Journal of Neurotrauma

Striking gender differences have been reported in the pathophysiology and outcome of acute neurological injury. Greater neuroprotection in females versus males may be due, in part, to direct and indirect sex hormone-mediated antioxidant mechanisms. Progesterone administration decreases brain levels of F₂-isoprostane, a marker of lipid peroxidation, after experimental traumatic brain injury (TBI) in male rats, and estrogen is neuroprotective in experimental neurological injury. In this study, we evaluated the effect of gender on lipid peroxidation, as assessed by cerebrospinal fluid (CSF) levels of F₂-isoprostane, after severe TBI in humans. Lipid peroxidation was assessed in CSF from 68 adults enrolled in two randomized controlled trials evaluating the effect of therapeutic hypothermia after severe TBI (Glasgow coma scale [GCS] score ≤ 8). Patients treated with hypothermia (n = 41, 12 females, 29 males) were cooled to 32-33°C (within ∼6 h) for either 24 or 48 h and then re-warmed. F₂-isoprostane levels were assessed by ELISA in ventricular CSF samples (n = 199) on day 1, 2, and 3. The association between age, GCS score, time, gender, treatment, duration of treatment, core temperature at the time of CSF sampling, secondary hypoxemia, and CSF F₂-isoprostane level was assessed by multivariate and dichotomous analyses. F₂-isoprostane was ∼2-fold higher in males than females (145.8 ± 39.6 versus 75.4 ± 16.6 pg/mL, day 1 p = 0.018). An effect of time after injury (p = 0.007) was reflected by a marked early peak in F₂-isoprostane (day 1). CSF F₂-isoprostane was also associated with hypoxemia (p = 0.04). Hypothermia tended to decrease F₂-isoprostane levels only in males on d1 after TBI. To our knowledge, this is the first study showing gender differences in lipid peroxidation after clinical TBI. Lipid peroxidation occurs early after severe TBI in adults and is more prominent in males vs females. These results established that gender is an important consideration in clinical trial design, particularly in the case of antioxidant strategies.

The role of reactive oxygen-induced oxidative damage to lipids (i.e., lipid peroxidation, LP) and proteins has been strongly supported in previous work. Most notably, a number of free radical scavengers and lipid antioxidants have been demonstated to be neuroprotective in traumatic brain injury (TBI) models. However, the specific sources of reactive oxygen species (ROS), the time course of oxidative damage and its relationship to post-traumatic neurodegeneration in the injured brain have been incompletely defined. The present study was directed at an investigation of the role of the ROS, peroxynitrite (PON), in the acute pathophysiology of TBI and its temporal relationship to neurodegeneration in the context of the mouse model of diffuse head injury model. Male CF-1 mice were subjected to a moderately severe head injury and assessed at 1-, 3-, 6-, 12-, 24-, 48-, 72, 96- and 120-h post-injury for neurodegeneration using quantitative image analysis of silver staining and semi-quantitative analysis of PON-mediated oxidative damage to proteins (3-nitrotyrosine, 3-NT) and lipids (4-hydroxynonenal, 4-HNE). Significant evidence of silver staining was not apparent until 24-h post-injury, with peak staining seen between 72- and 120-h. This time-course of neurodegeneration was preceded by intense immunostaining for 3-NT and 4-HNE, which occurred within the first hour post-injury. The time course and staining pattern for 3-NT and 4-HNE were similar, with the highest staining intensity noted within the first 48-h in areas surrounding trauma-induced contusions. In the case of 3-NT, neuronal perikarya and processes and microvessels displayed staining. The temporal and spatial coincidence of protein nitration and LP damage suggests that PON is involved in both. However, lipid-peroxidative (4-HNE) immunoreactivity was broader and more diffuse than 3-NT, suggesting that other reactive oxygen mechanisms, such as iron-dependent LP, may also contribute to the more widespread 4-HNE immunoreactivity. This indicates that optimal pharmacological inhibition of post-traumatic oxidative damage in TBI may need to combine two functionalities: one to scavenge PON or PON-derived radicals, and the second to inhibit LP caused by multiple ROS species.

Statins administered postischemia promote functional improvement in rats, independent of their capability to lower cholesterol. We therefore tested the effect of statin treatment on traumatic brain injury(TBI) in rats. Atorvastatin was orally administered (1 mg/kg/day) to Wistar rats starting 1 day after TBI for 7 consecutive days. Control animals received saline. Modified Neurological Severity Scores and Corner tests were utilized to evaluate functional response to treatment. Bromodeoxyuridine (BrdU, 100 mg/kg) was also intraperitoneally injected daily for 14 consecutive days to label the newly generated endothelial cells. Rats were sacrificed at day 14 after TBI, and the brain samples were processed for immunohistochemical staining. Atorvastatin administration after brain injury significantly reduced the neurological functional deficits, increased neuronal survival and synaptogenesis in the boundary zone of the lesion and in the CA3 regions of the hippocampus, and induced angiogenesis in these regions. The results suggest that atorvastatin may provide beneficial effects in experimental TBI.

This study was designed to investigate the effects of intravenous administration of marrow stromal cells (MSCs) on the expression of growth factors in rat brain after traumatic brain injury (TBI). The fate of transplanted MSCs and expression of growth factors was examined by immunohistochemistry. In addition, the level of growth factors was measured quantitatively using enzyme linked immunosorbent assay (ELISA). Growth factors that were studied included nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and basic fibroblast growth factor (bFGF). For immunohistochemical studies, 12 male Wistar rats were subjected to TBI and then divided into three groups with the first group receiving no treatment, the second group receiving saline (placebo) and the third group receiving MSCs intravenously 1 day after TBI. The neurological function of rats was studied by using Rotarod motor test and modified neurological severity scores. The animals were sacrificed 15 days after TBI and brain sections stained by immunohistochemistry to study the distribution of MSCs as well as expression of growth factors NGF, BDNF, and bFGF. For quantitative analysis, a second set of male Wistar rats (n = 18) was subjected to TBI and then injected with either saline (n = 9) or MSCs (n = 9) 1 day after injury. These rats were sacrificed on days 2, 5, and 8 after TBI and brain extracts used to measure NGF, BDNF, and bFGF. We found that after transplantation, MSCs preferentially migrated into the injured hemisphere and there was a statistically significant improvement in the functional outcome of MSC-treated rats compared to control rats. NGF, BDNF, and bFGF were expressed in the injured brain of both treated as well as control rats; however, quantitative ELISA studies showed that expression of NGF and BDNF was significantly increased (p < 0.05) in the treated group. This study shows that intravenous administration of MSCs after TBI increases the expression of growth factors (NGF, BDNF), which possibly contributes to the improvement in functional outcome seen in these rats.

Cerebral contusions are one the most frequent traumatic lesions and the most common indication for secondary surgical decompression. The purpose of this study was to investigate the physiology of perilesional secondary brain damage and evaluate the value of hyperbaric oxygen therapy (HBOT) in the treatment of these lesions. Five groups of five Sprague-Dawley rats each were submitted to dynamic cortical deformation(DCD) induced by negative pressure applied to the cortex. Cerebral lesions produced by DCD at the vacuum site proved to be reproducible. The study protocol entailed the following: (1) DCD alone, (2) DCD and HBOT, (3) DCD and post-operative hypoxia and HBOT, (4) DCD, post-operative hypoxia and HBOT, and (5) DCD and normobaric hyperoxia. Animals were sacrificed after 4 days. Histological sections showed localized gross tissue loss in the cortex at injury site, along with hemorrhage. In all cases, the severity of secondary brain damage was assessed by counting the number of terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) and caspase 3-positive cells in successive perilesional layers, each 0.5 mm thick. Perilesional TUNEL positive cells suggested the involvement of apoptosis in group 1 (12.24% of positive cells in layer 1). These findings were significantly enhanced by post-operative hypoxia (31.75%, p < 0.001). HBOT significantly reduced the severity and extent of secondary brain damage expressed by the number of TUNEL positive cells in each layer and the volume of the lesion (4.7% and 9% of TUNEL positive cells in layer 1 in groups 2 and 4 respectively, p < 0.0001 and p < 0.003). Normobaric hyperoxia also proved to be beneficial although in a lesser extent. This study demonstrates that the vacuum model of brain injury is a reproducible model of cerebral contusion. The current findings also suggest that HBOT may limit the growth of cerebral contusions and justify further experimental studies.

Chondrodystrophoid breeds of dog are prone to explosive herniation of mineralized disc material into the thoracolumbar spinal canal. The resulting acute spinal cord injury may respresent an excellent spontaneous model of acute traumatic spinal cord injury. The aims of this study were to quantify the recovery of dogs following acute disc herniations, to evaluate external factors that influence recovery, and to identify a group of dogs suitable for use in clinical trials on neuroprotective drugs. The gait of 88 dogs with thoracolumbar disc herniations was scored at the time of injury and 2, 4, and 12 weeks after surgical decompression. Dogs were placed into four groups dependent on the severity of presenting signs; dogs in group 1 had the most severe injury severity. Group 1 dogs showed a variable but incomplete recovery by 12 weeks. Dogs in groups 2 and 3 recovered uniformly but more completely, while dogs in group 4 made a rapid and excellent recovery and were deemed unsuitable for clinical trials. Combining dogs in groups 1, 2 and 3 produced a population of dogs with incomplete recovery by 12 weeks. Power analysis revealed that 87 such dogs would be needed per treatment group to detect a 20% change in function with a power of 95%. The number needed reduced drastically to 19 by eliminating dogs in group 1, but this produced less room for functional improvement. External factors did not appear to influence outcome. We conclude that dogs with spontaneous disc herniations provide a useful model of acute spinal cord injury for clinical trials.

We use a new in vitro model to examine the effect of mechanical deformation on neurons. We examined acute changes in cytosolic calcium concentrations ([Ca²⁺]_i) caused by a rapid stretch of cultured hippocampal neurons, using mechanical loading conditions that mimic brain deformations during trauma. We found that stretch-injury of neurons induces a strain-dependent increase in [Ca²⁺]_i. Remarkably, the extent of this calcium response exceeded the levels initiated by chemical toxicity with NMDA (100 μM) or glutamate (5 mM) exposure. Propidium iodide labeling at 24 h following stretch showed neuronal death occurred only at the most severe level of mechanical injury. Although NMDA-induced toxicity could be inhibited in calcium free media or by treatment with MK-801, stretch-induced neuronal death was not similarly reduced with either treatment. Unexpectedly, reduction of the acute stretch-induced calcium transient with calcium-free media or MK-801 resulted in an increase in neuronal death at lower stretch levels. These data suggest that mechanical stretch can initiate calcium influx in hippocampal neurons, but substantially modulating the early calcium flux from the extracellular space or through the NMDA channel does not provide an effective means for improving neuronal survival.

Following transection of the optic nerve (ON) in the adult rat, 85% of axotomized retinal ganglion cells (RGCs) undergo degeneration within 14 days. Here, we examined the effects of various anesthetic and analgesic compounds on the number of RGCs surviving ON lesion. Five different protocols for rodent anesthesia were used (A, chloral hydrate; B, chloral hydrate/carprofen; C, chloral hydrate/buprenorphine; D, ketamine/xylazine; E, fentanyl/medetomidin/midazolam), and the numbers of RGCs surviving 14 days after ON axotomy were compared to evaluate if the agents used may affect numbers of surviving RGCs. In many laboratories, rodent ON surgery is performed with chloral hydrate anesthesia, and this condition was used as baseline, with 343.7 ± 29.1 RGCs/mm² surviving after 14 days. The addition of carprofen to chloral hydrate did not affect RGC numbers (382.7 ± 15.2 RGCs/mm²; n.s.), while chloral hydrate with buprenorphine (421.1 ± 25.1 RGCs/mm²; p < 0.05), ketamine and xylazine (403.6 ± 36.1 RGCs/mm²; p < 0.05), or fentanyl with medetomidine and midazolam (481.3 ± 10.4 RGCs/mm²; p < 0.05) all increased RGC survival. In a second series of experiments, ON axotomized rats were treated with an adenoviral vector expressing GDNF (Ad.GDNF) that rescues injured RGCs, to study if the anesthetics (A, B, E; see above) would influence the degree of RGC neuroprotection afforded by GDNF. Intravitreal injection of Ad.GDNF at a low titre rescued approximately 10% of RGCs that would have degenerated without treatment using either of the three different anesthesia protocols, yet GDNF did not exert synergistic neuroprotection with any of the anesthetics tested. Our results indicate that in combination carprofen and chloral hydrate, while affording safe and reliable anesthesia and analgesia for rat ON surgery, does not affect the numbers of surviving RGCs. Therefore, data obtained with this combination may be related to experimental data obtained previously with only chloral hydrate anesthesia. All other protocols afforded some degree of RGC neuroprotection that may be utilized for experimental therapies of neurodegeneration, yet needs to be taken into careful consideration when mechanisms of neurodegeneration or approaches towards neuroprotection of RGCs are examined.

Using a novel in vivo model for cerebral ischemia produced by short-lasting compression of a well-defined brain area of sensorimotor cortex we studied neuroprotective effects of the NMDA NR2B subunit selective antagonist, CP-101,606, in Sprague-Dawley rats. Cortical compression for 30 min produced a consistent and highly reproducible functional impairment, that is paresis of contralateral hind and fore limbs. The neurological deficit was accompanied by marked brain damage in cerebral cortex, hippocampus and thalamus as identified by Fluoro-Jade, a marker of general neuronal cell death. Using a daily performed beam walking test it was shown that untreated animals recovered from their functional impairment within 5-7 days following surgery. Intravenous administration of increasing doses (1, 5, 10, 20 mg/kg) of the NMDA NR2B subunit receptor specific antagonist, CP-101,606, dose-dependently improved the rate of functional recovery and protected against the ischemic brain damage in cerebral cortex, hippocampus, and thalamus as identified 2 days after the ischemic insult. Based upon these results, we conclude that NMDA NR2B receptor subunits represent potential targets to reduce not only the functional deficits, but also neuronal death in cortex and several midbrain regions produced by moderate, transient, cerebral ischemia.

Functional recovery after peripheral nerve injury depends on the amount as well as on the accuracy of reinnervation by regenerative axons. In this study, the rat sciatic nerve was subjected to crush injury or complete transection repaired by either (1) straight nerve suture, (2) crossed nerve suture of tibial and peroneal fascicles, or (3) silicone tubulization leaving a gap of 4 mm. The compound muscle action potentials (CMAP) of gastrocnemius, tibialis anterior and plantar muscles were recorded 90 days post operation to assess functional reinnervation and Fast Blue, Fluoro Gold and DiI were applied to the nerve branches projecting into these muscles to quantify morphological reinnervation. The CMAP amplitude achieved in gastrocnemius, tibialis anterior and plantar muscles was higher after nerve crush (86%, 82%, 65% of control) than after any surgical nerve repair (straight suture: 49%, 53%, 32%; crossed suture: 56%, 50%, 31%; silicone tube: 42%, 44%, 25%). The total number of labeled motoneurons, however, did not significantly differ between groups (control: 1238 ± 82, crush: 1048 ± 49, straight suture: 1175 ± 106, crossed suture: 1085 ± 84, silicone tube: 1250 ± 182). The volume occupied by labeled motoneurons within the spinal cord was larger after surgical nerve repair than in crush or normal control animals, and fewer neurons showed abnormal multiple projections after crush (2.5%) or straight suture (2.2%) than following crossed suture (5%) or silicone tube (6%). In conclusion, nerve repair with a silicone tube leaving a short gap does not increase accuracy of reinnervation.

Researchers studying the impact of treatments designed to facilitate recovery after neural injury face competing demands. On the one hand, because treatment effects often emerge slowly over days, and because researchers seek evidence of stable long-term effects, it is common practice to observe experimental subjects for many weeks after treatment. On the other hand, the cost of performing studies and the need to evaluate a multitude of alternative treatment procedures requires optimal efficiency, pushing researchers towards shorter test procedures. With these issues in mind, researchers have appeared to derive a test window based on previously published methodologies and inspection of their recovery curves, with testing terminated after the recovery curve reaches asymptote (approaches a slope of 0). An alternative procedure is introduced here that evaluates the stability of the data set over time. Using correlational techniques, researchers can determine whether (1) testing should be continued for additional days; or (2) equivalent statistical power can be achieved in fewer days. This provides a rational decision rule to help researchers balance competing demands. Applying these techniques to a procedure that evaluates the impact of acute treatments on recovery from spinal cord injury, it is shown that equal statistical power can be achieved in half the time, greatly increasing the efficiency with which alternative treatments can be evaluated.

Oligodendrocytes undergo apoptosis in the white matter tracts remote from the experimental cord injury, although its significance is not understood. Our ultrastructural study, however, enabled us to speculate on its neurobiological implications. The spinal cords of male Wistar rats (4 week-old) were transected at Th11 level. At 4, 5, and 7 days after surgery the animals were transcardially perfusion-fixed. The removed cord was embedded in epoxy resin and examined by electron microscopy. Post-embedding terminal deoxynucleotidyl transferase-mediated biotinylated deoxyuridine triphosphate nick end labeling (TUNEL) method was also performed. In the degenerative dorsal column above the transection, profiles of apoptotic oligodendrocytes were always found, embedded in a space formed by dilated degenerative myelin lamellae. Often, the dilated space in the myelin sheath lacked any apparent background proteinaceous matrix. In the electron microscopic TUNEL method, these apoptotic cells were electron dense in accordance with nuclear heterochromatinization. In conclusion, in the process of Wallerian degeneration, we observed the apoptosis of oligodendrocytes in a space formed by the split myelin sheath. These degenerative cells, which were enclosed in an ultrafiltrate-filled space formed by split myelin lamellae, were reminiscent of "anoikis."