivo in a 7.0 Tesla MRI scanner.
Rare and spin-echo DTI sequences were used to acquire anatomical MRI and diffusion tensor images, respectively. The following metrics were computed from the DTI images using tractography-defined regions of interest (ROIs): fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD), and mean diffusivity (MD). FA is the degree of directionality per voxel and ranges from 0 to 1, with 0 being completely isotropic (moving equally in all directions) and 1 being completely anisotropic. AD measures diffusion parallel to the axon, with decreased AD corresponding with axonal damage. RD averages diffusion along the two minor perpendicular axes and MD computes a global value based on the average diffusivity in all 3 directions.
In the corpus callosum, FA significantly decreased by 3% in the RTBI4 group relative to sham at the acute time point. It is important to note that the entire corpus callosum fiber bundle was traced; future analyses dividing the corpus callosum into smaller regions (genu, body, and splenium) will likely produce a more robust anisotropy decrease, particularly in the region nearest the impact site. There was a trend towards increased radial diffusivity in the RTBI4 group relative to sham, suggesting the occurrence of demyelination after multiple injuries. Consistent with our hypothesis, there were no significant group differences 6 months post injury for any of the outcome measures.
Repeat mTBI has been correlated with impaired cognition, decreased cerebral metabolic rate of glucose, and increased axonal injury in the juvenile rat. In this study, we used DTI to examine the cumulative acute and chronic effects of mTBI on white matter structures in vivo. DTI has been proposed as a potential biomarker for TBI based on its ability to non-invasively infer the microstructure of white matter tracts. These preliminary results indicate that although a single injury might not affect white matter anatomy, repeat mTBI within a short period of time decreases anisotropy and may perpetuate demyelination.
National Football League (NFL) Charities, UCLA Brain Injury Research Center
DTI, Repeat TBI, mild TBI
EFFECTS OF GLUCOSE ON BEHAVIORAL RECOVERY AND CONTUSION VOLUME IN RATS
Sima Ghavim, BA, University of California
Katsunori Shijo, MD, PhD, University of California
The impact of hyperglycemia (HG) after traumatic brain injury (TBI), and even the administration of glucose (Glc) to head injured patients, remains controversial. This study determined effects of Glc treatments on behavioral and histological outcomes in adult male Sprague-Dawley rats with sham injury or left cortical contusion injury (CCI).
Rats (n=12/group) were subjected to Sham injury or CCI and injection (i.p.) of saline (Sal; 8% or 0.9%) or 50% Glc (2 g/kg) at 0, 1, 3 and 6 h post-surgery. Over 2 weeks rats were assessed for beam-walk (BW) ability (traversal time, rating scores), left forelimb tactile placing (TP), retraction of the right hindlimb after posterior-lateral extension (HE), forelimb contraflexion (CF) and for cognitive ability (%4/5 alternation scores in a plus maze task). After euthanasia at 15 days post-surgery, Nissl-stained brain sections were analyzed for cortical contusion volume (left volume as percent of right) and for loss of polymorphic neurons in the dentate hilus (left as percent of right cell counts). Data (Mean±SEM if reported below) were analyzed using ANOVA, with repeated measures when appropriate. Sub-groups of Sham and CCI rats with different Sal (8% vs. 0.9%) treatment were compared in separate analyses.
Type of Sal treatment did not significantly affect any outcomes in Sham or CCI groups, so 8% and 0.9% sub-groups were collapsed for further analyses. Significant deficits in BW traversal time and rating scores occurred after CCI and recovered over test sessions from 1 to 13 days (Injury, p's<0.001, Days, p's<0.001, Day X Injury, p's<0.001), but were not affected by Glc or Sal treatment. Persistent post-CCI deficits in forelimb CF were not affected by Glc, and mild improvements in forelimb TP in the CCI-Glc group from 1 to 13 days were not significant compared to CCI-Sal. Deficits in right HE latency (Days, p<0.001) were significantly attenuated in the CCI-Glc group (p<0.05 versus CCI-Sal). Total arm entries during plus maze tests on days 4, 9 and 14 were not affected by Injury or Drug conditions. Although both CCI groups were significantly impaired on %4/5 alternation scores across testing days compared to sham injury controls (Injury, p<0.001), there was no effect of Glc versus Sal on these measures of working memory. The left/right cortical volumes (Sham-Sal=99.2±1.1%; Sham-Glc=98.1±1.3%) were significantly reduced by CCI (p<0.001), although the reduction was greater in the CCI-Sal group (64.3±1.9%) than in the CCI-Glc group (70.0±2.8%; p<0.05). Counts of hilar neurons are ongoing, but with n=7-8/group there is no apparent effect of Glc compared to Sal animals with CCI (∼40% neuronal loss ipsilateral to injury in both groups).
Administration of multiple Glc injections to induce HG acutely (0-6 h) after TBI in rats was found to significantly improve 1 of 3 neurological tests as well as significantly attenuate cortical injury volume. Results from use of hypertonic (8%) versus normal (0.9%) Sal indicate these benefits of Glc are not due to osmolarity effects of the 50% Glc treatments. As previously suggested for our results showing this Glc treatment regimen can significantly reduce dead/dying neurons in peri-contusional cortex and hippocampus at 24 h post-CCI, these benefits of Glc may be due to increasing available fuel supplies to satisfy increased energy demands acutely after TBI. Importantly, multiple injections of exogenous Glc did not adversely affect any sensorimotor, cognitive or histological outcomes studied in a 2 week period post-CCI. Further research is needed in this area given the recent trend towards tight glycemic control in TBI patients.
This work was supported by grant NINDS P01NS058489 and the UCLA Brain Injury Research Center.
Cognition; Contusion; Glucose; Neuroprotection; Sensorimotor
CHARACTERIZATION OF THE CONTROLLED CORTICAL IMPACT RAT TRAUMATIC BRAIN INJURY MODEL BASED ON POSITRON EMISSION TOMOGRAPHY IMAGING AND HISTOLOGY
Fiona Brabazon, Uniformed Services University
Sanjeev Mathur, MD, Uniformed Services University
Shalini Jaiswal, MS, Uniformed Services University
Haiying Tang, PhD, Uniformed Services University
Dr. Kimberly Byrnes, PhD, Uniformed Services University
Traumatic brain injury (TBI) diagnosis and treatment is a top priority for military medicine. The cerebral uptake of 18F-fluorodeoxyglucose (FDG), quantified using high-resolution positron emission tomography (PET), may provide a sensitive marker of temporal changes in metabolic activity after controlled cortical impact (CCI) TBI in rats.
Adult male Sprague Dawley rats were subjected to a craniotomy and CCI injury in the left motor cortex with a 5m/s impact speed, 150msec dwell time, and 2mm piston depth. All rats underwent a pre-injury FDG-PET scan (Siemens Inveon PET/CT) and a second PET scan at 3 hours, 1 day, or 7 days post-injury. The 30-minute PET scan was initiated following the injection of approximately 1.5mCi of FDG and a 45-minute uptake period. The uptake period and all imaging scans were conducted under isoflurane anesthesia with continuous heating and physiological monitoring. Images were analyzed using the Inveon Research Workplace 3.0 software and template regions of interest (ROIs) for the central brain region and cerebellum. Uptake quantities were normalized to the cerebellum. Twenty-four hours after scanning, brain tissue was processed for immunohistochemical labeling of markers for neurons, microglia and astrocytes to correlate with PET images.
The in vivo normalized cerebral glucose uptake (nCGU) showed a triphasic temporal response. A measurable increase in nCGU was observed at 3 hours post-injury (hyperacute phase) followed by a return to baseline at day 1 (acute) and a significant increase by day 7 (subacute). Histopathology showed a loss of tissue, myelin and axons but a subsequent increase in both microglia/macrophage and astrocytic activity in the perilesional region. This increase was demonstrated as not only an increase in microglia/macrophage and astrocyte number from baseline to 3 hrs to 7 days post-injury, but also as clear changes in the morphology of these cells indicative of increased activity.
Previous studies using the lateral fluid percussion TBI model in rats have shown a biphasic temporal response in glucose uptake, which were marked by a hyperacute increase followed by a significant depression from day 1 to day 7. However, this study showed a triphasic response without a depression in glucose uptake. The CCI injury model utilized in this study resulted in cell death, which should reduce glucose uptake, but also shows a time-dependent compensatory increase in glial activity, which should increase glucose uptake. Overall, it appears that cellular activation in response to focal brain injury outweighs the reduction in cellular activity due cell death or ischemia. Overall, this injury model and diagnosis technique may provide a sensitive platform for testing and evaluating TBI treatments.
This work was funded by the U.S. Department of Defense in the Center for Neuroscience and Regenerative Medicine.
FDG-PET, TBI, CCI, microglia
MILD TBI IN MICE WITH TRAUMATIC AXONAL INJURY IN THE ANTERIOR CORPUS CALLOSUM
Amanda Mierzwa, Ph.D., USUHS
Naruchorn Kijpaisalratana, Chulalongkorn University
Haiying Tang, Ph.D., Uniformed Services University of the Health Sciences
Reed Selwyn, Ph.D., DABR, Uniformed Services University of the Health Sciences
Regina C. Armstrong, Ph.D., Uniformed Services University of the Health Sciences
The majority of traumatic brain injuries (TBI) are mild with impact-acceleration forces resulting in traumatic axonal injury (TAI) to white matter tracts. A mouse model of mild TBI with reproducible TAI in the anterior corpus callosum (CC) facilitates analysis of neural stem cell regeneration from the subventricular zone (SVZ).
Closed head brain injury was produced in adult male C57BL/6J mice using an Impact OneTM Stereotaxic Impactor. The skull was impacted at the intersection of the sagittal suture and Bregma. Impact parameters tested multiple depths (1.0mm, 1.5mm, or 2.0mm) with 5m/s velocity and 100ms dwell time using a 3mm tip diameter. Mice were sacrificed post-injury on days 1, 3, 7, 14, or 42 for comparison with naïve or sham mice. Paraffin sections were stained with hematoxylin and eosin (H&E) or Luxol fast blue/periodic acid-Schiff reaction (LFB/PAS). Frozen sections were immunostained with antibodies against GFAP, CD11b, beta-APP, and Ki67. In vivo MRI with diffusion tensor imaging (DTI) was used for longitudinal analysis of the CC integrity at 1, 3, and 7 days post-injury. The CC region-of-interest was defined on the fractional anisotropy (FA) map in a 0.5cm coronal slice aligned using the midline crossing of the anterior commissure.
Histopathology showed an increase in injury severity with increased impact depth. The 1.0mm depth produced minimal pathology, whereas a 2.0mm depth produced complex pathology, including extensive cortical damage that confounded use as a model of mild TAI white matter damage. A 1.5mm impact depth was chosen for further characterization. Axon damage (beta-APP immunolabeling) within the CC was most evident during the first week post-injury. Axon damage was accompanied by astrogliosis (GFAP) and activation of microglia/macrophages (CD11b). Cell proliferation (Ki67) was increased at 1 and 3 days post-injury in the CC, particularly in areas adjacent to the SVZ. MR-DTI showed reduced FA values during the first week post-injury. Radial diffusivity values were unchanged. In contrast, axial diffusivity values were significantly decreased at 1, 3, and 7 days post-injury in comparison to pre-injury measures. Histological analysis after MRI revealed significant axon damage in the CC regions with decreased axial diffusivity.
This modified concussive model of mild TBI produces consistent TAI in the CC of adult mice. Yet, the acceleration-deceleration force is mild enough to limit damage predominantly to white matter tracts, particularly the CC. The anterior location of impact allows for analysis of damage within the CC as well as the regenerative response from endogenous neural stem cells within the adjacent SVZ, which is the largest pool of neural stem cells in the adult mammalian CNS. The impact location at the level of Bregma allows use of the midline crossing of the anterior commissure for coronal alignment during imaging and subsequent post-imaging tissue analysis. Longitudinal MR-DTI analysis detects reduced white matter integrity, with decreased axial diffusivity corresponding to histopathologically confirmed axon damage during the first week post-injury. This approach is advantageous for exploiting mouse genetics for mechanistic studies to prevent axon damage and promote regenerative responses from SVZ cells.
This work was funded by the U.S. Department of Defense in the Center for Neuroscience and Regenerative Medicine.
TBI, Corpus Callosum, MRI
POSTTRAUMATIC MOTOR AND COGNITIVE BEHAVIORAL IMPROVEMENTS ARE NOT ACCOMPANIED BY NEOCORTICAL NEUROPROTECTION IN CALPASTATIN OVEREXPRESSING TRANSGENIC MICE
Jennifer Brelsfoard, M.S., Unversity of Kentucky
Glenn C. Telling, Ph.D., Colorado State University
Dr. Kathryn Saatman, Ph.D., Department of Physiology; Spinal Cord and Brain Injury Research Center (SCoBIRC), University of Kentucky, Lexington, KY, USA
Prolonged activity of calpains early after traumatic brain injury suggests that endogenous action or levels of calpains' inhibitor, calpastatin, may be insufficient to fully inhibit its proteolytic activity, leading to neurodegeneration and behavioral dysfunction. Therefore, we hypothesized that calpastatin overexpression would reduce neuronal death and behavioral deficits after TBI.
Transgenic mice that overexpress human calpastatin (hCAST) under the ubiquitous prion protein promoter and their wildtype (WT) littermates were subjected to severe (1.0mm) controlled cortical impact (CCI) injury. Cleavage of the calpain substrate, collapsin-response mediator protein-2 (CRMP-2), was investigated at 6 and 24 hours post-injury due to the protein's close cytoskeletal association and its function in axonal growth. In an additional cohort, posttraumatic motor and cognitive behavior was assessed over the first week following injury. Assessments included a modified neurological severity score (NSS), Morris water maze post-injury learning paradigm, and the novel object recognition (NOR) task. At 10 days post-injury, neocortical tissue damage was quantified in Nissl-stained brain tissue.
Immunoblot analysis of neocortical and hippocampal homogenates demonstrated the appearance of CRMP-2 fragments following CCI (n=3-5/genotype/condition). Calpastatin overexpression resulted in a statistically significant decrease of CRMP-2 fragments in the neocortex at 6 and 24h post-injury (p<0.05 and p<0.0005, respectively). Within the hippocampus, CRMP-2 breakdown was reduced at 6h (p=n.s.), and completely inhibited at 24h post-injury (p<0.005). In behavioral tests, hCAST transgenic mice (n=20) exhibited a significant improvement in motor function compared to WT mice (n=19) across a 7d post-injury period (p<0.005). Brain-injured WT mice showed significantly impaired learning behavior at days 6–9 after CCI injury (p<0.0005), while injured hCAST mice did not (p=0.39). Furthermore, calpastatin overexpression attenuated injury-induced impairment in the ability to recognize a novel object at 10d post-injury (p<0.005). Despite these functional improvements, calpastatin overexpression did not affect neocortical contusion size.
Considering the structural and functional importance of calpain substrates, early preservation of these proteins with calpastatin overexpression may contribute to a reduction in behavioral dysfunction and neuronal cell death. Motor and cognitive testing revealed that calpastatin overexpression attenuated deficits associated with coordinated movement, post-injury learning, and novelty recognition. However, calpastatin overexpression had no effect on neocortical contusion size. Behavioral improvements may be due to neuroprotection in the hippocampus or other brain areas. In addition, calpain inhibition may mediate behavioral recovery through the preservation of axonal morphology and function. Given its beneficial effects on behavioral outcome, calpastatin may be a potential therapeutic avenue for the treatment of TBI.
Supported by NIH grants F31 NS071804 (KMS), T32 DA0222738, P30 NS051220, P01 NS058484, and Kentucky Spinal Cord and Head Injury Research Trust grant 6–12
Calpain, cognition, neuroprotection, proteolysis, transgenic
IMPROVING SURVIVAL FOLLOWING CEREBRAL EDEMA USING A HOLLOW FIBER-HYDROGEL DEVICE
Mike Hsu, M.S., University of California, Riverside
B. Hyle Park, Ph.D., University of California, Riverside
Victor G. J. Rodgers, D.Sc., University of California, Riverside
Devin K. Binder, M.D., Ph.D., University of California, Riverside
Cerebral edema is a significant cause of morbidity and mortality in many disease states. Current therapies of cerebral edema are often ineffective in treating severe edema. Here, we develop a hollow fiber-hydrogel device (HFHD) for direct surface contact-based treatment of severe cerebral edema.
Brain edema was induced in adult mice via water intoxication by intraperitoneal water administration (30% body weight, i.p.). Control mice received no treatment. A distinct group of mice were treated with craniectomy but no device application (craniectomy only). A third experimental group was treated with craniectomy and HFHD application. The HFHD contained a lumen solution of 350 g/L BSA in artificial cerebrospinal fluid at pH 7.4 and room temperature. Survival and brain water content were assessed as endpoints. A fourth experimental group was not water intoxicated but were treated with the HFHD to study the effects of the treatment on the underlying brain tissue.
Craniectomy and application of the HFHD enhanced survival in animals with severe cerebral edema. Animals treated with a craniectomy and HFHD (n=5) survived up to five hours longer than animals treated with craniectomy only (n=5) (p<0.001) or no treatment (n=5) (p<0.001). Animals treated with craniectomy and HFHD (n=5) had a survival rate of 80% within the observation period (360 minutes), whereas no animals treated with craniectomy only (n=5) or no treatment (n=5) survived longer than 50 and 33 minutes, respectively. Statistical significance was observed for survival rate between the animals treated with a craniectomy+HFHD (n=5) vs. craniectomy only (n=5) (p<0.001), and craniectomy+HFHD vs. no treatment (n=5) (p<0.001). Histological analysis demonstrated no significant cell loss in the cortex subjacent to HFHD application.
Here we demonstrate the feasibility of our HFHD to treat cerebral edema in this model. These results indicate that controlled water extraction from edematous brain tissue can be performed and lead to increased survival compared to craniectomy only. Further studies remain to be performed to further optimize the HFHD and to test it in more clinically relevant models, such as traumatic brain injury.
NIH K08 grant NS-059674 NSF IGERT Video Bioinformatics Fellowship (DGE 0903667) Richard Hausman and Isabel Escobar, University of Toledo Monica Carson, Jenny Szu, Jacqueline Hubbard, UC Riverside
Edema, Hollow Fiber-Hydrogel Device, Treatment
CAN STIMULATION OF BRAIN RECOVERY BY ENRICHED ENVIRONMENT PREVENT EPILEPTOGENESIS AFTER TBI?
Jukka Jolkkonen, MD, PhD, University of Eastern Finland
Asla Pitkanen, MD, PhD, University of Eastern Finland
Traumatic brain injury (TBI) is estimated to cause 10–20% of all symptomatic epilepsies. Previous studies show that enriched environment (EE) enhances motor and cognitive recovery. Here we hypothesized that improvement of post-traumatic recovery by EE is associated with reduced epileptogenesis.
Severe TBI was induced in 23 adult male Spraque-Dawley rats by lateral-fluid percussion injury. Six animals were sham-operated. At 4 weeks post-injury, 11 rats were placed into EE cages and 10 were housed in standard environment (SE). Motor recovery was assessed by Neuroscore and beam-walking tests at 1 d before TBI, and thereafter at 2 d, 7 d, 2 wk, 4 wk (before exposure to EE), and 8 wk (after EE) after TBI.The spatial learning was tested in Morris water-maze at 8 wk post-injury. To monitor the effect of EE on epileptogenesis,electrodes were placed over the parietal cortex at 12 wk post-TBI. Occurrence of spontaneous seizures was monitored with a continuous (24/7) video-EEG for 1 wk. Thereafter, seizure susceptibility was assessed by administering pentyleneterazol (PTZ, 25 mg/kg i.p.). Then, rats were perfusion-fixed for histology to assess axonal sprouting using Timm staining.
The EE protocol used did not result in any difference in motor recovery between the EE and SE groups. Spatial learning showed a trend towards improvement in the EE groups as compared to SE groups. No spontaneous seizures were detected during 1-wk video EEG monitoring. However, injured rats housed in EE had a higher number of spontaneous epileptiform spikes than injured rats in SE. Moreover, in PTZ test rats housed in EE showed a higher number of epileptiform discharges than SE group (p<0.05). Histological analysis revealed similar location and extent of cortical injury in the EE and SE groups. Interestingly, both the sham-operated controls and rats with TBI housed in EE had an increased density of mossy fibre sprouting in the infrapyramidal region of the CA3 ipsilaterally (p≤0.05, compared to SE).
Unexpectedly, EE appeared to facilitate rather than suppress epileptogenesis under the paradigm used. The relationship between the mechanisms of motor and cognitive recovery and epileptogenesis warrants further studies.
Academy of Finland, the Sigrid Juselius Foundation, Center for International Mobility (CIMO), Finland.
Enriched environment, Epileptogenesis, LFPI, Plasticity
DIFFUSE BRAIN INJURY INCREASES ACUTE QUANTITATIVE MEASURES OF REPARATIVE SLEEP IN THE MOUSE
Jordan Harrison, B.A., University of Kentucky/Spinal Cord and Brain Injury Research Center
Martin Striz, B.S., University of Kentucky/Department of Biology
Kevin Donohue, Ph.D., University of Kentucky/Electrical and Computer Engin.
Bruce O''Hara, Ph.D., Illinois State University
Dr. Jonathan Lifshitz, PhD, University of Kentucky/Barrow Neurological Institute at Phoenix Children's Hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Clinical observations report excessive sleepiness immediately following TBI, however, it is controversial whether one should sleep following brain-injury. We seek to associate acute inflammatory cytokine upregulation post-injury with immediate post-injury sleep. Our data will improve the understanding of the contribution of acute post-traumatic sleep to the natural course of outcome.
Diffuse TBI was induced by midline fluid percussion injury (mFPI) in young adult, male C57Bl/6 mice. Groups of mice were sham (n=16), mild (n=16; 0.8 atm; 1–3 min righting reflex time) or moderate (n=15; 1.2 atm; 6–10 min righting reflex time) brain-injured at 9AM or 9PM to evaluate injury-induced sleep behavior at different time points of the light/dark cycle. After brain injury, mice were returned to individual, non-invasive sleep monitoring cages. Sleep profiles (quantitative aspects of sleep/wake) were measured using a system that incorporates piezoelectric materials in the cage floor as highly-sensitive pressure detectors for 7 days post-injury. A separate cohort of mice received moderate or sham injury and mice were killed at specific time points for tissue collection. Inflammatory cytokines in the neocortex were quantified by MSD multiplex immunoassay and associated with the increases in acute post-traumatic sleep.
Results indicated that immediately after diffuse TBI there are injury-dependent changes in quantitative measures of sleep in the mouse characterized by a significant increase in sleep for the first 6 hours post-injury observed following both mild and moderate injury (suggesting the observable increase in post-traumatic sleep is independent of injury severity) and at 9 AM and 9 PM (suggesting the increase in post-traumatic sleep is also independent of the time of injury). Results also showed an increase in bout length (length of one sleep episode) that is temporally related to the increase in sleep and is significantly increased compared to the uninjured sham over the first 4 hours post-injury. Data indicated a significant increase in pro-inflammatory cytokines (IL-1β, IL-6) measured in the cortex of brain-injured mice in comparison to shams.
The pathophysiology of TBI consists of primary and secondary injuries. Secondary injury processes result in post-traumatic glial activation and consequently an increased production in pro-inflammatory cytokines, which can include cytokines also identified as sleep regulatory substances (SRSs). These cytokines concomitantly contribute to the reparative processes after neurotrauma and act on sleep regulatory circuits of the brain. Prevailing hypotheses suggest sleep is restorative in function in which sleep may mitigate injury following diffuse TBI. This study shows a significant increase in acute injury-induced sleep, which may be driving the natural course of recovery through cellular repair. We conclude that diffuse brain injury initiates the upregulation of pro-inflammatory cytokines that contributes to the promotion of sleep in the initial post-injury period. Disrupting post-injury sleep or interfering with injury-induced cytokines may have negative consequences on long term outcome.
Support provided by KSCHIRT 10-5A, NIH R21 NS072611. R01 NS065052-S1 Studies were conducted at the University of Kentucky.
Sleep, TBI, mFPI, inflammation
IMMEDIATE POST-INJURY SLEEP DISRUPTION ALTERS THE EXPRESSION OF INFLAMMATION RELATED GENES AFTER DIFFUSE BRAIN INJURY IN THE MOUSE
Rachel Rowe, B.S., University of Kentucky/Spinal Cord and Brain Injury Research Center/Department of Anatomy and Neurobiology
Bruce O'Hara, Ph.D., Illinois State University
Dr. Jonathan Lifshitz, Ph.D., University of Kentucky/Barrow Neurological Institute at Phoenix Children's Hospital/University of Arizona College of Medicine/Neuroscience Program, Arizona State University
Immediately following traumatic brain injury (TBI), individuals report being drowsy and excessively sleepy. Controversy surrounds whether an individual should sleep after a concussion or be awoken frequently. We seek to identify genetic expression patterns relating to post-injury sleep and the functional consequences of post-injury sleep disruption.
Adult male C57Bl/6 mice were subjected to either moderate midline fluid percussion injury (n=9; 1.2 atm; 6–10 min righting reflex time) or sham injury (n=6).These groups were further partitioned into subgroups receiving intentional sleep disruption using a minimally stressful strategy or no sleep disruption for the 6 hours following injury. At 6 hours, tissue biopsies were taken from selected brain regions for NanoString nCounter mRNA analysis of inflammation-related gene expression. A separate cohort of mice (n=25 injury, n=17 sham) received moderate or sham injury and were subjected to sleep disruption (n=9 injury, n=5 sham) or no disruption (n=16 injury, n=12 sham) for 6 hours post-injury. Post-injury learning was assessed using the Morris Water Maze (MWM) paradigm over 4 consecutive days (days 3–6 post-injury). Further functional evaluation was made using the Novel Object Recognition (NOR) paradigm at days 3, 7, and 14 post-injury.
In injured mice, sleep disruption significantly impacted cortical expression levels of genes associated with inflammation and the innate immune response (compared to disrupted sham group and non-disrupted injured and sham groups). Notably, interleukins (IL-2, IL-6, IL-10), cytokine signaling molecules that mediate cellular communication following the initiation of an immune response, were differentially expressed as a result of intentional sleep disruption post-injury. All isoforms of transforming growth factor beta (TGF-β), an immunoregulatory protein, were significantly increased in the sleep disrupted injured group. This gene expression data suggest that acute sleep disruption alters cellular immune responses in the brain and may influence the natural course of brain injury. Regardless of sleep disruption immediately post-injury, mice were able to learn the location of the hidden platform in the MWM task. Uninjured and brain-injured mice showed a time-dependent improvement in the latency to find the platform, (F(3,78)=14.66, p≤0.0001) even in sleep disrupted animals (F(3,36)=5.84, p=0.0023), but no group differences were observed. Performance on the NOR test also showed a time-dependent increase in familiar object recognition in sleep disrupted (F(3,36)=4.72, p=0.0071) and no sleep disrupted animals (F(3,39)=5.07, p=0.0046) but no group differences were observed.
Disrupting post-injury sleep significantly modifies the inflammatory response to brain-injury and highlights the importance of investigating acute post-injury sleep to determine its function and role in recovery from TBI. Despite this molecular response, these data indicate that acute sleep disruption following diffuse TBI does not elicit a profound cognitive-functional impairment. Sleep disruption was prescribed to offset a natural window of increased sleep following injury, which may not be sufficient to influence function. To date, the behavioral consequences of midline fluid percussion injury in the mouse are unknown, and may not include cognitive deficits in this time course. Disruption over a longer time period may alter cognitive outcome. Sleep disruption-linked changes in gene expression suggest that physiological interventions do interfere with the natural course of brain injury and recovery.
Support provided by NIH R21 NS072611, R01 NS065052-S1. Studies were conducted at the University of Kentucky.
TBI, mFPI, Behavior, gene expression
RESIDENTIAL NEURAL STEM/PROGENITOR CELLS PROMOTE TISSUE SPARING FOLLOWING TRAUMATIC BRAIN INJURY
Michelle H. Theus, PhD, University of Miami
Jose Mier, University of Miami
Steven G. Kernie, MD, Columbia University
Daniel J. Liebl, PhD, University of Miami Miller School of Medicine
TBI is a devastating disability that leads to tissue destruction and cell losses. Reducing and restoring cell losses through enhancing residential neural stem/progenitor cell (NSPC) replacement is a promising strategy. Ephrins and Eph receptors are key regulatory of residential NSPC function, and play important roles in promoting recovery following TBI.
Wildtype, Wildtypeδ-nestinTK, EphB3-/- or EphB3-/-δ-nestinTK mice carrying the viral thymidine kinase (TK) gene were administered cytovene or ganciclovir (10–100 mg/kg S.C. injection or mini-osmotic pumps), or vehicle for 4 weeks to ablate NSPCs and tested for optimal dosage. At 2 weeks of treatment the mice underwent moderate controlled cortical impact injury (CCI: 6.0 m/s velocity, 0.5 mm deep, 150 ms impact duration) or sham-operation. Motor function was tested on the accelerating rotarod at post-injury days 3, 5, 7 and 14 and expressed as a percentage of pre-injury function. At 4 weeks mice were perfused and the SVZ and injury site analysed using immunohistochemistry. In the SVZ, proliferating Ki-67 positive cells, as well as nestin positive cells were counted, and at the cortical injury site infarct size and extent of GFAP staining was measured. Statistical analysis was performed using students T-test and ANOVA.
Comparison of ganciclovir and cytovene showed that cytovene was a more effective compound at reducing NSPC numbers with less overall toxicity. Four-week treatments of cytovene (50 mg/kg S.C. mini-osmotic pumps) significantly ablated nestin-positive NSPCs in the SVZ in a dose-dependent manner. Comparison of the proliferative capacity between treatment groups showed that wildtype mice had greater numbers of Ki-67 positive cells compared to mice carrying single or double allele TK genes (i.e. wildtypeheterozygousNestinTKand wildtypehomozygousNestinTK mice, respectively). Losses in NPC numbers were associated with increased motor behavioural deficits using rotarod assessment, as well as increased infarct size. These findings support a role for nestin-positive stem/progenitor cells in the stabilization of the injured motor cortex. Analysis of EphB3−/− mice showed that EphB3 functions to limit neurogenesis following TBI, and eliminating these prohibitive signals leads to enhanced tissue sparing and motor function.
These data demonstrate that residential NSPCs play an important role in stabilizing/repairing the adult brain following traumatic injury, and that eliminating EphB3 signals may be beneficial to enhancing NSPC functions.
This work was supported by the Miami Project to Cure Paralysis, NIH/NINDS NS049545 (DJL) and NS30291 (DJL).
TBI, EphB3, stem cell, rotarod, thymidine kinase
PERFLUOROCARBON EMULSION IMPROVES CEREBRAL TISSUE OXYGENATION AFTER PENETRATING BALLISTIC-LIKE BRAIN INJURY IN RAT
Shoji Yokobori, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Shyam Gajavelli, Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Christine Bomberger, University of Miami, The Miami Project to Cure Paralysis
Michelle Zeidan, University of Miami, The Miami Project to Cure Paralysis
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Frank C Tortella, Ph.D., Walter Reed Army Institute of Research
M. Ross. Bullock, M.D., Ph.D., Miami Project to Cure Paralysis, Department of Neurosurgery, University of Miami Miller School of Medicine
Cerebral ischemia is a common secondary sequela of traumatic brain injury (TBI). The aim of this study is to analyze the pharmacological effect of the normobaric perfluorocarbon (PFC) emulsion (PerftecTM) in the rat penetrating ballistic-like brain injury (PBBI) on cerebral oxygenation using continuous brain oxygen tension (PbtO2, LICOX) monitoring.
Adult SD rats received unilateral PBBI and were assigned into one of three groups (n=7/group). Under anesthesia and controlled ventilation, a small burr hole was made at the right parietal skull and a PbtO2 monitor was inserted into the parietal brain tissue. Another burr hole was made at the right frontal skull for PBBI. After stabilization of the PbtO2 monitor (>20min), the PBBI probe was inserted and rapidly inflated to induce injury. (1) PFC group: Fifteen minutes post-injury, PFC emulsion (10ml/kg) was infused at a rate of 0.5 ml/min through the femoral vein catheter. (2) Vehicle group: An equivalent volume of saline was infused at a rate of 0.5 ml/min. (3) Sham control group: Received craniotomy and equivalent saline infusion without PBBI. PbtO2, MAP, and arterial blood gases were compared among the three groups. Data was expressed as percentage change from baseline.
There were no significant differences in MAP and arterial blood gas data among the three groups. Compared to the sham control group, cerebral oxygen tension was significantly decreased in the PBBI vehicle group. Tissue oxygenation was significantly improved in the PFC group (p<0.05) but not in the vehicle group. At the end of the recording, PbtO2 in the PFC group was significantly higher than in the vehicle group (last 10 minutes; 23.54±0.65 vs 20.05±0.38 mmHg).
This study suggests the efficacy of PFC as a therapeutic strategy targeted to abolish/reduce the hypoxia observed in PBBI. Further, volumetric histopathology and cell count studies are under way.
Supported by DOD Grant PTO74521-W81XWH-08-1-0419
TBI, PBBI, PbtO2, perfluorocarbon, neuroprotection
BEHAVIOR, MOTOR, AND COGNITION ASSESSMENTS IN NEONATAL PIGLETS
Sarah Sullivan, BS, MS, University of Pennsylvania
Stuart Friess, BS, MD, The Children's Hospital of Philadelphia
Jill Ralston, BS, University of Pennsylvania
Colin Smith, BSc, MBChB, MD, FRCPath, University of Edinburgh
Kathleen Propert, BS, MS, ScD, University of Pennsylvania
Paul Rapp, BS, PhD, Uniformed Services University
The alterations of animal behavior following traumatic brain injury (TBI) can be subtle and their quantitative characterization can present significant methodological challenges. Quantitative behavior outcome measures are required to compare the efficacy of different clinical interventions in animal models of TBI.
Neonatal piglets underwent open field, T-maze and inclined beam behavioral testing prior to experiencing either a sagittal rapid rotational brain injury (N=8) (Eucker 2011) or sham instrumentation (N=7). The testing was then repeated on Days +1 and +4 post-injury. Using the outcomes from this testing we developed a battery of assessments to quantify behavioral, motor, and cognitive changes in neonatal piglets with good sensitivity and specificity to the detection of long-term post-TBI deficits. Furthermore, the composite porcine disability score developed was correlated to overall axonal injury severity and axonal injury in particular anatomical regions of the brain, as assessed by neuropathology on Day +6 post-injury. Outcomes from behavioral testing were also compared across previous studies to show behavioral differences that arise based on the plane of rotational injury (sagittal or axial) (Eucker 2011, Naim 2010, Friess 2009).
The battery of measures developed included open field behaviors of sniffing and moving a toy, novel locomotion measures of Lempel-Ziv complexity and the probability of remaining in current location, and a novel metric for evaluating motor performance. Our composite porcine disability score was able to detect brain injury with a sensitivity of 100% and specificity of 85.7% at Day +4 post injury and significantly correlated with the percent axonal injury (Day +4: ρ=0.76 p=0.0011). Furthermore, the porcine disability score had significant correlations to percent injury in the regions evaluated (anterior and posterior cerebrum, internal capsule, fimbria fornix, and corpus callosum) with the highest correlation in the fimbria fornix (ρ=0.74 p=0.0018).
Comparisons across studies showed that for axial and sagittal rotations with similar levels of axonal injury at 6 hours post-injury, the percent injury was significantly decreased by 6 days post-injury in axial plane injuries, but unchanged in sagittal plane injuries. Additionally, at Day +1 post-injury significant differences were found between axial and sagittal plane injuries in the locomotion measures of Lempel-Ziv complexity and the probability of remaining in current location. These differences showed two distinct presentations of TBI based on plane of rotation, one characterized by stationary and highly patterned movement (sagittal) and the other by frequent and more random movements (axial).
We have developed a novel battery of cognitive, motor, and behavioral assessments for the neonatal piglet that has enhanced sensitivity and specificity compared to our previous assessments (Friess 2009, Friess 2007). This new porcine disability score has potential use in a wide variety of porcine disease and injury models, specifically in the evaluation of clinical interventions in the pediatric population.
Supported by NIH grants R01NS039679, R01NS0698545, K08NS064051, and by the Traumatic Injury Research Program of USU and the Defense Medical Research and Development Program.
TBI, Cognition, Pediatric, Porcine
FK-506 AND AMANTADINE: A COMBINED TREATMENT STRATEGY TARGETING DARPP-32 FOLLOWING AN EXPERIMENTAL TRAUMATIC BRAIN INJURY
James W. Bales, M.D./Ph.D., University of Pittsburgh/School of Medicine/Neurosurgery
Kristin L. Macfarlane, B.S., University of Pittsburgh/School of Medicine/Neurosurgery
C. Edward Dixon, Ph.D., University of Pittsburgh, School of Medicine/Neurosurgery
This study tested the hypothesis that treatments targeting TBI induced dysfunction in DARPP-32 phosphorylation can ameliorate chronic cognitive deficiencies. We examined the role of a combined therapy of FK-506, a calcineurin inhibitor, and Amantadine hydrochloride, a partial NMDA antagonist, on DARPP-32 signalling post injury.
Controlled Cortical Impact (CCI) was performed with an impact velocity of 4 m/s and deformation of 2.4 mm. The striatum was excised and frozen until use for protein or animals were perfused and brains extracted for sectioning. Rats with sham surgery had a craniotomy performed, but no CCI. Eighty rats (n=20/treatment; 10 sham and 10 injured) underwent CCI. At 5 minutes post injury FK-506 (1.0 mg/kg i.p.) was administered to the FK-506 alone and combined groups. Beginning at 24 hours post injury AMH (5 mg/kg i.p.) was administered to the AMH alone and combined group rats daily for 18 days. Combination therapy included the injection of FK-506 with chronic AMH. Saline treated animals were used for controls. At days 1–5 post injury animals underwent motor testing. At days 14–19 post injury animals underwent Morris water maze testing. Animals were sacrificed on day 21.
Previously we demonstrated that a lateral controlled cortical impact injury on rats (isoflurane anesthetic during injury) produced a chronic decrease in DARPP-32 phosphorylation at threonine-34 (p≤0.05; ANOVA), subsequent increase in protein phosphatase-1 activity (p≤0.01; ANOVA), and decrease in the phosphorylation of nuclear ERK and CREB. A combined treatment strategy of an acute administration of FK-506 (1.0 mg/kg i.p.) at 5 minutes post injury with two weeks of Amantadine given daily (5.0 mg/kg i.p.) showed no significant difference between sham and injured animals receiving the combined treatment on Morris water maze testing (p>0.05; ANOVA, Bonferroni post-hoc) or on beam walk (p>0.05; ANOVA, Bonferroni post-hoc) or beam balance tasks (p>0.05; ANOVA, Bonferroni post hoc). Both FK-506 and Amantadine were able to restore DARPP-32 Thr-34 phosphorylation and ERK phosphorylation in injured animals. Animals treated with FK-506 alone or in combination with Amantadine had significantly reduced lesion volumes compared to animals treated with vehicle (p<0.05; ANOVA).
These results indicate that a combined treatment strategy aimed at DARPP-32/Thr34 phosphorylation provided functional benefit to rats following a controlled cortical impact, assessed utilizing Morris water maze, beam walk, and beam balance tasks. Future studies will utilize DARPP-32 knockout mice to evaluate the role of DARPP-32 in this treatment strategy.
Support: NIH P50NS030318, VA RR&D#B6761R, NIH R01NS060672
TBI; dopamine; striatum; cognition
USING ROBUST MORRIS WATER MAZE METRICS TO DISSOCIATE IMPLICIT AND EXPLICIT LEARNING IN THE CCI MODEL OF TBI
Max Hurwitz, BS, Philadelphia College of Osteopathic Medicine
Michelle Carter, BS, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Jay Fuletra, UPMC Department of Physical Medicine and Rehabilitaton
Chris Henderson, UPMC Department of Physical Medicine and Rehabilitaton
Scott Ketchman, UPMC Department of Physical Medicine and Rehabilitaton
Huichao Zou, MD, PhD, UPMC Department of Physical Medicine and Rehabilitaton
Amy Wagner, MD, University of Pittsburgh - Department of Physical Medicine & Rehabilitation
Distinct networks exist for implicit and explicit learning/memory, and non-spatial pre-training (NSPT) can dissociate their use in the Morris water maze (MWM). Previous work suggests that NSPT is effective in reducing latencies for sham and controlled cortical impact (CCI) model TBI rats, without elucidating the specific behavioral strategies.
We evaluated swim strategy selection and time allocation measures to better understand behavioral mechanisms behind latency reductions in the MWM. 76 adult male Sprague-Dawley rats [sham N=32, injured (parasagittal CCI 2.8mm; 4m/s, parietal cortex) N=44] were divided into eight groups. CCI and sham rats were exposed to NSPT before surgery, and subjected to place learning (acquisition: d14–18) and retention trials (probe/visible platform: d19, reversal: d20–21), both with/without extra-maze cues. NSPT was performed across four days with a rotating hidden platform, extra-maze cues covered, and static entry point. Swimming strategy selection was analyzed to determine how NSPT impacts navigational behavior and incorporation of extra maze cues on latency reduction. Swim strategies where characterized as spatial (most efficient), non-spatial, or thigmotaxic (least efficient) by an observer blinded to treatment groups. Pool zone maps were utilized to preclude from analysis incidental time spent in the target quadrant during thigmotaxic behavior.
Acquisition Trials: Within injured groups, NSPT rats utilized more non-spatial strategies than No-NSPT groups, regardless of extra-maze cues. Sham/NSPT rats in the cued condition utilized more spatial strategies than No-NSPT rats. Cues had little effect on thigmotaxis for Injured NSPT rats. Sham/NSPT rats, by contrast, responded positively to cues, exhibiting less thigmotaxis. Probe Trial: NSPT groups had higher target zone time allocation in both injured and sham groups regardless of cues, compared to no-NSPT controls. Cues, however, had little effect on target zone time allocation within injured/NSPT and sham/NSPT groups. No group utilized a spatial strategy during this “memory retention” task. Visible Platform (VP) Trial: Injured and sham NSPT groups performed better than No-NSPT on target zone allocation measures in the cued condition. NSPT animals relied heavily on spatial strategies, and exhibited fewer thigmotaxic swim patterns. NSPT was beneficial for injured No-Cues rats, and enhanced swim strategy by decreasing thigmotaxis and increasing spatial strategies. NSPT had little effect on time allocation of non-cued shams, but increased non-spatial swim strategy use in this proximal cue only condition. Reversal Trials: NSPT led to decreased thigmotaxis and increased spatial and non-spatial strategy utilization in the injured/cued groups, and had similar effects in the non-cued condition. In shams, NSPT eliminated thigmotaxis and led to higher frequencies of spatial search strategies in cued groups, but had little effect in the non-cued condition.
Our results indicate that implicit/nonspatial and explicit/spatial components of place learning and memory are dissociable in the MWM. With NSPT, both sham and injured rats show less thigmotaxic behavior. Latency reductions in platform location for Sham/NSPT rats occur because these rats choose more efficient spatial swimming strategies to locate the platform. In contrast, injured/NSPT rats make improvements in platform latencies by using predominantly non-spatial strategies, but are largely unable to incorporate extra-maze cues into their swimming strategy selection. Data also suggest low sensitivity with standard probe trials as a spatial memory retention task, given that spatial strategies are not utilized in this condition. Even in the VP condition, NSPT can measurably improve performance, despite the presence of the proximal spatial cue, solely by reducing thigmotaxis. These results support clinical literature regarding relative preservation of implicit learning networks after TBI and have implications for cognitive training strategies in experimental TBI.
UPMC Department of Physical Medicine and Rehabilitation
TBI, Learning/Memory, Water Maze, Thigmotaxis
EFFECTS OF MILD BLAST-INDUCED NEUROTRAUMA ON BLOOD-BRAIN BARRIER PERMEABILITY IN A COMPLEX BLAST RODENT MODEL
Margaret Parsley, University of Texas Medical Branch
Debbie Boone, BS, University of Texas Medical Branch
Douglas DeWitt, PhD, University of Texas Medical Branch
Donald S Prough, MD, UTMB
The blood-brain barrier (BBB) regulates the passage of molecules between the cerebral vasculature and central nervous system. Although much is known of the integrity of the BBB in other TBI models, the effects of blast-induced neurotrauma (BINT) on the BBB are unknown.
We evaluated the integrity of the BBB at acute time points following sham injury or BINT by extravasation of Evan's blue dye and quantification via spectrophotometer. Rats were injected with 2% Evan's blue dye at 30 minutes, 120 minutes or 24 hours after BINT or sham injury. After 60 minutes of dye circulation, animals were perfused and brains removed. Brains were divided into hemispheres, weighed, homogenized with 50% trichloroacetic acid in 1X PBS and centrifuged. 2μl of the supernatants were analyzed at 620nm via spectrophotometry.
At 30 minutes post-injury, Evan's blue concentrations in the mild BINT or sham groups were 20.1+1.8 SEM or 5.5+3.9 SEM μg/g tissue, respectively. At 2 hrs post-injury, mild BINT Evan's blue concentrations were 11.8+0.7 SEM μg/g tissue and 24 hrs post-injury, mild BINT Evan's blue concentrations were 15.4 μg/g tissue. Injury group does affect the result at all time points (p-value=0.0792).
The increases in brain tissue concentrations of Evan's blue after mild BINT indicates that blast injury is associated with increased BBB permeability. Blast-induced BBB damage may be due physical damage to cerebral blood vessels, disruption of endothelial cell-cell interactions and/or changes in regulatory genes associated with maintaining the trans-endothelial resistance of tight junctions. Understanding the timing and extent of permeability changes as a result of BINT will facilitate the development of more effective preventative or therapeutic applications.
This research is funded by The United States Department of Defense.
Blood-brain barrier, traumatic brain injury
INHIBITION OF TBI-INDUCED HIPPOCAMPAL NEUROGENESIS IMPAIRS COGNITIVE RECOVERY
Teresa Daniels, B.S., Virginia Commonwealth University
Andrew Rolfe, B.S., Virginia Commonwealth University
Michael Waters, B.S., Virginia Commonwealth University
Traumatic brain injury (TBI) induces a robust cellular proliferative response among neural stem/progenitor cells in the dentate gyrus (DG) of the hippocampus. This proliferative response is thought to contribute to the innate cognitive recovery observed following TBI. Furthermore, enhancement of injury-induced hippocampal neurogenesis via intraventricular administration of basic fibroblast growth factor (bFGF) improves cognitive function in animals following TBI.
In this experiment, we investigated the direct association of injury-induced hippocampal neurogenesis to cognitive recovery utilizing an antimitotic agent, arabinofuranosyl cytidine (Ara-C). In this study, adult rats received moderate lateral fluid percussion injury (LFPI). Immediately following injury, Ara-C with or without bFGF was administrated into the lateral ventricle via an osmotic mini-pump for 7 days. Animals received daily single injections of 5-bromo-2′-deoxyuridine (BrdU) to label dividing cells at 2–7 days post-injury. To examine the effect of Ara-C on cell proliferation, a group of animals were sacrificed at 1 week following injury. Brain sections were immunostained for BrdU and cell type specific markers, and the number of BrdU+ cells in the hippocampus was assessed by stereology. To examine the outcome of inhibition of injury-induced cell proliferation on cognitive recovery, animals were assessed on Morris water maze tasks (MWM) either at 21 to 25 days or 56–60 days post-injury
We found that post-injury Ara-C treatment significantly reduces injury-induced cell proliferation in the DG and abolishes the innate cognitive recovery on MWM performance at 56–60 days post-injury. Additionally, Ara-C diminishes bFGF enhanced cell proliferation in the DG and cognitive recovery following TBI.
These results support the causal relationship between injury-induced hippocampal neurogenesis and cognitive functional recovery. Our studies suggest that the post-TBI neurogenic response is an endogenous repair mechanism that contributes to the restoration of hippocampal function post-injury.
Supported by NIH/NINDS RO1-NS055086 (Sun).
Neurogenesis, hippocampus, morris water maze
QUALITATIVE AND QUANTITATIVE ASSESSMENT OF THE RETINAL GANGLION CELL RESPONSE TO DIFFUSE TRAUMATIC AXONAL INJURY(TAI) IN THE OPTIC NERVE: IMPLICATIONS FOR NEURONAL SURVIVAL AND OPTIC NERVE REPAIR AND REORGANIZATION
Prof. John T. Povlishock, PhD, Anatomy & Neurobiology, Virginia Commonwealth University
Visual damage/dysfunction has recently been identified as a feature of human traumatic brain injury (TBI). TAI and axonal dieback have also been observed in the optic nerve following experimental TBI. However, it is unclear whether this TAI is associated with retinal ganglion cell death and its downstream consequences.
YFP-16 transgenic mice were subjected to mild central fluid percussion injury and allowed to survive from 24h to 3months. Qualitative evaluation of retinal ganglion cells (RGC) apoptosis and death were conducted by immunostaining targeting cleaved caspase-3 as well as TUNEL approaches interfaced with electron microscopy (EM). Quantification of RGC survival was based upon retinal sector counting of Brn3a immunoreactive RGCs. Phospho-C-jun immunolabeling was performed to assess RGC recovery and axonal reorganization which was also evaluated by EM. Additionally, optic nerve sections were reacted with antibodies targeting microglia / macrophages (Iba1) and oligodendrocytes (CC1).
YFP positive cells in the RGC layer revealed no overt loss at 2d, 7d, 14d and 28d post TBI. No cleaved caspase-3 immunoreactive cells were observed within the RGC layer from 2 days to 3 months post-TBI. Further, no TUNEL+ cells were observed within the RGC layer at any time point assessed post-TBI. Ultrastructural analysis of cells within the RGC layer did not show evidence of apoptosis or necrosis, with no chromatin condensation or apoptotic bodies visualized within RGC layer cells via electron microscopy. Quantitative analysis of the number of RGCs, based upon Brn3a immunostaining, revealed no significant loss of RGCs at 7d, 14d and 28d post TBI. Concomitant with the survival of RGCs, phospho-c-Jun immunoreactivity was significantly elevated within the RGC layer from 2d to 3months post-TBI. Consistent with the survival of RGCs and potential axonal reorganization/repair, the axonal fibers originating from the RGC revealed improved structural detail and reorganization at 28d post TBI in comparison to that seen at 24h and 48h post TBI. In parallel with these axonal changes, numerous activated microglia/macrophages were recognized in the optic nerve at 7d, 14d and 28d post TBI. Specifically, many of the distal disconnected swollen axonal segments, in the process of dieback, were encompassed by the processes of activated microglia/macrophage, consistent with their role in the clearance of the degenerating axonal segments. In contrast, the microglia/macrophages appeared less active in the proximal axonal segments, with their processes paralleling while not enveloping these axonal profiles.
1.Using multiple qualitative and quantitative endpoints to assess the potential for RGC death and/or recovery over an extended time period post injury, we could find no evidence of significant RGC death following diffuse traumatic axonal damage in the optic nerve. 2.Overtime, the reorganization of axonal fibers in the optic nerve is consistent with the survival of RGCs following axonal damage. 3.Parallel studies probing the activation and morphology of microglia/macrophages demonstrate that they closely approximate the damaged and disconnected axons over time, suggesting that they are responsible for the observed distal axonal degeneration and dieback. 4.The morphological transformation of macrophage, in the distal, disconnected axonal segment versus the proximal axonal segment in continuity with RGC, suggests that the activity of microglia/macrophage maybe modified by the microenvironment, creating a conduit for potential axonal reorganization and repair.
This study is supported by HD055813 and NS047463.
TAI, RGC survival, axonal reorganization
TRANSIENT MEMBRANE DISRUPTION AFTER BLAST EXPOSURE AND PHARMACOLOGICAL INTERVENTIONS FOR PROTECTION
Rania Abu-Taleb, BS, Walter Reed Army Institute of Research
Manojkumar Valiyaveettil, PhD, Walter Reed Army Institute of Research
Ying Wang, MD, Walter Reed Army Institute of Research
Joseph Long, PhD, Walter Reed Army Institute of Research
Madhusoodana Nambiar, PhD, Walter Reed Army Institute of Research
Currently, neurocellular mechanisms of blast-induced traumatic brain injury (blast TBI) and the subsequent neurobehavioral abnormalities are still not completely understood. We have reported earlier that blast exposure transiently releases cytosolic proteins without cell death and proposed that compromised cell membrane integrity is a mechanism leading to neuronal injury.
The integrity of cell membrane after blast exposure was assessed by fluorescent dye uptake/release techniques in SH-SY5Y human neuroblastoma cells using our newly established in vitro model of blast TBI with shock tube. Additionally, the possibility of cell membrane disruption after blast exposure was explored using our mouse model of single and repeated blast exposures.
Our data using the in vitro model indicate that blast exposure results in transient changes in neuronal cell membrane integrity leading to abnormal bidirectional transport of molecules across the cell membrane. These findings support the hypothesis that transient loss of cell membrane integrity is a mechanism for blast-induced neuronal injury. Additional experiments with mice showed that, blast exposure leads to overpressure dependent rapid release of tissue enzymes into the blood, which was independent of cell necrosis. Treatment with CDP-choline, a cell membrane stabilizer, preserved neuronal cell membrane integrity and inhibited neurobiological effects induced by blast exposure.
These results suggest that transient cell membrane disruption after blast exposure might be a potential mechanism of blast TBI. The effect of CDP-choline in the preservation of neuronal cell membrane integrity against blast exposure imply that pharmacological intervention with membrane stabilizers serve as a potential therapeutic strategy against blast TBI.
N/A
Blast TBI, membrane disruption, CDP-choline
TEMPORAL AND SPATIAL PROFILE OF HISTOPATHOLOGICAL CHANGES CAUSED BY HEMORRHAGIC SHOCK AFTER PENETRATING BALLISTIC-LIKE BRAIN INJURY (PBBI)
Shawn McLoughlin, Walter Reed Army Institute of Research
Guo Wei, Ph.D., Walter Reed Army Institute of Research
Deborah A Shear, Ph.D., Walter Reed Army Institute of Research
Frank Tortella, Ph.D., Brain Trauma Neuroprotection & Neurorestoration Branch, Center of Excellence for Psychiatry & Neuroscience, WRAIR
Traumatic brain injury (TBI) is often accompanied with hemorrhagic shock (HS). Evidence suggests that HS may worsen the brain insult and corresponding outcome. However, the complex neuropathological cascade of the combined injuries(TBI+HS) is poorly understood. This study examined whether HS following PBBI exacerbated histopathological outcome in different brain regions.
Rats were randomly assigned into four groups: Sham control(craniotomy only); PBBI; HS; PBBI+HS (n=6/group/time point). Unilateral frontal PBBI was produced in the right hemisphere of isoflurane anesthetized rats. HS was induced by drawing blood from the tail artery to reduce mean arterial pressure to 40–45 mmHg. This hypotensive state was maintained for 60 min. Lactated Ringer's solution was given as fluid resuscitation after HS. In the PBBI+HS group, HS was initiated 5 minutes after PBBI. Rats were sacrificed by transcardial perfusion at 3 and 7 days post-injury. Brains were harvested, post-fixed in 4% paraformaldehyde and cryoprotected in 20% sucrose solution. The samples were processed for Hematoxylin-Eosin (HE) staining, silver staining and immunostaining for glial fibrillary acidic protein (GFAP) for astrocyte activation. Lesion volume was quantified on HE-stained slices. Positive stained areas in silver-stained and immunostained slices were quantified using threshold analysis in cortical and subcortical regions of interest.
Lesion volume in the PBBI+HS group was comparable to the PBBI group at 3 and 7d post-injury. At 3d post-injury, upregulation of GFAP in the hippocampus was significantly (3-fold) higher in the PBBI+HS vs. PBBI group but no significant differences in GFAP expression were detected between these 2 groups in the ipsilateral cerebral cortex, corpus callosum, striatum and thalamus. At 7d post-injury, GFAP expression in these brain regions remained elevated (compared to sham and HS) but was more pronounced in the corpus callosum and thalamus of the PBBI+HS group (compared to PBBI alone; p<.05). At 3d post-injury, the PBBI+HS and PBBI groups showed similar profiles of neuronal degeneration (measured by silver staining). Particularly in the corpus callosum, significantly (2-fold) more degenerated axonal fibers was observed in the PBBI+HS vs. PBBI group at 7d post-injury (p<.05). Notably, neurodegeneration was also evident in the contralateral hemisphere of PBBI and PBBI+HS animals.
Overall, the differential profiles of astrocyte activation and neurodegeneration in the PBBI+HS vs. the PBBI alone groups indicates that HS exacerbates the effects of PBBI producing more severe neuropathological outcome. Critically, the combined effects of PBBI+HS increased over time and were more apparent at 7 days post-injury. It has been well-established that HS reduces cerebral blood flow, depriving tissue of oxygen and nutrients and promoting further damage to already compromised neurons and glia. Based on the current results, lesion volume analysis alone may not be sensitive enough to capture this aggravated effect. On the contrary, GFAP expression and silver staining may provide sensitive measures to detect the pathological differences between PBBI and the combined injury. Chronic effects of combined HS and PBBI injuries as well as other neuroinflammatory responses involving microglia remain to be determined.
This work is supported by CCCRP. LYL is sponsored by National Research Council Research Associateship Program. The authors thank Xiaofang Yang for technical assistance.
Polytrauma, PBBI, Hemorrhagic Shock, Histopathology
LONGITUDINAL PROFILE OF GAIT DISTURBANCES FOLLOWING PENETRATING BALLISTIC-LIKE BRAIN INJURY USING THE CatWalk GAIT ANALYSIS SYSTEM
Lai Yee Leung, Ph.D., Walter Reed Army Institute of Research
Rebecca Pedersen, Walter Reed Army Institutes of Research
Deborah A. Shear, Ph.D., Walter Reed Army Institute of Research
Frank C. Tortella, Ph.D., Walter Reed Army Institute of Research
Traumatic brain injury results in enduring motor and cognitive dysfunction. Although gait disturbances have been documented among TBI patients, few studies have profiled gait abnormalities in animal models of TBI. We sought to obtain a comprehensive longitudinal analysis of gait function following severe penetrating ballistic-like brain injury in rats.
Unilateral frontal PBBI (n=10) was induced in the isoflurane anesthetized rat by stereotactically inserting a perforated steel probe through the right frontal cortex and rapidly inflating the elastic tubing of the probe tip into an elliptical shaped balloon with a volume equal to 10% of total rat brain volume. Control groups included insertion of the probe alone (probe control) or sham surgery (n=10/group). Sensorimotor performance of sham, probe, and PBBI rats was assessed using the CatWalk automated gait analysis system. Baseline measurements were taken 3 days prior to injury and detailed analysis of gait was performed at 1, 3, 7, 14, and 28 days post-injury. Individual footprints were identified and processed using the CatWalk software, which generates comprehensive static and dynamic gait parameters. Mean changes of statistically significant gait abnormalities were analyzed between all groups.
Our primary objective was to identify a longitudinal profile of gait abnormalities following PBBI using the CatWalk. Both PBBI and probe-inserted rats displayed altered static and dynamic gait parameters that were primarily evident during the acute (<7 days) post-injury phase and were resolved by 1 month post-injury. As expected, PBBI produced more severe deficits compared to probe-alone which were reflected in the number, magnitude, and resolution time of abnormal gait parameters. In both PBBI and probe-inserted rats, altered parameters were detected in all four paws but were more apparent on the contralateral side. Gait parameters including paw pressure, print area, swing speed, and stride length were significantly decreased whereas stance, swing, and step cycle duration were increased compared to sham. Injured rats showed an altered distribution of limb support that appeared immediately following injury and resolved by two weeks.
Altered gait patterns detected using the CatWalk system in the PBBI model appear similar to those reported in severe TBI patients. However, significant motor recovery occurred by 1 week post-PBBI on the CatWalk task, whereas severe TBI patients show more enduring motor deficits. Collectively, these results indicate that the CatWalk may be useful for short-term neuroprotection studies focused on the acute post-injury recovery period after PBBI but not for chronic studies that require repeated testing at later time points.
This research was funded by the Army Combat Casualty Care Research Program and performed while the authors (AM, LYL) held an NRC Research Associateship Award at Walter Reed Army Institute of Research.
TBI, gait, Catwalk
INCREASE IN CELL-FREE DNA LEVEL AND BRAIN DNA DAMAGE AFTER BLAST EXPOSURES
Yanling Wei, M.D, Walter Reed Army Institute of Research
Samuel Oguntaya, Walter Reed Army Institute of Research
Irene Gist, Walter Reed Army Institute of Research
Joseph Long, Ph.D., Walter Reed Army Institute of Research
Madhusoodana Nambiar, Ph.D., Walter Reed Army Institute of Research
Blast exposures often result in head injuries with heterogeneous pathology. Shockwaves can degrade cellular DNA and eject them out by disrupting the membrane. We have evaluated the effect of repeated blast exposures on the integrity of DNA in brain and the level of cell-free DNA (CFD) in the plasma.
C57BL/6J mice were anesthetized and exposed repeated (three times with 1–30 min intervals) blast overpressures (20.6 psi) using a shock tube and euthanized at different time points (2 h, 6 h, 1, 2 and 7 days) (Wang et al., 2011). Cerebellum, cortex and hippocampus were dissected immediately and neural cells were separated and subjected to comet assay. The plasma samples were collected for measurement of CFD by using SYBR fluorescence detection.
Our data show that blast exposure results in time-dependant DNA breakdown. Minimal DNA damage was observed immediately after blast where as maximum DNA breaks was observed at 24 h. Consistent with the earlier reports that cerebellum is more vulnerable to blast exposure, the percentage of cells with DNA damage showed significantly higher in the cerebellum compared to the cortex and hippocampus. Compared to the sham controls, the plasma CFD level increased significantly at 2 and 6 h and no difference at 24 h after injury. A positive correlation was demonstrated between CFD level and the righting reflex time in 6 h after blast exposures.
The data indicate that brain injury after primary shockwave transmission may be a significant factor contributing to the DNA damage in addition to shcokwaves. Although the source of CFD is not clear, it may be an early predictor of neurological outcomes. Therapeutic treatments aimed at enhancing DNA repair after neurotrauma can be beneficial for long-term positive outcomes in blast-induced traumatic brain injury.
RAD III
Blast, Neurotrauma, DNA damage, Cell-free-DNA
EFFECTS OF ACUTE ETHANOL TREATMENT ON SYNAPSIN PHOSPHORYLATION IN THE DELAYED POST-TRAUMATIC PERIOD
Emily L. Henson, BS, Wayne State University
Jennifer L. Lowing, BS, Wayne State University
Laura L. Susick, PhD, Wayne State University
Ramesh Raghupathi, PhD, Drexel University College of Medicine
Traumatic brain injury (TBI) is associated with an increase in alcohol-use disorders. Previously we have shown that non-contusive TBI increases ethanol-induced sedation in mice. Here, we examine the role of phospho-synapsin, a known mediator of ethanol action in the brain, in the mechanism of TBI-induced increases in ethanol sensitivity.
Male C57BL/6J mice (6–7 wks) were anesthetized with isoflurane and subjected to an impact over the midline suture of the intact skull; sham-injured animals were surgically prepared but were not injured. At 14 days post-injury/surgery, mice were treated once with either saline or ethanol (4 g/kg, i.p.). At 30 minutes after saline or ethanol administration, the cortex, hippocampus and striatum were rapidly dissected and whole cell lysates were subjected to immunoblotting for phosphorylated synapsin I and II proteins.
Non-contusive brain injury along with saline administration appeared to increase the signal intensity of phospho-synapsin in a region- and isoform specific manner compared to saline-treated sham-injured mice. For example, in the striatum, phospho-synapsin I and II signals were intensified compared to sham controls, while in the hippocampus, this intensity increase appeared restricted to phospho-synapsin II. Interestingly, the cortex did not demonstrate any alterations in phospho-synapsin levels. Although sham-injured animals treated with ethanol demonstrated intensified phospho-synapsin I and II signals in all regions analyzed compared to uninjured controls treated with saline - as reported in previous studies - the response to ethanol in brain-injured mice was dependent on the region and the isoform. Ethanol appeared to intensify the signal of phospho-synapsin I in the injured striatum and hippocampus, but not in the cortex. In contrast, ethanol administration did not appreciably affect the TBI-induced alterations of phospho-synapsin II in any region examined.
These observations of altered phosphorylation states of presynaptic vesicle-related proteins in brain-injured mice are suggestive of sustained derangements of synaptic function, which may underlie the altered sensitivity to ethanol in the chronic post-traumatic period.
Supported by resources from Department of Neurosurgery, WSU Office for Vice President for Research and John D. Dingell VA Medical Center, Detroit, MI (ACC).
Ethanol, plasticity, synapsin, TBI
RECOVERY OF STRESS RESPONSE COINCIDES WITH RESPONSIVENESS TO VOLUNTARY EXERCISE AFTER MILD TRAUMATIC BRAIN INJURY
Delia Tio, BA, UCLA
Shyama Nair, BA, UCLA
Voluntary exercise increases levels of brain derived neurotrophic factor (BDNF) after fluid percussion injury (FPI) when it occurs at a delayed time window. Here we determine the stress response with delayed exercise. We also explored if the type of exercise influences the stress and BDNF responses.
Four weeks after injury, animals were randomly assigned to a sedentary (Sed), voluntary running (vRW), or forced running (fRW) condition. Rats were exercised for 10 days. Blood collections were taken before and during the exercise period to determine levels of corticosterone (CORT) and adrenocorticotropic hormone (ACTH). At postinjury day (PID) 39, rats were sacrificed. Hippocampal BDNF levels were determined.
All animals under the fRW condition presented higher levels of both CORT and ACTH compared to either the vRW or Sed conditions (p<0.005). These results differ from the acute exercise paradigm in that ACTH and CORT increases were more pronounced in the FPI rats; particularly those that endured a longer period of unconsciousness following injury. The relationship between the duration of unconsciousness and increases in CORT was not observed when exercise was delayed. Likewise, exercise-induced BDNF increases were observed after delayed vRW in both the FPI and Sham rats (p<0.005). No increases in BDNF were observed after delayed fRW.
These findings suggest that the heightened stress response observed during the subacute postinjury period contributes to the inability of exercise to increase BDNF. These findings also suggest that exercise paradigms that elicit a pronounced stress response may be counterproductive following TBI.
Support: NS06190 and the UCLA Brain Injury Research Center
exercise, corticosterone, stress, brain-derived-neurotrophic-factor
PROGESTERONE FOR NEUROPROTECTION IN A RAT MODEL OF PEDIATRIC TRAUMATIC BRAIN INJURY
Manda Saraswati, PhD, Johns Hopkins School of Medicine
Bindu Balakrishnan, PhD, Johns Hopkins School of Medicine
Sujatha Kannan, MD, Johns Hopkins School of Medicine
Progesterone has been studied extensively in preclinical models of adult traumatic brain injury (TBI), and has advanced to clinical trials in adults with moderate to severe TBI. However, there are no preclinical studies in pediatric TBI models investigating progesterone for neuroprotection.
Immature male and female rats underwent controlled cortical impact (CCI) to the left parietal cortex at 2 different developmental ages (postnatal day, PND 16–17 and 22–23). Rats received either progesterone (10mg/kg) at 1h ( i.p.) and 6h (s.c.) after TBI or vehicle (22.5% cyclohexdrin), and were compared to naïve, age-matched littermates receiving vehicle. At 24h after CCI, brain mitochondria were isolated from the hemisphere ipsilateral to injury. Active (State 3) and resting (State 4) mitochondrial respiration were measured in the presence of pyruvate and malate. Mitochondrial respiratory control ratio (RCR, State III/State IV) was determined. A separate group of male rats (PND 17) were studied for histologic analysis, and received progesterone or vehicle every 24h (s.c.) for 7 days. In this group, sections were stained with cresyl violet for measurement of lesion volume, and Iba-1 for evaluation of microglial activation.
In PND 17 male rats, TBI significantly reduced mitochondrial RCR by ∼25% ipsilateral to injury, and post-injury progesterone treatment preserved mitochondrial RCR (p<0.05 TBI+PROG vs TBI+vehicle). This improvement of RCR was predominantly through significant decreases in State 4 respiratory rate with progesterone. In female rats (PND 17 & 23) and PND 23 male rats, there was a trend toward preserved mitochondrial RCR (p=0.1). Lesion volume was reduced in progesterone treated rats at 7d after CCI. Activated microglia were seen throughout the brain following TBI. Progesterone treatment decreased the density of Iba-1 stained microglia in several brain regions studied. Furthermore, progesterone treatment appeared to change the morphology of microglia from an amoeboid “activated’ form to a more ramified “quiescent” form.
Progesterone prevents mitochondrial dysfunction early after pediatric TBI in immature rats, and reduces lesion volume at 7d. Progesterone also appears to reduce post-traumatic neuroinflammation in the developing brain. Future studies will be directed at additional developmental ages (PND 10–15), and correlation with neurologic outcome testing. Comparisons across developmental ages and between male and female animals could provide valuable information for planning future clinical trials of progesterone treatment in children with TBI.
Department of Anesthesiology and Critical Care Medicine, Johns Hopkins School of Medicine
neuroendocrinology, progesterone, mitochondria, pediatric
Footnotes
Abstract Author Index
Numbers refer to abstract numbers, not page numbers.