The 23rd International Symposium on Cerebral Blood Flow,Metabolism and Function

ASTROCYTES PROVIDE A SIGNALING CONDUIT FOR UPSTREAM (PIAL ARTERIOLAR) DILATION DURING CORTICAL NEURONAL ACTIVATION IN THE RAT

Dale Pelligrino, Hao-Liang Xu

Neuroanesthesia Research Department, University of Illinois at Chicago, Chicago, IL, USA

BACKGROUND: The tight coupling between neuronal activity and blood flow is fundamental to brain function. Astrocytes play an important role in that process, via their capacity to “sense” neuronal activity and, in turn, transmit that information to parenchymal arterioles. In this study, we examined another aspect of astrocyte-related neurovascular coupling?the ability of astrocytes, in vivo, to act as a signaling conduit for the vitally important process of upstream vasodilation (represented by pial arterioles). Pial arterioles are segregated from the underlying cortical neuropil by the presence of the astrocytic processes comprising the glia limitans (GL). We, therefore, postulated that, irrespective of the strength of cortical neuronal activation, whether excessive (bicuculline-induced seizure [BIS]) or physiologic (sciatic nerve stimulation [SNS]), the associated pial arteriolar dilation involves an astrocytic and GL-related signaling conduit. To test this, we topically applied (via a closed cranial window) L-alpha-aminoadipic acid (L-AAA), an astroglial toxin which selectively damages the GL (without affecting vascular smooth muscle, endothelium, or neurons). Two other hypotheses were addressed: a) inter-astrocytic signaling is dependent upon the gap junctional protein, connexin-43 (Cx43); and b) signaling from the GL to pial arterioles involves a diffusible mediator.

METHODS: Pial arteriolar diameter increases during BIS or SNS, in anesthetized rats, were measured in the absence or presence of L-AAA or Cx43 blockade (via topically-applied gap-27 peptide) and during increased rates of cortical suffusion (tests influence of diffusible mediator). In BIS experiments, EEG power increases were not altered by any experimental intervention, indicating that the changes in pial arteriolar responses were not due to changes in the level of cortical activation.

RESULTS: The predominance of an astrocytic signaling conduit was indicated by data showing that pial arteriolar dilating responses to both neuronal activation paradigms were completely blocked following selective disruption of the superficial GL (fig. 1). Gap-27 elicited a 60–75% reduction in BIS and SNS-induced dilations (fig. 2). Increasing the aCSF suffusion rate led to 40–50% reductions in BIS and SNS-related dilations (fig. 3).

CONCLUSIONS: Astrocytes and the GL play a key role in signaling upstream pial arteriolar dilation during both moderate and excessive cortical neuronal activation. The signaling process may involve Cx43-related gap junctions and/or hemichannels and a diffusible mediator arising from the GL.

CHARACTERIZATION OF A NOVEL POTASSIUM CHANNEL, TWIK-2, CLONED FROM THE RAT MIDDLE CEREBRAL ARTERY

Eric Lloyd, Sean Marrelli, Robert Bryan

Department of Anesthesiology, Baylor College of Medicine, Houston, TX, USA

Background and aims: recently a new family of ion channels, termed two-pore domain potassium channels (K2P), has been discovered and cloned. Although little is known about the expression and function of these channels in the vascular system, we have demonstrated that TWIK-2, a member of the K2P family, is highly expressed in rat middle cerebral artery (MCA) (AJP 291:H770, 2006). The aim of this study was to characterize TWIK-2 and determine if its activation could dilate rat MCAs.

Methods: TWIK-2 was cloned from rat MCA and expressed heterologously (COS and CHO cells) as a fusion protein to green fluorescent protein (GFP) for characterization. Electrical properties of TWIK-2 were determined using patch clamp methods in the whole cell configuration. MCAs were removed from rat brain, mounted on micropipettes in a vessel bath, pressurized, and perfused. MCAs were magnified and the diameter recorded using image analysis software.

Results: Direct fluorescence analysis by deconvolution microscopy in transiently transfected COS cells revealed that GFP-rTWIK-2 localized to the plasma membrane in a punctate pattern by 48 hours after transfection. Stable transfectants of GFP-rTWIK-2 were selected in CHO cells and analyzed by whole-cell electrophysiology. Whole cell currents were 30-fold greater than the nontransfected control group, were non-rectifying in physiological K+, but were inwardly rectifying in symmetrical K+. The observed reversal potentials in different concentrations of extracellular K+ were similar to the theoretical reversal potentials. In physiological K+, membrane potential was −81 +/− 0.5 mv (n=13) for transfected and −51 +/− 2.5 mv (n=16) for control cells. 1 mM BaCl2 inhibited currents by 90 +/− 1% with a calculated IC50 of 86 uM. The TWIK-2 currents were minimally affected by 10 mM tetraethylammonium chloride (TEA), 3 mM 4-aminopyridine (4AP), and 10 uM glibenclimide (glib). 100 uM arachidonic acid increased the currents by 88 +/− 15% (n=6). In isolated pressurized rat MCA, arachidonic acid elicited robust dilations that were not blocked by 10 mM TEA, 3 mM 4AP, or 10 uM glib.

Conclusions: TWIK-2 is a novel potassium channel expressed in cerebral arteries. It is characterized by a non-rectifying current in physiological K+ gradients and is resistant to most potassium channel inhibitors. We conclude that TWIK-2 is an important regulator of vascular tone. (RO1 NS 46666, NS 038660, and AHA Bugher Award 027011N)

TISSUE-TYPE PLASMINOGEN ACTIVATOR CONTRIBUTES TO NEUROVASCULAR COUPLING

Laibaik Park¹, Gang Wang¹, Josef Anrather¹, Sidney Strickland², Costantino Iadecola¹

¹Division of Neurobiology, Weill Medical College of Cornell University, New York, NY, USA, ²Laboratory of Neurobiology & Genetics, The Rockefeller University, New York, NY, USA

NOC/OFQ ANTAGONIST POST INJURY TREATMENT PROTECTS AGAINST CEREBRAL CIRCULATORY COLLAPSE AND HISTOPATHOLOGY AFTER COMBINED PROLONGED HYPOTENSION AND BRAIN INJURY

William M. Armstead, Xiao-Han Chen, Douglas H. Smith

University of Pennsylvania, Philadelphia, PA, USA

Introduction: In humans, traumatic brain injury (TBI) is often accompanied by hypovolemia and hypotension resulting from damage to other organ systems, which increases morbidity and mortality. Basic science animal studies have observed disturbed cerebral autoregulation during acute hypotension after TBI but few have investigated the effects of prolonged hypotension. The opioid NOC/oFQ contributes to impaired cerebral hemodynamics after fluid percussion brain injury (FPI) in the pig. In the context of the neurovascular unit concept, this study addressed the hypothesis that NOC/oFQ contributes to the transition from reversible cerebrovascular dysregulation to the cerebral circulatory collapse and histopathology that occurs during combined prolonged hypotension and FPI.

Methods: FPI (2 atm) was produced in pigs (4 weeks old) equipped with a closed cranial window and CBF determined by radiolabeled microspheres. Acute (10 min) hypotension (45% decrease in arterial blood pressure) was induced at 60 and 240 min post FPI, prolonged hypotension (225 min) was induced at 15 min post FPI, while the NOC/oFG antagonist [F/G]NOC/oFQ(1-13)NH2 (NA) (1 mg/kg iv) was administered 30 min post FPI.

Results: CSF NOC/oFQ was increased significantly more by prolonged hypotension and FPI compared to acute hypotension and FPI or normotensive FPI. CBF and pial artery diameter decreased after normotensive FPI. Pial artery dilation (PAD) was blunted at 60,120, and 240 min during acute hypotension and at 60 min post FPI for prolonged hypotension. In contrast, at 120 and 240 min post FPI and prolonged hypotension, PAD during hypotension was reversed to vasoconstriction. In the pons, CBF was decreased at 60, but not at 120, 240 min post FPI during acute hypotension, but decreased at 60, 120, and 240 min post FPI during prolonged hypotension. In the cerebrum, CBF was decreased more during prolonged compared to acute hypotension at all timepoints post FPI. Hypercapnia (PaCO2 55 or 75 mm Hg) and topical acetylcholine (10-7, 10-5 M) induced PAD was blunted significantly greater during prolonged hypotension compared to acute hypotension or normotension at 60 min post FPI, but was reversed to vasoconstriction during prolonged hypotension at 120 min post FPI. Post FPI administration of the NOC/oFQ antagonist NA blocked the reversal of PAD to vasocontriction, the CBF dysregulation in the cerebrum and pons, and the reversal of PAD to vasoconstriction with hypercapnia and acetylcholine administration during prolonged hypotension at 120, 240 min post FPI. H + E staining of the CA3 hippocampus demonstrated normal neurons in sham control animals, considerable degenerating neurons in FPI and prolonged hypotension animals, but no degenerating neurons in injured animals post treated with NA. In the cortex, significant hemorrhage with edema and high density of degenerating neurons were noted in untreated FPI and prolonged hypotension animals, but small hemorrhage and edema with few degenerating neurons in injured animals that received NA.

Conclusions: These data indicate that NOC/oFQ contributes to the transition from reversible dysregulation to circulatory collapse and histopathology after combined prolonged hypotension and FPI. These observations suggest a novel pharmacotherapeutic intervention for the treatment of CNS disorders resulting from complex hypovolemic TBI origin.

IN VIVO MULTIPHOTON MICROSCOPY OF OLFACTORY GLOMERULI REVEALS ASTROCYTIC CONTROL OF NEUROVASCULAR COUPLING VIA MULTIPLE GLUTAMATERGIC PATHWAYS

Gabor C. Petzold, Dinu F. Albeanu, Tomokazu Sato, Venkatesh N. Murthy

Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA, USA

Background and aims: Glutamate is centrally involved in the signaling pathways underlying neurovascular coupling. However, the relevance of presynaptic release, postsynaptic actions and extrasynaptic activity of glutamate have remained controversial. Recent studies have suggested that astrocytic metabotropic glutamate receptors (mGluR) are involved in stimulation-evoked changes of cerebral blood flow (CBF), but the physiological relevance of this and other astrocytic pathways have remained undefined. The aim of this study was to noninvasively investigate these pathways in vivo using physiological stimuli.

Methods: We imaged mice expressing a reporter of neurotransmitter release called synaptopHluorin (spH) in synapses made by the sensory neurons of the olfactory system in vivo using two-photon microscopy. With this probe, presynaptic glutamate release by olfactory sensory synapses in specialized regions called glomeruli can be visualized following physiological odor stimulation (Fig. 1a). Simultaneously, we monitored erythrocytic velocity, flow and linear density within glomeruli by injecting Texas Red dextran into the tail vein and obtaining line scans along the axis of glomerular capillaries (Fig. 1b-d).

Results: We found that CBF increases in glomeruli activated by physiological stimulation with odorants. The CBF increase showed a wide dynamic range (~5–70%), and was highly correlated with the amount of presynaptic glutamate release (Fig. 1e). Surprisingly, odor-evoked CBF increase was not affected by the suppression of postsynaptic activity of the principal output neurons using antagonists of AMPA and NMDA receptors. Glutamate is likely to mediate neurovascular coupling by acting on astrocytes since blockers of astrocytic mGluR-5 reduced the odor-induced CBF increase. Cyclooxygenase (COX) inhibitors partially prevented the effects of mGluR-5 antagonists, indicating the involvement of the previously described phospholipase-A signaling pathway. Unexpectedly, the CBF increase was also inhibited by blockade of glutamate uptake by the broad-spectrum glutamate transport inhibitor DL-threo-β-Benzyloxyaspartic acid and by the astrocyte-specific glutamate transport inhibitor dihydrokainate. This inhibition remained unchanged after pretreatment with COX inhibitors, indicating that astrocytic mGluR and glutamate uptake affect neurovascular coupling via separate pathways.

Conclusions: We have shown that neurovascular coupling in olfactory glomeruli involves presynaptic release of glutamate, but not its postsynaptic actions. Our data also indicate that glutamate increases CBF by acting on astrocytes via separate pathways: stimulation of mGluR and subsequent COX activation, and uptake by astrocytic glutamate transporters.

THE RESPONSE OF THE CEREBRAL MICROVASCULATURE TO REGIONAL CHANGES IN CEREBRAL BLOOD FLOW

Mark G.F. Catherall, Alexander B. Rowley, Stephen John Payne

Department of Engineering Science, University of Oxford, Oxford, UK

EFFECTS OF EC-IC BYPASS SURGERY ON CEREBRAL PERFUSION IN MOYAMOYA DISEASE AND NON-MOYAMOYA ISCHEMIC STROKE EVALUATED BY OPTICAL TECHNIQUES

Tatsuya Hoshino, Noriaki Yokose, Takashi Awano, Shin Nakamura, Norio Fujiwara, Yoshihiro Murata, Tsuneo Kano, Kaoru Sakatani, Yoichi Katayama

Department of Neurological Surgery, Nihon University School of Medicine, Itabashi-Ku, Tokyo, Japan

Background: EC-IC (extracranial-intracranial) bypass such as superficial temporal artery-middle cerebral artery (STA-MCA) anastomosis has become a standard surgical therapeutic option in moyamoya disease (MD) to prevent recurrent ischemic events. In addition, EC-IC bypass surgery has been performed to prevent stroke in patients with transient ischemic attack caused by hemodynamic compromise. Several studies have demonstrated beneficial effects of bypass surgery on cerebral hemodynamic status in both MD and non-moyamoya ischemic strokes (non-MD). However, little is yet known regarding the difference between these groups of patients in the extent to which the bypass contributes to maintaining adequate cerebral blood oxygenation (CBO), or the temporal changes after surgery. In the present study, we evaluated the CBO changes induced by bypass blood flow in patients with MD and non-MD during the perioperative periods employing optical spectroscopy including visual light spectroscopy (VLS) and near infrared spectroscopy (NIRS).

Methods: We developed a VLS monitoring system which permits continuous monitoring of CBO changes during surgery. Employing the VLS, we evaluated the CBO changes in the MCA territory on the lesion side in 13 patients who underwent STA-MCA anastomosis, including 5 MD and 8 non-MD patients. In addition, employing NIRS, we evaluated the effects of STA blood flow on the CBO at one week after surgery. We measured the changes in CBO during compression of the anastomosed STA; when decreases in oxyhemoglobin concentrations were observed, the bypass was considered to maintain CBO in the MCA territory because decreases in oxyhemoglobin indicate cerebral ischemic changes.

Results: In 4 MD patients and one non-MD patient, the STA blood flow increased the oxyhemoglobin and cortical oxygen saturation (CoSO2), indicating that the bypass supplied blood flow to the ischemic brain; the CBO changes were observed more frequently in MD than in non-MD (p<0.02). The preanastomosis CoSO2 (65.4±5.4%) in MD was significantly lower than that (72.8±7.6%) in non-MD (p<0.05). Postoperative NIRS demonstrated that the bypass began to supply blood flow to the brain in 5 non-MD patients whose bypass did not supply blood flow during surgery.

Conclusion: Although MD has vessels of small diameter as compared to non-MD, the bypass begins to supply blood flow to the ischemic brain earlier in MD than in non-MD after anastomosis. The fact that the CoSO2 in MD was lower than that in non-MD suggested that the perfusion pressure in MD was lower than that in non-MD, and this might account for the difference in the bypass blood supply after anastomosis between MD and non-MD. Our data suggest that, even if the bypass does not supply blood to the brain during surgery in non-MD, the bypass blood flow gradually increases after surgery. The optical spectroscopy is considered useful for evaluating bypass function and facilitates safe and accurate bypass surgery.

ABNORMAL DIAZOXIDE INDUCED VASODILATION IN CEREBRAL ARTERIES FROM ZUCKER OBESE RATS WITH INSULIN RESISTANCE

Prasad Katakam, James A. Snipes, Anna Busija, David W. Busija

Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston Salem, NC, USA

Diazoxide, a mitochondrial KATP channel activator, elicits arterial vasodilation by increasing the generation of mitochondria-derived reactive oxygen species (ROS) that stimulate Ca2+ sparks and Ca2+ activated K+ channels (KCa). We evaluated the diazoxide induced vasodilation in serotonin preconstricted isolated cerebral arteries from 11-13 weeks old Zucker obese (ZO) and lean (ZL) rats by videomicroscopy. The ZO rats were insulin resistant with hyperinsulinemia but were euglycemic and normotensive. Diazoxide induced vasodilation (% maximal relaxation) was diminished in ZO (5.4±5%, p<0.05) compared to ZL (43.2±14%). 3-nitropropionic acid, an activator of mitochondrial ROS generation via inhibition of succinate dehydrogenase, has no affect on vasodilation in ZL (42.7±17%, p=NS), but enhances diazoxide induced vasodilation in ZO (71.2±11, p<0.05). Diazoxide induced vasodilation was partially inhibited by iberiotoxin in ZL (23.6±7%, p<0.05) while it was totally abolished in ZO (1.1±4%). ZO arteries exhibit reduced pressure induced constriction compared to ZL (14.4±4% in ZO versus 28.4±3% in ZL, p<0.05). However, vasodilation to nitroprusside was similar in both ZO and ZL (80.4±6% in ZO versus 87.2±4% in ZO, p=NS). Thus, cerebral arteries from ZO exhibit abnormal diazoxide induced vasodilation mediated by reduced generation of mitochondrial ROS and/or Ca2+ sparks or impaired coupling of Ca2+ sparks with KCa channels.

PRESSURE AND FLOW REGULATION WITHIN THE BRAIN -FACTS, MYTHS AND MISSING LINKS

Mare Czosnyka, Ivan Timofeev, Piotr Smielewski, Andrea Lavinio, Peter Hutchinson, Peter Kirkpatrick, Jonn Pickard

Academic Neurosurgery, Addenbrooke's Hospital, Cambridge, UK

Objective: Continuous assessment of cerebral autoregulation using time- or frequency- domain correlation, phase shift- and transmission between slow waves of MCA blood flow velocity and cerebral perfusion pressure (CPP) or arterial pressure (AP) have been introduced a decade ago. We intend to review clinical applications of two indices describing cerebral autoregulation (1) and pressure reactivity (2) following head injury.

Material and Method: We studied nearly 300 ventilated head injured patients. TCD was recorded digitally along with ICP and AP daily within baseline periods (no interventions) from 10 minutes to 2 hours. ICP and AP were recorded and analysed continuously during whole period of intensive care. Moving correlation coefficient (3 min window), called Mx index, was calculated between low-pass filtered (0.05 Hz) signals of FV and CPP or AP. Similar index, expressing Pressure Reactivity (PRx), was calculated as moving correlation coefficient between AP and ICP.

Results:Facts: Mx correlated strongly with PRx index. Both indices correlated significantly with static autoregulation. Mx showed good agreement with transient hyperaemic response test, leg-cuff test, phase-shift methods, and CO2 reactivity. Poor autoregulation and pressure-reactivity proved to be independent predictor of fatal outcome following head injury. Both indices were significantly disturbed by intracranial hypertension (ICP>25 mmHg). Autoregulation was worse for low CPP (CPP<55 Hg) and for too high CPP(>105 mmHg) than for normal CPP (between 60 and 100 mmHg). Asymmetry of Mx (left-right) was associated with asymmetry of CT scanautoregulation was worse at the side of contusion and at the side of brain expansion causing midline shift. Both Mx and PRx after head injury worsens with age, probably contributing to worse outcome in elderly patients. Mx showed deterioration of cerebral autoregulation in large (>4 ug/ml) plasma concentration of propofol. PRx correlated with a state of low CBF and CMRO2 revealed using PET. PRx is more robust than Mx (no need of fixation of external probes) and seems to be helpful in optimization of CPP-oriented therapy.

Myths: It is probably not true that Mx in head injury patients can be calculated using only AP instead of CPP. Both indices are able to reflect transient changes of autoregulation, like during plateau waves of ICP. Indices can be quite noisy and decent time for averaging (30 minutes minimum) in individual cases is required to assess whether given patient at given time autoregulates or not. Thresholds for functional and disturbed autoregulation dramatically depend on arterial tension of CO2 ?therefore comparison between patients cannot be performed without comparing their PaCO2 concentrations.

Missing Links: Discrepancies between PRx and Mx are still poorly understood. Whether restoration of failing autoregulation can improve outcome, remains to be demonstrated prospectively. Can ‘optimal CPP’ therapy be understood as a consensus between CPP-protocol and Lund-concept ? Can autoregulation after head injury be improved with statins or EPO, like in SAH?

Conclusion: PRx and Mx can be used in many clinical scenarios for continuous monitoring of cerebrovascular regulation, being possibly helpful in predicting outcome and optimizing therapeutical strategies.

NEURONAL REPLACEMENT AND ELECTROPHYSIOLOGICAL CHANGES OF THE NEWLY GENERATED NEURONS IN THE ADULT STRIATUM AFTER ISCHEMIC BRAIN INJURY

Gakushi Yoshikawa^1,4, Toshihiko Momiyama^2,5, Soichi Oya^1,4, Keisuke Takai^1,4, Junichi Tanaka^1,4, Nobuhito Saito¹, Takaaki Kirino^3,4, Nobutaka Kawahara^1,4

¹Department of Neurosurgery, The University of Tokyo Graduate School of Medicine, Tokyo, Japan, ²Laboratory of Cerebral Structure, National Institute for Physiological Sciences, Okazaki, Japan, ³Research Institute International Medical Center of Japan, Tokyo, Japan, ⁴SORST, Japan Science and Technology Corporation (JST), Osaka, Japan, ⁵CREST, Japan Science and Technology Corporation (JST), Osaka, Japan

<Introduction> Recent findings have provided evidence that neurons can be generated in damaged regions in which neurogenesis does not occur normally, e.g., hippocampal CA1, striatum, and neocortex. In the present study, we elucidated whether neural progenitors in the subventricular zone (SVZ) are stimulated in situ by the treatment of growth factors, migrate into the damaged striatum, and replace lost neurons following global ischemia. We also followed the time course of the number of these replaced neurons and their neural phenotypes, and investigated the electrophysiological properties using whole-cell patch-clamp recording to characterize the functional development of newly generated neurons in the adult brain. <Materials and methods> Male Wister rats were subjected to 9 minutes global ischemia under the condition of 4-vessel occlusion and hypotension, and the EGF and FGF-2 were infused into the bilateral ventricles for 7 days from Day 2. For labeling the proliferating cells, BrdU was injected intraperitoneally. We also injected GFP-expressing recombinant retroviruses into the ventricles of growth-factor treated animals at Day 2, to determine the migration of proliferating progenitors in the SVZ. <Results> At Day 2 when the treatment of growth factors had not started, severe neuronal loss in the dorso-lateral striatum was observed, where no more than 2.3% of neurons remained. Immunostaining for BrdU and neural progenitor markers such as Pax 6, Mash 1 at Day 7 showed a marked increase of the progenitor cells in the SVZ in the treated animals. At Day 42, a significant increase in the number of striatal neurons, stained with both cresyl-violet and NeuN, was observed in the treated animals (approximately 15% of normal controls, p< 0.05). We observed neurons co-expressing BrdU and NeuN in the lesioned striatum, most of which also expressed phenotypic markers of striatal medium spiny neurons (DARPP-32). Furthermore, we detected that the GFP+−infected cells in the striatum were positively stained by NeuN, indicating that these neurons were derived from dormant neural progenitors residing in the SVZ. These immunohistochemical characteristics of neurons did not change afterwards till Day 70-84, suggesting that the morphological differentiation had matured by 6 weeks. Electrophysiologically, on the other hand, newly generated striatal neurons recorded at 6 weeks showed immature patterns of membrane and firing properties, whereas surviving neurons in the medial striatum of the same specimen showed the mature patterns. Developmental changes in electrophysiological properties were also observed in newly generated striatal neurons until 20 weeks or later. <Conclusion> The treatment of growth factors potentiated proliferation of neural progenitors in the SVZ and enhanced the neuronal replacement in the damaged striatum of the adult brain. Functional development required much longer term than morphological, immunohistochemical maturation. Our present data show that recruitment of progenitors in the SVZ for the repair of ischemic injury may provide a new therapeutic approach for the treatment of stroke.

VASCULAR ENDOTHELIAL GROWTH FACTOR (VEGF) INCREASES PRODUCTION, MIGRATION AND SURVIVAL OF NEURAL STEM CELLS DURING HYPOXIA AND ISCHEMIA

Hugo H. Marti¹, Katja Bieber¹, Yaoming Wang², Kunlin Jin², David A. Greenberg²

¹Institute of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany, ²Buck Institute for Age Research, Novato, CA, USA

VCAM-1 TARGETED INTRACAROTID DELIVERY OF NEURAL STEM CELLS IMPROVES FUNCTIONAL OUTCOME AFTER EXPERIMENTAL STROKE

Raphael Guzman, Alejandro De Los Angeles, Samuel Cheshier, Jason Liauw, Stanley Huang, Raymond Choi, Tonya Bliss, Gary Steinberg

Department of Neurosurgery, Stanford University, Stanford, CA, USA

Background: In acute injury of the brain, transplantation of neural progenitor cells (NPC) is aimed at replacing damaged neural cells. Intravascular delivery of NPC after stroke has been limited by the low efficiency of transendothelial migration. VCAM-1 is one of the endothelial adhesion molecules known to be upregulated early after stroke and is responsible for the firm adhesion of CD49d (VLA-4) expressing inflammatory cells. We hypothesized that selecting CD49d positive NPC would improve their homing to the injured brain and improve functional outcome after stroke.

Methods: Mouse NPC's were analyzed for the expression of the surface integrin CD49d by FACS analysis, immunohistochemistry and static adhesion assay. On the day of intraarterial injection, a CD49d enriched (>90%) and depleted (<1%) NPC population was obtained by magnetic bead enrichement and double FACS sorting. Bl6 mice (20-25g, n=18) were subjected to a left side hypoxia ischemia (20min 8% O2 + temporary right common carotid artery (CCA) occlusion). All mice were tested by rotarod for behavioral deficits and assigned to receive 3×105 CD49d enriched (n=6) or CD49d depleted (n=6) stem cells or vehicle (n=6) injection into the right CCA 48 hours after stroke. Animals were tested by rotarod at 7 and 14 days after stroke and subsequently sacrificed and the brains processed for immunohistochemistry.

Results: FACS analysis revealed approximately 25% CD49d positive NPC at passage six in vitro. Analysis of CD49d cluster like immunoreactivity by confocal microscopy showed a 3-fold (p=0.03) decrease of the integrin expression between passage 6 and 8. In the static adhesion assay NPC adhered to VCAM-1 in a dose dependent manner and adhesion was significantly greater than in uncoated control wells (p<0.005). Two weeks after intracarotid injection significantly more NPC were found in the cortex, hippocampus and subventricular zone adjacent to the stroke in animals receiving CD49d positive as compared to CD49d negative NPC's (1066/cm2 vs. 583/cm2, p<0.05). Behavioral recovery was significantly better for the CD49d positive cells injected group as compared to CD49d negative and vehicle injected animals (p=0.031). The CD49d negative group had a better recovery than the vehicle injected group (ns.).

Conclusions: We show that enrichment of NPC by FACS sorting for the surface integrin CD49d and intracarotid delivery promotes cell homing to the area of stroke in mice and improves sensorimotor recovery. We present two concepts which potentially improve the efficiency of intravascular delivery of stem cells for stroke treatment.

SUPPRESSED BRAIN ANGIOGENESIS DURING MODERATE HYPOXIA IN CYCLOOXYGENASE-2 DEFICIENT MICE

Joseph C. LaManna, Xiaoyan Sun, Mary P. Modesitt, Constantinos P. Tsipis

Department of Anatomy, Case Western Reserve University, Cleveland, OH, USA

Background and Aims: Exposure to chronic moderate hypoxia induces angiogenesis in tissues through cooperation of the HIF-1 transcription factor-vascular endothelial growth factor and the Cyclooxygenase-2(COX-2)-angiopoietin 2 signaling pathways(1, 2, 3). This study demonstrated the significance of COX-2 in hypoxia-induced brain angiogenesis.

Methods: male, 3-month old COX-2 knockout (KO, -/−), wild type (WT) and heterozygote (HET) mice were exposed to hypobaric hypoxia (0.4 ATM, ~8% oxygen) for 3 weeks. Their littermates were kept normoxic in the same location. Mice were fixed for immunohistochemical staining of GLUT-1, an indicator of brain capillaries. Capillary density (N/mm2) was estimated through the full depth of the frontal cortex from 5 sections (4 µm).

Results: Under normoxic conditions there was no difference in capillary density of KO mice compared to their WT and HET littermates. As expected, hypoxic exposure significantly (*) increased the capillary density in the WT and HET mice, compared to that of their normoxic littermates (491 ± 45, n = 4 vs. 416 ± 39, n = 3, mean ± SD, and 496 ± 41, n = 5 vs. 393 ± 16, n = 3, respectively). However, the capillary density in COX-2 -/− mice was unchanged (424 ± 24 vs. 401 ± 13, n = 3 each).

Conclusions: Our data suggest that the COX-2 signaling pathway is required for cerebrovascular adaptation to prolonged moderate hypoxia. However, COX-2 is apparently not necessary for establishing capillary density under normoxic conditions.

ISOLATION AND CHARACTERIZATION OF INJURY-INDUCED NEURAL STEM/PROGENITOR CELLS FROM POST-INFARCT AREA IN MICE

Takayuki Nakagomi¹, Akihiko Taguchi², Orie Saino¹, Masatoshi Fujikawa¹, Yoshihiro Fujimori¹, Tomoyuki Nishizaki³, Tomohiro Matsuyama¹

¹Institute for Advanced Medical Science, Hyogo College of Medicine, Nishinomiya, Hyogo, Japan, Hyogo, Japan, ²Department of Cerebrovascular Disease, National Cardiovascular Center, Osaka, Japan, Osaka, Japan, ³Department of Physiology, Hyogo College of Medicine, Nishinomiya, Hyogo, Japan

INHIBITING DEGRADATION OF HYPOXIA-INDUCIBLE FACTOR INCREASES NEURAL STEM CELLS IN MOUSE FOCAL ISCHEMIA

Tsuyoshi Uesugi¹, Shunya Takizawa¹, Yuko Izuhara², Yuko Morita¹, Shigeharu Takagi¹, Toshio Miyata²

¹Division of Neurology, Department of Internal Medicine, Tokai University School of Medicine, Kanagawa, Japan, ²Institute of Medical Science, Tokai University School of Medicine, Kanagawa, Japan

Background and purpose: Hypoxia-inducible factor (HIF) is a regulatory factor of transcription that is induced in hypoxic cells. It activates target genes involved in compensation for ischemia, including erythropoietin, VEGF, and glut-1. We examined the cerebroprotective effects of TM6008, a novel inhibitor of the HIF-degrading enzyme prolylhydroxylase (PHD), on infarct volume and intrinsic neural stem cells in mice with permanent focal ischemia.

Methods: The experiments were performed on C57BL/6J mice weighing 20 grams. Briefly, animals were anesthetized using 2.0% isoflurane anesthesia with a mixture of 30% oxygen and 70% nitrous oxide, and the left middle cerebral artery was permanently occluded by cauterizing. Mice were divided into two groups: vehicle (0.2 ml/day of milk alone) and TM6008-treated (200 mg/kg/day of TM6008 + 0.2 ml/day of milk) groups. Mice in both groups were given 50 mg/kg/day of BrdU for 2 days before sacrifice on day 7 or day 14 after occlusion. In all animals, the rotor rod test and blood sampling were performed. Coronal brain sections were fixed with 4% paraformaldehyde, and stained with hematoxylin and eosin. Infarct volume was measured, and immunohistochemical reactions for HIF-1 and BrdU were performed.

Results: No significant difference of rotor rod test results was seen between the TM6008-treated and vehicle groups at any time point. At day 7 after occlusion, there was no statistically significant difference of infarct volume, HIF-1 or BrdU-positive cells between the two groups. At day 14, however, TM6008 significantly reduced brain atrophy and the number of BrdU-positive cells, compared with the vehicle group.

Conclusions: Inhibition of PHD by TM6008 increases intrinsic neural stem cells in the infarcted hemisphere of mice with permanent focal ischemia, accompanied with a reduction of infarct volume. Increasing HIF-1 by blocking its degradation is a novel strategy to treat cerebral infarction, at least in the subacute phase, by promoting neurogenesis.

CIRCULATING ENDOTHELIAL PROGENITOR CELLS AS THE PATHOGENETIC MARKER OF MOYAMOYA DISEASE

Kon Chu, Keun-Hwa Jung, Soon-Tae Lee, Eun-Cheol Song, Dong-In Sinn, Jeong-Min Kim, Hee-Kwon Park, Manho Kim, Jae-Kyu Roh

Stroke and Neural Stem Cell Laboratory, Department of Neurology, Stem Cell Research Center, Seoul National University Hospital, Seoul, South Korea

Backgrounds: Endothelial progenitor cells (EPCs) derived from bone marrow are believed to support the integrity of the vascular endothelium. It has reported that EPC mobilization increases following cerebral ischemia, whereas some data have suggested that the number and function of EPCs correlate inversely with cardiovascular risk factors, outcome and the presence of stroke. Moyamoya disease is an unusual from of chronic cerebrovascular occlusive disease followed by the formation of characteristically abnormal vessel. Objectives: In this study, we attempted to examine the EPC number and function in moyamoya disease. Methods: The stage of steno-occlusive lesions on angiogram was determined according to the modified staging system. MMD patients (stage II-IV) undergoing chronic stroke was included. EPCs were isolated from the peripheral venous blood of 20 patients with moyamoya disease and 20 age-matched healthy volunteers. EPC was counted as colony-forming units (CFU) at the 7-day culture, and the yield of outgrowth endothelial cells were measured during the 3 months of culture. EPC function was measured by Matrigel plate assay.

Results: The number of CFU after 7-day culture was significantly reduced in patients with moyamoya disease versus control subjects (12.34±8.66 versus 70.6±15.35, p < 0.0001). During the long-term culture, OECs was isolated from 5% of healthy volunteer, but more efficiently from 50% of patients with moyamoya disease. Tube formation was reduced in MMD, more in advanced stage. The CFU and OECs yield was not correlated with the angiographic stage.

Conclusions: The presence of moyamoya disease was associated with a reduced number of CFU together with an impairment of functional activity, but higher yield of OECs. These parameters of circulating EPCs might represent the pathogenetic process of Moyamoya disease.

HYPOXIA-INDUCED DAMAGE AND REPAIR IN ADULT MOUSE BRAIN

Marc Uy, Aaron Hirko, Renee Dallasen, Nicole Brandon, Yan Xu

Department of Anesthesiology, University of Pittsburgh, Pittsburgh, PA, USA

Background and aims: Hypoxia is associated with a number of clinical conditions such as asthma, altitude sickness, carbon monoxide poisoning, and stroke. It has been postulated that mild hypoxia alone does not cause neuronal death, and hypoxia preconditioning in rodents has been shown to significantly reduce injury in focal and global ischemia. A possible protective mechanism through preconditioning is the pre-activation of neuronal stem cells in the adult brain tissue. We report here an investigation of the long-term physiological responses to a 30-min episode of respiratory hypoxia with 5% O2 in mice, with the ultimate goal of understanding the mechanisms of brain injury and recovery after hypoxia.

Method: Male CD1 mice, weighing 20-30 g, were anesthetized with pentobarbital. Animals were transorally intubated and mechanically ventilated. Hypoxia was induced by adjusting pO2 to 38 mmHg while the animals were paralyzed with pancuronium to prevent spontaneous gasping. Control animals were administered a normal pO2 =138 mmHg for 30 minutes. Body temperature was held constant at 36.8 ± 0.4°C. After the hypoxia, ventilation with room air (pO2 =138 mmHg) was continued until spontaneous breathing was restored. The heart rate, femoral artery pulse distention, blood O2 saturation (spO2), breath distention, and rectal temperature were recorded continuously before, during, and after the hypoxia. Diffusion-weighted MRI, apparent diffusion coefficient (ADC) maps, and angiography were acquired at 14.1 T. Final histology included Nissl staining and NeuN, GFAP, and vimentin immunohistostaining. The percent of damaged neurons and the density of GFAP+ and vimentin+ cells were analyzed.

Results and conclusions: Normal spO2 with room air ventilation was 96 ± 3%. During the 30-min hypoxia, spO2 steadily declined from 82 ± 6% (averaged the first 5 min) to 65 ± 7 % (averaged for the last 5 min). While most animals had normal ADC in the hippocampal CA1 region, some did show a significant decrease in ADC values the second day after the hypoxia, followed by an overshoot in the subsequent days. This is a typical response to ischemia explained by the initial cytosolic edema followed by the interstitial edema in response to injuries. The animals with MRI abnormalities also showed visible neuronal damage in Nissl stained sections. There is a rather strong correlation between the percent of damaged neurons in the CA1/CA3 regions and the density of vimentin+ cells in the same and adjacent regions (r2 =0.90). This result might indicate an influx of neural and astrocytic precursors in response to the cellular damage. At the short and intermediate time points examined after hypoxia, the vimentin+ cells have astrocytic morphology and are NeuN^—. At a later stage, these cells are NeuN+ and resemble neurons in structure and orientation. Thus, our results suggest that similar to developing brain, the radial glia in the adult brain from the subventricular zone and dentate gyrus also migrate and transdifferentiate into neurons in response to injury after prolonged hypoxia and reoxygenation. (Funded by NIH R01NS/HL036124).

TRANSPLANTATION OF MOUSE AND PRIMATE ES CELLS INTO ISCHEMIC MOUSE BRAIN

Yasushi Takagi, Motoaki Fujimoto, Makoto Hayase, Junya Hayashi, Jun Takahashi, Nobuo Hashimoto, Kazuhiko Nozaki

Department of Neurosurgery, Kyoto University, Graduate School of Medicine, Kyoto, Japan

PET IMAGING OF AMYLOID DEPOSITIONS IN MCI PATIENTS USING PIB

Anton Forsberg¹, Henry Engler⁴, Gunnar Blomquist⁵, Ove Almkvist^1,2, Goran Hagman², Anders Wall³, Anna Ringheim³, Bengt Langstrom^3,6, Agneta Nordberg^1,2

¹Neurotec Department, Karolinska Institutet, Stockholm, Sweden, ²Department of Geriatric Medicine, Karolinska University Hospital Huddinge, Stockholm, Sweden, ³Uppsala Imanet, GE Healthcare, Uppsala, Sweden, ⁴Department of Nuclear Medicine, Uppsala University Hospital, Uppsala, Sweden, ⁵Department of Oncology, Radiology and Clinical Immunology, Uppsala University, Uppsala, Sweden, ⁶Department of Organic Chemistry, Uppsala University, Uppsala, Sweden

NORMAL DATABASE OF DOPAMINERGIC NEUROTRANSMISSION SYSTEM IN LIVING HUMAN BRAIN MEASURED BY PET

Hiroshi Ito, Hidehiko Takahashi, Ryosuke Arakawa, Harumasa Takano, Tetsuya Suhara

Clinical Neuroimaging Team, Molecular Neuroimaging Group, Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan

Introduction: The central dopaminergic system is of interest in the pathophysiology of schizophrenia and other neurodegenerative diseases. Both pre and postsynaptic dopaminergic functions can be estimated by PET with several radiotracers. However, the integrated database of both pre and postsynaptic dopaminergic functions including receptors, transporter, and endogenous neurotransmitter synthesis has not been reported. In the present study, we have built the normal database of pre and postsynaptic dopaminergic functions in the living human brain using PET.

Methods: To measure the dopamine D1 receptor binding, striatal and extrastriatal dopamine D2 receptor binding, dopamine transporter binding, and endogenous dopamine synthesis, PET scans were performed on healthy men after intravenous injection of [C-11]SCH23390, [C-11]raclopride, [C-11]FLB457, [C-11]PE2I, or L-[C-11]DOPA, respectively (10 healthy men for each tracer). The binding potential (BP) was calculated by the reference tissue model method on a voxel by voxel basis with use of the cerebellum as a reference region. For L-[C-11]DOPA study, the dopamine synthesis index (I) was calculated on a voxel by voxel basis with use of the occipital cortex as a reference region from the ratio of area under the time-activity curve of brain regions to a reference region (I = ratio ? 1). All PET images were anatomically standardized using SPM2, and database was built for each radiotracer. Gray matter images were segmented and extracted from all anatomically standardized MR images using the SPM2 and used for partial volume correction.

Results: These databases allow to compare regional distributions of the striatal and extrastriatal dopamine D1 and D2 receptor binding, dopamine transporter binding, and endogenous dopamine synthesis. The BP and I values of each region-of-interest are given in table 1. In the striatum, greatest bindings to dopamine D1, D2 receptors and transporters and greatest dopamine synthesis were observed. For the limbic system, relatively high bindings of dopamine D1 and D2 receptors and relatively high dopamine synthesis were observed in the amygdala and anterior cingulate. Among the neocortical regions, highest binding to dopamine D2 receptors was observed in the temporal cortex, while D1 receptor binding was uniform. The binding to dopamine transporter was very low in the neocortical regions. These distributions were in good agreement with those in the human post mortem studies. The partial volume correction did not change regional distribution of pre and postsynaptic dopaminergic functions in cerebral cortices.

Conclusions: Although a limited spatial resolution of PET scanner might hamper observation of pre and postsynaptic dopaminergic functions in detail, such database must be useful to understand physiology of neurotransmission in the living human brain. These database can also be used to investigate regional abnormality of dopaminergic neurotransmission in neuropsychiatric diseases.

HYPEROXIA AND HYPERCAPNEA INCREASE THE CONCENTRATION OF OXIDISED CYTOCHROME C OXIDASE IN THE HEALTHY HUMAN ADULT BRAIN

Ilias Tachtsidis¹, Martin M. Tisdall², Terence S. Leung¹, Caroline A. Pritchard², Martin Smith², Clare E. Elwell¹

¹Department of Medical Physics and Bioengineering, University College London, London, UK, ²Department of Neuroanaesthesia and Neurocritical Care, The National Hospital for Neurology and Neurosurgery, London, UK

PITFALLS IN NIR-SPECTROSCOPY AND FUNCTIONAL MRI FOR MEASURING BRAIN OXIDATIVE METABOLISM WITH SPECIAL REMARKS ON A-V SHUNTS

Minoru Tomita¹, Yutaka Tomita², Takashi Osada¹, Miyuki Unekawa¹, Haruki Toriumi¹, Norihiro Suzuki¹

¹Department of Neurology, Keio University School of Medicine, Tokyo, Japan, ²Department of Neurology, Institute of Brain and Blood Vessels, Mihara Memorial Hospital, Isesaki, Gunma, Japan

The interplay between function and oxidative metabolism of the brain has been extensively investigated by spectroscopy (NIRS) and fMRI of metabolic changes responding to somatosensory stimuli. However, concerning optical spectroscopy, we have pointed out that the Lambert-Beer law does not hold during brain activation, since the extinction coefficient of the blood alters due to flow-dependent changes in light scattering by RBC (flow effect: Tomita et al. (2006) NeuroImage 33,1; ibid., 13). In addition, the extinction coefficient per se varies with RBC oxygenation: the concave shape of RBC tends to be spherical with a diameter of ca. 7.2 micrometer, whereas the same RBC become flat and flaccid with a diameter of ca. 7.7 micrometer when deoxygenated. Light scattering at such membranes is significantly affected. The inconstant extinction coefficient jeopardizes quantification of oxy-/deoxyhemoglobin in the tissue. In addition, the Bohr effect in arterial blood produced by respiratory changes in human subjects, e.g., breath-holding or hyperventilation with powerful somatosensory stimuli, shifts the oxygen dissociation curve of hemoglobin, changing cerebral blood oxygenation without any changes in brain metabolism. This paper examines the effect of arterio-venous shunting blood, which could augment dissociation between blood oxygenation and brain oxygenation in both NIRS and fMRI, since the channels transport oxygenated blood from the arteries to the veins without gas exchange. We used 50 rats with a closed cranial window and measured velocities of FITC-labeled RBC in the 50 micrometer deep cortical microvasculature with high speed laser scanning confocal fluorescence microscopy. We found labeled RBC velocity varied widely in capillaries selected with the aid of FITC-dextran staining. The RBC velocity distribution revealed a group of outlier higher velocities (p<0.01) from the normal group which must belong to a different category of vessels: the RBC velocities were extraordinarily high, nevertheless the channels had the same diameter as capillaries (Unekawa, Brain 07). Dynamic observation in vivo revealed the channels run straight and are richly invested with astrocyte endfeet, which were stained in vivo with sulforhodamine 101. The amount of shunted blood without gas exchange through these channels was estimated from the velocity, number of capillaries, and theoretical extraction as 22% of the total perfusion, which agrees with our previous observation of 20% based on the compartmental analyses of inert gas clearance from human brain. In the human study we found that the shunted blood held in reserve was recruited with CO2 inhalation. With CO2, cerebral tissues became equally perfused by blood. When the brain is activated, CO2 in the tissue increases, and the direct shunting blood decreases. NIRS and fMRI falsely estimate the oxidative metabolism to be increased. In conclusion, the shunting blood constitutes a serious pitfall for quantifying brain oxidative metabolism. However, this paper does not intend to deny the possible clinical usefulness of NIRS and fMRI as tools to investigate brain function, since their signals are surely biological, but rather, to emphasize that the interpretation must be cautious.

ACUTE FUNCTIONAL RECOVERY OF CEREBRAL BLOOD FLOW FOLLOWING FOREBRAIN ISCHEMIA IN THE RAT

Chao Zhou¹, Tomakazu Shimazu², Turgut Durduran^1,3, Daniel Y. Kimberg², Guoqiang Yu¹, Xiao-Han Chen⁴, John A. Detre³, Arjun G. Yodh¹, Joel H. Greenberg²

¹Department of Physics and Astronomy, University of Pennsylvania, Philadelphia, PA, USA, ²Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA, ³Department of Radiology, University of Pennsylvania, Philadelphia, Philadelphia, PA, USA, ⁴Department of Neurosurgery, University of Pennsylvania, Philadelphia, Philadelphia, PA, USA

Background: Following complete cerebral ischemia the post-ischemic blood flow response to functional activation is severely attenuated for several hours. However, little is know about the spatial and temporal extent of the blood flow response in the acute post-ischemic period following incomplete cerebral ischemia.

Methods: We used laser speckle flowmetry, which provides two-dimensional high spatial and temporal resolution blood flow images, to examine the response of relative cerebral blood flow (rCBF) to controlled vibrissae stimulation during the first few hours following temporary incomplete ischemia. Male Sprague-Dawley rats (n=10) were prepared under halothane anesthesia and then switched to α-chloralose. The bone overlying the right somatosensory cortex was thinned for the laser speckle imaging. Prior to ischemia the left vibrissae (B1-3, C1-3, D1-3) were stimulated for 3 sec at 5 Hz every minute over a 10-minute period while the rCBF data was collected every 200 msec. Fifteen minutes of transient forebrain ischemia was induced by bilateral carotid arterial occlusion along with controlled hypotension (mean arterial blood pressure lowered to 45-50 mmHg by slowly withdrawing blood from a femoral arterial catheter). This reduced cerebral blood flow to 6.6±1.8% (mean±SE) of the pre-ischemic level. The shed blood was re-infused, the carotid ties were released, and the rCBF response to vibrissae stimulation was repeated at 20, 30, 60, 120, and 180 mins after the end of the ischemia. rCBF images were analyzed using a modified general linear model (GLM) with VoxBo (www.voxbo.org). Statistical maps from the analysis were used to determine the spatial extents of the activation at different time points with a fixed Bonferroni-corrected threshold (pth=0.01).

Results: The activation area was significantly reduced 20 mins into reperfusion compared to prior to ischemia (49±12%, p<0.01), and recovered to 77±17% at 2 hours (p=0.36, Figure.1.A). The maximum rCBF response in the activation area determined from the statistical analysis did not change significantly up to three hours post ischemia [ΔrCBF=10.6±0.8% (pre-ischemia); 9.0±0.3% (20 min, p=0.06); 11.1±0.7% (3 hrs, p=0.62)]. However, the time when rCBF reached maximum was significantly delayed from 2.4±0.2 sec prior to ischemia to 3.6±0.1 sec at 20 mins into reperfusion (p<0.001); the delay was reduced to 3.0±0.2 sec after 3 hours, which was still significantly greater than that observed prior to the insult (p<0.05, Figure.1.B).

Conclusions: We have incorporated the GLM method into laser speckle imaging, and showed that although the magnitude of the response to functional stimulation was preserved in the early hours following incomplete cerebral ischemia, the area of activation and the temporal response to activation were significantly altered.

IMAGING BETA AMYLOID DEPOSITION IN VIVO: QUANTITATIVE COMPARISON OF [18F]FDDNP AND [11C]PIB

Nelleke Tolboom^1,2, Maqsood Yaqub¹, Wiesje van der Flier², Ronald Boellaard¹, Gert Luurtsema¹, Albert D. Windhorst¹, Frederik Barkhof³, Philip Scheltens², Adriaan A. Lammertsma¹, Bart N.M. van Berckel¹

¹Department of Nuclear Medicine and PET Research, VU University Medical Centre, Amsterdam, The Netherlands, ²Department of Neurology and Alzheimer Centre, VU University Medical Centre, Amsterdam, The Netherlands, ³Department of Radiology, VU University Medical Centre, Amsterdam, The Netherlands

EFFECTS OF EARLY LIFE STRESS ON 5-HT1A RECEPTOR AVAILABILITY IN TWO YEARS OLD RHESUS MONKEYS MEASURED BY [18F]FPWAY AND PET

Simona Spinelli¹, Richard E. Carson², Cristina S. Barr¹, Stephen J. Suomi³, Markus Heilig¹, J. Dee Higley¹, Lixin Lang⁴, Elliot Stein⁵, Svetlana I. Chefer⁵

¹Laboratory of Clinical and Translational Studies, National Institute On Alcohol Abuse and Alcoholism, NIH, Bethesda, MD, USA, ²Yale PET Center, New Haven, CT, USA, ³Laboratory of Comparative Ethology, National Institute of Child and Human Development, NIH, Bethesda, MD, USA, ⁴PET Radiochemistry Group, National Institute of Biomedical Imaging and Bioengineering, NIH, Bethesda, MD, USA, ⁵Neuroimaging Research Branch, National Insitute On Drug Abuse, NIH, Baltimore, MD, USA

Traumatic childhood experiences are associated with increased risk for developing depression, anxiety, posttraumatic stress and substance abuse disorders. Rearing infant macaques with same-aged peers in the absence of adults, has been established as a model of early adversity. This condition has been shown to induce high levels of anxiety-like behavior in monkeys, particularly among females. Chronic early life stress exposures affect 5-HT1A receptor expression and function in various animal models (Leonard, 2005), and dysregulation of the serotonin system is believed to increase vulnerability to developing mood and anxiety disorders. As such, we hypothesized that nursery raised (peer-reared, PR) macaques would exhibit differences in [18F]FPWAY binding potential (BP) relative to those raised with their mothers (mother-reared, MR) and that these changes would not only appear early in life but may be particularly marked among females. To assess the effects of rearing condition on the development of the serotonin system, we compared MR (1 male and 5 female) to PR (5 male and 5 female) rhesus monkeys, assessed at 24-29 months, an age corresponding to 6–7 years in humans. PET data were acquired on an ECAT HR+ scanner over 90-120 min, under isoflurane anesthesia using [18F]FPWAY (Lang et al.,1999;2000) and a bolus-plus-infusion technique (Kbol=60 min) in order to attain constant radioactivity level in all brain regions. PET images were co-registered to individual MRIs using PMOD. A standard region of interest (ROI) template, including prefrontal cortex (PFC), anterior and medial cingulate cortex (ACC and MCC), hippocampus, amygdala, midbrain (including dorsal raphe nucleus) and temporal cortex was constructed from the MRI images for each monkey. BP* was calculated in the ROIs under equilibrium assumption as (CROI?CCER)/CCER. Using sex as a covariate, ANOVA was performed, revealing a main effect of rearing condition (P<0.001). BP* values were elevated in the PFC and hippocampus (increase of 40% and 10%, respectively, P<0.05) in the PR group (BP*=1.85+0.52 in PFC and 3.12+0.35 in hippocampus, mean+SD) compared to MR group (BP*=1.34+0.36 in PFC and 2.83+0.24 in hippocampus, mean+SD). Comparison of only MR and PR female monkeys revealed a 70% (P<0.05) increase in PFC (BP*=2.12+0.58 and 1.24+0.29 in PR and MR group, respectively) and a 20% (P<0.05) elevation in BP* values in ACC in the PR group (BP*=2.94+0.38 and 2.44+0.27 in PR and MR group, respectively). Previous human studies show that depressed subjects have decreased BP (Sargent et al.,2000). Increased BP* values in our study may have been driven mostly by female monkeys whereas the majority of the human studies used male subjects. Studies in humans suggest higher levels of specific distribution volumes in women compared to men (Parsey et al.,2002). In conclusion, these findings suggest that early stress affects the density of 5-HT1A receptors available for radioligand binding during early macaque development. This effect is particularly evident in some sub-regions of the frontal cortex. Elevated BP* values could be related to either an up-regulation of 5-HT1A receptors or to a decrease in tonic serotonin levels.

Support Contributed By: NIDA-, NIAAA-, NICHD- and NIBIB-IRP

IMPROVEMENT OF CEREBRAL OXYGEN METABOLISM AFTER BYPASS SURGERY IN PEDIATRIC MOYAMOYA DISEASE

Satoshi Kuroda¹, Tatsuya Ishikawa¹, Tohru Shiga², Naoki Nakayama¹, Kiyohiro Houkin³, Nagara Tamaki², Yoshinobu Iwasaki¹

¹Department of Neurosurgery, Hokkaido University Hospital, Sapporo, Japan, ²Department of Nuclear Medicine, Hokkaido University Hospital, Sapporo, Japan, ³Department of Neurosurgery, Sapporo Medical University, Sapporo, Japan

Background and aims- Previous studies have clarified that bypass surgery improves cerebral hemodynamics and prevents further ischemic stroke in patients with moyamoya disease. However, the effect of bypass surgery on cerebral metabolism is still unclear. In this study, therefore, the authors evaluated it in both pediatric and adult patients with moyamoya disease, using [15] O-gas positron emission tomography (PET).

Methods- This study included 20 patients who were admitted to our hospital due to moyamoya disease. There were 8 children and 12 adults. All pediatric patients experienced TIA or cerebral infarction. Of 12 adult patients, 11 developed TIA or cerebral infarction and another did intracranial bleeding. Using [15] O-gas PET, cerebral hemodynamics and oxygen metabolism were measured in all patients before and 3 to 4 months after STA-MCA anastomosis and pial synangiosis that widely cover the cerebral hemisphere. By placing 10-mm diameter ROIs on the frontal or temporal cortex, the effects of bypass surgery on cerebral hemodynamics and metabolism were quantitatively analyzed.

Results- In pediatric patients (n=8), cerebral blood flow (CBF) significantly increased from 37.3±9.4 mL/min/100g to 42.8±6.1 mL/min/100g after surgery (P<0.01). Cerebral blood volume (CBV) significantly decreased from 6.2±2.5 mL/100g to 3.6±0.8 mL/100g after surgery (P<0.05). Cerebral metabolic rate for oxygen (CMRO2) significantly increased from 2.84±0.63 mL/min/100g to 3.56±0.53 mL/min/100g after surgery (P<0.01). Pre- and postoperative values of oxygen extraction fraction (OEF) did not differ; 0.43±0.10 and 0.47±0.03, respectively. In adult patients (n=12), CBF significantly increased from 29.4±7.6 mL/min/100g to 37.7±7.8 mL/min/100g after surgery (P<0.01). CBV significantly decreased from 5.3±2.1 mL/100g to 3.6±1.2 mL/100g (P<0.05). OEF significantly decreased from 0.46±0.11 to 0.41±0.05 after surgery (P<0.05). However, CMRO2 did not significantly change after surgery. Thus, pre- and postoperative CMRO2 were 2.37±0.61 mL/min/100g and 2.57±0.54 mL/min/100g, respectively.

Conclusions- These findings strongly suggest that cerebral oxygen metabolism may be depressed in response to prolonged ischemia in pediatric, but not adult, patients and that effective bypass surgery may have the potential to improve cerebral oxygen metabolism by improving cerebral hemodynamics in pediatric patients with moyamoya disease. Improvement of cerebral oxygen metabolism may contribute to not only prevent ischemic stroke but also improve their intellectual outcome.

VALIDITY AND VALUE OF QUANTITATIVE SPECT RECONSTRUCTION IN A MULTI-CENTER CLINICAL STUDY USING I-123 RADIOPHARMACEUTICALS

Iida Hidehiro, Watabe Hiroshi, Nakazawa Mayumi, Hayashida Kohei

Quantitative SPECT Working Group, Japan

Introduction: SPECT has a potential to provide parametric functional images in a “quantitative” manner, as has been widely done with PET in vivo, using several tracers typically labeled with I-123. Due to availability of equipments, as well as well-established delivery/transportation of radio-ligands, SPECT has an advantage in applying it to large-scale clinical studies. It has however been considered that accuracy and inter-institutional reproducibility of SPECT are limited for quantitative assessment of functional parametric images, largely attributed to a lack of general consensus of reconstruction procedures (attenuation/scatter correction) as well as, in case of I-123, significant amount of penetration through collimator from high-energy photons. We have recently developed a novel method to reconstruct SPECT images from existing projection data including appropriate corrections for scatter and attenuation in the object, head contour detection, as well as for collimator-dependent penetration. This study was aimed at evaluating validity and impact of our approach for Multicenter clinical study. We particularly investigated consistency and reproducibility of quantitative values for I-123, at different SPECT systems from different manufacturers.

Materials and methods: The Windows program can handle original projection data to analyze and/or DICOM format obtained from 5 manufactures with essentially 7 different operation systems. Twenty-five institutions provided a series of phantom data for 28 collimators according to a given protocol. Nine are symmetric fanbeam, and rest parallel beam collimators. Experiments were done for a radioactivitycalibrated point source of I-123 in air to determine the absolute sensitivity of the system and to determine the penetration factor for each collimator. Additional experiments were done using a uniform cylindrical phantom filled with the radioactivity mentioned above, and single/multiple rod sources placed in cylindrical water. The scatter/penetration correction was based on the previously proposed transmission dependent convolution subtraction method, and the attenuation correction was implemented during the ordered-subset EM reconstruction. Dynamic SPECT studies were carried out in 9 institutions on 64 patients using I-123 iodo-amphetamine (IMP) to assess quantitative CBF images at rest and during Diamox challenge. Inter- and intra-institutional reproducibility was then evaluated.

Results and discussion: Scatter fraction was consistent among different collimators from different manufacturers as predicted, but the penetration varied largely dependent on the collimator design and manufacturers. Our approach successfully compensated for the penetration and the scatter as well as the attenuation, as demonstrated in reconstructed images of the water-filled single/multiple rod source phantoms, except for one fanbeam collimator. Parallel beam collimator tends to have small penetration than fanbeam. Absolute concentration of the uniform cylindrical phantom was also consistent among institutions, although it was significantly smaller than the true value (0.88+/−0.05). Clinical data also demonstrated that CBF values obtained with I-123 IMP at rest and during Diamox were reproducible among institutions. Normal CBF values obtained in 2 institutions showed also no significant difference.

Conclusion: These data suggested that QSPECT could be a useful tool for quantitative mapping using I-123 compounds, allowing a large scale clinical evaluation using SPECT.

MICROGLIA PROVIDE NEUROPROTECTION AND ENGAGED IN CLOSE CELL-CELL CONTACT WITH NEURONS AFTER ISCHEMIA

Jens Neumann¹, Matthias Gunzer², Oliver Ullrich³, Klaus Dinkel⁴, Klaus G. Reyman^1,4

¹Project Group Neuropharmacology, Leibniz Institute for Neurobiology, Magdeburg, Germany, ²German Research Centre for Biotechnology, Immunodynamics, Braunschweig, Germany, ³Institute of Immunology, University Hospital Magdeburg, Magdeburg, Germany, ⁴Institute for Applied Neuroscience (FAN GGmbH), Magdeburg, Germany

Many neurological insults are accompanied by a marked acute inflammatory reaction, involving the activation of microglia. Using a model of exogenous application of fluorescence-labeled BV2 microglia in pathophysiologically relevant concentrations onto organotypic hippocampal slice injury. Neuronal cell death after oxygen-glucose deprivation (OGD) was determined by propidium iodide incorporation and Nissl staining. Migration and interaction with neurons were analyzed by time resolved 3-D two-photon microscopy. We show that microglia protect against OGD-induced neuronal damage and engage in close physical cell-cell contact with neurons in the damaged brain area. Neuroprotection and migration of microglia were not seen with integrin regulator CD11a-deficient microglia or HL-60 granulocytes. The induction of migration and neuron-microglia interaction deep inside the slice was markedly increased under OGD conditions. Lipopolysaccharide-prestimulated microglia failed to provide neuroprotection after OGD. Pharmacological interference with microglia function resulted in a reduced neuroprotection. Microglia proved to be neuroprotective even when applied up to 4 h after OGD, thus defining a ??protective time window.?? In acute injury such as trauma or stroke, appropriately activated microglia may primarily have a neuroprotective role. Anti-inflammatory treatment within the protective time window of microglia would therefore be counterintuitive.

OVEREXPRESSION OF STAT3 PROMOTES NEURONAL CELL DEATH THROUGH ACTIVATION OF MICROGLIA

Ryuji Hata, Pengxiang Zhu, Fang Cao, Masahiro Sakanaka

Department of Functional Histology, Ehime University Graduate School of Medicine, Toon, Ehime, Japan

CORTICAL SPREADING DEPRESSION AND PERI-INFARCT DEPOLARISATION IN HUMAN ISCHEMIC STROKE

Christian Dohmen^1,2, Oliver Sakowitz³, Martin Fabricius⁴, Bert Bosche⁵, Thomas Reithmeier⁵, Ralf-Ingo Ernestus⁵, Gerrit Brinker⁵, Jens Dreier⁶, Johannes Woitzik⁷, Antony J. Strong⁸, Rudolf Graf²

¹Department of Neurology, University of Cologne, Cologne, Germany, ²Max-Planck Institute for Neurological Research, Cologne, Germany, ³Department of Neurosurgery, University of Heidelberg, Heidelberg, Germany, ⁴Department of Clinical Neurophysiology, Glostrup Hospital, University of Copenhagen, Copenhagen, Denmark, ⁵Department of Neurosurgery, University of Cologne, Cologne, Germany, ⁶Department of Neurology, Charite University of Berlin, Berlin, Germany, ⁷Department of Neurosurgery, University of Mannheim, Mannheim, Germany, ⁸Department of Clinical Neurosciences, King's College London, London, UK

Objective: Cortical spreading depression (CSD) and peri-infarct depolarisation (PID) have been shown in various experimental models of stroke to cause infarct expansion and secondary neuronal damage. For many years it has been speculated that CSD/PID occur in stroke patients. Recently, CSD and PID have been documented in patients with traumatic brain injury and subarachnoid hemorrhage. In the present study, we report for the first time on CSD/PID in human ischemic stroke measured by subdural electrocorticography (ECoG) in patients with malignant MCA infarction.

Methods: Patients were recruited by centers of the Co-operative Study on Brain Injury Depolarisations (COSBID). The study was approved by the local ethics Committees. All patients suffered from complete or subtotal infarction of the MCA territory and underwent decompressive hemicraniectomy. During the operation, subdural electrode strips with 6 linear contacts were placed over the peri-infarct region of 16 patients. From these electrodes, 4 ECoG channels were aquired in a bipolar fashion; one electrode was used as ground. CSD was defined as gradually developing depression of ECoG activity with simultaneous slow potential changes (SPC) spreading between at least 2 channels. If no local ECoG activity occurred up to the time of the SPC, the event was defined as PID. If some electrodes recorded a local ECoG signal before the SPC, while others did not, the event was defined as CSD/PID.

Results: Monitoring was started between 24 and 109 h after stroke and performed for up to 120 h. The first patient had to be excluded from data analysis due to technical failure during the recording. From the following 15 patients, a total recording time of 1547 h with good data quality was scanned for depolarisation events. In 3 patients, no depolarisation events occurred. In these patients, the electrode strip was placed mainly over infarcted tissue i.e. hypodense tissue on computer tomography images, and no local ECoG activity occurred throughout the measurement. In the remaining 12 patients (80 %), a total of 164 episodes were recorded (127 CSD, 20 CSD/PID, 16 PID). Seizure activity was not apparent in any of the patients.

Conclusion: Similar to findings in experimental stroke models, CSD and PID are found with high incidence in patients with MCA infarction suggesting that this phenomenon is a relevant factor also in the pathophysiology of ischemic tissue damage in humans and thus a target for treatment development.

CEREBRAL BLOOD FLOW, OXYGEN CONSUMPTION AND GLUCOSE UTILIZATION IN HUMANS:A STEREOTAXIC CORRELATION STUDIED WITH PET AND SPM

Kazuhiko Hayashi, Shinichi Takeda, Keisuke Miyake, Kentaroh Fujii, Taku Fujimoto, Yasuhiro Ohsaki, Ashik Bin Ansar, Katsufumi Kajimoto, Yasuyuki Kimura, Hiroki Katoh, Makiko Tanaka, Eku Shimosegawa, Jun Hatazawa

Department of Nuclear Medicine and Tracer Kinetics, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

We studied local differences in the coupling among cerebral blood flow (CBF), oxygen consumption (CMRO2) and glucose utilization (CMRGlc) by means of statistical parametric mapping (SPM).

RECOMBINANT HUMAN C1-INHIBITOR PROTECTS AGAINST BRAIN ISCHEMIA/REPERFUSION INJURY WITH A WIDE TIME-WINDOW OF EFFICACY

Claudio Storini¹, Raffaella Gesuete¹, Emanuela Rossi², Alessandro Fantin¹, Luigi Bergamaschini², Bertjan Ziere³, Helene Vietsch³, Maria-Grazia De Simoni¹

¹Mario Negri Institute for Pharmacological Research, Milan, Italy, ²Geriatric Unit, University of Milan, Ospedale Maggiore IRCCS, Milan, Italy, ³Pharming Technologies B.V., Leiden, The Netherlands

We have previously shown that plasma derived C1-INH (pdC1-INH), the endogenous inhibitor of complement and contact/kinin system, has potent neuroprotective effects in ischemic mice, although with a narrow time-window of efficacy(1)–(3). This inhibitor, currently used to manage angioedema attacks in patients with hereditary or acquired C1-INH deficiency (angioedema, HAE or AAE), is purifed from human plasma. Using a plasma derived molecule can be disadvantageous, due to risks for human pathogenic viruses contamination and to difficulties in supply and purification. Due to these limitations, a recombinant approach was chosen to overcome these problems.

Recombinant human C1-INH (rhC1-INH, Pharming Technologies B.V.) has been administrated intravenously to C57Bl/6 ischemic mice at different doses (5, 10 and 15U/mouse) and time-points (0, 3, 6, 18 and 24 hours) from transient ischemia. Forty-eight hours after ischemia, rhC1-INH, at the dose of 15U/mouse, was able to markedly improve the cerebral damage when administrated 0, 3 and 6h from ischemia (13.67±2.59 mm3, 11.71±0.63mm3 and 20.38±2.37mm3, respectively) compared to saline-treated ischemic mice (44.43±5.94mm3). Also when administrated 18 hours after ischemia, rhC1-INH was still significantly effective (27.13±2.58mm3). Fluoro jade staining showed that amelioration of cerebral damage was due to an effective sparing of neurons. Immunohistochemistry for macroglia/macrohages (anti-CD11b) and astrocytes (anti-GFAP) showed that rhC1-INH was able to counteract the microglial and astrocytic activation, as well as macrophage infiltration.

Analysis of rhC1-INH localization in the brain by confocal microscopy revealed that rhC1-INH was confined to cerebral vessels and that it was able to bind endothelial cells as shown by colocalization with Claudin-5 and ICAM-1.

These data demonstrate that rhC1-INH has an important neuroprotective action with a wider time-window of efficacy compared to pdC1-INH.

This favorable profile of rhC1-INH is probably related to a different glycosylation of the native versus recombinant molecules.

EVIDENCE THAT PROTECTION BY HSP70 IS ANTI-INFLAMMATORY IN MODELS OF BRAIN ISCHEMIA

Zhen Zheng, Hualong Ma, Xiannan Tang, Midori A. Yenari

Department of Neurology, VA MED CTR, University of California, San Francisco, CA, USA

Previous studies from our lab and others show that overexpression of Hsp70 in brain cells of transgenic mouse is neuroprotective following transient middle cerebral artery occlusion (tMCAO). While the mechanism of protection has largely been thought to be due to its chaperone functions, preventing protein aggregation and assisting nascent protein folding, other work has shown that Hsp70 is also capable of modulating inflammatory responses, by either potentiating or inhibiting them. Prior work in our lab suggested an anti-inflammatory role for Hsp70 in tMCAO, but since Hsp70 probably protects via several mechanisms, we now directly test the hypothesis that Hsp70 overexpression in inflammatory cells is protective using a bone marrow chimera model. Hsp70 bone marrow chimeras were created by transplanting marrow from male CB6 Hsp70 transgenic mice into female wildtype littermates (Hsp70 chimera). In controls, marrow from male wild-type littermates (Wt chimera), or lacZ transgenic mice were used. Chimeras were generated by lethally irradiating recipients with the heads shielded. Bone marrow cells isolated from donor mice were injected via the tail vein into the irradiated recipient mice (100 × 106 bone marrow cells/recipient). Pilot studies of lacZ chimeras showed that donor marrow engrafted by 1–2 months post transplant (>70% of cells in blood assessed by FACS), and no transplanted cells were visible in the brain. However, 3 months post transplant, a few perivascular transplanted cells were visible. Therefore, 2 months post transplant, mice were subjected to 2 h tMCAO using an intraluminal suture. 48 h later, brains were harvested and stained with cresyl violet. Infarct size was measured using an image analysis system. MHC-class П staining was performed to identify activated inflammatory cells, and the positive cells were counted in 3 random non-overlapping fields within the cortex adjacent to the outer boundary of the infarct. To determine whether Hsp70 expression in microglia may also have an anti-inflammatory effect, cultures of either primary wildtype astrocytes alone or astrocytes with microglial cells (collected from Hsp70 transgenic or wildtype mice) underwent oxygen and glucose deprivation (OGD) for 6 h followed by 24 h reperfusion, then assessed for cell death. Infarct size among Hsp70 chimeras was reduced by 22 % compared to Wt chimeras (P<0.05). MHC Class II expression was significantly decreased in Hsp 70 chimeras (25%) compared to Wt chimeras (37.6%) (P<0.01). In vitro, microglia nearly doubled the extent of cell death due to OGD. The addition of Hsp70 transgenic microglia to astrocytes completely inhibited microglia potentiated cell death (P<0.001). Furthermore, Hsp70 transgenic microglia showed more than 50% decrease in the number of MHC II positive cells following LPS stimulation compared to wildtype microglia using FACS analysis (p<0.005). Our data indicate that Hsp70 overexpression in inflammatory cells is anti-inflammatory, and is protective following brain ischemia. This suggests another novel mechanism of protection by Hsp70.

RBC EXCLUSION WITHOUT CAPILLARY DIAMETRIC CHANGE DURING PASSAGE OF K+−INDUCED CORTICAL SPREADING DEPRESSION IN RATS

Takashi Osada¹, Minoru Tomita¹, Yutaka Tomita², Miyuki Unekawa¹, Haruki Toriumi¹, Norihiro Suzuki¹

¹Department of Neurology, School of Medicine, Keio University, Tokyo, Japan, ²Institute of Brain and Blood Vessels, Department of Neurology, Mihara Memorial Hospital, Gunma, Japan

[Introduction]In a previous communication we reported K+−induced cortical spreading depression (CSD) caused a few seconds' cessation of capillary flow in association with the wave-ring spread of CSD based on an analysis of tissue hemodilution curves (Neurosci Lett 322: 157, 2002; J Cereb Blood Flow Metab 25: 742, 2005) and segmental constriction/dilatation of arterioles (Neuroreport 17: 1365, 2006). In this paper, we examined whether such changes in RBC movement were related to diametric changes of the pertinent capillaries presumably affected by astrocytic swelling during CSD in rats. [Material and Methods]Nine urethane-anesthetized rats were used. To confirm CSD, DC potential in the cerebral cortex was measured. Images of the movements of individual FITC-labelled RBCs in intraparenchymal single capillaries at an approximate depth of 50 µm in the cortical region of interest (ROI) were acquired employing either a conventional video camera (30 frames/s) or a high-speed camera (250-500/s frame rate) with laser scanning confocal fluorescence microscopy, and analysed with Matlab domain homemade software (KEIO-IS2) to document the RBC tracking and calculate the number and velocity of labelled RBCs appearing in the ROI during a period of CSD wave passage after microinjection of 150 mM K+ solution on the brain surface. In vivo staining of single capillaries with FITC-dextran injected intravenously, and astrocytes stained in vivo with direct application of sulforhodamine 101 on the brain surface (Nat. Methods 1: 31, 2004), respectively. [Result]FITC-labelled RBCs moving in the capillary network appeared like fireflies, and the mean RBC velocity was 2.1 ± 1.8 mm/sec. Within a few minutes after K+ application to the periphery of the ROI, fluctuations of DC potential were initiated with a darkening of the background of the video frames. In 8 out of 9 cases we observed transient RBC exclusion from single capillaries in the ROI during K+-induced CSD: both the number and velocity of RBC were decreased until RBC disappeared from the ROI completely. The disappearance lasted for 3.3±2.3 seconds and the occurrence was statistically significant (p< 0.05). This transient RBC exclusion was followed by recovery of capillary flow to its initial state. The RBC exclusion from capillaries was never observed under control conditions. However, no detectable changes in the diameter of single capillaries were observed in association with the RBC exclusion during CSD passage. Endfeet of astrocytes were detected stained in a brown colour and changed their morphology slowly, but no direct compression effect was observed during CSD passage.[Conclusion]We conclude that RBCs were excluded (de-recruitment) from the capillary network in the ROI during CSD passage with little capillary diametric change. Astrocytes may participate in RBC flow changes but not in the form of direct compression of the capillaries.

SLOWLY EVOLVING INFARCTION IN THE ISCHEMIC PERIPHERY: MRI, BEHAVIORAL AND PATHOLOGICAL ANALYSIS OF RODENT BRAIN AFTER TRANSIENT ISCHEMIA

Toshihiko Kuroiwa¹, Liyuan Sun², Naoyuki Miyasaka³, Ichiro Yamada⁴, Satoru Ishibashi², Shu Endo⁵, Kikuo Ohno⁶

¹Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA, ²Department of Neurology, Tokyo Medical and Dental University, Tokyo, Japan, ³Department of Obstetrics and Gynecology, Tokyo Medical and Dental University, Tokyo, Japan, ⁴Department of Radiology, Tokyo Medical and Dental University, Tokyo, Japan, ⁵Animal Research Center, Tokyo Medical and Dental University, Tokyo, Japan, ⁶Department of Neurosurgery, Tokyo Medical and Dental University, Tokyo, Japan

Infarction evolves slowly in the ischemic periphery after mild ischemia. We examined the time course of post-ischemic peripheral tissue with disseminated selective neuronal changes, reactive astrocytosis and infarction and analyzed the correlation with MRI and behavioral findings in the experimental unilateral forebrain ischemia of rodent.

[Methods] Transient unilateral forebrain ischemia was induced by occlusion of the left common carotid artery for 15 ? 60 min with 8% oxygen inhalation in SD rats (n=57) and by two times 10 min occlusion of the left common carotid artery in Mongolian gerbils (n=36). Intraischemic blood flow was monitored in gerbils by laser Doppler flowmetry (LDF). MRI apparent diffusion coefficient (ADC) mapping and T2 mapping during ischemia and post ischemia was performed using an experimental MRI (4.7T). Neurological deficit was evaluated in gerbil by applying a battery of behavioral tests. The brain removed after perfusion fixation at 1d ? 16w post ischemia was prepared for histological examination. [Results] Intra-ischemic LDF decreased to 22.1±9.5% of baseline in the ischemic core and 37.5±3.5% in the ischemic periphery. Areas with eosinophilic neurons (ENs) appeared in the ischemic center of the cortex at 1d post-ischemia, extended to the peripheral cortex by 4d. The area of high EN density (>80 EN/mm2) transformed into infarction during 2 weeks of post ischemia. Reactive astrocytes (RAs) in this area were involved in the infarction process, indicating occurrence of delayed astrocytic death. Areas with low EN density (<80 EN/mm2) was not transformed into infarcts. Behavioral examination revealed neurological deficit during ischemia, transient recovery shortly after recirculation and secondary neurological deterioration. ADC mapping revealed rapid ADC decrease (6.97±0.4×10-4 mm2/s to 3.97±0.21×10-4 mm2/s, P<0.05) of the ischemic area after onset of ischemia in rats. ADC decrease was transient in the animals with 15 min ischemia. Only selective neuronal death evolved in this group. Animals showing a biphasic ADC decrease ? intra-ischemic decrease followed by a transient recovery and subsequent decrease ? and permanent ADC decrease developed cerebral infarction. [Conclusion] Areas of high, but not low, EN density transformed slowly into infarction after transient forebrain ischemia. Delayed astrocytic death took place in the areas with high EN density. Areas of the slowly evolving infarction were predictable by early repeated ADC mapping. Secondary neurological deterioration was observed in the animals with slowly evolving infarction.

Figure. ADC map 4d post ischemia (left), histological changes 4d (middle) and 16w (right) post ischemia in gerbils

TOWARD A FUNCTIONAL METABOLOMICS OF BRAIN ENERGY METABOLISM AND NEUROTRANSMITTER TRAFFICKING

Paul K. Maciejewski

Magnetic Resonance Research Center, Yale University, New Haven, CT, USA

Although it is widely accepted that glucose is the primary substrate that supports brain function, fundamental questions about pathways and compartmentalization of metabolic processes that accomplish glucose catabolism in brain tissue remain unanswered. Furthermore, a basic understanding of the role of glucose in brain function must recognize that glucose not only serves as a basic energy substrate for brain but also as a substrate for the de novo synthesis of chemical neurotransmitters such as glutamate and GABA.

Metabolomics, an emerging field of biological investigation, is interested in not only the identification and characterization of small-molecule metabolites that are present in biological systems, but also in the identification and characterization of their role within the context of those systems (i.e., what might be called a “functional metabolomics:). Metabolites are material elements that are consumed and produced in metabolic processes. Fluxes of metabolites into and out of metabolic processes provide an indication of the function of metabolites within the context of living systems. Consequently, a functional metabolomics of brain energy metabolism and neurotransmitter trafficking ought to consist of the identification and characterization of metabolic processes that participate in these basic metabolic functions in brain quantified in terms of fluxes of metabolites through those processes.

In principle, brain energy metabolism and other brain metabolic functions (e.g., neurotransmitter synthesis and neurotransmitter transport processes) can be investigated in vivo using isotopic label-tracer methods implemented using Mass Spectroscopy (MS) or Magnetic Resonance Spectroscopy (MRS). However, unlike other metabolomics technologies that identify metabolites and characterize systems in terms of profiles of metabolites, label-tracer methods depend on detailed a priori knowledge about metabolic pathways, knowledge which is used to explain the transfer of label from one metabolite to another. This a prior knowledge about metabolic pathways is provided by metabolic models, that is, by detailed, mathematical descriptions of processes that are presumed to explain the transfer of label from label-enriched precursor substrates to other metabolites. However, existing metabolic models tend to focus on single pathways and do not accommodate the complexity of metabolism and the likelihood that specific metabolic functions may be accomplished via a multiplicity of functionally redundant pathways. This narrow focus limits the design and interpretation of existing in vivo label-tracer studies of metabolism.

Here I will introduce a new “functional metabolomics” approach to the investigation of biochemical pathways for brain energy metabolism and neurotransmitter trafficking in vivo that consists of: 1) a rational method for the specification of alternative, hypothetical biochemical pathways that in principle accomplish these basic metabolic functions in brain, and 2) a method for the design of isotopic label-tracer experiments to determine the extent to which these alternative, hypothetical pathways actually accomplish these specific metabolic functions in vivo. The approach integrates information derived from cellular and molecular biology, metabolomics and proteomics with concepts and techniques derived from engineering, mathematics, and computational sciences.

(Supported by NIH NS-044316)

PROPYLENE GLYCOL AFTER HUMAN TRAUMATIC BRAIN INJURY: NOVEL BIOMARKER OF ALTERED GLUCOSE METABOLISM

Daniel Hirt¹, Mujtaba Kadri¹, Neil A. Martin¹, Shane Que Hee², Matthew Eliseo¹, Jon Berg¹, Paul Vespa¹, Thomas C. Glenn¹

¹Division of Neurosurgery, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA, ²Department of Environmental Health Sciences, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA

INHIBITION OF THE MYOSIN LIGHT CHAIN KINASE PREVENTS HYPOXIA/STROKE-INDUCED BLOOD BRAIN BARRIER DISRUPTION IN VITRO AND IN VIVO

R. Tamaki^2,3, C. Kuhlmann¹, V. Lessmann¹, H. Luhmann¹, Oliver Kempski²

¹Institute of Physiology and Pathophysiology, Johannes Gutenberg University, Mainz, Germany, ²Institute for Neurosurgical Pathophysiology, Johannes Gutenberg University, Mainz, Germany, ³Department of Neurosurgery, Nara Medical University, Nara, Japan

Background: Increased mortality after stroke is associated with development of brain edema. The aim of the present study was to examine the contribution of endothelial myosin light chain (MLC) phosphorylation to hypoxia/stroke-induced blood brain barrier (BBB) opening.

Methods: Measurements of Trans-Endothelial-Electrical-Resistance (TEER) were performed to analyze BBB integrity in an in-vitro co-culture model (bovine brain microvascular endothelial cells (BEC) and rat astrocytes). Brain water content was analyzed in rats 24 hrs after venous stroke induction using a two-vein occlusion (TVO) model (1). Dihydroethidium was used to monitor intracellular generation of reactive oxygen species (ROS) in BEC. MLC phosphorylation was detected by immunostainings using a confocal laser-scanning microscope.

Results: Hypoxia induced in-vitro by a combined oxygen-glucose deprivation caused a decrease of TEER values by more than 40 %, which was prevented by the application of the MLC-kinase inhibitor ML-7 (10 µmol/l). ML-7 also significantly reduced the elevated brain water content invivo after TVO (80.184± 0.540%, n=8) to 79.714 ± 0.337% (n=10) (sham-op: 78.793± 0.326%, n=8). The NAD(P)H-oxidase inhibitor apocynin (500 µmol/l) prevented the hypoxia-induced TEER decrease. Hypoxia-dependent ROS generation was completely abolished by apocynin and by the Ca2+ chelator BAPTA (10 µmol/l). Interestingly, the hypoxia-dependent decrease of TEER was also blocked by BAPTA. Inhibition of Ca2+ signaling (BAPTA), ROS-formation (apocynin), or the MLC-kinase (ML-7) completely abolished the hypoxia-induced increase of MLC-phosphorylation.

Conclusions: Our data demonstrate for the first time, that MLC phosphorylation contributes to hypoxia/stroke-induced BBB disruption in-vivo and in-vitro. The hypoxia-induced increase of intracellular Ca2+ increases the activity of the NAD(P)H-oxidase. The resulting ROS generation is responsible for an increase of MLC phosphorylation that activates the endothelial contractile machinery leading to paracellular gap formation.

INDUCTION OF AQUAPORIN EXPRESSION IN ISCHEMIC MOUSE BRAIN AFTER THROMBIN PRECONDITIONING

Jerome Badaut¹, Melanie Price², Karine Wiegler², Nabil Mastour¹, Jean-Francois Brunet¹, Lorenz Hirt²

¹Neurosurgical Research Group, Pavillon 3, CHUV, UNIL, Lausanne, Switzerland, ²Neurology Department, CHUV, UNIL, BH19, Lausanne, Switzerland

PROBABILISTIC CORTICAL SURFACE MAP OF THE MIDDLE CEREBRALARTERY TERRITORY FOR SINGLE-PHOTON EMISSION COMPUTED TOMOGRAPHY STUDIES

Kohkichi Hosoda, Eiji Kohmura

Department of Neurosurgery, Kobe University Graduate School of Medicine, Kobe, Japan

[Background and aims] The middle cerebral artery (MCA) is the intracranial vessel most frequently affected by cerebrovascular diseases. A more accurate knowledge of the topography of this vessel may have an impact on treatment strategies for ischemic cerebrovascular diseases. The aim of this study was to construct a topographic map of the MCA territory for SPECT using statistical brain mapping. [Methods] The margin of the perfusion deficit associated with infarction due to arterial occlusion, as seen on SPECT imaging, is presumed to approximate the borders of the territory of the artery. Basing the study on this hypothesis, SPECT images obtained in 12 patients with large MCA infarctions due to angiographically confirmed MCA trunk occlusion were selected, anatomically standardized, and compared with SPECT images of healthy volunteers to construct probabilistic cortical surface maps of the MCA territory. Crossed cerebellar diaschisis was used as a primary cutoff marker for creation of the map. This MCA map (Method C) was compared with the conventional ROI method (Method A) and previously reported predefined cortical templates (Method B) for preliminary clinical application. [Results] The probabilistic cortical surface map of the MCA territory showed that regions with the highest ratio of MCA territory included the transverse temporal gyrus (100%), supramarginal gyrus (100%), and inferior parietal lobule (91–92%). For preliminarily clinical application, this map (Method C) was compared with the conventional ROI method (Method A) in predicting hyperperfusion after carotid endarterectomy by performing a receiver operating characteristic (ROC) analysis, which demonstrated the statistically significant superiority of the MCA map (area under the ROC curve [Az] = 0.91) to the ROI method (Az = 0.75; p = 0.025). The ROC analysis also demonstrated a diagnostic value of the MCA map (Az = 0.95) that equaled predefined cortical templates (Method B) (Az = 0.93). [Conclusions] The probabilistic cortical surface map of the MCA territory used for SPECT, which was created using statistical brain mapping techniques, would be useful for an objective assessment of the cerebral perfusion status of patients with cerebrovascular diseases.

LATE VASCULAR ABNORMALITIES FOLLOWING SUBARACHNOID HEMORRHAGE IN THE RAT CORRELATE WITH THE SEVERITY OF ACUTE ISCHEMIA

Radoslaw Michalik¹, Anna Domasiewicz¹, Roman Gadamski², Piotr Piotrowski², Janina Rafalowska², Zbigniew Czernicki¹, Ewa Kozniewska¹

¹Department of Neurosurgery, Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland, ²Department of Experimental and Clinical Neuropathology, Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland

BACKGROUND AND AIMS. Subarachnoid hemorrhage (SAH) results in acute decrease of cerebral blood flow which is fatal for about 30% of patients. In about 30% of survivors delayed persistent vasospasm develops resulting in morbidity. Despite years of experimental and clinical research on SAH no effective treatment of the delayed vasospasm was found. Majority of experimental studies on SAH aimed to prevent or reverse delayed vasospasm concentrated on this late phase. The aim of this study was to find out whether dysfunction of microcirculation and structural changes in extraparenchymal large, conduit arteries following SAH correlate with severity of acute post-SAH ischemia

METHODS. Adult, anaesthetized male Sprague-Dawley rats were subjected to SAH or sham surgery. SAH was induced by endovascular puncture of the bifurcation of the internal carotid artery. Microflow (LDF) in cerebral cortex was measured bilaterally before, during, and for, at least, 30 min after SAH using laser Doppler flowmeter. Reactivity of microcirculation to CO2, intravenous papaverine, acetylcholine, or thromboxane receptor antagonist (SQ 29,548) administration was assessed at 3, 24, 48, and 72 after SAH. At the same time points ipsi- and contralateral middle cerebral artery and basilar artery were harvested and processed for histological examination and measurement of wall thickness.

RESULTS. Following SAH microflow in the ipsilateral cortex did not respond to the tested vasodilators up to 48 hours. At 72 hours after SAH in some animals the recovery of the response to CO2 and papaverine was noted. This recovery correlated with the severity of acute ischemia and was characteristic for the cases with small SAH (microflow in acute phase below 30% of control for 15-20 min). In all cases with large SAH (microflow in acute phase below 30% of control for more than 40 min) no recovery was observed. In addition, in the animals with large SAH, at 72 hours after the bleeding cerebrovascular resistance decreased after blocking of TXA2/PGH2 receptor. The response to acetylcholine was ebolished irrespective of the severity of SAH and did not recover. Degree of corrugation of the ipsilateral MCA at 3 days after SAH which was chosen as a measure of vasospastic changes also correlated with severity of acute ischemia

CONCLUSIONS. This study demonstrates that duration of the impairment of microvascular vasodilatory capacity following SAH correlates with the severity of acute ischemia. It also shows that critical underperfusion in the acute phase after SAH results in progressive functional deterioration of microcirculation leading to the increase of arachidonic acid metabolites-dependent tone of microvessels in the late phase after SAH.

The study suggests that the therapy of post-SAH states should concentrate on the improvement of microvascular functions.

Supported by grant 3P05A 02421 from the KBN.

NORMOBARIC HYPEROXIA REDUCES BBB OPENING BY INHIBITING MATRIX METALLOPROTEINASE-MEDIATED DEGRADATION OF TIGHT JUNCTION PROTEINS IN A FOCAL ISCHEMIC RAT MODEL

Wenlan Liu, Zhongrui Yuan, Ke Jian Liu

Department of Pharmaceutical Sciences, College of Pharmacy, University of New Mexico, Albuquerque, NM, USA

Background and aims: Normobaric hyperoxia treatment has recently been shown by us and other groups to be an effective strategy in reducing stroke lesion volumes and improving neurological function following ischemic stroke in rats. However, the mechanisms of normobaric hyperoxia-induced neuroprotection remain elusive. The disruption of blood brain barrier (BBB) is known to be a critical event during focal cerebral ischemia, which leads to vasogenic edema and hemorrhage. Occludin and claudins are important tight junction proteins regulating paracellular permeability of the BBB by sealing the tight junctions. We have recently shown that matrix metalloproteinases (MMPs) mediate the BBB disruption by degrading tight junction proteins following focal cerebral ischemia in rats. In the present study, we tested the hypothesis that normobaric hyperoxia reduces the opening of blood brain barrier by inhibiting matrix metalloproteinase-mediated degradation of tight junction proteins.

Methods: Rats were subjected to hyperoxia (95% O2) or normoxia (30% O2) during 90 min filament occlusion of the middle cerebral artery, followed by 3-hr reperfusion. At the end of reperfusion, BBB damage was assessed by measuring Evan's blue extravasation in ischemic cortex and caudate. Real-time RT-PCR, immunohistochemistry and western blot were used to analyze the mRNA and protein levels of tight junction proteins in isolated cerebral microvessels. MMP-2 and ?9 were measured by in situ and gel zymography.

Results: We found that extensive BBB opening occurred in ischemic cortex and caudate after 3-hr reperfusion, which was significantly reduced by hyperoxia treatment. The mRNA and protein levels of occludin and claudin-5 were decreased in ischemic cerebral microvessels, which concurrently occurred with increased gelatinolytic activity. Gel zymography showed a 4-fold increase in MMP-9, while no MMP-2 was detected in ischemic microvessels. The reduction of occludin and claudin-5, as well as the increase of MMP-9, was reversed by hyperoxia treatment. To further investigate the involvement of MMP-9 in the mechanism of hyperoxia-mediated neuroprotection, we also carried out parallel experiments using a broad-spectrum MMP inhibitor, BB1101. It was found that treatment of the ischemic rats with BB1101 produced the similar results on MMP-9, occludin, and claudin-5.

Conclusions: These results suggest that normobaric hyperoxia treatment protects early BBB opening by inhibiting MMP-mediated degradation of tight junction proteins. Furthermore, these findings support the notion that hyperoxia treatment could potentially be used as a promising interventional strategy to treat stroke patients.

A VASORELAXANT PEPTIDE IN THE BRAIN AND DELAYED CEREBRAL VASOSPASM AFTER ANEURYSMAL SUBARACHNOID HEMORRHAGE

Masayuki Fujioka¹, Kenji Nishio², Shiro Ueda², Atsushi Kubo³, Naoto Minamino³, Kenji Kangawa³, Kazuo Kitamura³, Kenichi Mishima⁴, Takashi Mori⁵, Takao Asano⁵, Toshisuke Sakaki⁶, Kazuo Okuchi², Akio Asai¹, Keiji Kawamoto¹, Naoki Otani⁷, Yasuhiro Yonekawa⁷, Bo Siesjo⁸

¹Department of Neurosurgery, Kansai Medical University, Hirakata, Osaka, Japan, ²Emergency Medical Center, Nara Medical Univeristy, Nara, Japan, ³National Cardiovascular Center, Suita, Osaka, Japan, ⁴Department of Neuropharmacology, Faculty of Pharmaceutical Sciences, Fukuoka University, Fukuoka, Japan, ⁵Department of Neurosurgery, Saitama Medical School and Center, Kawagoe, Saitama, Japan, ⁶Department of Neurosurgery, Nara Medical University, Kashihara, Nara, Japan, ⁷Department of Neurosurgery, Zurich University Hospital, Zurich, Switzerland, ⁸Neuroscience Center, Lund University, Lund, Sweden

PROSTAGLANDIN D2 PROTECTS NEONATAL MOUSE BRAIN FROM HYPOXIC ISCHEMIC INJURY

Hidetoshi Taniguchi¹, Ikuko Mohri², Hitomi Arahori¹, Kosuke Aritake³, Kazuko Wada¹, Takahisa Kanekiyo¹, Masahiro Nakayama⁴, Shuh Narumiya⁵, Keiichi Ozono¹, Yoshihiro Urade³, Masako Taniike²

¹Department of Pediatrics, Osaka University Graduate School of Medicine, Osaka, Japan, ²Department of Mental Health and Environmental Effects Research, The Osaka-Hamamatsu Joint Research Center for Child Mental Development, Osaka, Hamamatsu, Japan, ³Department of Molecular Behavioral Biology, Osaka Bioscience Institute, Osaka, Japan, ⁴Division of Clinical Laboratory Medicine and Anatomic Pathology, Osaka Medical Center and Research Institute for Maternal and Child Health, Osaka, Japan, ⁵Department of Pharmacology, Kyoto University Faculty of Medicine, Kyoto, Japan

Prostaglandin (PG) D2, the most abundant prostanoid in the brain, is synthesized by hematopoietic PGD synthase (HPGDS) or lipocalin-type PGDS (L-PGDS) and binds to specific receptors, DP1 or DP2. We investigated the role of PGD2 in the pathogenesis of hypoxic-ischemic encephalopathy (HIE) in neonatal mice at postnatal day 7. In wild-type mice, hypoxia-ischemia increased PGD2 production in the brain up to 90-fold compared with the level in sham-operated brains at 10 min after cessation of hypoxia. Whereas the size of the infarct was not changed in L-PGDS- or DP2-knockout mouse brains compared with that in the wild-type HIE brains, it was significantly increased in HPGDS-L-PGDS double knockout or DP1-knockout mice. The PGD2 level in L-PGDS-, HPGDS-, and HPGDS-L-PGDS-knockout mice at 10 min of reoxygenation was 46%, 7%, and 1%, respectively, of that in the wild-type, indicating the infarct size to be in inverse relation to the amount of PGD2 production. Thus, PGD2-DP1 signaling protected brains from hypoxia-induced ischemic damage in neonatal mice. DP1 receptors were exclusively expressed in endothelial cells after 1 hr of reoxygenation, and cerebral blood flow decreased more rapidly after the onset of hypoxia and did not return to the baseline level after reoxygenation in HPGDS-L-PGDS-knockout mice. Endothelial cells were severely damaged in HPGDS-L-PGDS- and DP1-knockout mice after 1 hr of reoxygenation. In the human neonatal HIE brain, HPGDS-positive microglia were remarkably increased in number. In conclusion, PGD2 protected the neonatal brain from hypoxic-ischemic injury via DP1 receptors by preventing endothelial cell degeneration.

STABILITY AND ABRUPT FALL OF BRAIN TISSUE POTASSIUM IN RAT FOCAL CEREBRAL ISCHEMIA AND THE BLOOD-BRAIN BARRIER

Stephen C. Jones^1,2, Weixing Hu³, Yang Wang⁴, Alejandro D. Perez-Trepichio⁵, Victor E. Yushmanov¹, Alexander Kharlamov¹, Fernando E. Boada⁶

¹Department of Anesthesiology, Allegheny Singer Research Institute, Pittsburgh, PA, USA, ²Department of Neurology, Allegheny Singer Research Institute, Pittsburgh, PA, USA, ³Nanjing Medical University, Nanjing, China, ⁴Intel Corporation, Portland, OR, USA, ⁵Medical Surgical Specialists, Naples, FL, USA, ⁶Department of Radiology, MR Research Center, University of Pittsburgh, Pittsburgh, PA, USA

ROLE OF MAPK IN NEUROPROTECTION BY CART IN ISCHEMIC STROKE

Yun Xu¹, Jia Jia², Zichun Hua³, Patricia Hurn⁴, Nabil J. Alkayed⁵

¹Department of Neurology, The Affiliated Drum Tower (Gulou) Hospital of Nanjing University, Nanjing, China, ²State Key Laboratory of Pharmaceutical Biotechnology of Nanjing University, Nanjing, China, ³Department of Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA

BACKGROUND AND SIGNIFICANCE: Our previous work indicated that CART (cocaine and amphetamine regulated transcript) plays an important role in estrogen-mediated neuroprotection in ischemic stroke (Xu Y et al. 2006). Specifically, we previously demonstrated that CART protects primary cultured cortical neurons from cell death induced by oxygen-glucose deprivation (OGD) and decreases infarct size after ischemic stroke induced in mice by middle cerebral artery occlusion (MCAO). The aims of this study are: 1) to further investigate the neuroprotective role of CART in vitro and in vivo using small interfering RNA (siRNA) to downregulate endogenous CART, and 2) to explore a potential mechanism of neuroprotection by CART. CART and its mechanism of action may serve as a novel therapeutic strategy for neuroprotection in ischemic stroke.

METHODS: Five double-stranded RNA fragments corresponding to selected mouse CART mRNA sites were produced using plasmid siRNA expression vector under control of U6 promoter. The vector also co-expresses the coral GFP marker (cGFP) under CMV promoter, which is used to track transfection efficiency. Cortical neurons and HEK293T cells were transfected with CART siRNA or control vector, and interference efficiency was assayed by RT-PCR and ELISA. HEK293 cells were also co-transfected with CART cDNA, since they do not express CART. Finally, CART or vehicle, CART siRNA or control vector in vehicle (cationic polymer polyethylenimine, PEI) were injected into mouse lateral ventricles in vivo 24 hours prior to MCAO. Brains were removed, stained with 2,3,5-triphenyltetrazolium chloride (TTC) for infarct size measurement. Neuronal cell death in vitro was induced by OGD, and assayed by propidium iodide (PI)/Calcein staining, and survival assayed by MTT. The phosphorylation of ERK1/2 was measured by western blot with anti-ERK anti-p(44/42)ERK antibodies.

RESULTS: CART expression was effectively suppressed by siCART. Selected siRNA reduced CART RNA and protein levels in vitro by approximately 80% and 90%, respectively. CART administration reduced infarct size at 24 hrs after MCAO from 39·6% to 22·4% (n=6 per group, p<0.05), while administration of siCART increased infarct size to 48·4% compared to saline-injected mice (23.5·4%, n=6 per group, p<0.05). Neuronal survival by MTT after 2-hr OGD was higher in CART-treated cultures (49±6%, n=3, each in triplicate) compared to saline-treated cultures (35±%, n=3 in triplicate, p<0.05), while survival was significantly reduced in cultures treated with siCART (21±4%, n=3 in triplicate) compared to saline-treated cultures. Similar results were obtained using PI/Calcien staining. Finally, CART increased ERK phosphorylation in cultured neurons after OGD, while siCART inhibited ERK activation after OGD.

CONCLUSIONS: The findings suggest that CART is an important endogenous neuroprotectant against ischemic neuronal cell death in vitro and in vivo, and that mechanism of neuroprotection is in part linked to ERK activation. CART may serve as a novel neuroprotectant against stroke-related brain damage.

INTRATHECAL INJECTION OF EGF AND FGF2 PROMOTES PROLIFERATION OF NEURAL PRECURSOR CELLS IN THE SPINAL CORDS OF ALS MODEL MICE

Yasuyuki Ohta, Makiko Nagai, Tetsuya Nagata, Tetsuro Murakami, Isao Nagano, Hisashi Narai, Tomoko Kurata, Mito Shiote, Yasushi Takehisa, Nobutoshi Morimoto, Kazunori Miyazaki, Tatsushi Kamiya, Koji Abe

Department of Neurology, Graduate School of Medicine, Dentistry and Pharmacy, Okayama University, Okayama, Japan

Background: Endogenous neural precursor cells reside close to the ependyma of the central canal (CC). In injured spinal cord, they are activated to proliferate, but most of them differentiate into glial cells and rarely into neurons. Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease, which characterized by selective death of motor neuron in the spinal cord. Here, we investigated three steps of their activation, proliferation, migration and differentiation, in the spinal cord of symptomatic ALS model mice treated with intrathecal infusion of epidermal growth factor (EGF) and fibroblast growth factor 2 (FGF2).

Methods: EGF and FGF2 or vehicle were continuously injected for 7 days using an osmotic minipumps into the lumbar spinal cord region of normal and symptomatic transgenic (Tg) mice with a mutant G93A SOD1 gene at 100 days of age, when they display paralysis. After placement of the minipump, intraperitoneal injections of 5-bromodeoxyuridine (BrdU) began every 6 h (four times per day) for 7 days. 21 days after minipump placement, the lumbar cord (L4-L5) was removed. In order to determine the character of BrdU-labeled cells, we performed immunofluorescence studies using BrdU, nestin, highly polysialylated neural cell adhesion molecule (PSA-NCAM), ionized calcium-binding adapter molecule-1 (Iba1), neuronal nuclear antigen (NeuN), and glial fibrillary acidic protein (GFAP).

Results: We observed that BrdU-labeled cells and BrdU+nestin double-labeled neural precursor cells increased in the spinal cords of Tg mice as compared with non-Tg mice, with a much greater increase produced by EGF and FGF2 treatment. The number of BrdU+nestin double-labeled cells was larger than that of BrdU+Iba1, GFAP or PSA-NCAM double-labeled cells, but none expressed NeuN. On further analysis of the gray matter of Tg mice, the number of BrdU+nestin and BrdU+PSA-NCAM double-labeled cells increased more in the ventral horns than the dorsal horns, which was again greatly enhanced by EGF and FGF2 treatment.

Discussion and conclusions: The present study suggests that cell proliferation becomes active in symptomatic Tg mice and is further enhanced by EGF and FGF2 treatment. Furthermore, the neural precursor cells preferentially differentiate into neuronal precursor cells in Tg mice by EGF and FGF2 treatment.

ALLOPREGNANOLONE REDUCES DAMAGE TO CEREBELLAR PURKINJE CELLS FOLLOWING EXPERIMENTAL CARDIAC ARREST AND CPR IN MICE

N. Taguchi, M.H. Kelley, P.D. Hurn, R.J. Traystman, Paco S. Herson

Department of Anesthesiology and Peri-Op Medicine, Oregon Health and Science University, Portland, OR, USA

Background and Aims: Progesterone has recently been demonstrated to be neuroprotective in animal models of transient cerebral ischemia (MCAO). Using an in vitro neuronal culture model we recently found that progesterone neuroprotection of cerebellar Purkinje cells (PCs) requires metabolism to its GABA-active metabolite allopregnanolone (ALLO). We chose to investigate PCs because they are particularly sensitive to ischemia and surprisingly understudied. The current study evaluated the ability of ALLO to protect cerebellar Purkinje cells in vivo, using a mouse cardiac arrest and CPR (CA/CPR) model of global ischemia. In addition, we utilized an in vitro cerebellar culture model to investigate the mechanism of ALLO protection of PCs.

Methods: Male C57Bl/6 mice were subjected to 10 min. cardiac arrest and resuscitated by ventilation with 100% oxygen in combination with, epinephrine and chest compressions (CPR). Vehicle (20% β-cyclodextran) or ALLO (2 or 8 mg/kg) was administered by i.p. injection 30 min. prior to arrest, and 6 and 24 hours following resuscitation. 48 hours after resuscitation, mice were sacrificed and brains analyzed by FluorJade B staining to identify damaged PCs. The percentage of damaged PCs was determined by counting the number of fluorescent/non-fluorescent PCs in the vermis of each cerebelli.

Neurons were cultured from the cerebellum of embryonic day 18 Sprague-Dawley rats. The culture media was replaced with glucose-free de-oxygenated saline and the cells were placed in an anaerobic incubator for 2 hours in order to perform in vitro ischemia (OGD) experiments. Following OGD, cells were returned to normoxia by incubation in normal culture media for 1–3 hours. In order to determine the effects of ischemia on PC GABA-A receptor activity, whole cell voltage-clamp recordings were made from the somas of Pcs using an Axopatch 200B amplifier interfaced to a computer running pCLAMP9.2. In separate experiments, protein was isolated from cultures and quantitative western blot analysis was performed to determine changes in GABA-A receptor protein.

Results: Our data indicate that ALLO protects cerebellar Purkinje cells following global ischemia (CA/CPR) in a dose-dependent manner. While 2 mg/kg ALLO had no effect on PC damage, 8 mg/kg ALLO provides significant protection of cerebellar PCs, decreasing damage by over 50%. CPR time and epinephrine dose used for resuscitation in vehicle and ALLO treated groups were not different.

In vitro, OGD results in a significant reduction in GABA-activated current in PCs recorded within the first hour after OGD, reducing the average current density by approximately 40%. ALLO (10 nM) prevented the decrease in current after OGD and interestingly, high levels of ALLO (1 µM) caused a 50% increase. Similarly, OGD causes a significant reduction of GABA-A receptor protein 1 hr after OGD, decreasing by approximately 70%. Treatment with 1 µM ALLO maintained the levels of GABA-A receptor protein at control levels following OGD.

Conclusions: These findings indicate that ALLO protects cerebellar PCs in vivo and in vitro and that ALLO prevents OGD-induced loss of GABA-A receptor activity, making it an exciting therapeutic agent that has the ability to both potentiate GABA-A receptors and maintain their activity following ischemia.

IN VIVO STAINING OF ASTROCYTIC ELEMENTS SURROUNDING MICROVESSELS WITH FLOWING RBCS IN MICE

Yutaka Tomita^1,2, Minoru Tomita¹, Takashi Osada¹, Miyuki Unekawa¹, Haruki Toriumi¹, Ban Mihara², Norihiro Suzuki¹

¹Department of Neurology, Keio University School of Medicine, Shinjuku-Ku, Tokyo, Japan, ²Institute of Brain and Blood Vessels, Department of Neurology, Mihara Memorial Hospital, Isesaki, Gunma, Japan

Background and aims: We have previously visualized the astrocytic-vascular interface with sulforhodamine 101 1) in rats (unpublished observation). This study aims to examine the structural relationship of astrocytic endfeet stained with sulforhodamine 101, and endothelial cells stained with FITC dextran in the microvasculature of mouse brain in vivo during continuous recording of FITC-labeled red blood cells (RBCs) flowing through the microvasculature, as observed with a confocal fluorescence microscope.

Methods: The head of a C57BL/6J mouse (n=5) was fixed to a stereotaxic apparatus under isoflurane anesthesia, a skull window was opened over the left parietal cortex 2), and the dura mater was removed carefully. A tail vein was catheterized for injection of FITC-labeled RBCs and FITC-dextran. Images of the movements of individual FITC-labelled RBCs in intraparenchymal single capillaries at a depth of approximately 50µm in the cortical region of interest (ROI) were acquired employing either a conventional video camera (30 frames/s) or a high-speed camera (500 frames/s) through a laser scanning confocal fluorescence microscope. Single capillaries were stained with a small amount of FITC-dextran injected intravenously, and astrocytes were stained with direct application of sulforhodamine 101 on the brain surface in vivo 1). Astrocytic attachment to the intraparenchymal microvasculature and its morphological changes were examined in vivo.

Results: We observed no toxic effects following sulforhodamine application, and the physiological parameters remained unchanged throughout the dye-staining experiments (ca. 3 h). Circulating FITC-labeled RBCs, which had been introduced from the tail vein, were observed like fireflies on the dark background of the cerebral cortex. The astrocytes and their endfeet were visualized in brown, in good contrast to FITC-dextran-stained endothelial cells and flowing, brightly FITC-labeled RBCs in the microvasculature. Arterioles were found to be enveloped densely by a thin endfeet sheath, like a sleeve. The endfeet locally saddled capillaries with occasional protrusions, which pushed the capillary wall towards the lumen and caused marked elongation of RBCs flowing through the narrowed gap. The endfeet tended to cluster at the branching site of capillaries, forming a structure like an ants' nest tunneled with capillaries for RBC traffic. This astrocytic structure changed its morphology slowly but dynamically. In contrast, the astrocytic endfeet had no contact with the venous system. These findings are similar to the results obtained previously in SD rats, except that the capillaries and astrocytic elements appeared to be more clearly visualized in mice.

Conclusions: We successfully visualized astrocytic elements and capillaries of the mouse microvasculature in which FITC-labeled RBCs were actually flowing in vivo. This technique provides a promising tool to investigate neuro-astrocytic control of capillary blood flow.

MODULATION OF THE FUNCTIONAL CEREBRAL BLOOD FLOW RESPONSE BY REDUCTION IN BASELINE BLOOD FLOW

Joy Liau¹, Thomas T. Liu²

¹Department of Bioengineering, University of California at San Diego, San Diego, CA, USA, ²Department of Radiology, University of California at San Diego, San Diego, CA, USA

CEREBROVASCULAR EFFECTS OF INHALED NITRIC OXIDE

Nicole Angela Terpolilli, Seong-Woong Kim, Serge Thal, Nikolaus Plesnila

Department of Neurosurgery and Institute for Surgical Research, University of Munich Medical Center, Grosshadern, Munich, Germany

Background and aim: Inhaled nitric oxide (iNO) induces selective pulmonary vasodilatation and has been used as treatment for various lung and heart diseases associated with pulmonary hypertension.(1) Though at first thought to act selectively on the lung increasing evidence for extrapulmonary iNO effects emerged.(2),(3) However, direct evidence for iNO mediated cerebrovascular effects is missing yet. The aim of the current study was therefore to investigate the effect of iNO on cerebral blood vessels by using intravital microscopy.

Results: iNO led to a prompt and lasting dilatation of cerebral venules (10±1% increase vs. baseline, p<0.05) without affecting arteriolar diameter and cerebral blood flow. This effect gradually subsided upon discontiuation of iNO. Direct in vivo imaging of NO with DAF showed signifiantly increased NO concentrations in cerebrovascular endothelial cells (Fig. 1). To further prove that the observed change in vessel diameter was NO dependent we blocked guanylate cyclase with 1H-[1,2,4]Oxadiazolo[4,3-a]quinoxalin-1-one (ODQ). This completely abolished the observed effect.

Results: iNO led to a prompt and lasting dilatation of cerebral venules (10±1% increase vs. baseline, p<0.05) without affecting arteriolar diameter and cerebral blood flow. This effect gradually subsided upon discontiuation of iNO. Direct in vivo imaging of NO with DAF showed signifiantly increased NO concentrations in cerebrovascular endothelial cells (Fig. 1). To further prove that the observed change in vessel diameter was NO dependent we blocked guanylate cyclase with 1 abolished the obs H-[1,2,4]Oxadiazolo[4,3-a]quinoxalin-1-one (ODQ). This completely erved effect.

Discussion: Inhaled NO is transported from the lung to the brain and leads to dilatation of cerebral venules. Accordingly, this is the first report demonstrating a cerebrovascular effect of inhaled NO. Further studies need to investigate the mechanisms of NO transport from the lung to the brain and additional effects of iNO on the cerebrovasculature during physiological and pathological conditions.

Figure 1: DAF fluoresence in cerebral vessels

BRAIN MODULATION WITH MCS & SCS FOR INTRACTABLE DEAFFERENTATION PAIN; PET ANALYSIS

Haruhiko Kishima¹, Youichi Saitoh¹, Azuma HIrayama¹, Amami Kato¹, Masayuki HIrata¹, Satoru Oshino¹, Naoki Tani¹, Masahiko Sumitani², Jun Hatazawa³, Toshiki Yoshimine¹

¹Department of Neurosurgery, Osaka University Medical School, Suita, Osaka, Japan, ²Department of Anesthesiology, Osaka University Medical School, Suita, Osaka, Japan, ³Department of Tracer and Kinetics, Osaka University Medical School, Suita, Osaka, Japan

[Background and aims] The mechanisms underlying deafferentation pain are not well understood. motor cortex stimulation (MCS) and spinal cord stimulation (SCS) are useful in the treatment of this kind of chronic pain; however, the detailed mechanisms underlying its effects are also unknown.

[Methods] Six patients with intractable deafferentation pain in the left hand were treated with MCS, and nine patients with this kind of pain in the leg were treated with SCS. For the MCS, 4-array strip stimulating electrode (Resume II, 3587A; Medtronic, Inc., Minneapolis, MN, USA) was placed subdurally at the surface of the right precentral gyrus. For the SCS, 4-array stimulating lead electrode (Pisces Quad, 3487A; Medtronic, Inc.) was placed epidurally in the thoracic-lumbar lesion. These electrodes were controlled by stimulator (Itrel3, Medtronic, Inc.). Both MCS and SCS reduced pain; visual analog scale (VAS) values for pain were 82 ± 20 (mean ± SE) before MCS and 39 ± 20 after MCS, and 76 ± 5 before SCS and 41 ± 5 after SCS. rCBF was measured by PET with H215O before and after MCS and SCS. The obtained images were analyzed with statistical parametric mapping software (SPM).

[Results] Significant rCBF increases were identified after MCS in the left posterior thalamus (PT), left insula, and right orbitatofrontal cortex(OFC), and the right precentral gyrus and right prefrontal cortex (PFC) showed rCBF decrease. In the late post-MCS phase, a significant rCBF increase was detected in the left caudal part of the anterior cingulate cortex (ACC). After SCS, rCBF is increased in the right PT, right OFC, bilateral parietal association cortices. and left PFC.

[Conclusions] These results suggest that MCS modulates the posterior insula and OFC to PT pathways to upregulate the pain threshold and the posterior insula to caudal ACC pathway to control emotional perception, resulting pain relief. Like as MCS, SCS might also modulate brain activation to control of pain threshold and emotion.

ELECTROPHYSIOLOGICAL BASIS OF FUNCTIONAL MRI SIGNALS IN RAT MODELSOF PERMANENT AND TRANSIENT CEREBRAL ISCHEMIA

Maurits Pieter Adriaan Van Meer^1,3, Petrus Johannes Stienen², Kajo van der Marel¹, Arie Doornenbal², Jan Willem Berkelbach Van Der Sprenkel³, Richard Michiel Dijkhuizen¹

¹Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands, ²Department of Clinical Sciences of Companion Animals, Utrecht University, Utrecht, The Netherlands, ³Department of Neurosurgery, Rudolf Magnus Institute of Neuroscience, University Medical Center, Utrecht, The Netherlands

ANATOMICAL STANDARDIZED STATISTICAL MAPPING OF 123I-IMP SPECT IN BRAIN LESIONS

Yasushi Shibata, Manabu Akimoto, Akira Matsushita, Tetsuya Yamamoto, Shingo Takano, Akira Matsumura

Department of Neurosurgery, Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Ibaraki, Japan

CORRELATION OF INTRINSIC OPTICAL SIGNAL AND CEREBRAL BLOOD FLOW DURING ACTIVATION OF RAT SOMATOSENSORY CORTEX

Michael M. Haglund¹, Joseph R. Meno², Al C. Ngai², H. Richard Winn³

¹Department of Neurological Surgery, Duke University, Durham, NC, USA, ²Department of Neurosurgery, University of Washington, Seattle, WA, USA, ³Department of Neurosurgery, Mount Sinai School of Medicine, New York, NY, USA

Background and Aims: Optical imaging of intrinsic signals has been accomplished in vitro and in vivo. The source of the in vitro signal is thought to be related to postsynaptic activation and cellular swelling, while the in vivo signal has been suggested as arising from blood volume changes and oxygen delivery from capillaries. To better define the origin of the optic signal, we decoupled CBF from neuronal activation using adenosine receptor antagonists during contralateral sciatic nerve stimulation (SNS).

Methods: Rats were anesthetized, intubated, ventilated and a cranial window created. The contralateral sciatic nerve was exposed and stimulated as previously described (Ngai et all, 1986). We studied the relationship between arteriolar diameter, intrinsic optical signals (690 nm) and somatosensory evoked potentials (SSEPs), during cortical activation prior to and during application of adenosine receptor antagonists: Theophylline, a non-specific adenosine receptor antagonist (A1 and A2) and CPX (8-cyclopentyl-1, 3-dipropylxanthine), a selective A-1 antagonist.

Results: In mock CSF, pial arterioles in the hindlimb somatosensory cortex dilated 124.7+4.7% during contralateral SNS. Topical application of theophylline (THEO; 5µM, had no effect on resting arteriolar diameter but significantly attenuated SNS-induced vasodilatation (110.3±2.4%; p<0.05 versus mock CSF; n=4). Topical application of (CPX; 1 uM), also had no effect on resting arteriolar diameter but significantly potentiated vasodilatation during SNS (151.9±19.4%; p<0.05; n=5). In the same animals, the optical changes were potentiated during SNS by topically applied THEO, when compared to corresponding mock CSF controls. Topical application of CPX, however, caused a large increase of the intrinsic signals obtained during SNS. In a separate group of animals, SSEP recordings revealed that topical application of THEO had no effect on either the N-1 or P-1 wave. Topical application of CPX significantly attenuated the N-1 wave (n=8; p<0.01).

Conclusion: By adenosine receptor manipulation, we were able to independently alter cortical activity and vessel diameter changes. We demonstrate that change in intrinsic signal, imaged at 690 nm, is not related to the vascular compartment, but rather correlates with changes in parenchymal activity of the brain. Moreover, the present study provides further evidence for a neuromodulatory role of adenosine in regulating cerebral blood flow during increased cortical activity.

[MMH suuppt. by Klingenstein Foundation; HRW suppt. by NS21076]

FOCAL FUNCTIONAL METABOLIC ACTIVITY DURING ACOUSTIC STIMULATION IS UNDERESTIMATED WITH [1- OR 6-14C]GLUCOSE DUE TO METABOLITE TRAFFICKING AND RELEASE

Nancy Cruz, Kelly Ball, Gerald Dienel

Department of Neurology, University of Arkansas for Medical Sciences, Little Rock, AR, USA

Background/Aims. Labeled glucose and its analogs are widely used in imaging and metabolic studies of brain function, astrocyte-neuron interactions, and neurotransmission. Metabolite shuttling among astrocytes and neurons is essential for cell-cell transfer of neurotransmitter precursors, distribution of molecules synthesized or metabolized in one cell type, and elimination of metabolic by-products. However, dispersion and release of labeled compounds from activated cells would reduce signal registration in metabolic labeling studies, causing underestimation of focal functional activation. Processes involved in metabolite trafficking and release were, therefore, assessed in the auditory pathway of conscious rats.

Methods/Results. Unilateral monotonic stimulation increased glucose utilization (CMRglc) in tonotopic bands in the activated inferior colliculus by 35–85% compared to contralateral tissue when assayed with [14C]deoxyglucose (DG), whereas only 20–30% increases were registered with [1- or 6-14C]glucose. Tonotopic bands were not evident with [1-14C]glucose unless assayed during halothane anesthesia (which also blocks gap junctions) or pre-treatment with probenecid (an inhibitor of monocarboxylic acid transport), but were detectable with [6-14C]glucose, consistent with loss of 14C via the pentose shunt pathway during activation. Extracellular lactate levels transiently doubled during acoustic stimulation, suggesting release to interstitial fluid, so metabolite spreading was assessed by microinfusion of [14C]tracers directly into the inferior colliculus from a point source. The volume of tissue labeled by [1-14C]glucose exceeded that by [14C]DG by 3.2- and 1.4-fold during rest and acoustic activation, respectively. During activation, the tissue volume labeled by U-14C-labeled glutamine and lactate rose, whereas that by glucose fell 50% and that by DG was unchanged. Because [14C]DG-6-P is retained intracellularly, [14C]DG trapping represents the volume of tissue within which glucose is phosphorylated by hexokinase, and the higher label distribution by [14C]glucose must be due to spreading of downstream metabolites. Increased release of [14C]glucose metabolites from tissue would cause the low autoradiographic registration of activation, and lower retention of labeled metabolites probably underlies diminished label spreading during activation. Glutamine and lactate, two metabolites known to be released from cells to interstitial fluid, are distributed within tissue more widely during activation compared to rest and contribute to metabolite trafficking from activated cells. Dispersion of [1-14C]glucose and its metabolites during rest was also reduced 50% by pre-infusion of two gap junction blockers, implicating the astrocytic syncytium as a pathway for metabolite diffusion through gap junctional channels along their concentration gradients.

Conclusions. Loss of glucose-derived label from activated cells is due to a number of factors, including decarboxylation reactions in the pentose shunt and, perhaps, other pathways, metabolite spreading within tissue, in part, via the astrocyte syncytium, and metabolite release from activated tissue. Lactate release probably contributes substantially to reduced signal registration during activation for the following reasons, (i) inhibition of monocarboxylic acid transporters by probenecid enhances visualization of tonotopic bands, (ii) extracellular lactate levels rise, (iii) lactate spreading increases, and (iv) labeled lactate has a specific activity half that of glucose. To summarize, dispersion and release of various labeled metabolites contribute to the substantial underestimation of focal CMRglc in conscious subjects when labeled glucose is the tracer.

CEREBRAL PERFUSION PATTERN OF IDIOPATHIC NORMAL PRESSURE HYDROCEPHALUS STUDIED BY SPECT AND STATISTICAL BRAIN MAPPING

Kazunari Ishii¹, Hiroki Sasaki¹, Atsushi K. Kono¹, Naokazu Miyamoto¹, Kenichi Shimada², Shingo Ohkawa², Tetsuro Kawaguchi³, Etsuro Mori⁴

¹Department of Radiology, Hyogo Brain and Heart Center, Himeji, Hyogo, Japan, ²Institute for Aging Brain and Cognitive Disorders, Hyogo Brain and Heart Center, Himeji, Hyogo, Japan, ³Department of Neurosurgery, Hyogo Brain and Heart Center, Himeji, Hyogo, Japan, ⁴Department of Behavioral Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan

Background and aims: Normal pressure hydrocephalus (NPH) is characterized by gait disturbance, mental deterioration and urinary incontinence, in association with normal cerebrospinal fluid (CSF) pressure. Although frontal cerebral blood flow (CBF) reduction has been emphasized in the majority of NPH cases, no report has been able to demonstrate a typical CBF pattern because the inclusion of idiopathic normal pressure hydrocephalus (iNPH) and secondary NPH in previous reports has led to heterogeneous CBF patterns. The purpose of this study was to investigate the specific pattern of CBF in subjects with iNPH using voxel-based analysis. Methods: N-isopropyl-p-[I-123] iodoamphetamine (IMP) single photon emission computed tomography (SPECT) images were performed in 30 iNPH patients, who met probable iNPH criteria, 30 Alzheimer disease (AD) patients and 15 normal control (NC) subjects. Inter-group comparisons between iNPH patients and NC subjects and between AD patients and NC subjects were performed using three-dimensional stereotactic surface projection (3D-SSP) analysis. Individual 3D-SSP images of the iNPH patients were assessed by visual inspection. Results: On the Z-score maps, areas of relative hypoperfusion were recognized around the corpus callosum in all 30 iNPH patients, as well as in the Sylvian fissure regions in 19 of 30 iNPH patients which included artifacts by dilated ventricles and the Sylvian fissures. Ten frontal dominant, eight parietotemporal dominant, and 12 diffuse hypoperfusion types were demonstrated. Inter-group comparison between iNPH and NC subjects showed relative hypoperfusion in the frontal and parietotemporal areas and severe hypoperfusion around the corpus callosum and Sylvian fissure regions, while parietotemporal and posterior cingulate CBF reduction was demonstrated between the AD and NC groups (Figure). Conclusion: The voxel-based statistical mapping method revealed an iNPH CBF reduction pattern. Diffuse or frontal dominant CBF reduction was shown and, additionally, severe relative hypoperfusion around the corpus callosum and Sylvian fissures was demonstrated, which probably reflects dilatation of the ventricle systems and Sylvian fissures and may represent characteristic patterns of iNPH.

QUANTIFIED T1 AND T2 AS USEFUL ADJUNCTS TO DWI FOR EARLY INFARCT DETECTION IN PERMANENT AND TRANSIENT MCAO

Jaspreet Kaur^1,2, Ursula Tuor^1,2,3, Zonghang Zhao¹, Hisham Assem⁴, Jodie Lee-Petersen^1,2, Tadeusz Foniok³, Phillip Barber^1,2

¹Hotchkiss Brain Institute and Clinical Neurosciences, University of Calgary, Calgary, AB, Canada, ²Experimental Imaging Centre, University of Calgary, Calgary, AB, Canada, ³MR Technology, Institute for Biodiagnostics West, Calgary, AB, Canada, ⁴O'Brien Centre for The Bachelor of Health Sciences Program, University of Calgary, Calgary, AB, Canada

CYTOKINE BIOLOGY, LIPID METABOLISM AND CITICOLINE IN STROKE

Rao Muralikrishna Adibhatla, James F. Hatcher, Eric C. Larsen

Department of Neurological Surgery, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA

Background and aims: Phosphatidylcholine (PC) is a major membrane phospholipid (~50% of the total phospholipid) and even 10% loss is sufficient in itself to induce cell death. PC homeostasis is regulated by a balance between hydrolysis and synthesis. PC is hydrolyzed by phospholipase A2 (PLA2), PC-phospholipase C (PC-PLC) and phospholipase D (PLD). PC synthesis is rate-limited by cytidine triphosphate:phosphocholine cytidylyltransferase (CCT) that makes CDP-choline in the brain. TNF-α and IL-1 (IL-1a/s) are pro-inflammatory cytokines that are up-regulated in the brain after stroke. Phospholipases are induced by TNFa/IL-1 and mediate their cytotoxicity. TNF-α/IL-1 also down-regulate CCT. Thus, TNF-α/IL-1 alter lipid metabolism by increasing PC hydrolysis (stimulate phospholipases) and inhibiting PC synthesis (decrease CCT). The aim of the study is whether citicoline counteract the lipid alterations caused by cytokines after stroke.

Methods: Focal cerebral ischemia was induced for 1 hr in male spontaneously hypertensive rat (SHR) using middle cerebral artery occlusion (MCAO) by intraluminal suture technique. Citicoline liposomes were prepared by Lipex thermobarrel extruder pressurized with argon. Treatments: TNF-α ab (0.36 mg/kg i.v.), IL-1ra (20 µg/4 µL icv), free CDP-choline (500 mg/kg 0 & 3 h i.p.) and CDP-choline liposomes (18 mg/kg 0 & 3 h i.v.). TNF-α and IL-1s levels were measured by ELISA. mRNA expression by RT-PCR; Protein expression by Western blotting and infarction by TTC staining. PLA2 activity by radioactive assay, PC-PLC activity was by AMPLEX Red PC-PLC assay.

Results: tMCAO significantly increased sPLA2 mRNA and protein levels. CDP-choline, TNF-α antibody (TNF-α ab), or interleukin-1 receptor antagonist (IL-1ra) significantly attenuated sPLA2 protein levels. tMCAO resulted in loss of CCT protein as well as activity that were partially restored by citicoline. tMCAO up-regulated PLD1 and PLD2 over 7 d reperfusion. However, there were no changes in the cPLA2 and iPLA2 protein expression after tMCAO. PC-PLC activity was increased after tMCAO. The up-regulation of phospholipases (PLA2, PC-PLC, PLD) and down-regulation of CCT collectively resulted in the loss of PC and total phospholipids. Citicoline attenuated sPLA2 up-regulation and CCT down-regulation after tMCAO. TNF-α antibody and IL-1ra also significantly attenuated sPLA2 expression. Citicoline attenuated stroke-induced up-regulation of TNF-α and IL-1s levels. Liposome encapsulated CDP-choline significantly decreased infarction (by 62%) compared to equivalent dose of free CDP-choline (26% reduction) after tMCAO.

Conclusions: These results suggest that citicoline restores PC levels by differentially affecting phospholipases and CCT through cytokine alterations after stroke. Since all cells (neurons, astrocytes, endothelial cells etc., called the ‘neurovascular unit’) contain PC as a membrane building block, citicoline protection may arise from stabilizing the ‘neurovascular unit’. Advantages of citicoline liposomes in stroke include: 1) effective at low doses (18 mg/kg citicoline liposomes vs 500 mg/kg free citicoline), 2) the presence of GM1 ganglioside confers long-circulating properties by suppressing phagocytosis, 3) brain uptake of the drug is increased to ~23% of the injected dose, and 4) Citicoline liposomes deliver the agent intact to the brain, circumventing the rate-limiting CCT step. Our studies provide an integration of citicoline effects with cytokine biology and lipid metabolism in cerebral ischemia.

PHARYNGEAL COOLING DECREASES BRAIN TEMPERATURE AND INTRACRANIAL PRESSURE DURING RESUSCITATION IN MONKEYS

Yoshimasa Takeda, Motomu Kobayashi, Hideki Taninishi, Toshihiro Sasaki, Minako Arai, Kiyoshi Morita

Department of Anesthesiology and Resuscitology, Okayama University Medical School, Okayama, Japan

Introduction The development of a technique that enables brain temperature to be immediately and selectively reduced without affecting systemic circulation is needed to perform resuscitative hypothermia in a clinical situation. Since bilateral common carotid arteries exist at about 1 centimeter (C) from the pharynx, cooling the pharyngeal region decreases the carotid arterial temperature first and subsequently decreases brain temperature before lowering cardiac temperature. The present study was designed to evaluate the effects of pharyngeal cooling on brain temperature, systemic temperature, intracranial pressure and extracellular glutamate concentration during resuscitation in monkeys.

Methods Japanese monkeys (7.5+−2.4 kg) were divided into a control group (n=6) and a pharyngeal cooling group (n=3) and were anesthetised with 1% isoflurane. Cardiac arrest (12 minutes) was initiated with electrical stimulation. Changes in intracranial pressure (parietal region) and epidural (parietal region), sub-cortical (parietal region, 3 cm below the cortical surface) and rectal temperatures were monitored during a 30-minute resuscitation period. A micro dialysis probe was inserted into the parietal cortex for the measurement of glutamate concentration. In the pharyngeal cooling group, a pharyngeal cuff was inserted into the pharynx and was perfused with water (5 C degree) at the rate of 500 ml/minute from the onset of resuscitation for 30 minutes.

Results Cortical (33.3+−0.7 C degree) and sub-cortical (35.8+−0.5 C degree) temperatures decreased during cardiac arrest in both groups. In the control group, however, cortical and subcortical temperatures increased with the initiation of resuscitation and reached 35.1+−0.5 C degree and 36.5+−0.2 C degree, respectively, 30 minutes later. In the pharyngeal cooling group, cortical and sub-cortical temperatures decreased and reached 31.1+−1.5 C degree and 32.4+−1.3 C degree, respectively, 30 minutes later. Rectal temperature was unchanged in both groups. During the resuscitation period, the intracranial pressure increased to 30+−16 mmHg (control) and 14+−6 mmHg (pharyngeal cooling) and took 23+−9 minutes (control) and 6+−1 minutes (pharyngeal cooling) to become less than 10 mmHg. Extracellular glutamate concentration was unchanged during cardiac arrest in both groups and increased after the onset of resuscitation. The maximum concentration of extracellular glutamate was decreased in the pharyngeal cooling group (13 micro mol/l) compared to that in the control group (39 micro mol/l). The periods required for recovery of mean arterial blood pressure (>50 mmHg) were the same (4+−1 minutes after onset of resuscitation) in the two groups.

Conclusions Cortical and sub-cortical temperatures were decreased with the initiation of pharyngeal cooling without lowering of rectal temperature during the resuscitation period in monkeys. Intracranial pressure and extracellular concentration of glutamate were lower in the pharyngeal cooling group without affecting time required for recovery of mean arterial blood pressure (>50 mmHg).

NEUROPROTECTIVE EFFECTS OF INTRANASAL ERYTHROPOIETIN AFTER PERMANENT UNILATERAL ISCHEMIA IN MONGOLIAN GERBILS, WITH A THERAPEUTIC WINDOW UP TO 18 HOURS

Jorge Daniel Garia Salman¹, Iliana Sosa Teste², Yuneidys Mengana Tamos², Nelvis Subiros¹, Adriana Munoz⁴, Yanier Nunez⁴, Yamila Rodriguez Cruz⁵, Albertina Thomas², Barbara Gonzalez², Julio Cesar Garia Rodriguez², Rosa Maria Coro Antich¹, Katiuska Rodriguez²

¹INN Neurology National Institute, Havana, Cuba, ²CENPALAB, Havana, Cuba, ⁴CIDEM, Havana, Cuba, ⁵ISCM, Havana, Cuba

Human recombinant erythropoietin with low content of sialic acids was administered intranasally after permanent unilateral common carotid occlusion in Mongolian gerbils. Four ischemic treated groups were formed, in which the treatment began immediately, at 6, 12 and 18 hours after surgery. Treatment (120 IU/animal), lasts for 4 days, 30 IU/day. Habituation behavior of the exploratory activity was measured one day before and seven days after surgery. Survival rate at day 7 and behavioral test were compared with ischemic untreated and sham-operated animals. Survival rate was higher in immediately treated (68%), 6h (72%), and 12h (65%) than in ischemic untreated (46%) gerbils. Comparison between the habituation behavior before and after surgery revealed an increased habituation in shame-operated animals, as well as in all ischemic treated groups, indicating an erythropoietin-induced protection of learning and long term memory in survivors. When comparing the behavioral performance among groups seven days after surgery, was observed a loss of habituation in the ischemic-untreated group only. Results suggested that therapeutic window of intranasal erythropoietin with low content of sialic acids, could be extended to the first 12 hours of ischemia, although neuroprotective properties were observed even at 18h in less injured animals.

LONG-TERM OBSERVATION OF FREE RADICAL SCAVENGER EFFECTS ON CEREBRAL INFARCTION INDUCED BY NON-INVASIVE PHOTOTHROMBOSIS IN MICE

Shinji Nakajima^1,2, Syojiroh Sawada¹, Yuhji Morimoto³, Masashi Nibuya², Souichiroh Nomura²

¹Departments of Neurology and Psychiatry, National Defence Medical College, Tokorozawa, Saitama, Japan, ²1st Department of Internal Medicine, National Defence Medical College, Tokorozawa, Saitama, Japan, ³Department of Integrative Physiology and Nano-Medicione, National Defence Medical College, Tokorozawa, Saitama, Japan

Background and aims: Various drugs are used to remedy adverse events with spatiotemporal evolution in the acute phase of stroke, and recent clinical studies set the therapeutic assessment to the chronic stage with recovery. Longer experimental observation is also desirable in animal studies to assess therapeutic effects for the restoration of damaged tissues and neural functions. The aim of this study is to examine effects of free radical scavenger (edaravone, MCI-186) administered in the hyperacute stage of stroke using relatively small number of animals longitudinally.

Methods: C57BL/6J mice were anesthetized by the intra-peritoneal injection of 1% chloralose and 10% urethane mixture, keeping rectal temperatures at 37.5 degrees on the heating pad (Nihon Kohden, ATB-1100). Pphotothrombosis (PT) was created in the right parietal lobe by the stereotactic transcranial irradiation of He-Ne laser light (532 nm, 5 mW) after the injection of rose-bengal solution via the tail vein. Cerebral blood flow was measured using the laser-Doppler flow meter (Advance, ALF21). Animals were divided into the ischemic control (C) group, in which the laser was irradiated for 3-, 5, 10-, 15-, and 30-minute (N=6, in each group) to compare time-dependencies of PT formation, and edaravone (E) group (N=6) with 10-minute irradiation, where edaravone was administered twice, immediately before and 30-minute after PT (3 mg/kg IV, each time). Cerebral images were taken by the 9.4 T MRI (Bruker, AVANCE 400WB) at 1hour, 1- and 3-day, followed by 1- and 12-week sequentially in individual mice. Ischemic changes were recorded in T1+Gd-DTPA (PWI), Diffusion (DWI), and T2WIs (20mm FOV, 256×256 matrix, coronal slices), and sizes of ischemic loci were measured by the ROI analysis program. Microscopic observations were performed through the lubricated skull intermittently to analyze microvasculature changes after the acute ischemic load.

Results: MRI showed ischemic changes in PWI and DWI, and T2WIs at 1-hour with enlarged lesions dependent to irradiation time over 5-minute. T2WI lesions in 10-minute irradiation occupied 20.4±1.7 % (M±SEM) of the ipsilesional hemisphere in the C group, and decreased to 4.8±0.5 % (p<0.005) in the E group at 1-hour. Ischemic changes enlarged with edema to 27.7±4.5 % in the C group and 8.1±1.0 % (p<0.005) in the E group at 1-day. These edematous changes subsided at 1-week, and shrunk to small infarctions at 12-week. Microscopic observation showed extensive occlusions including arterioles, capillaries, and venules by platelets-thrombi immediately after PT with very gradual resolution and recanalization thereafter.

Conclusions: Laser irradiation in the present study created stereotypic ischemic lesions by multi-vessel PT in an irradiation-time dependent manner, which might be different from selective arterial PT with variable MRI lesions in cases of early recanalization. Free radical scavenger, edaravone, significantly decreased the size of photothrombotic lesions when observed by sequential MRI, showing the direct scavenging effect against singlet oxygen formed by photochemical reactions, as we already demonstrated in vitro and in vivo studies. Suppression of cerebral edema may be due to the reduction of the postischemic inflammatory process around venules through blocking the enhanced dissociation of NFkB and IkB by free radical oxygen species.

ISCHEMIC PRECONDITIONING RECOVERS NERVE GROWTH FACTOR, BRAIN-DERIVED NEUROTROPHIC FACTOR AND THEIR HIGH AFFINITY RECEPTORS AFTER FOREBRAIN ISCHEMIA

Tsong-Hai Lee¹, Jen-Tsung Yang², Yu-Shien Ko³

¹Department of Neurology, Chang Gung Memorial Hospital, Linkou Medical Center and Chang Gung University School of Medicine, Taoyuan, Taiwan-R.O.C., ²Department of Neurosurgery, Chiayi Chang Gung Memorial Hospital and Chang Gung University School of Medicine, Chiayi, Taiwan-R.O.C., ³Department of Cardiology, Chang Gung Memorial Hospital, Linkou Medical Center and Chang Gung University School of Medicine, Taoyuan, Taiwan-R.O.C.

OXIDATIVE STRESS CONTRIBUTES TO MEMORY IMPAIRMENT AND NEURONAL DAMAGE IN EXPERIMENTAL CEREBRAL HYPOPERFUSION

Eszter Farkas¹, Eva Mracsko², Monika Krecsmarik^1,2, Eszter Karg³, Botond Penke⁴, Ferenc Bari²

¹Department of Anatomy, School of Medicine, University of Szeged, Szeged, Hungary, ²Department of Physiology, School of Medicine, University of Szeged, Szeged, Hungary, ³Department of Pediatrics, School of Medicine, University of Szeged, Szeged, Hungary, ⁴Department of Medical Chemistry, School of Medicine, University of Szeged, Szeged, Hungary

The permanent, bilateral occlusion of the common carotid arteries of rats (2VO) has been widely used as a model for chronic cerebral hypoperfusion as it occurs in aging and dementia. 2VO in experimental animals leads to memory dysfunction and a progressive loss of hippocampal pyramidal neurons. However, the potential mechanisms of neuronal damage after the onset of 2VO have not been identified. The present investigation set out to determine whether oxidative stress may contribute to the 2VO-induced memory impairment and neurodegeneration. Treatments with the antioxidant alpha-tocopherol were employed as pharmacological means to counteract and, thus, confirm the incidence of oxidative stress in the 2VO model.

2VO was imposed on male Wistar rats (n=70). Sham-operated animals (SHAM) served as controls (n=66). Half of the animals were pre- or post-treated repeatedly with alpha-tocopherol (5 × 100 mg/kg daily, i.p.), the other half received soybean oil, the vehicle of alpha-tocopherol. One week after the onset of 2VO, the spatial learning capacity of the animals was assessed in the Morris water maze (n=65). After testing, hippocampal slices were stained with cresyl violet in order to examine the pyramidal cell layer integrity. The density of MAP-2-positive dendrites and the level of OX-42-labeled microglial activation were determined immunocytochemically. The amount of nitrotyrosine (generated by the free radical peroxinitrite) was determined with Western blotting (n=32). Alpha-tocopherol concentration was determined in the peripheral tissues, blood and brain with HPLC (n=39).

Alpha-tocopherol attenuated the 2VO-induced learning impairment. The treatments with alpha-tocopherol, and particularly the post-treatment, prevented the 2VO-induced pyramidal cell death and the activation of microglia in the hippocampus CA1 region, and the degeneration of MAP-2-positive dendrites in the CA3 region, determined 2 weeks after the onset of 2VO. The amount of nitrotyrosine in the hippocampus was moderately elevated as compared with SHAM controls on days 3 and 7 (relative optical density: 0.87±0.11 vs. 0.75±0.10, and 0.98±0.03 vs. 0.86±0.10, respectively), but was not affected on day 1 after the initiation of 2VO. The alpha-tocopherol concentration was considerably elevated in the peripheral tissues and the blood, while only moderate accumulation of alpha-tocopherol was detected in the hippocampus (31.99±1.79 vs. 28.98±1.36 nmol/g).

The data suggests that oxidative neuronal injury plays a role in the 2VO-related neurodegenerative processes. Further, alpha-tocopherol appears beneficial in combating the cerebral hypoperfusion-related neuronal damage. Finally, the observation that the post-treatment achieved better neuroprotection than the pretreatment may imply that the increased availability of alpha-tocopherol at the time of an ongoing ischemic injury may be required for a better neurological outcome from the insult.

Grant information: This work was supported by the Hungarian Scientific Research Fund (OTKA) grants F042803 and K063401, and the Hungarian National Office for Research and Technology (NKTH).

ISCHEMIC PRECONDITIONING CONFERS NEUROPROTECTION VIA GABA SYNAPSES

Richard Anthony DeFazio, Ami P. Raval, Miguel A. Perez-Pinzon

Department of Neurology, UM Miller School of Medicine, Miami, FL, USA

Preconditioning activates endogenous neuroprotective mechanisms against ischemia. Mild ischemic insults (ischemic preconditioning) confer protection against subsequent insults that would otherwise be lethal. Understanding the mechanisms of these endogenous protective mechanisms could lead to novel therapeutic strategies. Previous studies showed ischemic preconditioning increased the release of GABA during ischemia (Dave et al. 2005). We tested the hypothesis that preconditioning acts via GABA synapses. To define the changes in GABA synapses, whole cell patch clamp recordings of GABA-A receptor mediated postsynaptic currents were obtained 48 hours after preconditioning in the presence of tetrototoxin and glutamate receptor antagonists (miniature postsynaptic currents). Ischemic preconditioning significantly decreased the interval between synaptic currents (29.4 +/− 4.0 msec vs. 65.8 +/− 8.4 msec; p<0.01, N=8), suggesting an increase in the number of active synapses. Ischemic preconditioning depends on activation of the epsilon PKC signaling pathway (Raval et al. 2003). Preconditioning can be mimicked by epsilon PKC activators (pharmacological preconditioning) and abolished by antagonists of epsilon PKC. In contrast to ischemic preconditioning, pharmacological preconditioning increased the amplitude of GABA synaptic currents (−92.8 +/− 14.1 pA vs. −58.0 +/− 5.1 pA; p<0.05, N=7) without affecting the frequency. Thus, both forms of preconditioning appear to converge on the GABA synapse. We confirmed this conjecture by testing the hypothesis that partial blockade of GABA-A receptors abolishes neuroprotection conferred by preconditioning. Co-application of the GABA-A receptor antagonist bicuculline (100 nM; IC50 3 uM) during lethal ischemia abolished both forms of preconditioning (N=5). We conclude that modifications of GABA synapses represent a major target of endogenous neuroprotective signaling pathways evoked by preconditioning.

Supported by NIH NINDS R01 NS34773, R01 NS54147, R01 NS45676, P01NS005820 and the Stanely J. Glaser Research Foundation.

REDUCTION OF SECONDARY NEURONAL DAMAGE BY THE IMMUNOSUPPRESSANT MYCOPHENOLATE MOFETIL AFTER CONTROLLED CORTICAL IMPACT INJURY IN RATS

F. Dehghani², Beat Alessandri¹, M. Nazari², N.P. Hailer³, H.-W. Korf¹, Oliver Kempski¹

¹Insitute for Neurosurgical Pathphysiology, Johannes Gutenberg-University of Mainz, Mainz, Germany, ²Institute of Anatomy II, Johann Wolfgang Goethe-University, Frankfurt a.M., Germany, ³Institute for Surgical Sciences, University Hospital Uppsala, Department of Orthopaedics, Uppsala, Sweden

Background/Aim: Lesions of the CNS induce specific morphological and functional changes of glial cells such as proliferation, cell shape transformation and secretion of proinflammatory cytokines and NO. It has been suggested that suppression of glial cell activation in the aftermath of the neuronal injury inhibits the development of secondary damage and contributes to improved neuronal survival. In previous studies with organotypic hippocampal slice cultures and glial single cell cultures we have shown that mycophenolate mofetil (MMF) reduced the number of degenerating neurons after excitotoxic injury, inhibited the proliferation of glial cells and suppressed the activation state of microglial cells. To prove that MMF exerts equally potent neuroprotective effects in vivo, we tested this compound in a model of traumatic brain injury in rats.

Method: To inflict a focal injury on brain tissue, we used a model of controlled cortical impact (CCI). 8-12-week-old Sprague-Dawley rats (350g-450g) received a unilateral CCI injury (5 m/sec velocity, 2.5mm depth, 200 ms dwell time) and vital parameters were monitored for up to 2h after the injury. Injured animals were treated with vehicle or MMF up to 14 days. Animals were sacrificed 4h (n=4/group), 1 (n=4/group), 3 (n=4/group) or 14 days (n=8/group) after the trauma. The lesion volume was calculated by quantitative morphometry on serial sections stained with hematoxylin/eosin. Sections were also stained with GFAP and NeuN for glial and neuronal cell types, respectively, with IB4 (isolektin B4) for microglia and with ki-67 for proliferating cells.

Results: In vehicle treated animals CCI caused a moderate cortical contusion, swelling (1 day) and massive neuronal cell loss underneath the impact site. The lesion volume showed an average of 31.41 mm3 and the lesion edges were poorly defined. Immunohistochemistry showed microgliosis, astrogliosis, massive invasion of blood borne monocytes and a high number of proliferating glial cells. In contrast, MMF treatment led to a significantly reduced lesion volume of 19.41 mm3. The lesion was sharply demarcated against neighboring NeuN immunoreactive neurons. Immunohistochemically a strong suppression of glial activation and proliferation was visible and the number of proliferating glial cells was attenuated.

Conclusion: Our findings show that the immunosuppressant MMF protects in vivo neurons from secondary neuronal damage, inhibits the proliferation of astrocytes and reduces the activation state of glial cells resulting in a smaller lesion volume. MMF may be used as a neuroprotective agent in treatment of acute traumatic CNS pathologies. This study was supported by Dr. August Scheidel-Stiftung.

ERK1/2-P90RSK-MEDIATED PHOSPHORYLATION OF NA+/H+ EXCHANGER ISOFORM 1 IN ISCHEMIC NEURONAL DAMAGE

Jing Luo^1,2, Douglas B. Kintner², Gary E. Shull³, Dandan Sun^1,2

¹Department of Physiology, University of Wisconsin Medical School, Madison, WI, USA, ²Department of Neurosurgery, University of Wisconsin Medical School, Madison, WI, USA, ³Department of Molecular Genetics, Biochemistry and Microbiology, University of Cincinnati, Cincinnati, OH, USA

Background and Aims: We recently reported that Na+/H+ exchanger isoform 1 (NHE1) activity plays a role in ischemic cell damage and inhibition of NHE1 activity is neuroprotective in both in vitro and in vivo ischemic models (J. Neurosci., 2005). However, the cellular mechanisms underlying activation of NHE1 activity remain unknown. In this study, we investigated ERK1/2-mediated stimulation of NHE1 in neurons following in vitro ischemia.

Methods and Results: Two hours of oxygen and glucose deprivation (OGD) did not change expression of phosphorylated ERK1/2 (p-ERK1/2). In contrast, 10-60 min reoxygenation (REOX) triggered a significant increase in p-ERK1/2 levels. The increase in p-ERK1/2 was accompanied by translocation of p-ERK1/2 from cytosol to nuclei following OGD/REOX. The MEK inhibitor U0126 (10 µM) abolished elevation of p-ERK1/2 expression and its nuclear translocation. Moreover, NHE1 activity was significantly increased at 10-60 min REOX after 2 h OGD. OGD/REOX not only increased Vmax of NHE1 for H+ but also shifted Km toward a higher pHi. OGD/REOX-mediated changes in NHE1 kinetics were absent when MEK was inhibited by U0126. We speculated that changes in NHE1 kinetics result from phosphorylation of NHE1 (p-NHE1) via ERK-p90RSK signaling pathways. After 10 min REOX following 2 h OGD, not only was p-NHE1 increased, but there was a concurrent increase in phosphorylation of p90RSK (p-p90RSK), a known NHE1 kinase. Inhibition of MEK activity with U0126 abolished elevation of both p-NHE1 and p-p90RSK. Either genetic ablation or pharmacological inhibition of NHE1 activity reduced neuronal death following OGD/REOX. U0126 provided a similar level of neuroprotection.

Conclusions: Taken together, these results suggest that NHE1 is phosphorylated via activation of ERK1/2- p90RSK pathways and contributes to ischemic neuronal damage.

Key words: oxygen and glucose deprivation, HOE 642, p90RSK, protein phosphorylation, U-0126

GREATER OXIDATIVE STRESS SELECTIVELY IN CA1 ASTROCYTES IS CENTRAL TO THE SELECTIVE VULNERABILITY OF CA1 NEURONS AFTER TRANSIENT FOREBRAIN ISCHEMIA

Yi-Bing Ouyang, Ludmila A. Voloboueva, Li-Jun Xu, Rona G. Giffard

Department of Anesthesia, Stanford University School of Medicine, Stanford, CA, USA

Background: Transient global ischemia, as with cardiac arrest, causes loss of CA1 hippocampal neurons 2–4 days later, while nearby dentate gyrus (DG) neurons are relatively resistant. For many years selective neuronal vulnerability has been presumed to reflect the inherent vulnerability and connectivity of the neuronal population. Whether differential astrocyte vulnerability contributes to CA1 neuronal death has been much less studied, and the role of astrocytes in the different hippocampal regions in ischemic injury is uncertain. We investigated the role of astrocytes in selective neuronal loss using in vivo and in vitro preparations. Methods: Rat 2-vessel occlusion plus hypotension to 45 mm Hg for 10 minutes forebrain ischemia was used. Following different reperfusion intervals from 2 hr to 72 hr, brains were harvested for immunohistochemistry, or for western analysis. Primary cultures of astrocytes isolated from either CA1 or the DG region, and hippocampal slice cultures were also used. Protein expression was determined by western blot or immunohistochemistry, and live cell imaging was performed to assess ROS with hydroethidine or mitochondrial membrane potential with TMRE. Results: CA1 astrocytes were found to be more sensitive to ischemia than DG astrocytes. In rats subjected to transient forebrain ischemia, CA1 astrocytes lose immunoreactivity for GFAP, S100a and glutamate transporter GLT-1, as well as glutamate transport activity, within a few hours of reperfusion, but without astrocyte cell death. This occurs well before any evidence of neuronal injury. Using subregion specific astrocytes cultures we observed that CA1 astrocytes show early increases in mitochondrial free radicals and reduced mitochondrial membrane potential after stress, while similar changes were not observed in DG astrocytes. Thus greater oxidative stress may contribute to the selective CA1 astrocyte changes, including early loss of GLT-1, which is sensitive to oxidation. The respiratory chain inhibitor rotenone was used to increase mitochondrial production of ROS in culture and this resulted in rapid decrease of GLT-1 immunostaining in CA1 astrocytes. To determine whether the early loss of GLT-1 contributes to CA1 neuronal loss we increased GLT-1 expression in astrocytes with ceftriaxone. This significantly reduced the extent of CA1 neuronal loss from forebrain ischemia in vivo and in slice cultures. Discussion: Oxidative stress is a well recognized contributor to postischemic injury. In the case of forebrain ischemia, while some ROS may originate from neurons, it also appears that some ROS may originate from astrocytes. An important consequence of the toxicity of the ROS produced from either source is regionally selective astrocyte impairment. A clear difference was seen both in vivo and in vitro between the vulnerability of CA1 and DG astrocytes. We suggest that greater oxidative stress and loss of GLT-1 function selectively in CA1 astrocytes is central to the well known selective vulnerability of CA1 neurons.

IN VIVO DETECTION OF SUPEROXIDE-ANION RADICAL BY A NEWLY DEVELOPED ELECTROCHEMICAL BIOSENSOR IN BRAIN ISCHEMIA-REPERFUSION MODEL OF RATS

Motoki Fujita¹, Susumu Yamashita¹, Kenji Fujimoto¹, Kazumi Kumagai¹, Kotaro Kaneda¹, Yasutaka Oda¹, Ryosuke Tsuruta¹, Shunji Kasaoka², Makoto Yuasa³, Tsuyoshi Maekawa^1,2

¹Advanced Medical Emergency and Critical Care Center, Yamaguchi University Hospital, Ube, Yamaguchi, Japan, ²Department of Emergency and Critical Care Medicine, Yamaguchi University Graduate School of Medicine, Ube, Yamaguchi, Japan, ³Department of Pure and Applied Chemistry, Faculty of Science and Technology, Tokyo University of Science, Tokyo, Japan

Introduction: The reactive oxygen species (ROS) play an essential role in vivo, but the excessive production of it leads to an oxidative stress and a tissue injury. In ROS, a superoxide anion radical (O2-?) is the key radical because the generation of almost ROS derives from O2-?. Recently, we exploited an all-synthetic biochemical sensor that can detect O2-? specifically. This sensor is coated with Bromoiron (III) meso-tetra(3-thienyl)porphyrin ([FeBr(ttp)]) that mimics cytochrome c and is stable in vivo. To evaluate the production and the attenuation of O2-? in vivo, a newly developed electrochemical biosensor was used in forebrain ischemia-reperfusion rats with or without superoxide dismutase (SOD).

Materials and Methods: Fourteen Male Wister rats (260-280g) were randomly assigned to two groups, i.e., the reperfusion group (n=7) and the SOD group (n=7). A microdialysis probe for glutamate was embedded in the forebrain parenchyma on the previous day of the experiment under pentobarbital anesthesia. The rats were anesthetized by 4% of isoflurane and 100% of O2. They were subjected to tracheostomy and ventilated mechanically. The top of electrochemical O2-? sensor was inserted into the left internal jugular vein through the thyroid vein and the other venous branches were ligated. After the operation, the anesthesia was changed to 0.9% of isoflurane and 40% of O2 and the O2-? value was measured with Active O2-?TM (BORON JAPAN CO.LTD., Noda, Japan) continuously. After the stabilization of the blood gas and O2-? value, the forebrain ischemia-reperfusion was made by bilateral common carotid artery occlusion with hemorrhagic hypotension for 20 min and by releasing the occlusion with returning the shed blood. In the SOD group, SOD (5 mg/kg) was administrated intravenously at 20 minutes after reperfusion. The measurements were continued for 1 hour after reperfusion.

Results: The detection current of O2-? increased after the reperfusion in both groups. In the SOD group, the elevation of O2-? current attenuated after the administration of SOD (Figure 1). The electric charge of O2-? (Q value) was integrated in every 20 minutes after reperfusion. In the SOD group, Q value was decreased significantly during 40-60 minutes after the reperfusion (23.7±8.8 v.s. 13.1±5.9 micro C, p<0.05)

Conclusions: The excess production of superoxide anion radical was detected by a newly developed electrochemical biosensor in vivo and rapid reduction of superoxide by intravenous administration of SOD was confirmed. This is the first report of the real-time measurement of superoxide in vivo.

NEUROPILIN-1 IS A DIRECT TARGET OF THE TRANSCRIPTION FACTOR E2F1 IN ISCHEMIC NEURONAL DEATH

Susan Jiang¹, Melissa Sheldrick^1,2, Angele Desbois¹, Jackie Slinn¹, Sheng T. Hou^1,2

¹NRC Institute for Biological Sciences, National Research Council Canada, Ottawa, ON, Canada, ²Department of Biochemistry, University of Ottawa, Ottawa, ON, Canada

ROSIGLITAZONE PROTECTS BRAIN FROM ISCHEMIC INSULT BY ACTIVATING PEROXISOME PROFERATOR-ACTIVATED RECEPTOR GAMMA (PPAR-R) AND ATTENUATING APOPTOTIC CELL DEATH

Jui-Sheng Wu, Yi-Tong Chen, Wai-Mui Cheung, Teng-Nan Lin

Neuroscience Division, Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan-R.O.C.

Background and Aims: Peroxisome proliferator-activated receptor gamma (PPAR-r) is a nuclear membrane-associated transcription factor that governs various genes, including lipid biosynthesis, glucose metabolism and inflammatory. It has been shown that activation of PPAR-r attenuates myocardial ischemia/reperfusion damage. Rosiglitazone (Rosi), a TZD and synthetic agonist of PPAR-r, is a clinical drug for type 2 diabetes. Several lines of evidence suggest that TZDs may protect brain against ischemia/reperfusion injury; however, the underlying mechanisms are not clear. Previously, we have shown that intra-ventricular infusion of 15-deoxy-Delta (12, 14)-PGJ2 (15d-PGJ2; a stable PGD2 metabolite and physiological agonist of PPAR-r) reduced cerebral infarction, and this protective effect were abrogated by inhibitor of PPAR-r. The aim of this study is to investigate the cellular and molecular mechanisms underlying the protective effect of Rosi-PPAR-r against ischemic cerebral infarction.

Methods: Rats were subjected to 30 minutes of transient middle cerebral artery occlusion and intra-cerebral ventricular infusion of Rosi. For in vitro study, Rosi was added to Murine N2-A neuroblastoma cells either alone or in various combinations before subject to OGD. Western blot analysis was used to study the temporal profiles of proteins. Apoptotic cell death was analyzed by flow cytometry. PPAR-r activity was analyzed by luciferase reported assay. Plasmid and siRNA transfection were used to over-express or silence PPAR-r.

Results: Our results indicated that Rosi significantly reduced infarct volume, even when Rosi was administered 2 hours after a 30-min transient cerebral ischemia, via attenuating ischemia-induced caspase-3 activation and PARP-1 degradation. Moreover, Rosi also protects N2-A against oxygen-glucose deprivation (OGD)-induced cell death by reducing OGD-induced ROS production, depolarization of mitochondrial membrane potential (MMP), caspase-3 activation, PARP-1 degradation and apoptosis. The protective effects of Rosi were abolished in the present of PPAR-r antagonist, were mimicked by PPAR-r over-expression and were abolished in the presence of siRNA against PPAR-r.

Conclusions: In summary, our results suggest that Rosi, by activating PPAR-r, protects neuronal cells against ischemic-induced apoptosis.

HYPERLIPIDEMIA EXACERBATES ISCHEMIC BRAIN INJURY VIA PROMOTING CD36-MEDIATED FOAM CELL FORMATION

Sunghee Cho^1,2, Aaron Tolhurst¹, Maria Febbraio³, Eunhee Kim¹

¹Burke Medical Research Insitute, White Plains, NY, USA, ²Department of Neuroscience, Weill Medical College of Cornell Univesity, New York, NY, USA, ³Department of Cell BIology, Cleveland Clinic, Cleveland, OH, USA

Oxidative stress is implicated in the pathogenesis of ischemia-reperfusion injury. Oxidative modification of low-density lipoprotein (oxLDL) is a key step for the generation of pro-atherogenic ligands. Subsequent uptake of oxLDL by scavenger receptors leads to macrophage foam cell formation, atheroma, and a chronic pro-inflammatory state. Among multiple scavenger receptors, CD36 is one of the major scavenger receptors responsible for the uptake of oxLDL and foam cell formation. A recent study demonstrated that CD36 mediates free radical production and ischemic injury (Cho et al., 2005). It has been suggested that ischemic stroke shares common risk factors with coronary heart diseases. Among the risk factors, hypercholesterolemia is common in humans and a known risk factor for stroke. Based on the fact that CD36 i) is involved in ischemic injury and ii) functions in the uptake of oxLDL, we investigated whether the hyperlipidemic condition exacerbates ischemia-induced injury and whether the exacerbation occurs via up-regulating CD36 pathways.

To produce hyperlipidemic (HL) mice, 5 week-old ApoE null mice, which are spontaneously hyperlipidemic, were fed a high fat diet (TD88137, Halen Taklad) for 8 weeks. Age matched wild-type mice fed standard chows were used as normal lipid (NL) mice. Weight and cholesterol levels were determined at the end of diet and then mice were subjected to transient middle cerebral artery occlusion (MCAO) for 30 min using an intraluminal filament. Infarct volume and swelling were determined 3 days after ischemia in cresyl violet stained sections with correction for brain swelling. CD36 gene and protein expression were determined either in the post-ischemic brain tissue or mouse peritoneal macrophages by real-time RT-PCR and Western blot analysis. In addition, foam cell formation, a CD36-mediated function, was assessed in the post-ischemic brain sections.

Compared to the NL mice, significant increases in body weight and plasma cholesterol levels in HL mice were observed (weight, 35.8±0.6 vs 25.5±0.08 gr: Cholesterol: 725.80±54.82 vs 119.22±4.96 mg/dl, p<0.05, n=15). HL mice also exhibited larger infarct size and increased % hemispheric swelling (infarct, 64.9±2.9 vs 42.5±4.3 mm3 p<0.05: Swelling, 29.8±4.4 vs 8.5±1.3%, p<0.05 n=7). Because CD36 expression mainly occurs in microglia/macrophages within the infarct territory, CD36 expression was determined in peritoneal macrophages harvested from NL and HL mice. Compared to NL mice, CD36 expression was increased in HL mice prior to and 3 days after ischemia (Pre (n=4): 1.56±0.19 vs 1.00±0.13: Post (n=5): 2.74±0.41 vs 1.00±0.27, p<0.05). Importantly, in contrast to NL mice, numerous lipid-laden foam cells were found in the penumbra in HL mice (Figure), indicating that hyperlipidemic condition promotes foam cell formation in the post-ischemic brain.

The data suggest that hyperlipidemia exacerbates ischemia injury via upregulation of CD36 expression/function in post-ischemic brain. Targeting CD36 receptors may be a useful strategy to reduce ischemic injury in the hyperlipidemic condition. Supported by NIH grants HL82511.

IN VITRO PRECONDITIONING OF EMBRYONIC STEM CELLS PROMOTES CELL SURVIVAL AND FUNCTIONAL RECOVERY AFTER TRANSPLANTATION INTO THE ISCHEMIC RAT BRAIN

Ling Wei¹, Michelle Theus¹, Shan Ping Yu^1,2

¹Department of Pathology, Medical Unviversity of South Carolina, Charleston, SC, USA, ²Department of Pharmaceutical Sciences, Medical Unviversity of South Carolina, Charleston, SC, USA

Hypoxic preconditioning (HP) and stem cell transplantation have been studied as two individual therapies. We now combine these methods to achieve increased therapeutic effects after ischemic stroke. Cultured embryonic stem (ES) cells during neural induction with retinoic acid (RA) forms embryoid bodies (EBs). In HP group, EBs were exposed to low oxygen (1% O2, 8 h) during the final stage of RA induction. The sublethal in vitro hypoxia pre-treatment significantly enhanced the tolerance of ES cell derived neural progenitor cells (ES-NPCs) and bone marrow mesenchymal stem cells (BMSC) to serum deprivation-induced cell death, decreased caspase-3 activation, increased secretion of erythropoietin (EPO) and upregulated expression of bcl-2, glial-derived neurotrophic factor (GDNF), HIF-1a, erythropoietin receptor (EPOR), neurofilament (NF), and synaptophysin in ES-NPCs. One week after RA, expression of neurofilaments and synaptophysin were noticeably higher in HP-treated ES-NPCs compared to control cells. The HP-initiated protection lasted for at least 6 days; it was diminished by blocking EPOR. Instead of hypoxia, rhEPO pre-treatment also significantly reduced caspase-3 activation and cell death. In vivo experiments confirmed that transplanted HP-primed ES-NPCs survived better in the ischemic brain. Fourteen days later, transplanted cells were labeled by NeuN or GFAP. The HP treated, ES-NPCs-derived neuronal cells showed extended neurite processes and formed neuronal networks 2 weeks after transplantation. HP-primed ES-NPCs conferred accelerated recovery in a number of sensorimotor activities. Transplantation of ES-NPCs improved the rotarod test score 7 to 35 days after transplantation. The enhanced motor functional benefit lasted for at least 35 days. We propose that the HP/EPO-priming strategy presents a unique and beneficial method for improving clinical potential of cell transplantation therapy.

INTRAVITAL IMAGING REVEALES THAT TRANSIENT FOREBRAIN ISCHEMIA INDUCES LONG-LASTING INCREASE IN INTRACELLULAR CALCIUM ION CONCENTRATION IN RAT HIPPOCAMPAL CA1

Seiji Yamamoto, Yong Wang, Atsuo Miyakawa, Takashi Sakurai, Kyoko Ibaraki, Susumu Terakawa

Photon Medical Research Center, Hamamatsu University School of Medicine, Hamamatsu, Japan

INTRODUCTION: In cultured neurons, the increase in intracellular Ca2+ concentration ([Ca2+]i) plays a crucial role in excitotoxicity. However, in whole animals, calcium response following ischemia remains to be established. Especially, change in [Ca2+]i in the hippocampus upon ischemia remains unclear, since no method has been available to measure the [Ca2+]i in the deep structure of brain. In the present study, we sought to determine, in rats, the [Ca2+]i changes following transient forebrain ischemia in hippocampus.

METHODS: In anesthetized, spontaneously ventilated SD rats (300 g), bilateral vertebral arteries were coagulated, and transient forebrain ischemia was induced by occluding bilateral common carotid arteries with balloon occluders. To obtain the intravital fluorescence image, we employed an imaging fiber bundle (1.0 mm outer diameter) including 20,000 fibers (4.5 µm outer diameter) coupled to the microlens-attached Nipkow-disk scanner (CSU-21, Yokogawa, Japan) equipped with 10x objective lens. This fiber-coupled confocal microscope (FCM) is capable to observe the confocal image just under the fibers (1). FCM was inserted stereotaxically into CA1 and CA3 regions, respectively. Fluo-3/AM (440 µM, 2.5 µl), a [Ca2+]i sensitive fluorescent dye, was microinjected into the brain, and observed every 30 s before and after 10-min forebrain ischemia with FCM. Since the hemoglobin absorbs both excitation and emission light, ischemia may affect the fluorescent intensity of Fluo-3 inside the brain. To analyze the effects of ischemia on the fluorescence intensity, the fluorescent beads (0.2 µm diameter, similar excitation and emission to those of Fluo-3) were microinjected into hippocampus and the images were taken before and after 10-min ischemia in the same manner as did in [Ca2+]i imaging.

RESULTS: Four-vessel occlusion decreased CBF (mean±SD%) measured by laser-Doppler flowmetry, to 25±18 (n=6) in CA1 and to 20±11 (n=6) in CA3. They did not significantly differ. 10-min forebrain ischemia increased fluorescence intensity (mean±SD%) of the beads in the hippocampus up to 108±5 (n=4), which declined to the baseline level following reperfusion. In all measurements, the ischemia showed significant increase (102±2, n=12 to 106±4, n=84, p<0.01). In [Ca2+]i imaging, therefore, the change in fluorescence during ischemia was corrected. During 10-min ischemia, the fluorescence intensity of Fluo-3 slightly increased up to 110±9, and then increased to 134±19 at the end of 20-min reperfusion in CA1 (n=4). In contrast, in CA3 (n=5), the intensity did not change upon ischemia, and at reperfusion slightly elevated to 108±4 without long-lasting increase. During 14-20 min reperfusion, the [Ca2+]i increase in CA1 was statistically larger (p < 0.05) than that in CA3.

COMMENTS: In CA1, but not in CA3, transient forebrain ischemia induced long lasting Ca2+ increase in situ, whereas it produced similar CBF changes in these regions. Since the [Ca2+]i increase was significantly larger in CA1 during reperfusion period, it might reach to toxic levels and then mediate the vulnerability of CA1 against ischemia.

Supported by the grants #16390407(SY), #16047212(SY), #18053010(SY)

PROTEOMIC ANALYSIS OF APOPTOTIC CELL DEATH

Duncan Short, Ian Heron, Jui-Lee Birse-Archbold, Lorraine Kerr, John Sharkey, James McCulloch

ACE and Centre for Neuroscience Research, University of Edinburgh, Edinburgh, Scotland, UK

Apoptosis contributes to cell death after cerebral ischaemia. A quantitative proteomics approach has been employed to define alterations in protein levels in apoptosis induced with staurosporine. Human neuroblastoma derived SH-SY5Y cells were treated with staurosporine (500nM for 6 hours) to induce apoptosis. Cell viability, morphology and cytochrome C release was defined in this model system prior to proteomic analysis. Quantitative two dimensional gel electrophoresis was used to determine the altered protein levels with MALDI-TOF mass spectrometry identification of proteins. For each treatment condition(vehicle or staurosporine),8-9 gels were prepared independently and means+/−SD computed. Of the 154 proteins analysed, thirteen proteins were significantly altered as a result of the apoptotic stimulus. Ten of the proteins showed a statistically significant increase in level with staurosporine and these were identified as heat shock cognate 71 (Hsc71)(increased by 42%), two isoforms of heat shock protein 70 (Hsp70)(increased by 77&56%), glucose regulated protein 78 (GRP78)(increased by 81%), F-actin capping protein(increased by 151%), stress-induced phosphoprotein 1(increased by 64%), chromatin assembly factor 1 (CAF 1)(increased by 155%), protein disulphide isomerase A3 precursor (PDI A3)(increased by 69%), transitional endoplasmic reticulum ATPase (increased by 82%)and actin interacting protein 1 (AIP 1)(increased by 64%). Three proteins which displayed significant decrease in levels with staurosporine were identified as tubulin(decreased by 45%), vimentin(decreased by 62%) and glucose regulated protein 94 (GRP94)(decreased by 27%). The functional roles ascribed to the proteins which were significantly altered during staurosporine induced apoptosis were cytoskeletal (2 proteins), cytoskeletal assembly (2 proteins), protein folding (7 proteins), mitosis (1 protein) and chromatin assembly (1 protein) and by subcellular location into cytoplasmic (6 proteins), cytoplasmic and nuclear (2 proteins), endoplasmic reticular (4 proteins) and nuclear (1 protein). The functional roles and subcellular locations of these proteins collectively indicate that staurosporine induced apoptosis provokes induces an unfolded protein response involving molecular chaperones, co-chaperones and structural proteins indicative of Endoplasmic Reticulum (ER) stress.

VASOPROTECTIVE EFFECTS OF INOS-DERIVED NO AND NOX2-DERIVED SUPEROXIDE IN ISCHEMIC PRECONDITIONING

Alexander Kunz, Laibaik Park, Takato Abe, Josef Anrather, Ping Zhou, Costantino Iadecola

Division of Neurobiology, Weill Medical College of Cornell University, New York, NY, USA

Background: Ischemic preconditioning (PC) is a phenomenon in which a sublethal stimulus confers tolerance toward a subsequent lethal ischemic insult. For example, systemic administration of lipopolysaccharide (LPS) reduces the injury produced by focal cerebral ischemia (JCBF&M 25:493,2005). LPS-induced PC requires iNOS for its expression, but the mechanisms by which iNOS-derived NO mediates the tolerance remain unclear. In systemic vessels, LPS improves vascular reactivity (EJP 377:209,1999), raising the possibility that LPS acts by improving CBF. NO exerts many of its actions by reacting with the radical superoxide to form peroxynitrite (PN). Therefore, we investigated whether vasoprotective effects of NO play a role in LPS-induced PC and, if so, whether reactive oxygen species (ROS) and PN are involved.

Methods: Male C57Bl/6 mice (n=5-7/group) were treated with LPS (0.5 mg/kg; ip) and subjected to transient middle cerebral artery occlusion (MCAO) 24hrs later. Ischemic CBF was monitored by laser-Doppler flowmetry in the neocortex spared from infarction by PC. CBF reactivity to whisker stimulation or neocortical application of acetylcholine (ACh; 10µM), was tested in the post-ischemic neocortex 1h after reperfusion. Infarct volume was measured 72hrs after MCAO in thionine-stained sections.

Results: LPS improved CBF (Fig. A) and reduced infarct volume (Fig. B). Following MCAO, the increase in CBF induced by whisker stimulation (Fig. C) and Ach were attenuated (Ach: −56±8%, p<0.05) However, LPS prevented these cerebrovascular abnormalities (Fig. C). The beneficial effects of LPS on CBF, infarct volume and CBF reactivity were not observed in wild type (WT) mice treated with the iNOS inhibitor aminoguanidine (Fig. A-C) or in iNOS-null mice (e.g. infarct volume: Veh-iNOS-/−: 25±5; LPS-iNOS-/−: 28±2). LPS increased ROS production (53±24%; p<0.05), assessed with hydroethidine and upregulated the immunoreactivity for the PN marker 3-nitrotyrosine (69±14%; p<0.05). These effects were not observed in mice lacking the nox2 subunit of the ROS-producing enzyme NADPH oxidase (p>0.05). Furthermore, 3-nitrotyrosine immunoreactivity was abolished by treatment of WT mice with the PN decomposition catalyst FeTPPS (20mg/kg; i.v.). The beneficial LPS effects on CBF and infarct volume were not observed in nox2-null mice (p>0.05) and were attenuated in WT mice treated with FeTPPS (Fig. A-C).

Conclusions: LPS PC is associated with marked neuroprotection and preservation of penumbral CBF. The CBF improvement in regions at risk for infarction can be attributed to a beneficial effect of LPS on cerebrovascular reactivity, leading to improved intraischemic CBF. The vasoprotective effect of LPS is mediated by PN formed by iNOS-derived NO and NADPH oxidase-derived superoxide. Thus, PN, in addition to its deleterious roles, can also be protective by improving cerebrovascular function and ischemic CBF.

THE ROLE OF ACTIVATED MICROGLIA/MACROPHAGES IN NEONATAL FOCAL STROKE

Zinaida Vexler¹, Andra Dingman¹, Sarah Lee¹, Joel Faustino¹, Nikita Derugin², Takashi Wada⁴, Michael Wendland³

¹Department of Neurology, UCSF, San Francisco, CA, USA, ²Department of Neurosurgery, UCSF, San Francisco, CA, USA, ³Department of Radiology, UCSF, San Francisco, CA, USA, ⁴Kanazawa University, Kanazawa, Japan

Background and aims: Microglia - resident macrophages - are the main cell type that provide immuno-surveillance in the brain. In vivo data from cerebral ischemia and trauma models have suggested a general rule that more severe brain injury is associated with higher macrophage densities and that strategies that limit microglial activation in general, or that specifically inhibit synthesis of toxic molecules produced in these cells can be neuroprotective. These cells undergo a graded process of activation in response to injury, including morphological transformation, increased migratory activity, proliferatio, secretion of cytokines, and antigen presentation. Using a focal transient middle cerebral artery (MCA) occlusion model in postnatal day 7 (P7) rats, we have previously shown a major transient accumulation of inflammatory cytokines, rapid microglial activation, and minimal invasion of peripheral monocytes/macrophages in acutely injured regions (Fox et al JCBFM; 2005, Dingman et al JNC, 2006; Denker et al JNC 2007). The anti-inflammatory drug minocycline administered acutely, while protective, did not diminish the number of activated microglia/macrophages (ED1+ cells), or increased levels of IL-1b, MCP-1 and CINC-1 in ischemic-reperfused brain (Fox et al, JCBFM; 2005). We asked whether neuroinflammation after neonatal stroke can be attenuated and protection achieved by limiting migration of cells of monocyte lineage to injured region.

Methods: P7 rats were subjected to a 1.5 hr MCA occlusion, treated with either a neutralizing anti-MCP-1 antibody or propagermanium, a compound that attenuates monocyte migration by targeting of glycosylphosphatidylinositol (GPI)-anchored proteins closely associated with CCR2. A neutralizing anti-MCP-1 or vehicle was administered i.p. 30 min prior to occlusion. IL1a, IL1b, TNF-a, IL-6, IL-18, CINC-1 and MCP-1 were measured in the same sample (Multi-plex cytokine assay) in the plasma, injured tissue and matching tissue from contralateral hemisphere at 4 and 8 hours; histological outcome was determined at 24 hr post reperfusion. 8mg/kg propagermanium or the corresponding vehicle (1% gelatin) were administered orally 30 min prior to occlusion and daily for 3 days post-ischemia. Animals were sacrificed 7 days later for histology and quantification of macrophage densities.

Results: MCP-1 neutralization reduced ischemia-induced increases in IL1b, IL-6, and CINC-1 in the systemic circulation, but did not diminish the elevated cytokine production locally in the injured brain tissue. Administration of an anti-MCP-1 antibody did not affect the injury volume at 24 hr post-reperfusion (5511% Vs. 5611%, injured/contralateral hemisphere in anti-MCP-1 treated Vs. vehicle-treated group, respectively, n=5-6/group). Treatment with propagermanium substantially reduced accumulation of IB4+ cells in injured regions. The volume of remaining tissue was 7612% in treated Vs. 6915% in vehicle-treated group (n=4-6/group, p>0.05).

Conclusions: Our findings suggest the existence of a relatively independent pool of cytokine production in the injured brain that is not modulated by the systemically produced MCP-1, thus limiting the ability of peripheral MCP-1 to affect acute ischemic injury in neonatal brain. Our data also suggest that blocking the migration of peripheral macrophages may not be sufficient to protect the neonatal brain from stroke and that activated microglial cells play a profound role in neonatal stroke.

NIH NS44025, UCP R766-04, AHA GIA 0655236Y

INJURY TO CORPUS AND CINGULATE CALLOSUM DEVELOPMENT WITH HYPOXIA: ABEHAVIORAL DTI STUDY ON C57B/L6 MICE

Halima Chahboune^1,2,7,8, Laura R. Ment^2,3, William B. Stewart⁴, Douglas L. Rothman^1,6,7,8, Michael L. Schwartz⁵, Fahmeed Hyder^1,6,7,8

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA, ²Department of Pediatrics, Yale University, New Haven, CT, USA, ³Department of Neurology, Yale University, New Haven, CT, USA, ⁴Department of Neurosurgery, Yale University, New Haven, CT, USA, ⁵Department of Neurobilogy, Yale University, New Haven, CT, USA, ⁶Department of Biomedical Engineering, Yale University, New Haven, CT, USA, ⁷Core Center for Quantitative Neuroscience With Magnetic Resonance, Yale University, New Haven, CT, USA, ⁸Magnetic Resonance Research Center, Yale University, New Haven, CT, USA

ROLE OF CYCLOPHILIN D IN HYPOXIA-ISCHEMIA INDUCED BRAIN INJURY DURING DEVELOPMENT

Xiaoyang Wang¹, Michael A. Forte², Paolo Bernardi³, Catherine Rousset¹, Changlian Zhu⁴, Klas Blomgren⁴, Ylva Carlsson¹, Carina Mallard¹, Henrik Hagberg^1,5

¹Perinatal Center, Department of Physiology, Goteborg University, Goteborg, Sweden, ²Vollum Institute, Oregon Health and Sciences University, Portland, OR, USA, ³Department of Biomedical Sciences and Consiglio Nazionale Delle Ricerche Institute of Neuroscience, University of Padova, Padova, Italy, ⁴Center for Brain Repair and Rehabilitation, Institute of Neuroscience and Physiology, Goteborg University, Goteborg, Sweden, ⁵Perinatal Center, Department of Obstetrics and Gynecology, Goteborg University, Sahlgrenska University Hospital, Goteborg, Sweden

OBJECTIVE Mitochondria play a critical role in mediating both apoptotic and necrotic cell death. Cyclophilin D (CyD), a component of mitochondria membrane permeability transition pore (PTP) is important in Ca 2+ induced membrane permeability and necrotic cell death trigged by ischemia/reperfusion injury in adult brain and heart (1)–(4), whereas the role of CyD in hypoxic-ischemic (HI) injury in the immature brain is unknown. Caspase-dependent and non-dependent apoptotic mechanisms appear to be more critical in neonatal vs. adult brain (Zhu et al. 2005) and we hypothesized that CyD may play a different role in the developing brain. We crossed CyD deficient mice (KO) and wild-type (WT) mice to produce heterozygotes (Het), which were bred further to produce littermate animals with mixed genotypes.

METHODS At postnatal day (PND) 9 or 60, littermates of CyD KO, WT and Het were subjected to left carotid artery ligation and 10% O2 for 40 minutes or 30 minutes, respectively. Mice were sacrificed at 7 days after HI. Brain sections were stained with micro tube protein-2 (MAP-2) and the infarct volume and the total tissue loss was calculated.

RESULTS 1) At PND9, brain injury was more prononuced in CyD deficient mice than in WT mice. The total tissue loss was 20.6 ± 3.5 mm3 in CyD KO mice (n=19) vs 10.5 ± 1.7 mm3 in WT mice (n=16) (p=0.017). Regional neuropathology score analysis showed that injury in cortex, hippocampus, thalamus and striatum were 59%, 17%, 53% and 31% lower in WT compared to KO. 2) In PND60 adult brain, CyD deficiency resulted in a significantly reduced HI brain injury. The total tissue loss in PND60 mice was 2.3 ± 1.9 mm3 in CyD KO mice (n=13) vs 14.3 ± 4.8 mm3 in WT mice (n=18) (p=0.035). Brain injury was reduced in KO vs. WT by 68% in cortex, 50% in hippocampus, 90% in thalamus and 73% in striatum.

CONCLUSIONS 1) CyD-dependent PTP appears to play a different role in neontal and adult HI brain injury. 2) The present results might indicate that the known pro- and antiapoptotic actions of CyD do not have the same weight in immature as in the adult brain. These results agree with previous studies suggesting that cell death mechanisms especially those related to the apoptotic machinery are different in the developing CNS (5), (6) (7).

ROLE OF URACIL MISINCORPORATION FOR NEUROGENERATION AND BRAIN DYSFUNCTION INCUDED BY CHRONIC FOLATE DEFICIENCY

Golo Kronenberg^1,2, Christoph Harms¹, Robert W. Sobol³, Fernando Cardozo-Pelaez⁴, Heinz Linhart⁵, Mustafa Balkaya¹, Karen Gertz¹, Sarah Eckart², Rainer Hellweg², Heide Hoertnagl⁶, Rudolf Jaenisch⁵, Matthias Endres¹

¹Klinik Und Poliklinik Fuer Neurologie, Charite-Unversitaetsmedizin Berlin, Campus Mitte, Berlin, Germany, ²Klinik Und Hochschulambulanz Fuer Psychiatrie Und Psychotherapie, Charite-Universitaetsmedizin Berlin; Campus Benjamin Franklin, Berlin, Germany, ³University of Pittsburgh Cancer Institute, Hillman Cancer Center, Pittsburgh, PA, USA, ⁴Department of Biomedical and Pharmaceutical Sciences, Center for Environmental Health Sciences, University of Montana, Missoula, MT, USA, ⁵Whitehead Institute, Massachusetts Institute of Technology, Cambridge, MA, USA, ⁶Institut Fuer Pharmakologie Und Toxikologie, Charite-Universitaetsmedizin Berlin, Campus Mitte, Berlin, Germany

Background and aims: Folate deficiency contributes to neurodegenerative and psychiatric disease such as dementia and depression. We hypothesized that the pathogenetic mechanisms include uracil misincorporation.

Methods: We analyzed the effects of folate deficiency in mice lacking uracil DNA glycosylase (Ung-/−) vs. wildtype controls.

Results: Folate depletion increased nuclear mutation rates in Ung-/− embryonic fibroblasts, and conferred death of cultured Ung-/− neurons. Feeding animals a folate-deficient diet (FD) for 3 months induced degeneration of CA3 pyramidal neurons in Ung-/− but not Ung+/+ mice along with decreased hippocampal expression of brain-derived neurotrophic factor protein and decreased brain levels of antioxidant glutathione (Figure). Further, FD induced cognitive and mood alterations such as depressive and anxious behavior that were aggravated in Ung-/− mice. Independent of Ung genotype, FD increased plasma homocysteine levels, altered brain monoamine metabolism and inhibited adult hippocampal neurogenesis.

Conclusions: These results indicate that impaired uracil repair is involved in neurodegeneration and brain dysfunction induced by folate deficiency.

Fig legend: (A) Representative images of 10mm thick H&E-stained sections illustrating reduced neuronal density in CA3 of Ung-/− animals as a consequence of chronic folate deficiency. (B) Density of CA3 pyramidal neurons; ANOVA for genotype: F1,16=22,8; p<0.0005; treatment: F1,16=11.3; p<0.005, interaction: F1,16=3.8; p=0.068; 1-b=0.44. (C) Uracil misincorporation in DNA, ANOVA for genotype: F1,29=5.6, p=0.02 (D) BDNF protein levels in hippocampus, ANOVA for interaction: F1,31=16.6, p<0.0005. (E) GSH levels in whole cerebral hemispheres; ANOVA for interaction: F1,20=5.6, p=0.03. (F) GSH levels in cerebellum; ANOVA for interaction: F1,20=20.4, p=0.0002. *p<0.05 ND vs. FD, #p<0.05 Ung+/+ vs. Ung-/−, +p<0.05 for interaction; two-way ANOVA and Tukey post-hoc, n=5-9 animals per group

ERYTHROPOIETIN PROMOTES NEURONAL REPLACEMENT VIA ANGIOGENESIS AND NEUROGENESIS AFTER HYPOXIA/ISCHEMIA IN NEONATAL RATS

Masanori Iwai¹, Wei Yin¹, Guodong Cao¹, R. Anne Stetler^1,2, Jun Chen^1,2

¹Department of Neurology, University of Pittsburgh, Pittsburgh, PA, USA, ²Geriatric Research, Education and Clinical Center, Veterans Affairs Pittsburgh Health Care System, Pittsburgh, PA, USA

Background and aims: Erythropoietin (EPO) has shown potent neuroprotective effects against ischemic injury in both adult and neonatal brain via its anti-apoptotic actions. Recent evidence has suggested that EPO may have additional effects on post-ischemic adult brain by promoting neurovascular unit remodeling. Indeed, endogenously expressed EPO and its receptors appear to be essential for ischemia-induced neurogenesis in adult brain. The current study was conducted to determine the effect of exogenous administration of EPO on angiogenesis and neurogenesis, the two major events believed to contribute to neuronal replacement, in the neonatal brain after hypoxia/ischemia (HI).

Methods: HI was induced by left-CCA ligation followed by 130 minutes exposure to 8 % oxygen in seven-day-old rats. They were then treated with recombinant human EPO (rhEPO; 1000U/kg) at 20 minutes after HI, and again at 2, 4 and 6 days. Rats were euthanized on day 7 after HI for evaluation of infarct volume, angiogenesis and neurogenesis, and on day 28 for evaluation of neuronal replacement. Neurogenesis and angiogenesis were analyzed using bromodeoxyuridine (BrdU) labeling, and by tomato lectin labeling to visualize functional vessels. The newly generated neurons were also classified as either neuronal progenitors, using anti-doublecortin (DCX) antibody, or mature neurons, using anti-neuronal nuclei (NeuN) antibody. At 7 (n=28) and 28 days (n=19) after HI, neurogenesis was evaluated using double immunofluorescence for BrdU/DCX or BrdU/NeuN. Neurological outcomes were also evaluated until 28 days after HI.

Results: Treatment with rhEPO significantly reduced the infarct volume both in the cortex (47.8%, p<0.01) and in the ST (42.1%, p<0.01). Enhancement of angiogenesis with rhEPO treatment was observed in the ischemic boundary of cerebral cortex. Compared with vehicle treatment, rhEPO treatment significantly increased BrdU-labeled cells on the vessels (114 ± 45 vs 55 ± 15/mm2, p<0.01) as well as the number of vessels (178 ± 67 vs 141 ± 61/mm2, p<0.05). In addition, rhEPO treatment significantly increased the number of BrdU-labeled cells in the ipsilateral subventricular zone (486 ± 88 vs. 381 ± 87/section in the vehicle, p < 0.01) and the number of doublecortin positive cells in the cerebral cortex (29 ± 10 vs. 17 ± 6/mm2, p < 0.01) and striatum (55 ± 27 vs. 42 ± 13 / mm2, p < 0.05) of ischemic hemisphere. Finally, rhEPO treatment significantly increased the number of BrdU/NeuN-double positive cells in the ischemic boundary of cerebral cortex (16 ± 11 vs. 3 ± 4/mm2, p < 0.01) and striatum (16 ± 11 vs. 6 ± 7 / mm2, p < 0.01) at 28 days after HI. Treatment with rhEPO significantly improved neurobehavioral development.

Conclusions: Our data suggests that treatment with rhEPO contributes to neurovascular unit remodeling by promoting tissue protection, angiogenesis and neurogenesis in neonatal HI injured brain and produces improvement of neurobehavioral development. However, EPO appears to have limited role in supporting new born neuron survival after HI, as only modest amounts of newborn cells with a mature neuronal phenotype (NeuN positive) eventually survived.

IMAGING OF AMYLOID-PLAQUES AND CEREBRAL GLUCOSE METABOLISM IN SEMANTIC DEMENTIA AND ALZHEIMERS DISEASE

Alexander Drzezga¹, Timo Grimmer², Gjermund Henriksen¹, Isabelle Stangier¹, Robert Perneczky², Janine Diehl-Schmid², Chester A. Mathis³, William E. Klunk⁴, Hans-Jurgen Wester¹, Markus Schwaiger¹, Alexander Kurz²

¹Department of Nuclear Medicine, Technical University of Munich, Munich, Germany, ²Department of Psychiatry, Technical University of Munich, Munich, Germany, ³Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA, ⁴Department of Psychiatry, Western Psychiatric Institute & Clinic, Pittsburgh, PA, USA

Background and aims: Semantic dementia (SD) is a rare clinical syndrome, assigned to the group of frontotemporal lobar degenerations (FTLD). Histopathological analysis has not revealed the deposition of amyloid-plaques in the majority of SD-cases, in contrast to dementia of the Alzheimer type (AD). However, some overlap of clinical symptoms and imaging findings between AD and SD has been shown and a reliable differentiation of the underlying pathology can not be guaranteed based on clinical examination alone. Our aim was to determine, whether differences between AD and SD can be found by means of in vivo amyloid-plaque imaging, providing complementary information to results obtained by measurement of cerebral glucose metabolism.

Methods: Patients with AD (n=8, 4 female) and with SD (n=8, 4 female), matched for gender, age and overall degree of cognitive impairment, were recruited using established clinical criteria. Cerebral glucose metabolism was examined with [F-18]Fluorodeoxyglucose (FDG)-PET, cerebral amyloid-plaque load was assessed using [C-11]6-OH-BTA-1 (PIB)-PET. A volume-of-interest analysis (VOI), using the cerebellum as a reference region, and voxel-based statistical group comparisons (SPM2) were carried out between the patient groups (FDG- and PIB-PET data) and a group of healthy controls (FDG-PET data).

Results: Characteristic patterns of hypometabolism could be demonstrated in both clinically defined AD and SD (AD: bilateral temporoparietal and frontal cortex; SD: left>right temporal, frontal mesial cortex) with some regional overlap, particularly in temporal cortices. In contrast, strong [C-11]PIB amyloid-plaque tracer binding was observed only in patients with AD in bilateral temporoparietal, frontal and posterior cingulate cortex and the precuneus. This increased amyloid-plaque deposition, observed in AD, clearly extended the metabolic differences between the groups, even in regions with comparable metabolic deficits.

Conclusions: These findings support the common notion hat neuronal dysfunction is associated with amyloid-plaque deposition in AD. In contrast, in the examined cases of typical SD, neuronal dysfunction was apparently based on pathology other than amyloid-plaque deposition. Amyloid-plaque imaging may be valuable to clearly characterize dementias in vivo, based on the underlying pathology rather than on neuropsychological findings. This may be important for definition of individual prognosis and for the selection of patients for scientific trials.

OMENTUM TRANSPOSITION IN THE TREATMENT OF ALZHEIMER DISEASE

Harry Goldsmith

Department of Surgery, University of Nevada School of Medicine, Reno, NV, USA

Background and aims: It has been shown that the omentum is rich in biological substances and especially in the amount of angiogenic material incorporated in the tissue (VEGF). The aim of the presentation is to show that the omentum can introduce large amounts of blood, neurotransmitters, growth factors, and stem cells into the brain, which have been shown in some patients to reverse the cognitive changes of Alzheimer disease.

Methods: The omentum is separated from the transverse colon and the proximal portion of the stomach, leaving the gastroepiploic vessels within the omental apron. The omentum is then surgically lengthened and brought up through a subcutaneous tunnel developed along the chest and neck. A craniotomy is performed and the dura mater opened. The omentum is then laid directly on the brain.

Results: Several publications in neurosurgical journals have reported that in biopsy-proven Alzheimer-disease patients, of the twenty-five who were operated upon, six showed no improvement, nineteen showed some degree of improvement with nine patients showing significant reversal of their cognitive and neurological conditions.

Conclusions: It is possible to reverse cognitive symptoms of Alzheimer disease as a result of omental transposition to the brain.

CEREBRAL NICOTINIC ACETYLCHOLINE RECEPTORS IN PATIENTS WITH ALZHEIMERS DISEASE AND AMNESTIC MILD COGNITIVE IMPAIRMENT, COMPARISON TO GLUCOSE METABOLISM

Kai Kendziorra¹, Philipp Meyer¹, Henrike Wolf², Henryk Barthel¹, Swen Hesse¹, Anita Seese¹, Dietlind Sorger¹, Marianne Patt¹, Georg Becker¹, Andreas Schildan¹, Peter Brust³, Hermann Josef Gertz², Osama Sabri¹

¹Department of Nuclear Medicine, University of Leipzig, Leipzig, Germany, ²Department of Psychiatry, University of Leipzig, Leipzig, Germany, ³Institute of Interdisciplinary Isotope Research, Leipzig, Germany

EFFECTS OF CEREBRAL ISCHEMIA ON EVOKED CEREBRAL BLOOD OXYGENATION RESPONSES AND BOLD CONTRAST FUNCTIONAL MRI IN STROKE PATIENTS

Norio Fujiwara, Noriaki Yokose, Tatsuya Hoshino, Yoshihiro Murata, Tsuneo Kano, Kaoru Sakatani, Yoichi Katayama

Department of Neurological Surgery, Nihon University School, Tokyo, Japan

Background: Blood oxygenation level dependent (BOLD) contrast functional MRI (BOLD-fMRI) has been used in functional studies on stroke patients, under the assumption that these patients have normal neurovascular coupling. Recent studies have revealed, however, that BOLD-fMRI does not correctly image activation areas in stroke patients. It was suggested that impairments of neurovascular coupling may alter the evoked cerebral blood oxygenation (CBO) responses and hemodynamic changes, and this could result in failure of BOLD imaging in stroke patients. BOLD-fMRI alone, however, cannot elucidate the precise mechanisms in the failure of BOLD imaging, since BOLD-fMRI provides information mainly about concentration changes of deoxyhemoglobin (deoxy-Hb), which is paramagnetic. In this study, in order to evaluate these issues, we compared the evoked CBO responses measured by near infrared spectroscopy (NIRS) and BOLD imaging in chronic stroke patients with moderate and severe cerebral ischemia.

Methods: We measured the evoked CBO responses in the primary sensorimotor cortex (PSMC) by means of NIRS (NIRO-300, Hamamatsu Photonics KK, Japan) during contralateral motor tasks. We compared the AV of BOLD-fMRI in the PSMC on the non-lesion and lesion sides. SPECT was employed to classify the ischemic status as moderate (slight reduction of rCBF and cerebrovascular reserve capacity (CVRC)) or severe (marked reduction of rCBF and CVRC; i.e., misery perfusion).

Results: In age-matched controls, deoxyhemoglobin concentration decreased with concomitant increases in oxyhemoglobin and total hemoglobin concentrations during activation. The PSMC on the non-lesion side exhibited a normal CBO response pattern. On the lesion side, moderate cerebral ischemia did not affect the CBO response pattern, but severe cerebral ischemia caused an increase of deoxyhemoglobin during the task, associated with increases of oxyhemoglobin and total hemoglobin. Moderate cerebral ischemia induced only a slight reduction of the AV on the lesion side; however, severe cerebral ischemia markedly reduced the AV on the lesion side. The BOLD signal did not change in some areas of the PSMC on the lesion side in severe cerebral ischemia, while it tended to decrease in other areas during the tasks (Figure 1).

Conclusions: Misery perfusion caused a marked reduction of the AV in BOLD imaging, associated with an increase of deoxyhemoglobin concentration during activation. BOLD-fMRI investigations of stroke patients should be performed while giving consideration to the baseline circulatory status. Functional NIRS could be an alternative means to assess the CVRC.

CORTICAL SPREADING ISCHEMIA IS INDUCED BY COMBINED INHIBITION OF NA, K-ATPASE AND NO SYNTHASE AND DEPENDS ON INTRACELLULAR CA2+ STORES

Sebastian Major^1,2, Georg Royl^1,2, Ute Lindauer¹, Jens Dreier^1,2

¹Department of Experimental Neurology, Charite, Universitaetsmedizin Berlin, Berlin, Germany, ²Department of Neurology, Charite, Universitaetsmedizin Berlin, Berlin, Germany

Background: Delayed ischemic neurological deficit (DIND) is one of the major complications after subarachnoid hemorrhage (SAH). It occurs in up to 40% of patients who survive the initial bleeding. About 13% develop delayed cerebral infarcts on CT.

Based on experiments in rats, it has been proposed that cortical spreading ischemia (CSI) is a mechanism of DIND. CSI is a long-lasting, spreading wave of an ischemic blood flow change leading to widespread cortical infarcts. It represents an inverse hemodynamic response to cortical spreading depolarization. CSI occurs when the baseline extracellular K+ concentration ([K+]o) is increased and the NO concentration is reduced (e.g. due to nitric oxide synthase (NOS) inhibition by L-NNA) in the subarachnoid space.

In cell culture experiments increased baseline [K+]o is associated with a decrease of Na, K ATPase activity. To test whether this ubiquitous ATPase is involved in the mechanism of CSI, we investigated if CSI occurs after combined brain topical administration of L-NNA and the Na, K-ATPase inhibitor ouabain.

Methods: A closed cranial window was implanted in rats. Cerebral cortex was superfused with artificial cerebrospinal fluid (ACSF) containing L-NNA (1 mM). Subsequently, increasing concentrations of ouabain (0.5, 5, and 50 µM) were added to the ACSF. The CBF changes were monitored either by Laser-Doppler-Flowmetry or LAser Speckle Contrast Analysis imaging, the depolarization by subarachnoid Ag-AgCl electrode.

Results: Compared to L-NNA or ouabain alone the combination induced CSI in a dose dependent manner (duration of hypoperfusion under L-NNA and ouabain (5 µM): 3:50 (3:00 ? 5:26) min and L-NNA and ouabain (50 µM): 6:55 (5:20 ? 22:24) min). CBF fell to about 50% of baseline value (=100%). Using a digital camera we recorded a propagating progressive constriction of arteries simultaneously with the CBF decrease.

At least at the lower applied ouabain concentration (up to 5 µM), only the α2 and α3 isoforms of the Na, K ATPase are inhibited in rats. Their inhibition leads to increase of the cytoplasmic Ca2+ concentration and subsequently to filling of the Ca2+-stores in the sarco/endoplasmic reticulum. Thus we tested, if depletion of these internal stores has an influence on the CSI. Therefore we added 100 µM of thapsigargin, an inhibitor of the sarco/endoplasmic reticulum Ca2+-ATPase, to the ACSF 30 min before applying ouabain (50 µM). This did not prohibit hypoperfusion, but shortened it significantly to 1:16 (0:58 ? 3:18) min (vehicle control 4:31(2:07 ? 8:47)).

Conclusion: Combined NOS and Na, K-ATPase inhibition produce CSI in a dose-dependent manner. Ca2+ stored in the sarco/endoplasmic reticulum appears to be necessary for maintaining the vasoconstriction during CSI.

CEREBRAL HEMODYNAMICS AND CLINICAL COURSE AFTER SUPERFICIAL TEMPORAL ARTERY-MIDDLE CEREBRAL ARTERY ANASTOMOSIS IN ADULT ISCHEMIC-TYPE PATIENTS WITH MOYAMOYA DISEASE

Masaru Idei, Kanji Yamane, Hirohumi Okii, Kiyoshi Kumano, Hidetaka Onda

Department of Neurosurgery, Chugoku Rousai Hospital, Kure, Hiroshima, Japan

[Background and aims] We measured regional cerebral blood flow (rCBF) using xenon-enhanced CT and assessed overall outcome after superficial temporal artery-middle cerebral artery (STA-MCA) anastomosis in adult ischemic-type patients with moyamoya disease.

[Methods] Our subjects consisted of 16 (25 sides) cases who underwent STA-MCA double anastomosis from April, 1994 to February, 2006. Clinical features and rCBF were investigated before surgery, two weeks, half a year, one year, two years, three years and four years after surgery. rCBF was compared with normal control value (control rCBF) in our hospital. The region of interest was conducted in the anterior cerebral artery (ACA), middle cerebral artery (MCA) and posterior cerebral artery (PCA) territories.

[Results] Before surgery, rCBF in the ACA and MCA territories was found to be significantly lower than control rCBF, but in the PCA territory was preserved. During follow-up, rCBF in the ACA territory did not changed, but on the other hand rCBF in the MCA territory significantly increased. rCBF was preserved in the PCA territory and gradually decreased during follow-up. Only one patient had infarction in the ACA territory 9 years after surgery, but the remaining 15 patients had no recurrence of stroke.

[Conclusions] Our results showed that dominant posterior distribution of the rCBF before surgery gradually became to be the normal distribution after STA-MCA anastomosis. It suggested that STA-MCA anastomosis could prevent a recurrence of stroke in the MCA region, but not completely in the ACA region.

FACTORS OF HYPERPERFUSION SYNDROME AFTER CAROTID ENDARTERECTOMY AND INTRAOPERATIVE NEAR INFRARED SPECTROSCOPY

Norikazu Miura¹, Kenji Yoshitani², Masakazu Kuro²

¹Anesthesiology Department, Yokohama City University Graduate School of Medicine, Yokohama, Japan, ²Anesthesiology Department of National Cardiovascular Center, Osaka, Japan

Background: Postoperative cerebral hyperperfusion syndrome (HS) after carotid endarterectomy (CEA) causes miserable adverse cerebral complication, such as intracranial hemorrhage. Furthermore, the prognosis of patients with HS is poor. Previous studies reported that impaired cerebral vasoreactivity was implicated in HS (1). However, Powers' staging (2), one of clinical indices of cerebral hemodynamic impairment, has not been tested whether that is a factor of HS. In addition, near infrared spectroscopy (NIRS) said to reflected cerebral vasoreactivity (3). Therefore, we tested the hypothesis whether Power's staging was a factor of HS, or intraoperative NIRS measurement predicted HS.

Methods: The records of 116 consecutive patients with CEA were reviewed. Logistic regression analysis was used to assess the effects of Power's staging and following factors on HS; age, sex, history of hypertension, diabetes mellitus, smoking, atrial fibrillation, symptom for internal carotid artery stenosis, history of ischemic heart disease, severity of carotid artery stenosis, stump pressure, and systemic blood pressure after declamping. Tissue oxygen index (TOI, were measured using NIRO 200 (Hamamatsu Photonics, Hamamatsu) during CEA in 5 patients preliminary.

Results: Sixteen patients (18%) had HS after CEA. Significant predictors with univariate logistic regression analyses were Powers' staging, ipsilateral carotid artery stenosis (>80%), contralateral carotid artery stenosis (>50%), and systolic arterial pressure (>120mmHg) after declamping of carotid artery. Multivariate logistic regression analysis demonstrated that the Powers' staging (p=0.001), contralateral internal carotid artery stenosis more than 50% (p=0.003) and systolic arterial pressure (>120mmHg) (p=0.01) were significant predictive variables for postoperative HS. During CEA, tissue oxygen index (TOI) was measured in 5 patients of 16 patients who suffered from HS. The values of TOI after declamping of internal carotid artery were significant higher than that during clamping (61 ± 5 to 68 ± 8%, p<0.05).

Conclusion: The results in our retrospective study demonstrated that the Powers' staging, as an index of hemodynamic state, contralateral carotid artery stenosis more than 50% and systolic arterial pressure (>120mmHg) after carotid artery declamping were predictive factors for postoperative HS. NIRS monitoring, TOI could be effective to predict postoperative HS as an intraoperative index.

MAGNETIC RESONANCE MOLECULAR IMAGING OF ACTIVATED ENDOTHELIUM IN ACUTE STROKE USING A TARGETED IRON-OXIDE NANOPARTICLE CONTRAST AGENT

Albert Jin¹, David Rushforth², Rana Filfil³, Jaspreet Kaur¹, Ursula Tuor², Tadeusz Foniok², Kumud Deka², David Kirk², Jody Lee-Petersen¹, Boguslaw Tomanek², Feng Ni³, Philip Barber¹

¹Experimental Imaging Centre, University of Calgary, Calgary, AB, Canada, ²Institute of Biodiagnostics (West), National Research Council of Canada, Calgary, AB, Canada, ³Biotechnology Research Institute, National Research Council of Canada, Montreal, QC, Canada

Background: Because of the inherent complexity of neuroinflammation in acute stroke, critical aspects of the early evolution of stroke injury are poorly understood. Currently, MRI cannot discriminate pathological processes such as endothelial activation in neuroinflammation. The aim of this study was to use MRI to visualize endothelial activation in a mouse model of stroke using an iron oxide nanoparticle contrast agent (MNP-Psel) that was modified to bind with high affinity to P-selectin.

Methods: MNP-Psel was compared to the iron oxide agent Feridex (Berlex), which does not have any binding specificity. As a model for focal cerebral ischemia and reperfusion, the left middle cerebral artery (MCA) was transiently occluded with a silicon-coated nylon suture to effect a 75% or greater reduction in regional cerebral blood flow for 30 minutes in male C57 Black 6 mice. Normothermia was maintained throughout the procedure and afterwards. 24 hours post-reperfusion, mice received 2.8 Fe mg/kg of either Feridex or MNP-Psel. MRI scans at 9.4T acquired prior to and after contrast injection included: multi-echo T2 weighted images, T2*-weighted images, and a T1 map. Subtraction maps (pre-contrast ? post-contrast) were used to identify areas with differences in T1, T2 and T2*. The T2 effect in the contralateral side was subtracted from the effect in the stroke side to remove confounding from non-specific contrast agent accumulation: [1? Percent T2 decrease on contralateral side/Percent T2 decrease on stroke side] × 100. After imaging was complete, mice were euthanised by sodium pentobarbital overdose and the brain was perfusion fixed in 4% paraldehyde. The brain was then embedded in paraffin and sections through the cortex and striatum were taken for iron histochemical staining by the Prussian blue method.

Results: T1 values were unaffected by MNP-Psel or Feridex. T2 decreased nonspecifically in both hemispheres after Feridex injection but preferentially in the stroke hemisphere with MNP-Psel, indicating that MNP-Psel accumulation was mediated through an interaction specific to the infarct and peri-infarct regions. Prolonged lowering of T2* after MNP-Psel was seen preferentially in the stroke hemisphere. The presence of T2 and T2* effects of MNP-Psel in the peri-infarct regions indicated endothelial activation beyond the infarction boundary noted on T2 maps. Brain sections from MNP-Psel-injected mice showed iron accumulation on endothelium in the infarct and peri-infarct regions.

Conclusion: In this study, there is a prolonged shortening of T2 and T2* values within the infarct and peri-infarct region after injection of MNP-Psel that is not observed with Feridex. Additionally, histological examination of the mouse brain after injection with MNP-Psel showed an accumulation of iron on endothelium in the infarct and peri-infarct regions. These results suggest that MNP-Psel binds to P-selectin expressed on activated endothelium in the stroke hemisphere. This demonstrates the potential use of targeted iron oxide contrast agents to visualize in vivo endothelial activation in acute stroke.

NEUROVASCULAR COUPLING AND CEREBRAL METABOLIC RATE OF OXYGEN: EFFECTS OF BRAIN HYPOTHERMIA

Martina Fuechtemeier¹, Nikolas Offenhauser¹, Christoph Leithner¹, Jens Steinbrink¹, Matthias Kohl-Bareis², Ulrich Dirnagl¹, Ute Lindauer¹, Georg Royl¹

¹Department of Experimental Neurology, Charite Universitaetsmedizin Berlin, Berlin, Germany, ²University of Applied Sciences Koblenz, RheinAhr Campus, Remagen, Germany

Background: Neuronal activation is accompanied by a local increase in cerebral blood flow (CBF) and in cerebral metabolic rate of oxygen (CMRO2). From combined measurements of CBF and hemoglobin concentration, relative changes in CMRO2 during functional challenges can be calculated. The quantitative relationship between basal CMRO2 and brain temperature changes over a range of 10°C (Q10) has not yet been determined with this method and can serve as a validation. In addition, recent studies have raised the question whether neurovascular coupling serves local cooling of activated brain tissue. Then again, the influence of hypothermia on neurovascular coupling is not known. We addressed these issues by studying the effects of graded hypothermia in a rat model of neurovascular coupling in the somatosensory cortex.

Methods: Anesthetized Wistar rats (n=6) underwent surgical preparation of a closed cranial window over the somatosensory cortex. Using Laser Doppler flowmetry and optical spectroscopy, changes in CBF and in hemoglobin concentration were measured continously. Furthermore, an electrocorticogramm (ECoG) was simultaneously recorded from the measurement site. By the application of ice packs, whole-body hypothermia (masseter temperature from 35°C down to 25°C in 2 hours) was generated, followed by re-warming. While maintaining anesthesia with alpha-chloralose, spontaneous ongoing activity was measured, interleaved by blocks of functional measurements during electrical forepaw stimulation (4sec, 1.6mA, 4Hz).

Results: Hypothermia led to a decrease of spontaneous CBF and CMRO2. The calculated Q10 value for CMRO2 was 3.0 (95% Confidence Interval: 2.6 ? 3.4). Parallel to CBF and CMRO2, the power of the ECoG low-frequency-band decreased. Functional changes of CBF and CMRO2 were reduced during hypothermia as well as the amplitudes of somatosensory evoked potentials.

Conclusion: 1. With optical methods, spontaneous changes in relative CMRO2 can be quantified, revealing a Q10 for CMRO2 that is well in the range of literature values. 2. During hypothermia, neurovascular coupling is preserved. This applies to CBF changes accompanying spontaneous ongoing activity as well as CBF changes during functional activation. Thus, a possible role of activity-induced neurovascular coupling in local temperature regulation cannot be excluded.

Keywords: hypothermia; CMRO2; Q10 value; neurovascular coupling; rat

Supported by DFG and Hermann and Lilly Schilling Foundation.

MAGNETIC RESONANCE AND NEAR-INFRARED SPECTROSCOPY MEASUREMENTS OF ENERGY METABOLITES AND CEREBRAL OXYGEN CONSUMPTION FOLLOWING HYPOXIA ISCHEMIA

Jeff D. Winter^1,2, Ken M. Tichauer^1,2, Neil Gelman^1,2, R. Terry Thompson^1,2, Ting Y. Lee^1,2, Keith St. Lawrence^1,2

¹Imaging Division, Lawson Health Research Institute, London, ON, Canada, ²Department of Medical Biophysics, University of Western Ontario, London, ON, Canada

SUSTAINED ALTERATION OF NEUROMETABOLIC, -VASCULAR COUPLING BY CORTICAL SPREADING DEPRESSION IN RAT SOMATOSENSORY CORTEX

Nikolas Offenhauser, Georg Royl, Christoph Leithner, Ulrich Dirnagl, Ute Lindauer

Department of Experimental Neurology, Charite, Berlin, Germany

Neuronal activity increases blood flow and oxygen consumption in the brain. The mechanisms by which increased metabolic activity during local increases in neuronal activity are coupled to hemodynamic changes are incompletely understood. Especially, there is only limited knowledge, if pathophysiological events influence this coupling relationship and how baseline state alterations of brain blood flow or metabolism are related to activity induced changes in regional cerebral blood flow (rCBF), oxygen consumption and neuronal activity. We tested the hypothesis that this neurometabolic, neurovascular coupling relationship is variable.

Cortical spreading depression (CSD) was used to induce a transient alteration in neuronal activity and vascular hemodynamic responsiveness, to investigate the variability in the relationship between neuronal, metabolic and vascular responses under functional activation (forepaw stimulation, 5Hz, 10s). Alpha-chloralose-urethane anaesthetized, male Wistar rats where used. Activity induced changes in rCBF, regional blood oxygenation (rCHo: oxy-, deoxyhemoglobin concentration) and saturation, by a combined microfiber optical hemoglobin spectroscopy and laser-Doppler flowmetry probe, where recorded and oxygen consumption during activation was calculated. ECoG and SEPs were recorded by a silver ball surface electrode as indicators of baseline and evoked neuronal activity. Averaged responses prior to single (n=6) or repetitive (n=3) CSDs were compared with blocks of averaged responses up to 1.5h.

Baseline rCBF after a single CSD was initially significantly reduced (up to 35%, +/− 4, SEM) and accompanied by a significant increase in deoxy-hemoglobin concentration (Hbr; 16%, +/− 3) and decrease of oxy-hemoglobin concentration (Hbo; 36%; +/− 6). These changes gradually returned to baseline at the end of recording (90 min. post CSD).

After CSD, rCBF responses and SEPs were significantly attenuated (by 45%, +/− 5 and 20%, +/−4, respectively). Hbr responses after CSD ranged from biphasic increase-decrease patterns to entirely positive responses, mirrored by the concomitant Hbo response.

Most interestingly, the individual parameters showed a different recovery pattern after CSD. The Hbr response continuously increased after CSD. An effect found to be more pronounced after repetitive CSDs. In general, SEPs and activity induced CMRO2 changes after CSD where significantly different from the control condition only up to 45 min. after CSD, while activity induced rCBF responses remained significantly reduced up to the end of the recording period (90 min). At this time, baseline parameters of neuronal activity (indicated by ECoG), CBF and rCHo had returned to baseline. Additionally, was the change in the Hbr response not explained by a change in the CBF response onset timing.

The observed divergent modification of activity induced changes in rCBF, neuronal activity and oxygen consumption by CSD, indicates a long term change in neurometabolic, -vascular coupling and suggests a variable relationship between neuronal activity, oxygen consumption and CBF response in rat cerebral cortex. Taken together, these findings provide in vivo evidence for the concept of plasticity in neurometabolic -vascular coupling.

Supported by the DFG and the Hermann and Lilly Schilling Foundation.

ENDOTHELIAL NITRIC OXIDE SYNTHASE S1179 PHOSPHORYLATION IMPROVES CEREBRAL BLOOD FLOW AND REDUCES INFARCT SIZE IN FOCAL CEREBRAL ISCHEMIA

Hwa Kyoung Shin¹, Dmitriy N. Atochin², Sae-Won Lee¹, Paul L. Huang², Michael A. Moskowitz¹, Cenk Ayata^1,3

¹Stroke and Neurovascular Regulation Laboratory, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ²Cardiovascular Research Center and Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ³Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA

Background and aims: Endothelial nitric oxide synthase (eNOS) is an important enzyme regulating vascular tone, platelet aggregation and inflammatory cell migration in cerebral ischemia. eNOS activity is regulated by multiple mechanisms; among these, eNOS phosphorylation appears to be an important regulatory step. Phosphorylation of eNOS at serine 1179 (S1179) by protein kinase Akt increases its catalytic activity at all intracellular calcium concentrations. S1179 phosphorylation is induced by a number of physiological stimuli such as shear stress, and inhibited in diabetes, hyperlipidemia, and hypertension, and thus appears to be critical in both physiological and disease states. However, the impact of eNOS phosphorylation on vascular function and disease pathogenesis has not been studied, in vivo.

Methods: To determine the effect of eNOS S1179 phosphorylation on cerebral hemodynamic and metabolic response to ischemia, we subjected wild-type (WT, n=10), eNOS knockout (eNOS-/−, n=10), and transgenic mice expressing only phosphomimetic (S1179D, n=12) or unphosphorylatable (S1179A, n=11) mutant forms of eNOS (on eNOS-/− background) to distal or filament middle cerebral artery occlusion (dMCAO or fMCAO). Cerebral blood flow (CBF) was imaged non-invasively in ischemic core and penumbra using simultaneous laser speckle flowmetry.

Results: Total eNOS protein expression was similar between S1179A and S1179D lines, although both transgenic lines had reduced eNOS expression compared to wild type littermates by western blotting. The S1179 phospho-eNOS was not detectable in brain or heart in either transgenic line. Mean arterial blood pressures were elevated in eNOS-/−, S1179A, and S1179D mice compared to WT (p<0.05); arterial blood gas values did not differ. Upon dMCAO, the area of cortex with severe CBF deficit (i.e., ?20% CBF) was significantly greater in eNOS-/− mice compared to WT (p<0.05). Expression of the S1179A transgene in eNOS-/− mice did not improve the CBF deficit, while expression of the S1179D transgene reduced the CBF deficit close to that seen in WT (p<0.05 vs. eNOS-/−). Upon fMCAO, infarct volumes were smaller in S1179D and WT compared to S1179A transgenic and eNOS-/− mice (p<0.05).

Conclusions: These data provide the first in vivo evidence suggesting that phosphorylation of eNOS at S1179 is an important determinant of cerebral hemodynamic response to ischemia, and may be a useful target for the treatment and prevention of stroke.

This work was supported by the American Heart Association (0335519N, Ayata), NIH (P50 NS10828, Moskowitz; NS048426, Huang).

EFFECTS OF PROSTAGLANDIN E (PGE) ON REACTIVITY OF NON-PRESSURIZED CEREBRAL ARTERIOLES

JiSuk K. Yoo, David K. Kung, Leonid Groysman, Adam L. Sandler, H. Richard Winn

Department of Neurosurgery, Mount Sinai School of Medicine, New York, NY, USA

Background and Aims: Recent publications using in vitro brain slices have documented changes in intracerebral arterioles during astrocytic and neuronal activation. These in vitro intracerebral vessels studied in slices are non-pressurized and non-perfused. However, the physiological relevance of the observed changes in these arterioles is uncertain. We have previously shown that un-pressurized penetrating arterioles do not react to standard stimulants (changes in extra-luminal pH and adenosine) in an expected manner. Moreover, the addition of epinephrine (10-6M), a substance frequently used in cortical experiments, did not normalize the vascular response of un-pressurized arterioles. In the present study, we evaluate the potentiating effects of PGE.

Methods: Isolated penetrating rat arterioles were freshly harvested, cannulated, perfused, and pressurized as previously described (Ngai and Winn, 1995). Lumen diameter was measured with a video dimension analyzer. After a period of equilibration, we evaluated the response of these arterioles to change in the abluminal pH (6.8 & 7.6) and to adenosine (10-5 M). Between these challenges, the extra-luminal fluid was perfused with buffer and the vessel returned to baseline diameter. The reactivity of penetrating arterioles was then studied in absence and presence of PGE (7.5 ×10 −8 M) in two conditions: 1. Pressurized and perfused, simulating the in vivo state; group 2. Un-pressurized and non-perfused simulating conditions found in vitro in hippocampal and cortical slices.

Results: As indicated in Fig 1, pressurized and perfused arterioles reacted to stimulation in an expected manner. In the absence of intra-luminal pressure and flow, simulating conditions in a slice preparation, arterioles failed to dilate to a low pH or extraluminal Ado. The addition of PGE to the bath increased resting tone and restored reactivity in un-pressured and un-perfused vessels. We saw a similar trend in uncannulated, unpressurized vessels (data not shown).

Conclusion: In studying arteriole-neuronal interactions in brain slice, PGE can be used to simulate in vivo conditions.

[Supported by NS 021076]

PHOTOTHROMBOTIC OCCLUSION OF PENETRATING ARTERIOLES GENERATES SEVERE ISCHEMIA

Nozomi Nishimura¹, Chris B. Schaffer¹, Beth Friedman³, Patrick D. Lyden^3,4, David Kleinfeld^2,4,5

¹Department of Biomedical Engineering, Cornell University, Ithaca, NY, USA, ²Department of Physics, University of California at San Diego, San Diego, CA, USA, ³Department of Neurosciences, University of California at San Diego, San Diego, CA, USA, ⁴Graduate Program in Neurosciences, University of California at San Diego, San Diego, CA, USA, ⁵Center for Theoretical Biological Physics, University of California at San Diego, San Diego, CA, USA

Background and aims: Penetrating arterioles bridge the communicating arterioles on the surface of cortex to the subsurface microvascular bed that feeds the neurons. Optical methods were used to study ischemia due to clots in these vessels.

Methods: We used photochemically-triggered thrombosis to produce an occlusion in a single penetrating arteriole in the cortex of anesthetized rat. Two-photon microscopy was used to map the vasculature and measure red blood cell (RBC) velocity in vessels as deep as 400 µm before and after occlusion.

Results: We found that occlusions of penetrating arterioles caused severe ischemia and analyzed the spatial extent of the effect. Immediately downstream vessels dropped to 3% of initial speed. The blood flow increased with distance from the occluded arteriole and approached baseline values about 350 µm away from the occlusion. This suggests that the penetrating arterioles in rat parietal cortex primarily serve territories approximately 700 µm in diameter.

Conclusions: The ischemia produced by occlusions in penetrating arterioles is more severe than ischemia after occlusions in surface vessels and subsurface microvessels. The severe ischemia from occlusions of penetrating arterioles and arteries may be a cause of small lesions that are found in the brains of elderly humans. Such lesions are thought to play a role in neurodegenerative diseases.

Figure. (a) Schematic of photothrombosis of penetrating arterioles. (b) Fraction of the baseline speed is plotted versus the radial distance of the measured vessel from the occluded arteriole. Red line is average of 50 vessels, black lines show SEM.

POST-ISCHEMIC VASCULAR ADHESION PROTEIN-1 (VAP-1) BLOCKADE PROVIDES NEUROPROTECTION IN RATS SUBJECTED TO TRANSIENT MCAO

Lizhen Mao, Hao-Liang Xu, Dale Pelligrino

Neuroanesthesia Research Department, University of Illinois at Chicago, Chicago, IL, USA

BACKGROUND: There is no clear consensus regarding a role for leukocytes in the post-ischemic neuropathology associated with cerebral ischemia and reperfusion models. One example of this lack of agreement derives from studies employing neutrophil/leukocyte depletion paradigms, where both beneficial (1) and ineffectual (2) influences have been reported. However, the potential for “off-target” actions limits the usefulness of such strategies. We recently obtained evidence that VAP-1 (also called semicarbazide-sensitive amine oxidase [SSAO]), an enzyme concentrated in cerebral vascular cells, plays an important role in the process of post-forebrain ischemia neutrophil infiltration. In that study (1), we delayed treatment with the highly-selective inhibitor of VAP-1/SSAO (LJP-1207) until a time-point just prior to the onset of infiltration (6h reperfusion). Substantial reductions in post-ischemic neutrophil infiltration and significant neuroprotection (at 72h) were observed. However, the rats used in that study (diabetic, estrogen-treated ovariectomized females) exhibit an exaggerated post-forebrain ischemia neutrophil infiltration response. Such findings, therefore, do not necessarily indicate a beneficial role for VAP-1/SSAO inhibition in all cases of ischemia/reperfusion.

METHODS: In the present study, we examined the effects of LJP-1207 post-treatment (2h reperfusion) in a more commonly-used model; i.e., non-diabetic adult male rats subjected to 1h MCAO (intraluminal suture) and 3 days reperfusion. Prior to these experiments (in separate rats), we monitored, using cranial window-based intravital microscopy of pial venules overlying the ischemic cortex, patterns of adhesion and infiltration of neutrophils over the initial 10h of reperfusion following 1h MCAO.

RESULTS: Perivascular accumulation of leukocytes was seen to gradually rise up to 10h (expressed as the percentage of the viewed venular area-fig.1), with signs of extravasation seen (in 50% of the rats) after 8h. This represents a more modest neutrophil response than in the diabetic females cited earlier. Nevertheless, VAP-1/SSAO-inhibited rats displayed a 53% reduction in infarct volume, relative to controls (fig. 2A), and a significant improvement in neurologic outcome (fig. 2B).

CONCLUSIONS: These findings suggest that VAP-1/SSAO inhibition may also represent an effective intervention for limiting neuropathology following transient focal ischemia. The mechanism may relate to preventing generation of SSAO products like formaldehyde and hydrogen peroxide, thereby limiting leukocyte/vascular interactions. References: (1) J Pharmacol Exp Ther 317:19, 2006; (2) Stroke 37:1288, 2006.

DETECTION OF EXTRASTRIATAL DOPAMINE RELEASE IN HEALTHY HUMAN VOLUNTEERS

Rajendra Badgaiyan, Alan Fischman, Nathaniel Alpert

Division of Nuclear Medicine, Harvard Medical School and Massachusetts General Hospital, Boston, MA, USA

Detection of task-induced release of dopamine in the extrastriatal brain areas is critical for understanding of the brain mechanisms that control human cognition. The detection in these areas however has been a challenge because of relatively low receptor density. We have earlier demonstrated that task-induced release of striatal dopamine can be detected using a molecular imaging technique. This demonstration however, is of limited use because it cannot be used to detect dopamine released in the extrastriatal areas, where most of human cognitive processing takes place. Since the high affinity dopamine receptor ligand fallypride has been shown to bind in detectable quantities in the human extrastriatal areas, we used this ligand to detect task-induced release of extrastriatal dopamine. In this experiment, healthy volunteers performed an emotional memory task in the scanner after receiving a single intravenous bolus of 18F-fallypride. The task consisted of a control (no emotional memory) and a test condition. The control condition was started after the ligand administration. The test condition was started 25 min after the ligand injection. The ligand concentration was measured using an ECAT HR+ camera in 60 sec intervals. Analyses of behavioral data suggested that emotional memory was elicited in the test condition. Immediately after initiation of this condition the rate of ligand displacement increased significantly in the extrastriatal brian areas that have previously been implicated in the processing of emotional memory. These areas included amygdala, medial temporal lobe and ventral prefrontal cortex. The displacement rate did not change significantly in the reference region (cerebellum). Since the experiment found evidence of significant dopamine release in the extrastriatal areas that have previously been implicated in the processing of emotional memories, the results suggest that the ligand 18F-fallypride can be used to detect dopamine released in the extrastriate brain areas during performance of a cognitive task.

CILOSTAZOL MAINTAINS DOPAMINE AND SUBSTANCE P LEVELS AND IMPROVES SWALLOWING REFLEX IN A RAT CHRONIC CEREBRAL HYPO-PERFUSION MODEL

Ning Zhang^1,2, Terubumi Watanabe¹, Nobukazu Miyamoto¹, Ryota Tanaka¹, Nobu Hattori¹, Takao Urabe¹

¹Depertment of Neurology, Juntendo University School of Medicine, Tokyo, Japan, ²Research Institute for Disease of Old Age, Juntendo University School of Medicine, Tokyo, Japan

Background and aims: Pneumonia is a common complication with the highest attributable proportion of death in the entire stroke population. Dysphagia with aspiration and the depression of cough reflex caused by central nervous disorder are believed the most important factors contributing to the risk of post-stroke pneumonia. Cilostazol is a potent inhibitor of type III phosphodiesterase used as an anti-platelet agent for the treatment of chronic cerebral infarction. The present study was designed to assess the pneumonia prevent mechanism of Cilostazol through its effects on regulating the dopamine and substance P levels as well as on improving the swallowing reflex in rat model of chronic cerebral hypo-perfusion.

Methods: Rats were subjected to bilateral common carotid artery ligation (LBCCA) and divided randomly into the vehicle group (n=72) and the Cilostazol group (n=72). Saline elicited swallowing reflex was evaluated at baseline and at days 14, 28 and 42 days after LBCCA. The levels of dopamine and substance P in the striatum were detected. The immunohistochemistry and immunoblotting analyses for tyrosine hydroxylase (TH), cyclic adenosine monophosphate responsive element binding protein (CREB) and phosphorylated CREB (p-CREB) in the substantial nigra were also performed at the investigated time points.

Results: Cilostazol significantly improved the elicited swallow reflex by shortening the latent time (3.7±0.2 vs 4.3±0.24 sec; P< 0.05) and increasing the numbers of swallow reflex (8±0.3 vs 6±0.8; P<0.05) at 14 days of hypo-perfusion. Compared with the vehicle group, Cilostazol up-regulated CREB phosphorylation and increased TH expression in the substantial nigra; in the striatum, Cilostazol markedly maintained dopamine (84.71±2.26 vs 79.2±4.13 % control; P<0.05) and substance P levels (86.65±7.89 vs 73.88±6.52% control; P<0.001) at 14 days of hypo-perfusion.

Conclusions: Our results indicate that based on up-regulating the expression of TH through cell signaling pathway of CREB phosphorylation, Cilostazol modulates the striatal dopamine and substance P levels and improves the swallowing reflex, and suggest that Cilostazol is an adequate intervention to sufficiently prevent pneumonia in the patients of chronic cerebral ischemia.

INFLUENCE OF SURGICAL TREATMENT FOR BRAIN AND ITS FUNCTION IN PATIENTS WITH UNRUPTURED CEREBRAL ANEURYSMS

Yoshiaki Kumon, Hideaki Watanabe, Keiji Igase, Hiroaki Matsumoto, Takanori Ohnishi

Department of Neurosurgery, Ehime University Graduate School of Medicine, Toon City, Ehime, Japan

Background and aims: We evaluated the influence of direct surgical treatment in patients with unruptured cerebral aneurysms of the anterior circulation.

Methods: 133Xe-SPECT and Wechsler Adult Intelligence Scale - Revised (WAIS-R) examination were done in 30 patients (31 operations), before and 1 month after surgery.

Results: Deterioration of total IQ (more than 3 scores) was recognized in 8 operations (26%), and resting CBF decreased significantly in the ACA and anterior border zone (BZ) areas. In 10 operations for ACoA aneurysm, WAIS-R results deteriorated in 4, and resting CBF decreased in the ACA and anterior BZ areas. In 9 operations for MCA aneurysm, WAIS-R results deteriorated in 4, and cerebrovascular response decreased in the anterior and posterior BZ areas. In 12 operations for ICA aneurysm, no patient showed deterioration in WAIS-R results and CBF. In 4 patients older than 70 years old, WAIS-R results deteriorated in 3.

Conclusions: WAIS-R and CBF results deteriorated in considerable number of patients with ACoA or MCA aneurysm, and aged patients. We should recognize these baneful influences of direct surgery.

PLATELET AGGREGABILITY USING WHOLE BLOOD AFTER ANTIPLATELET AGENTS IN PATIENTS WITH NON-CARDIOEMBOLIC STROKE

Daisuke Furuya, Norio Tanahashi, Nobuo Araki, Tomokazu Shimazu, Harumitsu Nagoya, Yuji Kato, Yasuo Ito, Mikiko Ninomiya, Kunio Shimazu

Department of Neurology, Saitama Medical School, Moroyama, Saitama, Japan

Background and aims: The purpose of this study was to measure platelet aggregability using a new method of analysis, whole blood aggregation (WBA), and to compare inhibitory effects of antiplatelet agents.

Methods: Subjects comprised 52 consecutive patients with non-cardioembolic stroke. Mean age was 66±15 (mean±SD) years (range, 53-88 years). Platelet function was measured using WBA-neo (ISK Co, Japan) after pharmacotherapy. Four concentrations of adenosinediphosphate (ADP) and collagen were used as aggregating agents. Participants were retrospectively divided into three groups, receiving aspirin (100mg/day, n=30), cilostazol (200mg/day, n=13) and clopidogrel (75mg/day, n=9). Differences in platelet aggregations among three groups were compared using thresholds showing 50 % of maximum aggregating rate as inhibitory parameters.

Results: ADP-induced platelet aggregation showed significantly (p<0.05) greater inhibition in clopidogrel group (6.8±2.2 micromol/L) than in aspirin and cilostazol group (4.2±2.4, 2.8±1.9; figure A). Collagen-induced platelet aggregation displayed significantly (p<0.05) stronger inhibition in aspirin group than in cilostaazol and clopidogrel group (figure B).

Conclusions: Clopidogrel reduced ADP-induced platelet aggregation (PA) specifically, and aspirin reduced collagen induced PA.

INCREASING DOSES OF RT-PA REDUCE MMP-3 ACTIVITY FOLLOWING EXPERIMENTAL FOCAL ISCHEMIA

Dorothe Burggraf¹, Sebastian Schuhmacher¹, Martin Dichgans¹, Gerhard Hamann²

¹Department of Neurology, Ludwig-Maximilians University, Klinikum Grosshadern, Munich, Germany, ²Department of Neurology, HSK Dr. Horst Schmidt Klinik, Wiesbaden, Germany

Background and aims: Focal cerebral ischemia results in gradual breakdown of the extracellular matrix. Matrix metalloproteinases (MMPs) especially MMP-9 are important in this process. Recombinant tissue plasminogen activator (rt-PA) results in up-regulation of MMP-9. Pro MMP-9 is activated through active MMP-3 (stromelysin), to active MMP-9. We therefore hypothesized that rt-PA increases the activation of MMP-9 by activating MMP-3. To convert the precursor protein pro-MMP-3 into the active form, a proteolytic cleavage and the cleavage of the pro-form to lower molecular weight active froms (45, 28, and 21 kDa) is essential. Methods: In this study we quantified MMP-3 expression following ischemia (3h)/reperfusion (24h) in a suture model in rats. We observed changes of this activation after the treatment of the animals with rt-PA in different doses. Each group (n=6) received either saline, 0.9, 9 or 18 mg/kg bw rt-PA at the end of the ischemic period. MMP-3 was determined by western blot of the ischemic and non ischemic basalganglia and cortex separately. Zymography was used to confirm the results.

Results: Active MMP-3 (45kDa form) was slightly increased in the ischemic region compared to the contralateral area (cortex: 120·7%; basalganglia: 109·9%; p<0.05 versus sham operated rats). In contrast, the 21 kDa degradation fragment was significantly (p<0.05) decreased. With rising rt-PA doses a decrease of active MMP-3 was shown in ischemic cortical and basal ganglia region respectively (0.9?9-18mg: cortex: 86 ·7%; 85·7%;76·14%, p<0.001; basal ganglia: 97·8%;82·9%;77·10%; p<0.001) joint to a steady increase of the degraded form. Zymographic analysis showed that the active form of MMP-9 is rising with increasing rt-PA doses.

Conclusions: Our results suggest that active MMP-3 was depleted during the 3h ischemia and 24 h reperfusion period. The amount of MMP-3 seems to be a limiting factor in the activation cascade of MMP-9. So active MMP-3 could be a target to inhibit the MMP-9 destructive upregulation following rt-PA treatment.

ENERGY METABOLISM OF LYMPHOCYTES FROM ACUTE STROKE PATIENTS

Roberto Federico Villa¹, Antonella Gorini¹, Diego Geroldi², Guglielmo Pero³, Francesca Brera¹

¹Department of Physiological-Pharmacological Sciences, University of Pavia, Pavia, Italy, ²Department of Internal Medicine, University of Pavia, Pavia, Italy, ³Department of Neuroradiology, Niguarda Ca’ Granda Hospital, Milano, Italy

BACKGROUND AND AIMS - Systematic studies on energy metabolism of lymphocytes during ageing in male and female subjects are lacking, especially in patients with CNS Diseases. A study on cytoplasmic and mitochondrial enzymes of lymphocytes, as biological indicators of energy metabolism, was performed in healthy blood donors and in patients with acute stroke.

METHODS - The controls and patients were divided according to the: (a) sex, to study gender factor; (b) ages (from 20-30 to 60-70 up to 86 years groups) to study age factor; (c) patients with acute stroke (female: 31-91 years and male 53-86 years groups). On purified lymphocytes, the catalytic properties of the following enzymes were assayed: for glycolysis, hexokinase (HK), lactate dehydrogenase (LDH); for Krebs' cycle (TCA), citrate synthase (CS), malate dehydrogenase (MDH); for Electron Transfer Chain (ETC), NADH-cytochrome c reductase (Complex I, Complex I-III), succinate dehydrogenase (SDH, Complex II), cytochrome oxidase (COX, Complex IV); for glutamate and related amino acids metabolism, glutamate dehydrogenase (GlDH), glutamate-oxaloacetate- (GOT) and glutamate-pyruvate- transaminases (GPT); some enzyme activities were assayed also in presence of Triton X-100 (TX).

RESULTS - The enzyme activities of lymphocytes differ for gender, particularly during ageing (ANOVA test): HK increased (male); CS increased starting at 31 (male) and 51 (female) years, SDH at 51 (male) or 61 (female) years; ETC enzymes increased starting at 31 years, only in male. Amino acid linked enzymes were unaffected by age (male), only GPT increased at 31-50 years (female). In stroke male patients, LDH, CS activities increased, while COX, GlDH and GlDH-TX decreased; in female, HK, CS, MDH, Complex I, Complex I-III, GPT activities increased, while Complex IV, GlDH and GlDH-TX activities decreased and also proteins concentration.

CONCLUSIONS - This systematic study on functional proteomics of lymphocytes during physiological ageing, allows to elaborate a model to study the CNS Diseases (cerebral ischemia in this case) and, eventually, the drugs' actions. The most representative enzyme activities related to energy trasduction, were used as biological markers to monitor the energy metabolism on peripheral cells, reflecting that of the brain, as the literature suggests.

In control subjects, the cytosolic and mitochondrial enzyme activities are much more changed in their catalytic activities by ageing in female respect to male. Stroke strongly affected the energy-linked enzyme systems of lymphocytes, but differently for gender and age. Glycolysis, TCA, ETC and glutamate enzymes are particularly affected in female than in male, demonstrating that the glucose metabolism and ATP synthesis are primary affected in lymphocytes, suggesting the brain metabolic dysfunction. In fact, during the stroke, these results suggest an increased energy requirement that is not compensated by cytochrome oxidase activity, which decreased in male and female patients. In addition, in female patients, there is an increased metabolism of pyruvic acid via transamination reaction by GPT and a decreased capacity to metabolize glutamate (GlDH decrease), while in male patients there is an increased anaerobic metabolism (LDH increase). These metabolic modifications may play an important physiopathological role, according to the observed cerebral dysfunctions of patients with acute cerebral ischemia.

REGIONALLY INCREASED TEMPORAL LOBE EFFLUX OF R-[11C]VERAPAMIL OCCURS IN SOME PATIENTS WITH THERAPY REFRACTORY EPILEPSY

Martin Bauer¹, Oliver Langer^1,2, Alexander Hammers^3,4, Rudolf Karch⁵, Ekaterina Pataraia⁶, Matthias J. Koepp⁴, Aiman Abrahim¹, Gert Luurtsema⁷, Kurt Kletter⁸, Christoph Baumgartner⁶, Markus Muller¹

¹Division of Clinical Pharmacokinetics, Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria, ²Department of Radiopharmaceuticals, Austrian Research Centers, Seibersdorf, Austria, ³Division of Neuroscience, Faculty of Medicine, Imperial College, and MRC Clinical Sciences Centre, Hammersmith Hospital, London, UK, ⁴Department of Clinical and Experimental Epilepsy, Institute of Neurology, University College London, London, UK, ⁵Department of Medical Computer Sciences, Medical University of Vienna, Vienna, Austria, ⁶Department of Neurology, Medical University of Vienna, Vienna, Austria, ⁷Department of Nuclear Medicine and PET Research, VU University Medical Center, Amsterdam, The Netherlands, ⁸Department of Nuclear Medicine, Medical University of Vienna, Vienna, Austria

Background and aims: R-[11C]verapamil, a radiolabelled substrate of the multidrug transporter P-glycoprotein (P-gp), has been developed as a probe for in vivo evaluation of P-gp function at the human blood-brain barrier with PET (Lubberink et al., 2006). Because regional overexpression of P-gp in epileptic brain tissue is discussed as a factor that contributes to pharmacoresistance in epilepsy, we studied brain R-[11C]verapamil distribution in 7 patients with drug-resistant unilateral therapy-refractory temporal lobe epilepsy.

Methods: After i.v. bolus injection of about 400 MBq of R-[11C]verapamil (specific activity: >20 GBq/µmol) dynamic brain PET imaging and arterial blood sampling were performed for 60 min. Volumes of interest (VOIs) were defined on MRI scans by using a frequency-based anatomical atlas (Hammers et al., 2003). R-[11C]verapamil kinetics were compared in homologous brain VOIs located ipsilateral and contralateral to the seizure focus. PET data were modeled by a previously described 1-tissue compartment model using a partially metabolite-corrected arterial input function (Lubberink et al., 2006).

Results: Among different VOIs, radioactivity was highest in the choroid plexus. The hippocampal VOI could not be used for data analysis because it was contaminated by spill-in of radioactivity from the adjacent choroid plexus. In several other temporal lobe regions that are known to be involved in seizure generation and propagation ipsilateral influx rate constants K1 and efflux rate constants k2 of R-[11C]verapamil were increased as compared to the contralateral side. Parameter asymmetries were most prominent in parahippocampal and ambient gyrus (K1, range among patients: −3.8% to +22.3%; k2, range: −2.3% to +43.9%), amygdala (K1, range: −20.6% to +31.3%; k2, range: −18.0% to +38.9%) and medial anterior temporal lobe (K1, range: −8.3% to +14.5%; k2, range: −14.5% to +31.0%). In contrast to temporal lobe VOIs, asymmetries were minimal in a region not involved in epileptogenesis located outside the temporal lobe (superior parietal gyrus, K1, range: −3.7% to +4.5%; k2, range: −4.2% to +5.8%). In five of seven patients, ipsilateral efflux (k2) increases were more pronounced than ipsilateral influx (K1) increases, which resulted in prominent ipsilateral reductions (10–26%) of R-[11C]verapamil distribution volumes.

Conclusion: Our pilot data suggest that regionally enhanced P-gp activity might contribute to pharmacoresistance in some patients with temporal lobe epilepsy.

THE USE OF OXYGEN AS A CONTRAST AGENT AND METABOLIC BIOTRACER IN ACUTE STROKE PATIENTS

Krishna Dani¹, Keith Muir¹, David Brennan², I. Mhairi Macrae¹, Barrie Condon², Celestine Santosh³

¹University of Glasgow, Glasgow, Scotland, UK, ²Department of Clinical Physics, Institute of Neurological Sciences, Glasgow, Scotland, UK, ³Department of Neuroradiology, Institute of Neurological Sciences, Glasgow, Scotland, UK

COMPARISON BETWEEN CEREBRAL OXYGEN METABOLISM AND C-11 FLUMAZENIL PET IN THE PATIENTS WITH TRAUMATIC BRAIN INJURY

Tohru Shiga¹, Katsunori Ikoma³, Chietsugu Katoh⁴, Hirotaka Isoyama³, Yuji Kuge¹, Tomoya Kohno², Satoshi Terae⁵, Nagara Tamaki²

¹Department of Molecular Imaging, Hokkaido University, Sapporo, Japan, ²Department of Nuclear Medicine, Hokkaido University, Sapporo, Japan, ³Department of Rehabilitation, Hokkaido University, Sapporo, Japan, ⁴Department of Health Science, Hokkaido University, Sapporo, Japan, ⁵Department of Radiology, Hokkaido University, Sapporo, Japan

NMDA RECEPTOR ANTAGONIST SM-31900 PROTECTS AGAINST DELAYED NEURONAL DEATH AND REDUCES NEUROGENESIS INDUCED BY TRANSIENT GLOBAL ISCHEMIA IN RATS

Taku Sugawara¹, Suguru Yamaguchi¹, Masaya Oda¹, Tomoya Omae¹, Kazuo Mizoi¹, Hiroyuki

Kinouchi²

¹Department of Neurosurgery, Akita University School of Medicine, Akita, Japan, ²Department of Neurosurgery, University of Yamanashi, Faculty of Medicine, Kofu, Japan

The N-methyl-D-aspartate receptor (NMDAR) is known to play an important role in ischemic neuronal injury in the hippocampus, and also implicated in modulation of neurogenesis in the hippocampal dentate gyrus (DG). Recent evidence confirmed that amino acid neurotransmitter glycine acts as a coagonist in excitatory neurotransmission at NMDAR. The aim of this study is to investigate the role of glycine in ischemic injury and neurogenesis in the hippocampus in the rat global ischemia model. Global ischemia was induced with or without the treatment with NMDAR antagonist or NMDAR glycine site blocker. Neuronal injury was morphologically examined and neurogenesis was assessed by bromodeoxyuridine uptake. Global ischemia induced severe injury in the hippocampal CA1 neurons, but pharmacological antagonism of NMDAR or glycine site ameliorated the injury. Global ischemia enhanced neurogenesis in DG, but these pharmacological interventions resulted in attenuation of neurogenesis. These results suggest that glycine plays an important role in NMDAR-mediated neuronal injury and regulation of neurogenesis in the hippocampus.

NITRIC OXIDE PRODUCTION IN RAT BRAIN OF DIABETES MELLITUS TYPE 1 AND 2 MODELS DURING FOREBRAIN ISCHEMIA AND REPERFUSION

Tomokazu Shimazu, N. Araki, T. Ohkubo, Y. Asano, D. Furuya, H. Nagoya, Y. Ito, Y. Kato, M. Ninomiya, K. Shimazu

Division of Neurology, Deperment of Internal Medicine, Saitama Medical University, Saitama, Japan

Background and Aims: Experimental and clinical studies suggest that oxidative stress plays an important role in pathogenesis of diabetes mellitus type 1 and 2. Hyperglycemia leads to free radical generation and causes neuronal injury. The purpose was to investigate the nitric oxide (NO) production in rat brain of diabetes mellitus type 1 and 2 models during forebrain ischemia and reperfusion.

Methods: Wistar rats were divided into two groups: control (n=7) and type 1 diabetic group (n=7) which was induced by injection of streptozotocin (STZ) (40 mg/kg). Otsuka Long-Evans Tokushima Fatty (OLETF; n=6) and Long-Evans Tokushima Otsuka (LETO: n=5) rats were used as type 2 diabetic and control groups. The animals were anesthetized by pentobarbital sodium (i.p.). NO production was continuously monitored by in vivo microdialysis. Microdialysis probes were inserted into the left striatum and hippocampus. The probes were perfused with Ringer's solution at a constant rate of 2 µl/min. Laser Doppler probes were inserted into the right striatum and hippocampus. After a 2-hours equilibrium period, fractions were collected every 10min forebrain cerebral ischemia was induced by occlusion of both common carotid arteries, and systemic hypotension (MABP<50mmHg) which was induced by hemorrhage. After 20min, the loops around both common carotid arteries were released and the blood was reinfused. Levels of NO metabolites, nitrite and nitrate, in the dialysate were determined using the Griess reaction.

Results: (1) Cerebral Blood Flow (CBF): In all groups, CBF decreased rapidly after occlusion and systemic hypotension, and rapidly increased after reperfusion. There were no significant differences in CBF between the each two groups. (2) NO production during ischemia and reperfusion: during ischemia the changes in total NO (sum of nitrite and nitrate) in striatum in STZ group (91.4±4.7%; mean±SE, 63.7±2.0% of the baseline level) were significantly higher than those of the control group (64.4±6.8%, 48.6±6.5% of the baseline level) (p<0.05) (Fig.1.). There were no significant differences in hippocampus. In the OLETF groups, no significant differences were obtained in striatum. NO production during 10min after reperfusion the change in total NO in hippocampus of OLETF group was higher than that of LETO group (p<0.1) (Fig.2).

Conclusion: These data indicate that NO may be an important mediator in the exacerbation of post-ischemic brain injury in the striatum in STZ rat, and in the hippocampus in OLETF rat.

INCREASED BASELINE K+ DOES NOT CONTRIBUTE TO CORTICAL SPREADING DEPOLARIZATION (CSD)-INDUCED MICROVASCULAR SPASM VIA INCREASED PEAK K+ DURING CSD

Mesbah Alam, Sebastian Major, Jens P. Dreier

Department of Neurology and Experimental Neurology, Charite Hospital, Berlin, Germany

Introduction: It has been proposed that CSD-induced microvascular spasm (cortical spreading ischemia, CSI) is a mechanism of delayed ischemic neurological deficits after subarachnoid hemorrhage (Dreier et al., 1998; 2000; 2006). CSD/CSI is, e.g., induced by brain topical application of artificial cerebrospinal fluid (ACSF) containing increased K+ combined with either the NO scavenger hemoglobin or the NO synthase inhibitor N-nitro-L-arginine (L-NNA). The role of artificially increased baseline extracellular K+ concentration ([K+]o) for CSD/CSI has remained unclear. [K+]o rises to ~50mM during CSD and acts as a vasoconstrictor in concentrations above 20mM. Therefore, if artificial increase of baseline [K+]o caused increased peak [K+]o during CSD, it would provide a simple explanation how increased baseline [K+]o contributes to CSD-induced microvascular spasm. In the present study, we compared peak [K+]o between CSD/cortical spreading hyperemia (CSH) under physiological ACSF and CSD/CSI under L-NNA (1mM) with increased K+ (35mM).

Methods: An open cranial window was implanted in 26 rats under thiopental. Cerebral blood flow (CBF) was measured with laser-Doppler flowmetry, and the subarachnoid direct current (DC)-potential (DCsa) with an AgCl electrode. Intracortical DC-potential (DCic) and [K+]o were recorded with ion-sensitive microelectrodes (cortical depth, 100 µm).

Results: CSD/CSI was associated with a negative shift of DCsa (−4.0±1.8mV) and DCic (−19.7±1.9mV) lasting for 5.8±3.9min and 6.2±3.6min, respectively. Immediately before CSD/CSI, CBF was 94±27% of baseline (= 100%) caudally and 125±74% rostrally. CSI consisted of a sharp CBF decrease which spread between caudal and rostral laser probes with a latency of 21.1±13.3s. The minimal CBF value was 21±11% lasting for 4.4±4.2min. Immediately before CSD/CSI, [K+]o was 9.3±3.7mM and rose to 49.5±8.0mM during CSD/CSI. The negative shift of DCsa started 11.2±14.4s before the sharp CBF decrease in CSD/CSI whereas, after cardiac arrest, a simultaneous sharp CBF decrease at both laser probes preceded DCsa by 41.0±22.2s. CSD/CSH was associated with a negative shift of DCsa (?1.6±0.8mV) and DCic (−18.0±3.0mV) lasting for 1.6±0.8min and 1.6±0.5min, respectively. Immediately before CSD/CSH, CBF was 110±13% of baseline (=100%) caudally and 105±19% rostrally. CSH consisted of a rise in CBF to 285±65% spreading with a latency of 47.8±28.5s and lasting for 1.9±0.4min. Immediately before CSD/CSI [K+]o was 2.5±0.8mM and rose to 46.3±13.1mM during CSD/CSH. The negative shift of DCsa started 15.7±23.9s before the CBF rise in CSD/CSH whereas, after cardiac arrest, a simultaneous sharp CBF decrease at both laser probes preceded DCsa by 41.0±22.2s.

Conclusion: In contrast to intracortical baseline [K+]o before CSD, no significant difference of peak [K+]o was found between CSD/CSH and CSD/CSI. In conclusion, rise of baseline [K+]o before CSD/CSI does not contribute to CSD-induced microvascular spasm via increase of peak [K+]o during CSD.

BLOOD FLOW DEPENDENT CHARACTERISTICS OF OPTICAL INTRINSIC SIGNALS DURING CORTICAL SPREADING DEPRESSION IN RATS

Zhen Wang, Pengcheng Li, Weihua Luo, Qingming Luo

Wuhan National Laboratory for Optoelectronics, Britton Chance Center for Biomedical Photonics, Huazhong University of Science and Techonlogy, Wuhan, China

This study was aimed to investigate the potential influence of the residual cerebral blood flow (rCBF) on the spatiotemporal characteristics of optical intrinsic signals (OIS) induced by spreading depression (SD) in local ischemic rat cortex. For this purpose, SD waves were induced by K+ on rat cortex with small cerebral ischemia focal, and residual CBF and IOS were accessed by combining laser speckle flowmetry with optical intrinsic signal imaging. Adult male Sprague-Dawley rats (n=10) weighing 200 to 300 g were anesthetized with α-chloralose/urethane and the parietal-temporal cortex was exposed through craniectomy. Small branches of middle cerebral artery which serve targeted cortex were carefully ligated by passing short pieces of 11-0 suture through the dura and under the vessels to cause a small local cerebral ischemia (around 3 mm in diameter). A second cranial window located in the frontal bone was made for the induction of SD. For the CBF measurement using laser speckle flowmetry, a He-Ne laser (632.8nm, 17mW, Melles Griot, U.S.A) was used to illuminate the cortex, and a fixed wavelength (550±10 nm) light source was used for illumination when performing optical intrinsic signal imaging. Both the two techniques employed the same optics (Microscope SZX12, Olympus, Japan) and CCD camera (12-bit, 480×640 pixel array, Pixelfly VGA, Germany) for image acquisition. Six hours after ministroke, SD was induced by applying 3M KCl (ca. 50µl) into the cranial window located at the frontal cortex.

In the experiments, four kinds of temporal pattern of changes in OIS are observed at the cortical locations with different level of rCBF perfusion. At rCBF ranging from 40?60% of baseline, OIS expressed more significant initial rise in light reflectance than that observed at the cortical region with rCBF greater than 60% of baseline. More interestingly, for the cortical region where rCBF dropped to 30–40% of baseline, the duration of changes in OIS caused by SD was significantly prolonged and the number of SD waves decreased. For the region with rCBF below 30% of baseline, there were no SD occurred and also no significant changes in optical reflectance observed. Our experimental results indicate a tight coupling between the characteristics of OIS and CBF levels during cortical spreading depression. This new finding suggests that OIS induced by SD might be used as a reflection of blood flow levels to some extent.

Keywords: cerebral blood flow, spreading depression, optical intrinsic signals, laser speckle imaging

DELAYED NEURONAL CELL DEATH IN THE BRAINSTEM AFTER TRANSIENT ISCHEMIA IN GERBILS

Fang Cao, Ryuji Hata, Pengxiang Zhu, Masahiro Sakanaka

Department of Functional Histology, Ehime University Graduate School of Medicine, Toon, Ehime, Japan

INFLUENCE OF ENOS-GLU298ASP-POLYMORPHISM ON L-ARGININE INDUCED CEREBRAL BLOOD FLOW REACTIVITY

Karen Gertz¹, Peter Brunecker¹, Jorg Schultze¹, Stephan Schreiber¹, Florian Doepp¹, Jens Steinbrink¹, Constanze Jung², Karl Stangl², Christoph Naber³, Matthias Endres¹

¹Department of Neurology, Charite Universitatsmedizin Berlin, Berlin, Germany, ²Department of Cardiology, Charite Universitatsmedizin Berlin, Berlin, Germany, ³Department of Cardiology, University Essen, Essen, Germany

Background: Nitric oxide (NO), synthesized by endothelial nitric oxide synthase (eNOS), plays a key role in vascular regulation and is an important target for stroke protection. Epidemiological evidence links DNA variants in the eNOS gene to increased vascular risk in humans. Here, we studied the impact of the Glu298Asp (G894T) polymorphism which generates a protein with different susceptibility to cleavage and impaired enzymatic activity on cerebral blood flow (CBF).

Methods: The study was approved by institutional review boards and all participants gave informed consent. The study population consisted of 20 healthy volunteers which were matched according to Glu298Asp polymorphism determined by polymerase chain reaction (PCR): G/G (homozygous “normal” alleles), G/T (heterozygous) and T/T (homozygous polymorphism) genotype (5, 9 and 6 individuals, respectively). CBF was stimulated via infusion of the NOS substrate L-Arginine (500mg/kg body weight). Immediately before and after the infusion T2*-weighted dynamic susceptibility contrast (DSC) MRI was performed. CBF, cerebral blood volume (CBV), and mean transit time (MTT) were voxelwisely estimated by a common deconvolution analysis. Maps of time to peak (TTP) were calculated from the bolus data including the peripheral transport.

Results: To detect small changes in cerebral perfusion, we optimized the post-processing of DSC-MRI using a novel correction algorithm, which accounts for saturation effects in the arterial input curve (AIC) reducing variations in the estimated parameters (Brunecker et. al., ISMRM 2003). As a result of this analysis maps of tissue arrival time (TAT) as the first moment of an analytical tissue response function were determined. All above mentioned parameters were measured in different regions of interest (ROIs) including white matter, foramen semiovale, thalamus, caudatum and over a whole transversal slice.

TTP was significantly reduced after infusion with L-Arginine in individuals with G/G genotype in all measured ROIs (1,23; 1,42; 1,69; 2,12; 2,58 sec, respectively), but was not shortened in individuals with G/T and T/T alleles. Furthermore we found that in individuals with G/G or G/T genotypes TAT was significantly reduced after L-Arginine over the whole slice (0,24 and 0,21 sec, respectively) and in foramen semiovale (0,23 and 0,44 sec, respectively). In contrast, in individuals with T/T genotype L-Arginine had no significant effect on TAT.

Conclusion: Our study has the following major findings. 1) Perfusion changes in the brain induced by L-Arginine can be measured by DSC-MRI. 2) The effect of L-Arginine on different blood flow parameters measured by DSC-MRI is more pronounced in TTP and TAT as compared to MTT and CBF. Whereas TTP is more related to peripheral flow TAT reflects cerebral flow changes within the arterial vascular bed. In contrast, there was no apparent effect on CBF, which represents blood flow over the whole vessel supply from arterioles to venules. 3) Importantly, in individuals homozygous for the Glu298Asp polymorphism the blood flow response to L-Arginine was blunted or absent (ie, no reduction of TTP and TAT). In conclusion, our findings show reduced endothelial function in patients homozygous for Glu298Asp polymorphism.

MITOCHONDRIAL OXYGENATION IN THE RAT CEREBRAL CORTEX

Roger Springett, Michelle Abajian, Maureen Ripple

Department of Radiology, Dartmouth Medical School, Hanover, NH, USA

Introduction: The oxygen limitation model (OLM) is the prevailing model to explain why increases in cerebral blood flow (CBF) are much greater than increases in cerebral metabolic rate for oxygen (CMRO2) during functional activation in the brain. Within the framework of the OLM, oxygen diffuses from the capillary bead to the mitochondria where it is consumed. The rate of oxygen diffusion (DO2), which is equal to CMRO2, is given by:

DO2 = CMRO2 = L(PcO2-PmO2)

where PcO2 and PmO2 are the mean capillary bed and mitochondrial oxygen tension respectively and L is the effective permeability constant.

A fundamental assumption of the OLM is that PmO2 is zero, or near zero. Under these conditions, large increases in CBF are necessary to raise PcO2 and increase the oxygen gradient between capillary bed and mitochondria to support a modest increase in CMRO2. The major challenge in testing this model is that there are no methods of measuring PmO2.

We have developed and validated a visible spectroscopy system that can measure the oxidation state of mitochondrial cytochrome c (Cytc), which is sensitive to PmO2, as well as the absolute hemoglobin concentration and hemoglobin saturation (SmcO2) from the rat cerebral cortex using a thinned skull preparation. We have integrated this system with laser Doppler flowmetry and, using SmcO2 as a surrogate of local venous saturation, we are able to calculate CMRO2. The goal of this study is to use this system critically test the OLM.

Methods: Cortical Spreading depression (CSD) is a wave of depolarization that propagates across the cortex at a rate of 2-4mm/minute. At the wave front there is a large increase in CMRO2, presumably to generate ATP to restore membrane polarity, which drives PmO2 to low levels and results in a reduction of Cytc. Cytc cannot be used to measure PmO2 directly because the relationship between PmO2 and Cytc is not precisely known. Instead, we approximate PmO2 as zero at the maximum reduction in Cytc and use PcO2, derived from SmcO2 using the rat hemoglobin disassociation curve, to calculate L. Once L is known, baseline PmO2 can be calculated from the baseline PcO2 and CMRO2.

Sprague Dawley rats were anesthetized with isoflurane, the skull thinned for optical access and the animals artificially ventilated. CSD was induced by applying saturated KCL to the dura close to the measurement site.

Results: Baseline SmcO2 was 73.7±3.1% (mean±SD, n=6) giving a PcO2 of 54±3mmHg. CSD increased CMRO2 and CBF by 75±18% and 8±19% respectively, decreased SmcO2 and PcO2 to 59.7±3.0% and 42.8±2.0mmHg respectively and reduced Cytc by 29.6±8.9%. Assuming PmO2 was zero at this point results in a value for L of 0.260±0.028 mmHg/%CMRO2 and a baseline PmO2 of 28.4±4.6mmHg.

Discussion: The calculated value of PmO2 is not consistent with the fundamental assumption of the OLM. Furthermore, we observe an increase in CMRO2 of 75% at almost constant CBF in contradiction to the OLM. We conclude that the OLM cannot explain why CBF is always much greater than CMRO2 in functional activation studies.

CEREBRAL HEMODYNAMICS AND OXYGEN METABOLISM IN PATIENTS WITH MOYAMOYA VESSELS ASSOCIATED WITH ATHEROSCLEROTIC STENO-OCCLUSIVE ARTERIES

Hiroki Kato¹, Eku Shimosegawa¹, Yasuyuki Kimura¹, Katsufumi Kajimoto¹, Makiko Tanaka¹, Kazuo Kitagawa², Jun Hatazawa¹

¹Department of Nuclear Medicine and Tracer Kinetics, Osaka University Graduate School of Medicine, Osaka, Japan, ²Department of Internal Medicine, Cardiovascular Medicine, Osaka University Graduate School of Medicine, Osaka, Japan

EFFECTS OF GROUP II METABOTROPIC GLUTAMATE RECEPTOR ACTIVATION ON NEUROPROTECTION AND EXTRACELLULAR GLUTAMATE CONCENTRATION AFTER TRANSIENT FOREBRAIN ISCHEMIA

Hideyuki Yoshioka, Masao Sugita, Takashi Yagi, Toru Horikoshi, Hiroyuki Kinouchi

Department of Neurosurgery, Faculty of Medicine, University of Yamanashi, Chuou, Yamanashi, Japan

INHIBITION OF THE PI3K/AKT PATHWAY ABOLISHES THE PROTECTIVE EFFECT OF POSTCONDITIONING AGAINST FOCAL ISCHEMIA IN RAT

Xuwen Gao¹, Hanfeng Zhang¹, Heng Zhao^1,2

¹Department of Neurosurgery, Stanford University, Stanford, CA, USA, ²Stanford Stroke Center, Stanford University, Stanford, CA, USA

Backgrounds and Aims: We recently demonstrated that postconditioning with a series of mechanical interruptions of reperfusion significantly reduces ischemic damage 2 days after stroke. We extend these studies by asking if postconditioning improves neurological function and whether postconditioning has a long-term protective effect. In addition, we addressed whether the PI3K/Akt signaling pathway contributes to the protection afforded by postconditioning. The PI3K/Akt pathway plays a critical role in neuronal survival after stroke and is regulated by a cascade of phosphorylation events. PTEN is a negative regulator of the Akt pathway, and phosphorylation of PTEN blocks its pro-apoptotic effect. Upon activation, Akt phosphorylates GSK3β and downregulates its activity. Phosphorylation of β-catenin by GSK3β leads to degradation of β-catenin and apoptosis. By examining the activation state of the Akt pathway and modifying its activity during stroke with or without postconditioning, we found that postconditioning reduces infarction partially through the Akt pathway.

Methods: Focal ischemia was generated by permanent distal MCA occlusion plus 30 min of transient bilateral CCA occlusion in male SD rats. Ischemic postconditioning was performed by occluding CCAs for 10 seconds and releasing for 30 seconds for total of three cycles starting from 30 seconds after reperfusion. Behavioral tests using three standard methods were performed from 2 to 30 days after stroke, and ischemic damage was evaluated at 30 days. For Western blots, ischemic brains were harvested 1, 5 and 24 h after stroke. Whole cell homogenates from the ischemic core and penumbra were prepared for Western blots. Antibodies detecting Akt phosphorylated on Ser473 (P-Akt), P-PDK1, P-PTEN, pan-β-catenin and activated (non-phosphorylated on Ser37 and Thr 41) β-catenin were used. A PI3K/Akt inhibitor, LY 294002, was injected before ischemic onset to test the effect of inhibiting the PI3K/Akt in postconditioning. In vitro Akt kinase assays were also performed.

Results: We found that postconditioning not only improves neurological functions, but also reduces ischemic tissue loss up to 30 days after stroke. P-Akt (active Akt) increased from 1 to 24h in the ischemic penumbra, (P<0.05, n=5-7, Two way NOVA). In the ischemic core, P-Akt slightly decreased at 1h but increased at 24h. Postconditioning enhanced levels of P-Akt in both the penumbra and core (P<0.01). Akt kinase assays directly showed that postconditioning attenuated the decrease in Akt activity (P<0.05). Treatment with LY 294002 significantly enlarged infarct size in ischemic rat brains treated with postconditioning (P<0.05, n=7-8). The levels of P-PTEN (inactive PTEN) significantly decreased after stroke with or without postconditioning. Cleaved β-catenin was detected as early as 1h and the levels of cleavage products further increased at 24h in both the penumbra and core. Postconditioning attenuated β-catenin cleavage at 24h in the penumbra but not in the core. Changes in cleaved active β-catenin were similar compared with total β-catenin.

Conclusion: Postconditioning has a long-term protective effect on neurological function and the Akt survival pathway contributes to the neuroprotective effect of postconditioning.

INTEGRATING COMPREHENSIVE AND ALTERNATIVE MEDICINE INTO STROKE: COMBINATION OF CHINESE HERBS ATTENUATED DAMAGE AND ACCELERATED SELF-REPAIR FOLLOWING FOCAL CEREBRAL ISCHEMIA

Baowan Lin¹, Weizhao Zhao¹, Guillermo Fernandez¹, Xuming Ji², Lin Li¹, Raul Busto¹

¹Department of Neurology, Miller School of Medicine, University of Miami, Miami, FL, USA, ²Department of Neurosurgery, Xuanwu Hospital, Beijing, China

Background: Bedside herbal therapy has been using in stroke patients in China for generations. The present study used a specific combination of 9 herbs including Huang Qin (Secutellaria baicalensis) in animal experiment to investigate if and how it works. The therapeutic strategies are encouraging blood circulation and self-healing, inhibiting inflammation and scavenging free oxygen radicals.

Methods: Cerebral infarction was produced in overnight-fasted Male Sprague-Dawley rats (280-315g) by 2-h intraluminal occlusion in middle cerebral artery (MCA) with a suture. Head temperature was maintained at 36-36.5oC and body temperature at 37-37.5oC. All rats received behavioral test. Rats with full neurological deficit scores ranging10 −12 at 2h after MCAo participated in the study. Rats survived for 3 or 28 days before histopathological analysis. Treated rats had the herb cocktail-style (HC) soup, which was initiated at 2 hrs after reperfusion and daily administrated by gavage. Nontreated rats received drink water (DW).

THE EFFECT OF HYPERBARIC OXYGEN PRECONDITIONING ON EARLY APOPTOSIS IN A RAT MODEL OF TRANSIENT GLOBAL CEREBRAL ISCHEMIA

Robert P. Ostrowski¹, Gerhart Graupner², Elena Titova¹, Jennifer Zhang¹, Jeffrey Chiu¹, Neal Dach¹, Dalia Corleone¹, John H. Zhang^1,3,4

¹Department of Physiology, Loma Linda University, Loma Linda, CA, USA, ²Department of Pediatrics, Loma Linda University, Loma Linda, CA, USA, ³Department of Neurosurgery, Loma Linda University, Loma Linda, CA, USA, ⁴Department of Anesthesiology, Loma Linda University, Loma Linda, CA, USA

Background and aims: We hypothesized that the brain protective effect of hyperbaric oxygen (HBO) preconditioning is mediated by the inhibition of early apoptosis, and the anti-apoptotic effect is prolonged to the later phase post-ischemia. We also tested whether inhibition of apoptosis due to HBO preconditioning may be mediated by the suppression of p38 MAP kinase and increases in brain-derived neurotrophic factor (BDNF) early after ischemia.

Methods: We used one-stage four-vessel occlusion (4VO) global ischemic model (10 min. of occlusion). A total of 106 male Sprague-Dawley rats (300-350 g) were assigned into 4 groups: sham, global ischemia (4VO) untreated and ischemic group preconditioned with either 3 or 5 courses of hyperbaric oxygenation (3HBO+4VO or 5HBO+4VO). HBO was applied at 2.5 atmospheres absolute for 1 h per treatment. Five times HBO preconditioning included one treatment per day for five days; the last treatment was applied 24 h before induction of ischemia. Triple HBO regimen comprised of treatments at 24, 12 and 4 h before ischemia. Rats were sacrificed at 3, 6, 16, 24 and 72 h and 7 days after ischemia.

We developed in vivo fluorescent staining with annexin V ? an early apoptotic marker. The neuronal activation of p-38 was assessed by anti-phospho p38 (p-p38) and anti-NeuN double immunofluorescent stain. Cell counts were performed by observers blinded to the study. Brain BDNF levels were determined by ELISA. For evaluation of neurological outcomes sensorimotor and T-maze tests were used.

Results: The number of apoptotic neurons in the cerebral cortex was significantly greater in 4VO group than in the sham group. In parallel, the increased neuronal immunoreactivity of p-p38 was noted. HBO preconditioning provided reduction in early apoptosis in cerebral cortex and the hippocampus, associated with a suppression of p-p38 and increase in BDNF brain content. This protection was followed by a reduction in TUNEL-positive cells and good preservation of neurons (Nissl stain) at 24 h and 3 and 7 days.

However, more annexin-positive cells were observed in 3HBO+4VO group than in 5HBO+4VO group. Also p-p38 suppresion positively correlated with a degree of protection and was less remarkable in the 3HBO+4VO group. We observed multiple double annexin/TUNEL positive cells at 16 h after global ischemia in the untreated group but not in the preconditioned groups. Several annexin-positive cells in HBO groups were not TUNEL positive. Mortality and neurological outcomes were ameliorated in both HBO preconditioning groups.

Conclusions: This study provided evidence that neuronal apoptosis occurs in the early phase after global cerebral ischemia. Cells showing early signs of apoptosis displayed DNA fragmentation later on. Therefore, the HBO preconditioning-induced reduction of early apoptosis and the inhibited conversion of early to late apoptosis may prevent delayed cell injury from developing. The reduction of early post-ischemic apoptosis in the brain due to HBO preconditioning may be mediated by increasing BDNF and suppressing p38 activation. Our study also suggests that the fast track of HBO preconditioning can provide satisfactory protection against anticipated brain insults.

CARBON MONOXIDE PROTECTS AGAINST BRAIN ISCHEMIA

Emil Zeynalov¹, Sylvain Dore^1,2

¹Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, SOM, Baltimore, MD, USA, ²Department of Neuroscience, Johns Hopkins University, SOM, Baltimore, MD, USA

Background: Historically, carbon monoxide (CO) gas has been generally considered to be toxic. Here, we postulate that physiological levels or low amounts of CO can in fact be protective in the brain. Using both in vivo and in vitro approaches, we have previously demonstrated that the rate-limiting enzyme for the endogenous synthesis of CO, heme oxygenase, is highly neuroprotective against various types of insults. Our goal was to determine whether CO can prevent brain damage and neurological deficits following cerebral ischemia.

Method: To test this hypothesis, C57Bl/6 male mice (22-28g) were submitted to 90-min transient brain ischemia. Successful occlusion was confirmed by an 87–90% reduction in cerebral blood flow as measured by laser-Doppler flowmetry. The mice were then exposed to air or various levels of CO: 125, 250, or 500ppm (n/group = 11, 10, 8, and 6, respectively), amounts which can be present at street levels. Mice were survived 48 hours after reperfusion; brains were harvested, sliced into 2-mm thick sections, and stained with 1% triphenyltetrazolium chloride. To measure the brain water content, brains of sacrificed animals were harvested, weighed, and dried at 100°C for 72 hours. Brains were weighed again immediately after removal from the oven, and wet-to-dry ratio was calculated. Experiments were carried out in accordance with the guidelines of the NIH for the care and use of animals in research and were approved by the institutional animal care and use committee of Johns Hopkins University.

Results: We found that mice exposed to 125ppm CO developed a significantly smaller infarct size than those exposed to air only (p<0.05). Exposure to 250ppm appeared to produce the maximal beneficial effect. In mice subjected to unilateral transient focal ischemia and exposed to 250ppm CO after reperfusion, the ipsilateral hemisphere had lower brain water content at 48 hours than that of the control animals. Water content in the contralateral hemispheres remained unchanged in both CO- and air-exposed groups.

Discussion: To our knowledge, this is the first study reporting that CO can reduce infarct volume and limit brain edema. Various mechanisms of action are being postulated and investigated. Physiological amounts of CO have been shown to play an important role in many cytoprotective mechanisms. It has been shown that low concentrations of exogenous CO can have anti-inflammatory and anti-apoptotic effects, which would protect tissues and organs against various insults. In addition, CO can mediate vascular signaling that leads to vasodilatation, improving cerebral blood flow and brain perfusion. And by suppression of inflammatory and cell death pathways, CO permits protection of tissues and organs against various insults. Exposure of CO at 250ppm would be the optimal level for studying the neuroprotective effects and can be used for the treatment of experimental ischemic stroke in mice. Thus, development of a protocol to maintain CO within these optimal levels could be a novel therapeutic tool to fight brain damage and neurological outcome associated with stroke and other age- and vascular-related neurodegenerative disorders. (This work was supported by NIH grants.)

AN ELECTRIC FIELD INCREASES BRAIN-DERIVED NEUROTROPHIC FACTOR LEVELS IN THE BRAIN, ENHANCES SPATIAL LEARNING AND RESISTANCE TO FOCAL ISCHEMIA

Hiroji Yanamoto^1,2, Susumu Miyamoto², Yukako Nakajo^1,3, Yoshikazu Nakano¹

¹Laboratory for Cerebrovascular Disorders, Research Institute of National Cardiovascular Center, Suita, Osaka, Japan, ²Department of Neurosurgery, National Cardiovascular Center, Suita, Osaka, Japan, ³Laboratory for Basic Science, Rakuwakai Otowa Hospital, Kyoto, Japan

Background and aims: Brain-derived neurotrophic factor (BDNF) is known to enhance learning and memory acquisition or to promote neuronal survival protecting the brain from ischemic injuries. Recently, we reported that intracerebral infusion of recombinant BDNF protects the brain from focal ischemia, and that genetically engineered BDNF overexpression mice; heterozygous BDNF (+/−) transgenic mice demonstrate significantly enhanced spatial learning compared to the wildcard littermates. Development of a safe method that can increase endogenous BDNF levels has been expected to enhance learning, or to protect the brain against ischemic stroke. Specific types of electric stimulation that have been shown to be associated with elevated levels of BDNF in the brain include the following: a brief period of direct current electrification, an epileptiform discharge, and chronic electroconvulsive therapy (ECT). Here, we investigated the effects of whole-body exposure to alternating high voltage electric potential (HELP) on BDNF levels in the brain, spatial learning, and resistance to cerebral infarction development in mice.

Methods: An electric field generated with HELP; 5.8 kVrms or 0 Vrms (untreated control) was applied to normal adult mice (8-9-week old, C57BL/6J), for 5 hours a day for 1 or 3 weeks, using a mattress-type working electrode (HELP supplier) which was placed directly beneath the mouse cage, under the condition that the unit composed with the cage and HELP supplier was insulated from its surroundings to make indirect contact (through room air > 20 cm) with ground level. Estimated electric field generated in the cage was < 29 kV/m during the HELP supply. After the treatment, we analyzed alterations in BDNF levels in the brain using ELISA, or performance in a modified Morris water maze test (MWM). In a separate set of animals after the treatment, we also analyzed alterations in infarcted lesion volume caused by transient focal ischemia using the 3-VO technique.

Results: After HELP treatment for 1 or 3 weeks, the BDNF levels in the corticostriatum increased (205 % or 240 %, respectively), and escape latency in MWM improved (54 % or 31 %, respectively), compared with untreated the corresponding control animals (n=18-20 each, P < 0.05, one-way ANOVA). The infarcted lesion volume measured 24 hours after the induction of ischemia was decreased when ischemia was induced after HELP treatment for 3 weeks, compared with controls (15.7 ± 7.7 mm3 vs. 25.4 ± 8.7 mm3, mean ± SD, n=7 each, P < 0.05, one-way ANOVA).

Conclusions: It is here reported for the first time that repetitive HELP application increases BDNF levels in the brain, enhances spatial learning, and protects the brain from ischemic stroke. Use of repetitive HELP (< 30 kVrms) was approved 40 years ago by the Ministry of HLW (Japan) as an effective therapy for improving shoulder muscle stiffness, chronic constipation, insomnia, or headache in humans.

NORMOBARIC HYPEROXIA EXERTS ITS NEUROPROTECTIVE EFFECT IN ISCHEMIC RATS BY DELAYING AND ATTENUATING THE EARLY NOS-DERIVED NO PRODUCTION

Zhongrui Yuan, Wenlan Liu, Aaron Schnell, Ke Jian Liu

Department of Pharmaceutical Sciences, College of Pharmacy, University of New Mexico, Albuquerque, NM, USA

Background and aims: Neuronal release of nitric oxide (NO) plays a prominent role in excitotoxicity, an important trigger and executioner of tissue damage in focal cerebral ischemia. Our recent study has shown that normobaric hyperoxia (NBO) treatment during ischemia can maintain penumbral oxygenation at a physiological level, which leads to a significant reduction in stroke lesion volumes and improved neurological function in a rat model of transient focal cerebral ischemia. The mechanism for the observed neuroprotection of NBO treatment is, however, unclear. In the present study, we tested the hypothesis that NBO treatment can attenuate neuronal nitric oxide synthase (nNOS)-derived NO production.

Methods: NO production and the formation of 3-nitrotyrosine (3-NT) were measured in the ischemic cortex at 10 min., 30 min., 60 min., and 90 min. after onset of ischemia. Total amount of NO production, as determined by the concentration of nitrate and nitrite, was quantified by the Griess method. HPLC-EC analysis was performed to measure 3-NT.

Results: We found that in experimental groups with normoxia (30%O2:70%N2), ischemia caused a time-dependent production of NO in the ischemic cortex, reached peak at 10 min (831 ± 172 nmol/mg protein, n=6) after onset of ischemia, which then returned to baseline level at 60 min (327 ± 137 nmol/mg protein, n=6). On the other hand, the NO production in the NBO (95%O2:5%CO2) treatment group was significantly lower (691 ± 94 nmol/mg protein, n=6), and peaked much later (at 30 min). For the formation of 3-NT, compared with a normoxia group, NBO treatment reduced the formation of 3-NT by 30 ± 4% (p<0.05). In order to understand the source of the NO and 3-NT production, we also carried out parallel experiments in animals pre-treated with 7-nitroindazole (7-NI), which is an nNOS selective inhibitor. We found that 7-NI pre-treatment has similar effects on NO and 3-NT production as does NBO treatment. Combined treatments with both 7-NI and NBO did not reduce NO and 3-NT production further than the individual treatments alone. These findings indicate that that NO, and the resulting 3-NT, was derived from nNOS.

Conclusions: Taken together, our findings provide direct evidence that NBO exerts its neuroprotective effects through both delaying and attenuating the early nNOS-derived NO production, in a similar manner to the beneficial effect of nNOS selective inhibitors. These results also suggest that NBO could potentially be used to expand the therapeutic time window for stroke patients.

APOCYNIN REDUCES BLOOD-BRAIN BARRIER DISRUPTION BY INHIBITING NADPH OXIDASE ACTIVATION AND MMP-9 EXPRESSION IN EXPERIMENTAL STROKE

Xiannan Tang^1,2, Maya Koike^1,2, Lijun Xu², Nick Cairns³, Belinda Cairns³, Rona Giffard², Midori Yenari¹

¹Department of Neurology, UCSF and SF VAMC, San Francisco, CA, USA, ²Department of Anesthesia, Stanford University, Palo Alto, CA, USA, ³Combinix, Inc., Mountain View, CA, USA

Introduction: Inflammation accompanying stroke has received much attention in recent years, as it is now believed that this response contributes to secondary ischemic injury. We previously showed in vitro that microglia worsen injury to blood-brain barrier (BBB) components subjected to ischemia-like insults, and that this was largely due to the generation of superoxide by NADPH oxidase. Here we study whether NADPH oxidase contributes to BBB disruption and brain hemorrhage in vivo, and whether apocynin, an inhibitor of NADPH oxidase could reduce BBB disruption following experimental stroke. Methods: Male C57 mice were subjected to 2h transient middle cerebral artery occlusion (tMCAO) were treated with: apocynin 0 (n=5), 2.5 (n=7) or 5 (n=8) mg/kg IV dissolved in pyrogen-free 0.1% DMSO 30 minutes after reperfusion. Following 22h reperfusion, animals were evaluated for neurological deficits, and then sacrificed. Brains were assessed for infarct volume, hemorrhage, and Evans blue dye extravasation. To determine whether apocynin enters the brain after parenteral administration, 2 mice were intravenously injected with Texas Red conjugated apocynin. In order to further determine whether apocynin could reduce the superoxide generation, another 6 C57BL male mice were subjected to tMCAO and treated with apocynin (2.5 mg/kg) and 0.1% DMSO (n=3 respectively) In situ detection of superoxide anion was performed using hydroethidine (HEt). For MMP-9 expression study, also 6 C57BL male mice were subjected to tMCAO and treated with apocynin (2.5 mg/kg) and 0.1% DMSO (n=3 respectively). TTC staining was performed to delineate the infarct; brain tissue around the infarct core was sampled and used for Western blot assay. Results: Apocynin was found to be able penetrate the intact BBB. Apocynin treatment (2.5 mg/kg) reduced infarct volume by 37% (P<0.05) and improved neurological deficits (1.5 vs. 2.5, P<0.05). There was a trend of decreased cerebral hemorrhage rate (16.7% vs. 58.3%, P=0.09) and it reduced the extent of Evans blue dye extravasation by 74% (P<0.001). In contrast, 5 mg/kg apocynin did not decrease the hemorrhage rate (60% vs. 58.3%, NS), nor reduce infarct volume (140±35 vs. 144±36 cm3, NS), nor improve neurological function (2.48 ±0.42 vs. 2.5±0.56, NS). Apocynin markedly reduced superoxide generation. Superoxide generation as estimated by the relative fluorescence intensity (RFI) per microscopic field, increased 4.5 fold following tMCAO compared to sham operated controls. Apocynin treatment almost completed suppressed superoxide generation (4.5 fold less RFI compared to vehicle treated controls, P<0.01). Furthermore, number of cells positive for superoxide were increased by tMCAO (160±3 vs. 31±3 per 400X HPF in sham operated controls, P<0.05), and this was also decreased by apocynin treatment (3±1, P<0.05). Immunohistochemistry showed most of the superoxide were generated by neurons and in part by microglia, but very few by cerebrovascular endothelial cells. Compared to the vehicle treated animals, the expression of MMP-9 in treated animals (apocynin 2.5mg/kg) reduced significantly (P<0.05). Conclusion: NADPH oxidase plays an important role in worsening BBB disruption and hemorrhage following experimental stroke. Apocynin improves outcome, but appears to have a narrow therapeutic window. Apocynin reduces superoxide generation and MMP-9 expression.

CALCIUM CHANNEL BLOCKERS PREVENT NEURONAL DAMAGE AND HAVE AN ANTIOXIDATIVE EFFECT AFTER TRANSIENT FOCAL ISCHEMIA IN RATS

Violeta Lukic Panin, HanZhe Zhang, Takeshi Hayashi, Atsushi Tsuchiya, Yoshihide Sehara, Toru Yamashita, Kentaro Deguchi, Tatsushi Kamiya, Koji Abe

Department of Neurology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan

BACKGROUND: Cerebral ischemia is a third leading cause of death and at first place cause of disability all over the world. There are a lot of drugs that are in experimental stage for treatment of stroke. Among them are Calcium Channel Blockers (CCB) that have, in animal models, different effectiveness in healing of ischemic damage in brain. Mechanism of CCB action in cerebral ischemia is still unclear, but antioxidative property is supposed to be implicated. In the present study, we investigated antioxidative and neuroprotective properties of two CCB, azelnidipine and amlodipine.

METHODS: Male Wistar Kyoto rats were subjected to 90 minutes of transient middle cerebral artery occlusion (MCAO) by a nylon thread. Animals were divided into 3 groups, vehicle, azelnidipine and amlodipine group. In azelnidipine and amlodipine group, rats were treated with azelnidipine (1mg/kg) and amlodipine (1mg/kg) by gastric gauge for two weeks before MCAO. Vehicle group was treated by solution of methyl cellulose for two weeks. Rats were killed 24hs after MCAO. Clinical parameters (mean arterial pressure, heart rate, body weight), infarct volume, oxidative stress markers which are HEL, 4-HNE, AGE and 8-OHdG, and evidence of apoptosis by TUNEL, were investigated.

RESULTS: There were no significant difference among groups in mean arterial pressure, heart rate and body weight. Treatment with azelnidipine and amlodipine reduced infarct volume. Azelnidipine treated group showed more marked infarct volume reduction than amlodipine group (figure). The number of HEL, 4-HNE, AGE and 8-OHdG positive cells were significantly decreased in CCB treated groups. These molecules were again fewer in azelnidipine group than in amlodipine group. In TUNEL staining, the number of positive cells was smaller in the CCB treated group, especially in azelnidipine group.

CONCLUSIONS: Pretreatment of azelnidipine and amlodipine had a neuroprotective effect in ischemic brain. Antioxidative property is one of the important profiles of CCB that is implicated in brain protection.

CAROTID IMT AND RISK OF CARDIOVASCULAR EVENTS IN HIGH-RISK PATIENTS

Kazuo Kitagawa¹, Hidetaka Hougaku¹, Hiroshi Yamagami¹, Hiroyuki Hashimoto², Taku Hoshi¹, Shigetaka Furukado¹, Yuko Abe¹, Manabu Sakaguchi¹, Masayasu Matsumoto³, Masatsugu Hori¹

¹Stroke Division, Department of Medicine, Osaka University Hospital, Suita, Osaka, Japan, ²Division of Stroke and Hypertension, Department of Medicine, National Osaka Hospital, Osaka, Japan, ³Department of Clinical Neuroscience, Hiroshima Graduate School of Biomedical Sciences, Hiroshima, Japan

Background and Purpose: An increased carotid intima-media thickness (IMT) has proved to be associated with the presence of cardiovascular risk factors, cardiovascular disease (CVD), and atherosclerosis elsewhere in the arterial system. Prospective follow-up studies in the population at large have shown a positive correlation between increased carotid IMT and the risk of myocardial infarction and stroke. The purpose of this study was to determine the usefulness of carotid IMT measurement in such patients.

Methods: The study comprised 900 outpatients with cardiovascular risk factors or established atherosclerosis. Carotid IMT was calculated as the mean bilateral IMT of the common carotid artery (CCA), bifurcation, and internal carotid artery (ICA). Baseline vascular risk factors, medications, and history of CVD were recorded at the time of enrollment. The incidence of CVD events was determined prospectively.

Results: During a mean follow-up period of 2.6 years, there were 64 CVD events. The relative risk of a CVD event increased with increased IMT. Association between CVD events and carotid IMT was significant after adjustment for risk factors and history of CVD, showing an increased risk per IMT tertile from the middle tertile (relative risk, 2.5; 95% confidence interval [CI]: 1.0–6..3) to the highest (relative risk, 3.6; 1.4–9..0). When patients with a history of CVD were excluded (n=574), the predictive value of IMT was significant even after adjustment for risk factors (hazard ratio per 1 SD IMT increase was 1.57 [95% CI: 1.11 to 2.20]).

Conclusions: We have shown that carotid IMT independently predicts vascular events in high-risk patients. Carotid IMT measurement is of value for clinical stratification of patients at high risk of vascular events.

HYPOTHERMIA STRENGHTS INHIBITORS OF MATRIX DEGRADING PROTEASES FOLLOWING EXPERIMENTAL FOCAL CEREBRAL ISCHEMIA IN MICE

Jan Burk¹, Dorothe Burggraf¹, Milan Vosko¹, Martin Dichgans¹, Gerhard F. Hamann²

¹Department of Neurology, Ludwig-Maximilians University, Munich, Germany, ²HSK Dr. Horst Schmidt Klinik, Wiesbaden, Germany

The treatment of an ischemic stroke under hypothermic conditions resulted in a better outcome in clinical studies. In this experiment we investigated the microvascular damage and proteolytic systems caused by ischemia in mice using hypothermia.

Focal cerebral ischemia was induced by occlusion of the arteria cerebri media (MCAO) using an intraluminal filament technique. The hypothermia model was established in our laboratory in mice. A first set of hypothermia experiments contained 16 mice. Eight mice received normothermia (37°C, NT) and eight received hypothermia (32?34°C, HT) treatment during 24 hours of reperfusion.

The analysis of the hypothermic group in comparison with the normothermic group resulted in a significantly reduced infarct volume (NT: 56±6mm3 SEM, HT: 33±5mm3 SEM; p<0.05) and showed considerable less neurological deficits (Garcia score) after 24 hours (p=0.002). In addition, the degradation of the basal lamina collagen type IV in hypothermia was strongly reduced (p<0.05) as well as the extravasation of haemoglobin (p<0.05).

In the biochemical analysis we found in the hypothermia group a significant decrease of the urokinase plasminogen activator (uPA) (p<0.05) and its inhibitor PAI-1 (p<0.05) compared to normothermia group. Conspicuous, during hypothermia as opposed to normothermia, there was a shift in the plasminogen-plasmin system from the proteases to their inhibitors. The ratio of uPA/PAI-1 (NT: 11:1; HT: 5:1) decreased significantly (p<0.05) during hypothermia.

In conclusion we demonstrated that hypothermia reduces the severe consequences of cerebral ischemia, protects the basal lamina and reduces the infarct volume and hemorrhage. Furthermore, one possible explanation for the beneficial effects of hypothermia might be the shift of the matrix degradading proteolytic systems to the inhibitoric side.

COMPARISON BETWEEN ANTIPLATELET THERAPY AND COMBINATION OF ANTIPLATELET AND ANTICOAGULATION THERAPY FOR SECONDARY PREVENTION IN STROKE PATIENTS WITH ANTIPHOSPHOLIPID SYNDROME

Yasuhisa Kitagawa¹, Naoki Yuasa¹, Noriko Takahashi¹, Tatsuya Ishkawa¹, Kentaro Tokuoka¹,

Takashi Yasuda², Shigeharu Takagi²

¹Department of Neurology, Tokai University Hchioji Hospital, Tokyo, Japan, ²Department of Neurology, Tokai University, Kanagawa, Japan

Satisfactory results have not yet been obtained in therapy for secondary prevention in ischemic stroke patients with antiphospholipid syndrome (APS). We therefore compared single antiplatelet therapy and a combination of antiplatelet and anticoagulation therapy for secondary prevention in ischemic stroke patients with APS.

The subjects were 20 ischemic stroke patients with antiphosholipid antibody (10 males and 10 females, mean age 48 years), 13 with primary antiphospholipid syndrome and 7 with SLE-related antiphospholipid syndrome. Diagnosis of APS was based on the 1998 Sapporo Criteria. Eligible patients ware randomly assigned to either single antiplatelet therapy (aspirin 100 mg or ticlopidine 200 mg) or a combination of antiplatelet and anticoagulation therapy (INR: 2.0–3..0; mean 2.4±0.3) for the secondary prevention of stroke according to a double-blind protocol. There was no significant difference between the two groups in age, gender, NIHSS on admission, mRS at discharge, or rates of hypertension, diabetes mellitus, hyperlipidemia, and cardiac disease. We examined the Kaplan-Meier survival curves in each treatment. The primary outcome was the occurrence of stroke. The mean follow-up time was 3.9±2.0 years. The cumulative incidence of stroke in patients with single antiplatelet treatment was statistically significantly higher than that in patients receiving the combination of antiplatelet and anticoaglation therapy (log-rank test, p-value=0.026). The incidence of hemorrhagic complications was similar in the two groups. The recent APASS study did not show any difference in effectiveness for secondary prevention between single antiplatelet (aspirin) and single anticoagulant (warfarin) therapy. Our results indicate that combination therapy may be more effective in APS-related ischemic stroke.

EVALUATION OF SEE-JET METHOD ON HEMODYNAMIC CEREBRAL ISCHEMIA

Tetsu Kurokawa¹, Shoichi Kato¹, Kei Harada¹, Hisato Nakayama¹, Hirosuke Fujisawa¹, Hideyuki Ishihara¹, Hiroyasu Koizumi¹, Tatsuo Akimura¹, Nakagawara Joji², Michiyasu Suzuki¹

¹Department of Neurosurgery, Yamaguchi University School of Medicine, Ube, Yamaguchi, Japan, ²Department of Neurosurgery, Nakamura Memorial Hospital, Sapporo, Hokkaido, Japan

[Background and Aims] The indication of cerebral revascularization such as extracranial-intracranial bypass has been determined by the severity of hemodynamic ischemia. And the result of Japanese Extractranial-Intracranial Bypass Trial (JET study) suggests that appropriate selection of patient based on severity of cerebral hemodynamic ischemia is important. To evaluate the severity, the cerebral blood flow (CBF) and cerevrovascular reserve (CVR) had been evaluated by setting of region of interest (ROI). However, this ROI setting method highly depends on the examiner's bias. To avoid this, we have utilized the stereotactic extraction estimation (SEE) ?JET software. Using this software, the cortical CBF and CVR can be evaluated in the every pixel automatically. We evaluated whether the severe hemodynamic stage (stage II) in the motor area detected by SEE-JET correlate with the clinical symptoms and compared with the ROI setting method. [Methods] Between December 2004 and November 2005 in our institute, 210 (142 male and 68 female) patients aged 64.8±13.7 years (mean±s.d.) with cerebral major vessel disease were recruited in this study. Resting CBF and CBF after acetazolamid challenge were measured by the Dual table autoradiographic (DT-ARG) method. The ROI of a middle cerebral artery area was manually drawn on bilateral cerebral cortex of the basal ganglia level. In the SEEJET evaluation, the motor area was determined by automatically based on the anatomical map. We evaluated whether the stage II was included or not in the motor area. The sensitivity and the specificity were compared between the SEE-JET and ROI setting method. [Results] The sensitivity and the specificity to the clinical symptoms (motor deficits) was 0.17 and 0.91 respectively with ROI setting method. In the SEE-JET method, the sensitivity and the specificity to stage II with motor area and motor deficit were 0.55 and 0.81 respectively. [Conclusions] To compare the ROI setting method, stage II in the motor area detected by the SEE-JET method is more sensitive to the clinical symptom and therefore this modality can be strongly recommended to determine the indication of the cerebral revascularization.

RECURRENCE WITHIN 30 DAYS AFTER ISCHEMIC STROKE IN JAPANESE PATIENTS: THE JAPAN STANDARD STROKE REGISTRY STUDY

Kazunori Toyoda¹, Yasushi Okada², Shotai Kobayashi³

¹Cerebrovascular Division, National Cardiovascular Center, Suita, Japan, ²Department of Cerebrovascular Disease, National Hospital Organization Kyushu Medical Center, Fukuoka, Japan, ³Shimane University Hospital, Izumo, Japan

Background and aims: Stroke in Japan differs much from Western patterns in incidence and frequency of stroke subtypes, and has many unique characteristics. We determined the contributory factors and outcome of early recurrence of ischemic stroke in Japanese patients.

Methods: A multicenter stroke registration study (the Japan Standard Stroke Registry Study) based on a computerized database from 54 Japanese institutes, using 8,036 patients with brain infarction who were hospitalized within 48 hours after the symptom onset between January 2000 and March 2004.

Results: Within 30 days after the initial stroke, 395 patients (4.9%) suffered from recurrent stroke. Recurrence was most frequently occurred in atherothrombotic patients (6.6%), followed by cardioembolic patients (6.2%). Compared to patients who did not have a recurrent stroke, patients who had an early recurrence more frequently had: a previous ischemic stroke (p=0.0222); hypertension (p=0.0188); atrial fibrillation (p=0.0001); antiplatelet (p=0.0052) and antihypertensive (p=0.0017) medication prior to the initial stroke; and progression of symptoms within 48 hours (p<0.0001). After multivariate analysis, hypertension (OR 1.35, 95% CI 1.07 ? 1.70) and atrial fibrillation (OR 1.50, 95% CI 1.18 ? 1.92) were independently predictive of early recurrence of overall patients. In atherothrombotic patients, diabetes mellitus (OR 1.49, 95% CI 1.06 ? 2.09) and atrial fibrillation (OR 2.00, 95% CI 1.23 ? 3.24) were independently related to early recurrence; in cardioembolic patients, hypertension (OR 1.53, 95% CI 1.05 ? 2.22) and aortic aneurysm (OR 4.07, 95% CI 1.32 ? 12.57) were related. Among stroke subtypes, atherothrombotic stroke (OR 1.94, 95% CI 1.54 ? 2.45) and stroke of other etiology (OR 1.50, 95% CI 1.03 ? 2.19) were positive independent predictors of early recurrence after adjustment of risk factors. Cardioembolic stroke was no longer positively related to recurrence after adjustment for atrial fibrillation On discharge from our institute, modified Rankin Scale was higher in patients with than without early recurrence [3 (0 ? death) vs. 2 (0 ? death), p<0.0001]

Conclusions: This study using a large Japanese population reproduced the previous consensus that atherothrombotic stroke, hypertension, and atrial fibrillation contributed to early recurrence of ischemic stroke. In addition, analysis according to each stroke subtype clarified that diabetes mellitus was independently related to early recurrence for atherothrombotic patients.

REDUCTION IN CAROTID ARTERIAL PLAQUE AFTER FLUVASTATIN ASSESSED BY 3D ULTRASONOGRAPHY

Hideki Ohba, Kiyofumi Mori, Masato Kin, Masanori Mizuno, Yasuhiro Ishibasi, Yasuo Terayama

Department of Neurology, Iwate Medical University, Morioka, Iwate, Japan

NON-INVASIVE MEASUREMENT OF ADVANCED GLYCATION END PRODUCTS IN PATIENTS WITH CEREBRAL INFARCTION

Yuko Morita¹, Reiko Nagano¹, Yuko Izuhara², Shunya Takizawa¹, Shigeharu Takagi¹, Toshio Miyata²

¹Division of Neurology, Department of Internal Medicine/Tokai University School of Medicine, Isehara, Kanagawa, Japan, ²Institute of Medical Sciences, Tokai University School of Medicine, Isehara, Kanagawa, Japan

Background and purpose: Oxidative modification of carbohydrates and lipids enhances the formation of advanced glycation end products (AGEs). Although AGEs formation is markedly accelerated in hyperglycemia, its formation occurs in normoglycemia (e.g., during ischemic stroke) (1) and initiates a range of inflammatory responses (2). Recently, an AGE Reader (DIAGNOPTICS, USA) has become available to measure skin autofluorescence non-invasively. We used it to test our hypothesis that skin AGEs level increases in proportion to the progression of systemic atherosclerosis, even in patients with cerebral infarction, and is therefore available as a marker for predicting recurrence of ischemic stroke. In this study, we investigated the levels of skin AGEs in patients with chronic cerebral infarction, patients with silent infarction and age-matched controls. We also examined the effect of angiotension II receptor blocker (ARB) on the level of advanced glycation (3).

Methods: We measured skin AGEs levels in patients with chronic symptomatic cerebral infarction (n=107) and asymptomatic cerebral infarction (n=26), as well as age-matched controls (n=25), using an AGE Reader. Patients with chronic symptomatic cerebral infarction were separated into two groups; those who received ARB and those who did not.

Results: Skin AGEs levels in patients with symptomatic cerebral infarction were significantly increased compared with the control group (p<0.05; 2.1127±0.5355 v.s. 2.0202±0.4005). AGEs in patients with asymptomatic cerebral infarction were not increased versus the controls (p<0.05; 1.9781±0.3732 v.s. 2.0202±0.4005). However, AGEs in the symptomatic and asymptomatic groups were not significantly different. AGEs in the groups that did and did not receive ARB were also not significantly different.

Conclusion: Our data suggest that increased formation of AGEs may be an indicator of oxidative stress, not only in diabetes and renal failure, but also in chronic cerebral infarction. Further study will be required to establish whether skin AGE level is a useful parameter for predicting recurrence of ischemic stroke.

KNOCKING-DOWN AND KNOCKING-OUT OF NCKX2 GENE EXPRESSION EXACERBATE THE BRAIN INSULT INDUCED BY FOCAL ISCHEMIA

Ornella Cuomo¹, Giuseppe Pignataro¹, Rosaria Gala¹, Francesca Boscia¹, Anna Tortiglione¹, Valeria Valsecchi¹, Gianfranco Di Renzo¹, Jonathan Lytton², Lucio Annunziato¹

¹Division of Pharmacology, Department of Neuroscience, Federico II University of Naples, Naples, Italy, ²Department of Biochemistry and Molecular Biology, University of Calgary, Calgary, AB, Canada

Sodium/Calcium exchangers are neuronal plasma membrane transporters which, by coupling Ca2+ and Na+ fluxes across neuronal membranes, may play a relevant role in brain ischemia. The exchanger gene superfamily comprises two arms: the K+-independent (NCX) and K+-dependent (NCKX) exchangers. In the brain, 3 different NCX and NCKX family members have been described (Annunziato et al., 2004; Cai and Lytton, 2004, Visser et al; 2006).

Previous studies showed that more than 60% of calcium extrusion was mediated by sodium calcium exchangers and that 90% of this exchange was NCKX-mediated (Lee et al., 2002). Recently, it has been shown that NCX blockade induces a worsening in neuronal damage, whereas its activation may provide neuroprotection in an in vivo model of focal ischemia (Pignataro et al., 2004). By contrast, the function of NCKX in cerebral ischemia has not been determined. Thus, the aim of the present study was to investigate the role played by NCKX2 in rodents subjected to permanent or transient middle cerebral artery occlusion, pMCAO and tMCAO.

To this purpose, NCKX2 mRNA and protein expression were evaluated in the ischemic core and in the remaining ipsilateral non-ischemic area at different times after pMCAO in rats, by means of real-time RT-PCR, in situ hybridization, Western Blot, and immunohistochemistry analysis. The results showed that, both in the ischemic core and in the periinfarctual area, NCKX2 mRNA and protein were downregulated. Interestingly, immunohistochemical experiments revealed that, although the NCKX2 immunoreactivity was globally reduced, some surviving cells, mainly localized in the pheripheral zone of the ischemic core strongly expressed NCKX2 protein.

The role of this protein in the development of ischemic damage was also assessed in ischemic rats in which NCKX2 expression was knocked-down by a specific antisense oligodeoxynucleotide (ODN) intracerebroventricularly injected or in ischemic knocked-out nckx2-/− mice. Both knocking-down and knocking-out NCKX2 expression by antisense strategy or genetic deletion drammatically increased the extent of the ischemic lesion in rats and mice subjected to pMCAO and tMCAO, respectively, showing that NCKX2 plays a pivotal role in the development of cerebral ischemia.

Overall, these results suggest that NCKX2 is involved in the progression of the ischemic lesion and may represent a potential target in the development of new therapeutic strategies for stroke treatment.

SRC KINASE INHIBITION IMPROVES ACUTE OUTCOMES FOLLOWING EXPERIMENTAL INTRACEREBRAL HEMORRHAGE

Timothy D. Ardizzone, Xinhua Zhan, Brad Ander, Frank R. Sharp

Department of Neurology, MIND Institute, University of California at Davis, Sacramento, CA,

USA

• BACKGROUND AND AIMS. The mechanisms by which intracerebral hemorrhages produce changes of blood flow and metabolism, cell death, and behavioral abnormalities are complex. In this study we begin to test the hypothesis that intracerebral hemorrhage activates Src kinases that phosphorylate other molecules to produce cell injury and behavioral deficits following ICH. The study was prompted by observations that thrombin (Xi, Keep, Hoff and colleagues) appears to mediate the acute edema that occurs following intracerebral hemorrhage. One way in which thrombin might produce edema would be for thrombin to cleave thrombin receptors (called PAR receptors) that in turn activate Src kinase family members that phosphorylate target proteins like NMDA receptors, VEGF, HIF, metalloproteinases, tight junction proteins and others. Altered functions of these molecules could contribute to local changes of flow and metabolism around ICH; and to edema, cell death and behavioral abnormalities associated with ICH.

• METHODS. Intracerebral hemorrhage (ICH) was produced in adult male Sprague Dawley rats by direct injection of autologous blood (50 microliters) into striatum. Src kinase activity, glucose hypermetabolic areas around the ICH, TUNEL stained cells, and apomorphine induced rotational behaviors were assessed in animals with ICH pretreated with the Src kinase inhibitor, PP1, or with vehicle. These initial studies were aimed at determining whether a Src kinase inhibitor would improve outcomes following experimental ICH.

• RESULTS. PP1 (3mg/kg) blocked increases of Src kinase activity (5 fold) at 3h after ICH. PP1 also blocked the areas of glucose hypermetabolism (p<0.05) and decreased the numbers of TUNEL stained cells (p<0.05) surrounding the ICH at 24h. Finally, apomorphine (1mg/kg) induced rotation at 24h following ICH was markedly attenuated by prior treatment with PP1 (3mg/kg i.p.) (p<0.05).

• CONCLUSIONS. PP1 decreases Src kinase activation, glucose metabolic activation, cell death and behavioral abnormalities following ICH in striatum of adult rats. It is hypothesized that intracerebral hemorrhage, possibly via thrombin activation of PAR receptors, activates Src that phosphorylates NMDA receptors, MMPs and other proteins that mediate injury following ICH. Future studies will determine whether thrombin activates Src and whether this occurs via PAR receptors and whether ICH in fact stimulates Src mediated phosphorylation of NMDA and AMPA receptors, HIF and other potential target molecules following ICH.

DANTROLENE DECREASES ISCHEMIC BRAIN DAMAGE THROUGH REDUCING ENDOPLASMIC RETICULUM INJURY

Takeshi Hayashi, Feng Li, Guang Jin, Kentaro Deguchi, Shoko Nagotani, Hanzhe Zhang, Yoshihide Sehara, Atsushi Tsuchiya, Toru Yamashita, Violeta Lukic, Tatsushi Kamiya, Koji Abe

Department of Neurology, Okayama University Graduate School of Medicine and Dentistry, Okayama, Japan

Background: The endoplasmic reticulum (ER) plays an important role in ischemic neuronal cell death. Calcium depletion from the ER lumen is the key step in injury of this organelle, and therefore, prevention of this step may reduce the ischemic brain damage. Dantrolene is a ryanodine receptor antagonist, which prevents calcium release from the ER lumen. In order to determine the effect of dantrolene on ischemic brain damage and ER stress response, we investigated changes in ER stress-related molecules expression and brain infarction volume using rat middle cerebral artery (MCA) occlusion model.

Methods: Under anesthesia by isoflurane, the right MCA of male Wistar rat was occluded for 90 minutes by a nylon thread. Just after the restoration of cerebral blood flow (CBF), twenty µg of dantrolene, which was dissolved in dimethyl sulfoxide (DMSO) at 5 µg/µl, was slowly injected into left lateral ventricle. Control animals underwent the same surgery but only DMSO was injected. At 24 hrs after the CBF restoration, the animals were sacrificed and coronal brain sections of 2 mm thickness were prepared, with which infarction volume was measured by a 2,3,5-triphenyltetrazolium method. At 1, 4, and 24 hrs after the CBF restoration, the animals were sacrificed and frozen brain sections were prepared, with which ER stress responses were investigated. As ER stress markers, we used phosphorylated form of double-stranded RNA-activated protein kinase (PKR)-like ER kinase (p-PERK), phosphorylated form of eukaryotic initiation factor 2α (p-eIF2α), activating transcription factor-4 (ATF-4), and C/EBP-homologous protein (CHOP). We also performed western blotting study for these molecules using brain samples at the same time points.

Results: As shown in the figure, the infarction volume was significantly decreased by dantrolene (*p<0.05). Immunohistochemical and western blotting studies showed that p-PERK and p-eIF2α were induced in the ischemic peripheral region at 1 and 4 hrs, which were prevented by dantrolene. ATF-4 and CHOP were induced in the ischemic peripheral region from 1 hr and peaked at 24 hrs, but both of them were again inhibited by dantrolene.

Conclusions: It was indicated that dantrolene reduced ischemic brain damage through reducing ER stress response. Calcium depletion from ER lumen should be the key step in ischemic neuronal cell death in vivo.

THE ROLE OF BRAIN AND HEMATOMA-DERIVED COMPLEMENT C3 IN HEME OXYGENASE-1 UPREGULATION AND MITOGEN-ACTIVATED PROTEIN KINASE ACTIVATION FOLLOWING INTRACEREBRAL HEMORRHAGE

Y. He¹, S. Wan¹, Y. Hua¹, J.G. Younger², R.F. Keep¹, J.T. Hoff¹, Guohua Xi¹

¹Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA, ²Department of Emergency Medicine, University of Michigan, Ann Arbor, MI, USA

Background and Aims: Complement activation, heme degradation and iron toxicity contribute to intracerebral hemorrhage (ICH)-induced brain injury. Heme oxygenase-1 (HO-1), also called heat shock protein 32, is a key enzyme of heme degradation. Brain HO-1 levels are increased following ICH and inhibition of heme oxygenase reduces ICH-related brain damage. Activation of mitogen-activated protein kinases (MAPKs) occurs in the brain after cerebral ischemia and is associated with ischemic brain injury, but it is not clear if MAPKs are activated following ICH. The present study investigated if MAPK pathways in the brain are activated following ICH and whether hematoma-derived complement C3 plays a role in HO-1 upregulation and MAPK activation induced by ICH.

Methods: Male complement C3-sufficient and C3-deficient mice were used in this study. Mice received an intracerebral infusion of either 30-mL of autologous or donor blood. There were four groups: C3-sufficient blood injected into C3-sufficient brain, C3-deficient blood injected into C3-deficient brain, C3-deficient blood injected into C3-sufficient brain and C3-sufficient blood injected into C3-deficient brain. All mice were killed 1 or 3 days later and the brains were used to measure HO-1 and MAPKs (p44/42 and p38) by Western blot analysis and immunohistochemistry.

Results: Western blot analysis showed HO-1 levels in the ipsilateral basal ganglia after 3 days were lower in C3-deficient brains receiving C3-deficient blood (2081±987 pixels) compared to C3-sufficient brains injected with C3-sufficient blood (3939±505 pixels, p<0.05). There were also lower HO-1 levels in the ipsilateral basal ganglia of C3-deficient mice receiving C3-sufficient blood (1951± 1144 pixels) compared to C3-sufficient mice receiving C3-deficient blood (4618±515, p<0.05). Most of HO-1 positive cells in the perihematomal zone were glia-like cells.

For MAPK measurements, the ratio of phosphorylated MAPK to total MAPK levels was determined at 24 hours. C3-sufficient mice infused with their own blood had higher levels of activated p44/42 MAPK in the ipsilateral basal ganglia (0.69 ± 0.34) compared to C3-defficient mice infused with their own blood (0.04 ± 0.08, p<0.05). Also, there were significant higher activated p44/42 MAPK levels in C3-sufficient mice injected with deficient blood (1.23 ± 0.13) compared to C3-defficient mice infused with C3-sufficient blood (0.38 ± 0.33, p<0.05). The effect of C3 sufficiency/deficiency on p38 MAPK activation was similar to that on p44/42 MAPK.

Conclusion: These results indicate that complement factor C3 has a central role in HO-1 upregulation and p38 and p44/42 MAPK activation following ICH. The source of that complement does not appear to be the hematoma but rather local brain production or entry across the blood-brain barrier.

CONGENIC REMOVAL OF HYPERTENSION QTL ATTENUATES FOCAL ISCHEMIC INJURY

Hiroshi Yao¹, Zong-Hu Cui², Junichi Masuda³, Toru Nabika²

¹Center for Emotional and Behavioral Disorders, National Hospital Organization Hizen Psychiatric Center, Saga, Japan, ²Department of Functional Pathology, Shimane University School of Medicine, Izumo, Japan, ³Department of Laboratory Medicine, Shimane University School of Medicine, Izumo, Japan

Background and aims: A genome-wide screen found a blood pressure quantitative trait locus (QTL) on rat chromosome 1 in stroke-prone spontaneously hypertensive rats of a Japanese colony (SHRSP/Izm). Congenic strategy using inbred rat strains that display genetic hypertension is a powerful method to examine the effects of polygenetic traits on the stroke phenotype. In the present study, we investigated the effects of congenic removal of this QTL from SHRSP/Izm on infarct size produced by middle cerebral artery (MCA) occlusion. Methods: To establish the congenic strain (SHRSPwch1.0), the blood pressure QTL was introgressed from Wistar-Kyoto/Izm to SHRSP/Izm by repeated backcrossing. Male SHRSP/Izm (10-12 weeks old [young adult] n=8, 5 months old [adult] n=17) and SHRSPwch1.0 (young adult n=7, adult n=15) were randomly assigned to distal MCA occlusion as previously described (Stroke 1996;27:333-336). Briefly, a krypton laser operating at 568 nm (643-Y-A01, Melles Griot Inc.) was used to irradiate the distal MCA at a power of 20 mW for 4 min. The photosensitizing dye rose bengal (15 mg/mL in 0.9% saline; Wako) was administered intravenously to a body dose of 20 mg/kg over 90 sec starting simultaneously with 4 min of laser irradiation. Cerebral blood flow (CBF) was determined with laser-Doppler flowmetry 30 minutes after MCA occlusion in adult SHRSP/Izm (n=5) and SHRSPwch1.0 (n=5). A laser-Doppler flowmetry probe was laterally scanned, and CBF of the distal MCA cortical territory was measured at 5 points (2 mm posterior and 2.0, 2.5, 3.0, 3.5, and 4.0 mm lateral to the bregma) as previously described (Stroke 2003;34:2716-2721). Results: Resting mean arterial blood pressure (MABP) was 212±23 mmHg in adult SHRSPwch1.0, which was significantly lower than 241±22 mmHg in SHRSP/Izm. Infarct volume in the congenic rats was significantly decreased compared with that in SHRSP/Izm (66.4±21.5 mm3 vs. 103.4±24.8 mm3). CBF, determined at collaterally-perfused cortex with laser-Doppler flowmetry after MCA occlusion, was significantly greater in adult SHRSPwch1.0 compared with CBF in adult SHRSP/Izm. In young adult rats, there were no significant differences in MABP or in infarct volume between SHRSPwch1.0 and SHRSP/Izm. Conclusions: The congenic removal of a blood pressure QTL lowered blood pressure and caused a substantial reduction in infarct volume (−36%) with increased collateral CBF after MCA occlusion in the congenic rat. We demonstrated for the first time that the congenic strategy is useful to investigate the effects of genetic hypertension on focal ischemia or stroke.

DELETERIOUS EFFECTS OF DIHYDROTESTOSTERONE ON CEREBRAL ISCHEMIC INJURY

Jian Cheng, Nabil J. Alkayed, Patricia D. Hurn

Department of Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA

Outcome from cerebral ischemia is sexually dimorphic in many experimental models. Male animals display greater sensitivity to ischemic injury than do their female counterparts, however the underlying mechanism is unclear. The present study determined if the potent and non-aromatizable androgen, dihydrotestosterone (DHT), exacerbates ischemic damage in male rats and alters post-ischemic gene expression after middle cerebral artery occlusion (MCAO). Adult male rats were castrated and hormone-replaced with either testosterone (T) or DHT (50 mg subcutaneous implant) one week before MCAO. 2h MCAO was achieved by intraluminal filament techniques and uniformity of MCAO was confirmed by laser Doppler flowmetry. The effect of DHT on cerebral infarct damage was quantified at 22h reperfusion by 2% 2,3,5-triphenyltetrazolium chloride (TTC, Sigma) histology and compared to that of T. Micropunches (1 mm diameter) from cortical peri-infarct zones and corresponding contralateral tissue were isolated and analyzed for early gene expression at 6 h of reperfusion using DNA microarrays (Affymetrix). Quantitative real-time RT-PCR (qPCR) was used for subsequent confirmation of differential expression pattern of selected gene candidates, including anti-apoptosis gene bcl-2 and pro-apoptosis gene bax. As expected, castration reduced plasma T and DHT to below the level of assay detection (Radioimmunology Assay), while hormone replacement restored total T and DHT plasma levels (37±7 and 3±1 ng/ml, respectively) to the physiologically relevant ranges. Castrated males (CAST) sustained smaller damage than gonadally intact males (Intact) or DHT-treated animals. (Cortex: CAST 15±6% of ipsilateral structure; Intact 34±4%; DHT 42±4%; Striatum: CAST 43±6 %; Intact 75±5%; DHT 88±4%). T increased infarction in striatum (82±5%) as compared to CAST, but not in cortex. Microarray studies identified transcripts that were altered by DHT alone or DHT with MCAO/ 6 h reperfusion. In DHT-treated animals, 422 genes/ESTs were increased by MCAO and 358 were decreased. In castrated animals, 271 genes/ESTs were increased and 350 were depressed by ischemia. When ischemic cortex was compared between DHT and CAST groups (DI-CI), 421 genes were increased by DHT treatment and 239 genes were decreased. Functional analysis of the genes with differential expression patterns between DI-CI indicated that a larger number of genes were elevated by DHT vs castration after ischemia, many of which could account for cell death through enhanced inflammation, dysregulation of blood-brain barrier and the extracellular matrix, apoptosis, and ionic imbalance. We used qPCR and immunoblotting to confirm the microarray data. Seven of ten gene candidates were confirmed by qPCR and the differential expression pattern of cyclooxygenase-2 was additionally confirmed by immunoblotting. By using qPCR techniques we also found that intact and DHT groups demonstrated higher levels of cortical bax mRNA relative to CAST at 22h reperfusion, but bcl-2 was similar in all groups. Our data suggest that DHT is an important androgenic mediator of ischemic damage in male brain and that transcriptional mechanisms should be considered as we seek to understand innate male sensitivity to cerebral ischemia. Supported by NIH NS 49210, NS33668, N03521

EARLY INCREASE OF BLOOD OXYGEN TRANSPORT CAPACITY ALLEVIATES ISCHEMIC BRAIN DAMAGE

Clemens Heiser, Lothar Schilling

Division of Neurosurgical Research, Medical Faculty Mannheim, University of Heidelberg, Heidelberg, Germany

Background and aims: In the pathophysiological sequelae of cerebral ischemia a number of reaction pathways are thought to be involved including excitotoxic, apoptotic, and inflammatory mechanisms. However, the contribution of the individual pathways is still a matter of debate. In the present study we sought to characterize the role of tissue hypoxia on structural brain damage in a model of focal brain ischemia.

Methods: Male Sprague-Dawley rats received an intravenous bolus injection of pimonidazole (60 mg/100 g bw) and 1 h later underwent a permanent right-sided middle cerebral artery occlusion (MCAO) using an intraluminal filament. After establishing focal ischemia NaCl (1 ml/100 g bw, experimental groups I, II) or an artificial O2 carrier (Oxycyte, a second generation perfluorocarbone solution, 1 ml/100 g bw; group III) was injected followed by breathing air (group I) or pure oxygen (normobaric hyperoxygenation [NBH]), groups II, III) for a maximum of 8 h. The animals were killed 2, 4, 8 h (group I) or 2, 4, 8, 24 h (groups II, III) after MCAO, the brain was removed and serially cut in coronal slices for immunohistochemical detection of pimonidazole adducts indicative of cellular hypoxia or silver nitrate staining for outlining structural brain damage, respectively. The sections were scanned for volumetric analysis with appropriate correction for brain swelling.

Results: Even after 2h of MCAO areas of ischemic brain damage were clearly detectable and could be quantified in each animal. The results (mean+SD) are listed in the table with swelling given in % of contralateral hemisphere and damage and hypoxia indicated in mm3 (+p<0.05 vs. group II; *p<0.05 vs. group III; n.d., not determined).

Conclusions: Improving the O2 transport capacity by an artificial carrier early after onset of focal cerebral ischemia resulted in a significant decrease of tissue damage. This effect was sustained even after ceasing this therapy. These results suggest that an early tissue hypoxia plays a significant role in the development of post-ischemic brain damage.

NEURONS OF THE RODENT BRAIN EXPRESS HEMOGLOBIN

Dominik W. Schelshorn, Armin Schneider, Heinrich F. Burgers, Maria Harlacher, Mathilde Lorenz, Wolfgang Kuschinsky, Martin Henrik Maurer

Dept. of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany

The role of hemoglobin in oxygen transport of red blood cells is well known. In vertebrates the expression of neuroglobin and cytoglobin has been demonstrated in neurons, whereas the existence of hemoglobin in these cells has so far not been verified. While the presence of hemoglobin mRNA and protein has been assessed in brain homogenate, these findings have been interpreted as artifacts derived from blood residues in the brain samples.

In this study, we systematically investigated the in vivo expression of hemoglobin and its mRNA in the adult rodent brain. To exclude red blood cell contamination, we also assessed the basic in vitro expression of hemoglobin and its mRNA in rat neuronal cultures. The cultures were also treated with known inductors of hemoglobin synthesis to identify potential regulators of expression.

FOCAL CEREBRAL ISCHEMIA IN THE TNF-TRANSGENIC RAT: REGIONAL EXPRESSION OF TNF PROTEIN, TNF CONVERTING ENZYME, AND TNF RECEPTOR 1

Luther Creed Pettigrew^1,2, Xiao-Hong Song¹, Susan D. Craddock¹, Mark S. Kindy^3,4

¹Stroke Program, Sanders-Brown Center On Aging, and Department of Neurology, University of Kentucky Medical Center, Lexington, KY, USA, ²Department of Veterans Affairs Medical Center, Lexington, KY, USA, ³Department of Physiology and Neurosciences, Medical University of South Carolina, Charleston, SC, USA, ⁴Department of Veterans Affairs Medical Center, Charleston, SC, USA

Background and Aims: Tumor necrosis factor-alpha (TNFa) is an inflammatory mediator that becomes elevated in ischemic brain. TNF converting enzyme (TACE) cleaves active TNFa from a transmembrane precursor. Active TNFa will bind to its receptor, TNFR1, and may promote apoptosis through activation of caspase 8. We tested the hypothesis that overexpression of the murine TNFa gene in the brain of a transgenic (TNFa-Tg) rat will affect TNFa protein levels and alter regional expression of TACE and TNFR1 after focal ischemia.

Methods: TNFa-Tg and wild type (WT) rats underwent middle cerebral artery occlusion (MCAO) for 1 hr. Samples of ischemic core (IC) and penumbral (PN) tissue obtained after 30 min or 3/24 hr of reperfusion were assayed for TNFa (Linco multiplex) and expression of TACE and TNFR1 (western blot).

Results: The TNFa levels in IC and PN of TNFa-Tg rat brain were unchanged at 30 min and 3 hr but fell below non-ischemic control levels at 24 hr (p<0.02; n=3 ? 5 per group; Fisher's test; Table 1). The TNFa levels in IC and PN of WT rats could be measured only at 24 hrs (p=NS compared to non-ischemic WT controls). TNFa levels were significantly higher in IC (p<0.003) and PN (p<0.01) of the TNFa-Tg rat than in WT animals at all 3 sampling times.

Non-ischemic TNFa-Tg and WT rats were equivalent in expression of active TACE isotype and TNFR1. After MCAO, TACE expression increased in both IC and PN for TNFa-Tg and WT rats at 30 min (p<0.003) and at 24 hrs (p<0.03) but only in TNFa-Tg animals at 3 hrs (p<0.03). In ischemic TNFa-Tg rats, TNFR1 expression increased in both IC and PN at 30 min (p<0.004) and remained upregulated at 3 (p<0.05) and 24 (p<0.0001) hrs in IC. For ischemic WT animals, TNFR1 expression remained unchanged at 30 min but was increased in both IC and PN at 3 (p<0.01) and 24 hrs (p<0.006).

Conclusions: TNFa increased in both IC and PN after MCAO in TNFa-Tg rats and remained several fold higher than in WT rats 24 hrs afterward. TACE was not constitutively upregulated but became increased in response to ischemic stress more robustly than in WT animals. TNFR1 expression became elevated in the TNFa-Tg animal during early reperfusion and showed regional heterogeneity not observed in WT rats. We conclude that cerebral ischemic injury in the TNFa-Tg rat may amplify basal TNFa synthesis through upregulation of TACE and could augment cellular death through enhanced TNFR1 activation. This research was supported by NIH/NINDS NS047395 (LCP).

MICROCIRCULATORY RESPONSE TO FUNCTIONAL BRAIN STIMULATION

Bojana Stefanovic, E. Hutchinson, V. Yakovleva, L. Belluscio, A.P. Koretsky, A.C. Silva

NINDS, NIH, Bethesda, MD, USA

THE ROLE OF THE BLOOD FLOW RESPONSE DURING FUNCTIONAL ACTIVATION IN THE SOMATOSENSORY CORTEX

Roger Springett, Michelle Abajian, Maureen Ripple

Department of Radiology, Dartmouth Medical School, Hanover, NH, USA

Introduction: Optical spectroscopy has been extremely successful in measuring the vascular response to functional activation in the brain. However, current optical systems only measure the change in hemoglobin concentration and cannot measure the baseline state. We have developed and validated a visible spectroscopy system to measure the absolute concentration and saturation of hemoglobin as well as the oxidation state of cytochrome c (Cytc). Cytc is a component of the mitochondrial electron transport chain and is sensitive to oxygenation at the mitochondrial level. This study is examines mitochondrial oxygenation during forepaw activation in the rat.

Methods: Sprague Dawley rats were anesthetized with isoflurane, the skull thinned for optical access and electrodes were inserted in the forepaw. The inspired oxygen fraction (FiO2) was set to 0.30 and the animals artificially ventilated. The anesthetic was then switched to alpha-chloralose, the animals allowed to stabilize for 1 hour and then the forepaw was stimulated with a pulse train that consisted of 10 repeats of 4 seconds of 0.4ms pulses at 5Hz with amplitude of 1mA. FiO2 was then decreased to 0.16 and the pulse train repeated.

Results: Baseline hemoglobin saturation (SmcO2) was 75% under isoflurane when FiO2 was 0.30, it decreased to 55% under alpha-chloralose anesthesia and increased to 70% during stimulation. When FiO2 was decreased to 0.16, SmcO2 decreased to 45% and increased to 55% during stimulation. Switching to alpha-chloralose anesthesia resulted in small reduction of Cytc amounting to 3% of total Cytc. During activation, Cytc reoxidized to isoflurane baseline values. When FiO2 was decreased to 0.16, there was a further reduction in Cytc by an additional 6% and activation resulted in an oxidation back to 3%.

Discussion: Previously, we have observed that Cytc is independent of small changes in cerebral oxygenation under isoflurane anesthesia in the rat and that SmcO2 must fall below 60% before there is a reduction in Cytc. Alpha-chloralose is the preferred anesthetic for small-animal activation studies and it reduced Cytc by 3% consistent with the decrease in SmcO2. During activation, there was a small oxidation in Cytc as the blood flow response increased SmcO2 and relieved this reduction. This oxidation is consistent with the oxidation observed in cytochrome oxidase during activation in the human brain using near infrared spectroscopy and suggests that alpha-chloralose anesthesia in the rat accurately models cerebral oxygenation in the human brain. When the relationship between Cytc and SmcO2 was compared during activation and during hypoxia, it was observed that Cytc reacted passively to changes in SmcO2.

The results suggest that mitochondrial oxygen tension in the brain is in the range where the mitochondria have to adapt their metabolic state to maintain oxygen consumption and rate of ATP production. This adaptation is observed as a small reduction in Cytc under otherwise normoxic conditions. We hypothesize that the role of the blood flow response during activation is to increase mitochondrial oxygenation and drive the mitochondria out of this adaptive-range of oxygen tension into the range where the mitochondrial metabolic state is independent of oxygen.

SPECIFIC SUBSETS OF CORTICAL GABA INTERNEURONS ARE INVOLVED IN THE NEUROVASCULAR COUPLING RESPONSE TO WHISKER STIMULATION IN THE RAT

Priscilla Fernandes¹, Ara Kocharyan¹, Elvire Vaucher², Edith Hamel¹

¹Laboratory of Cerebrovascular Research, Montreal Neurological Institute, McGill University, Montreal, QC, Canada, ²Ecole D'Optometrie De L'Universite De Montreal, Montreal, QC, Canada

Introduction: The rat whisker-to-barrel pathway has a well-defined topography and displays well-delineated changes in contralateral cerebral blood flow (CBF) upon specific whisker stimulation. However, the cellular basis of this neurovascular coupling response remains largely unknown. In this study, we identified the neurons that are activated in the somatosensory cortex by whisker stimulation using double-immunocytochemistry. Further, we investigated the contribution of neuronal and glial messengers in this hemodynamic response by CBF monitoring with laser Doppler flowmetry (LDF).

Methods: Adult male Sprague-Dawley rats (280-320 g) were handled for one week prior to the experiment to habituate them to the stimulation procedure. Three days prior to the experiment all but three whiskers (C1-C3) on the right side of the snout and all whiskers on the left side were clipped under anesthesia. On the experimental day, unanesthetized rats were lightly restrained, and the C1-C3 vibrissae gently stroked (20 min). The identity of the activated cells (c-Fos positive) was determined by double-immunostaining for COX-2 (pyramidal cells), glial fibrillary acidic protein (GFAP, astrocytic marker), COX-1 (microglia) or, for the different subsets of GABA-interneurons, their co-localized peptides (somatostatin (SOM), neuropeptide Y (NPY), vasoactive intestinal polypeptide (VIP)) or enzymes (nitric oxide synthase (NOS), choline acetyltransferase (ChAT)) in 25 um coronal and 30 um tangential sections from perfusion-fixed brains 60 min after stimulation, in both the ipsilateral (control side) and contralateral (activated side) hemispheres. In another set of experiments in anesthetized rats, the mediators of the hemodynamic response evoked by 20 sec whisker stimulation were assessed by LDF monitoring of CBF after intracisternal injections (3 µl of a 10-4 M, pH 7.4 buffered solution) of either vehicle, a GABA-A receptor antagonist (picrotoxin), or selective inhibitors of COX-1 (SC-560) or COX-2 (NS-398).

Results: Whisker stimulation elicited CBF increases (35.3± 2.8%, from baseline, p<0.05, t-test, n=15), and a strong activation of c-Fos expressing cells in the contralateral barrel cortex corresponding to the activated whiskers. There was a selective activation of c-Fos positive nuclei in COX-2 positive pyramidal cells (25.4% of all labeled cells, p<0.01, t-test), and in GABA interneurons that contain VIP and/or ChAT (18.6 and 13.9% of all labeled cells, respectively, p<0.01, t-test), but not SOM, NPY or NOS. Astrocytes and microglia did not display c-Fos immunoreactivity. Tangential sections with cytochrome oxidase-delineated barrels demonstrated activation of VIP or ChAT interneurons and COX-2 pyramidal cells within the hollows of the activated barrels. The induced CBF response was significantly decreased by picrotoxin (−38.0%, p<0.01, n=7) and, in agreement with previous studies, by NS-398 (1) (−44.6%, p<0.05, n=4) but not by SC-560 (2) (−4.2%, ns, n=4).

Conclusions: These results show that GABA, most likely released from cortical GABA interneurons that specifically colocalize VIP and/or ChAT, contributes to the hemodynamic response to whisker stimulation, together with mediators released from COX-2 pyramidal cells (1). Whether the effects are direct through dilatory cerebrovascular GABA-A receptors or indirect via astroglial intermediaries is currently under investigation. Supported by CIHR grant (MOP-53334, EH).

EVIDENCE FOR CAPILLARY VASOCONSTRICTION IN CEREBRAL CORTEX OF MICE

Francisco Fernandez-Klett, Nikolas Offenhauser, Ulrich Dirnagl, Josef Priller, Ute Lindauer

Department of Experimental Neurology, Charite, Universitaetsmedizin Berlin, Berlin, Germany

The role of capillaries in the regulation of cerebral blood flow is still under debate. Pericytes, the second cell type present in capillaries beneath endothelial cells, possess contractile properties in vitro and have recently been shown to contract and dilate capillaries in cerebellar slices and the retina of mice (Peppiatt et al., 2006), but the physiological significance in vivo is still unclear.

Aim of our study was to investigate the ability of cortical capillaries to actively constrict in vitro and to elucidate the impact on blood flow in vivo by application of the thromboxane A2 (TXA2) mimetic U46619. TXA2 is a well known vasoconstrictor which can be released by platelets, endothelial cells and microglia and has been involved in the pathophysiology of stroke, subarachnoidal hemorrhage and CNS trauma. We have used the beta-actin-eGFP mouse, in which a large expression of GFP in endothelial cells and pericytes enables a detailed depiction of capillary morphology and morphometric changes by means of two photon fluorescence microscopy. In brain slices of the sensory cortex we observed a significant constriction in near all capillaries studied with 100 nM U46619 (74±8 % of outer diameter, n=7). The constriction was concentration-dependent. Prior incubation with the competitive TXA2-receptor antagonist SQ 29,548 (1 µM) totally prevented the constriction (n=5). In 4 out of 6 cases, SQ 29,458 also elicited a partial reversal of the reduction of capillary diameters induced by U46619. Hovewer, the nitric oxide donor spermine-NO (0,1 mM) failed to revert the constriction in 6 out of 7 cases. Morphologically, the constriction showed mostly a sphincter-like pattern and often occurred at capillary junctions. Occasionally, changes in the shape of the soma of pericytes were observed, suggesting that these cells are responsible for the constriction. This results demonstrate that cortical capillaries are capable of actively changing their lumen size, independently of perfusion changes at higher order vessels.

We have further used the closed cranial window preparation over the cerebral cortex of anesthetized, mechanically ventilated beta-actin-eGFP mice to study the vascular response to superfusion of U46619 in vivo. By means of two photon microscopy we have assessed the erythrocyte flow in single capillaries as well as diameter changes in capillaries and pial vessels. In parallel, laser-Doppler flowmetry was used to measure cerebral blood flow. Our preliminary results (n=3) show a decrease of blood flow in response to the TXA2 mimetic U46619 as measured by LDF. In single capillaries, flow reductions up to flow stops occured, together with a slight reduction in capillary diameters. Interestingly, this was not paralleled by a net constriction of neither pial arterioles nor venules.

Taken together, our data support the notion of capillaries as active contributors to blood flow changes in the cerebral cortex. This mechanism could be relevant in pathological conditions with increased TXA2 release.

Supported by DFG and Hermann and Lilly Schilling Foundation.

THE CEREBRAL SYMPATHETIC NORADRENERGIC SYSTEM AND REGIONAL CEREBRAL ENERGY METABOLITES IN AVM-PATIENTS - A MICRODIALYSIS STUDY

Carsten Stueer¹, Toshiki Ikeda², Michael Stoffel¹, Oliver Sakowitz³, Gerd Luippold⁴, Carlo Schaller², Bernhard Meyer¹

¹Neurosurgical Department, Munich Tech University, Munich, Germany, ²Neurosurgical Department, Bonn Medical Center, Bonn, Germany, ³Neurosurgical Department, University of Heidelberg, Heidelberg, Germany, ⁴Institute of Pharmacology, University of Tuebingen, Tuebingen, Germany

Background and aims: To test the hypothesis that the activity of the sympathetic nervous system (SNS) and regional metabolism around resection for arteriovenous malformations differs from those measured in surgery for nonvascular lesions.

Methods: Cortical microdialysis (MD) was performed intraoperatively before and after resection in n = 12 patients with small/medium-sized AVM and transsylvian dissection in n = 15 patients with nonvascular lesions (= controls). Serum Norepinephrine (NE) concentrations were assessed before and after resection/dissection to control for systemic NE alterations. All data were pooled according to groups and times, and compared via ANOVA (P<0.05) plus a post-hoc Dunnet correction for multiple comparisons. All data (given below as means ± SD) were collected in the context of a prospective study.

Results: Activity of the sympathetic nervous system as assessed with NE microdialysate concentrations were significantly higher in non-AVM patients (p<0.05) prior to sylvian transsection (5.4 ± 1.4 nmol/L) compared to post dissection (4.2 ± 1.1 nmol/L). NE levels were significantly lower in the AVM group, while there was no significant difference between pre- and post conditions (AVM pre: 3.3 ± 1.2 nmol/L, AVM post: 2.9 ± 1.7 nmol/L). Serum NE levels remained constant throughout surgery in both groups (Control pre: 1.0 ± 0.8 nmol/L, post: 0.9 ± 0.7 nmol/L; AVM pre: 0.7 ± 0.6 nmol/L, post: 1.1 ± 0.9 nmol/L), Concentrations of lactate, glutamate, glycerol as well as lactate/pyruvate ratio referred to as regional cerebral energy metabolites were without any significant differences.

Conclusions: Chronic arterial non-ischemic hypoperfusion in human AVM correlates with a down-regulation of the activity of the cerebral sympathetic nervous system. The postresection, protective capacity of the SNS seems sole impaired in AVM patients, which might be the basis for a hyperperfusion syndrome in larger and complex AVM.

INTRAMICROVASCULAR BEHAVIOR OF LEUKOCYTES IN MOUSE BRAIN OBSERVED BY HIGH SPEED LASER CONFOCAL FLUORESCENCE MICROSCOPY

Norio Tanahashi¹, Kimihiko Hattori¹, Nobuo Araki¹, Kunio Shimazu¹, Tatsuya Saitoh², Minoru Tomita³

¹Department of Neurology, Saitama Medical University, Saitama, Japan, ²Department of Neurosurgery, Saitama Medical Center, Saitama Medical University, Saitama, Japan, ³Department of Neurology, Keio University, Tokyo, Japan

PLATELET BEHAVIOR AND THROMBUS FORMATION IN MOUSE BRAIN MICROVASCULATURE OBSERVED ON HIGH SPEED LASER CONFOCAL FLUORESCENCE MICROSCOPE

Kimihiko Hattori¹, Norio Tanahashi¹, Nobuo Araki¹, Kunio Shimazu¹, Tatuya Saitoh², Minoru Tomita³

¹Department of Neurology, Saitama Medical University, Saitama, Japan, ²Department of Neurosurgery, Saitama Medical Center, Saitama Medical University, Saitama, Japan, ³Department of Neurology, Keio University, Tokyo, Japan

NON-INVASIVE HYPOTENSIVE FMRI STUDY IN RAT SOMATOSENSORY CORTEX

Peter Herman^1,2, Basavaraju Sanganahalli¹, Fahmeed Hyder^1,3, Andras Eke²

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA, ²Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Budapest, Hungary, ³Department of Biomedical Engineering, Yale University, New Haven, CT, USA

HYPERCHOLESTEROLEMIA WORSENS THE BLOOD FLOW DEFICIT IN FOCAL CEREBRAL ISCHEMIA IN THE ABSENCE OF ATHEROSCLEROSIS

Hwa Kyoung Shin¹, James A. Lapointe li², Dmitriy N. Atochin², Michael A. Moskowitz¹, Paul L. Huang², Cenk Ayata^1,3

¹Stroke and Neurovascular Regulation Laboratory, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ²Cardiovascular Research Center and Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ³Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA

Background and aims: Hypercholesterolemia may increase stroke risk by accelerating atherosclerosis and narrowing luminal diameter in cerebral vessels as well as disrupting vascular endothelial and smooth muscle function. We tested the hypothesis that that hypercholesterolemia impairs cerebrovascular function and worsens cerebral blood flow (CBF) deficit and tissue outcome after focal ischemia even in the absence of hemodynamically significant vessel narrowing.

Methods: We tested this hypothesis in ApoE knockout (KO) mice using high spatiotemporal resolution laser speckle flowmetry through intact skull noninvasively, and studied the hemodynamic impact of hypercholesterolemia on the evolution of CBF deficit in focal cerebral ischemia. Wild type (WT, C57BL/6), ApoE KO or ApoE/eNOS double KO mice were fed high-fat “western” diet for 4, 8, or 10 weeks starting at 4 weeks of age. Histopathological examination of extracranial arteries in 14 week-old ApoE KO showed only minimal atherosclerotic changes such as fatty streaks with subendothelial foam cells. Intracranial arteries did not develop atherosclerotic lesions even after 12 months, despite severe extracranial atherosclerosis.

Results: Mean arterial blood pressure was higher in ApoE KO compared to WT in all age groups (94+/−2 vs. 85+/−2 mmHg, respectively; n=17 each, p<0.05); arterial blood gas values did not differ. Distal middle cerebral artery occlusion (dMCAO) resulted in significantly larger area of severe CBF deficit (?20% residual CBF) in ApoE KO after 8 and 10, but not 4 weeks of high-fat diet compared to age- and diet-matched WT (n=4-7; see table). Ten weeks of high-fat diet also tended to increase the CBF deficit in WT mice compared to 4 or 8 weeks of high-fat diet (p=0.1). ApoE/eNOS double KO mice (8 weeks high-fat diet) developed 80% larger area of severe CBF deficit compared to age- and diet-matched ApoE KO (n=6 and 7, p<0.05). Unexpectedly, the frequency of peri-infarct depolarizations was significantly reduced in all groups of mice on high-fat diet compared to mice on regular diet (p<0.05), regardless of diet duration. Consistent with larger CBF deficits, infarct volumes (48h after 1h dMCAO) were 60% larger in ApoE KO mice compared to wild type after 8 weeks of high-fat diet (n=5 each, p=0.1).

Conclusions: These data suggest that hypercholesterolemia impairs cerebrovascular response to focal ischemia in the absence of hemodynamically significant atherosclerosis.

This work was supported by the American Heart Association (0335519N, Ayata), NIH (P50 NS10828, Moskowitz; NS048426, Huang).

REDUCTION IN 5HT2A BINDING POTENTIAL USING [11C] MDL 100907 IN SCHIZOPHRENIA

Dean F. Wong^1,2, Nicola Cascella², Yun Zhou¹, James R. Brasic¹, Mohab Alexander¹, Weiguo Ye¹, Ayon Nandi¹, Anil Kumar¹, Marika Maris¹, Suzana Corritori³, Daniel P. van Kammen³

¹Department of Radiology, Johns Hopkins Medical Institutions, Baltimore, MD, USA, ²Department of Psychiatry, Johns Hopkins Medical Institutions, Baltimore, MD, USA, ³ACADIA Pharmaceuticals Inc., San Diego, CA, USA

Background and Aims: Schizophrenia has been associated not only with altered dopamine receptors but also serotonin because of the possibility of serotonin's relationship in the therapeutic action of atypical antipsychotics. Some previous PET studies in drug naive patients using [18F]setoperone or in subjects at enhanced risk for schizophrenia using [18F]altanserin have demonstrated reductions in 5HT2A receptors in frontal and pre-frontal cortex, respectively. We now present the results of a study in drug free schizophrenia patients using the selective 5HT2A PET radioligand [11C] MDL 100907 (MDL).

Methods: Ten medically healthy subjects with schizophrenia or schizoaffective disorder (SS) (9 males, 1 female, average age 41 + 9) and nine healthy normal controls (CS) (4 females, 5 males average age 32 + 8) were studied. Exclusion criteria for the schizophrenia population included any other DSM-IV Axis I diagnosis other than schizophrenia/schizoaffective disorder, evidence of a current acute psychotic episode or any neurological disorders. The schizophrenia population was given a washout period of 2–3 weeks from any neuroleptic medication and any medication with a known effect on serotonin receptors.

In order to obtain measurements of 5-HT2A receptor levels, we administered 18-20 mCi IV bolus of the radiotracer [11C]MDL 100907 (MDL) followed by a 95-min PET scan. A parametric imaging approach (Zhou et al, 2003) was used to estimate MDL binding potential (BP)

Results: When the 10 SS were compared with the nine healthy CS, we found that SS had lower 5HT2A binding potential (BP) by up to 50% in all cortical regions as compared to CS (p = 0.001). However, since several studies have shown an age-related decrease in 5HT2A receptor levels (Adams, et al, 2004), we also examined an age-matched subset of the population, which included 6 patients and 6 controls, matched to an average age of 33 +7 years for both populations. In the age matched set, SS had significantly lower BP than CS in the following regions: cingulate cortex (p= 0.02), orbito-frontal cortex (p= 0.04), pre-frontal cortex (p=0.02), and temporal cortex (p=0.03).

Prior exposure to neuroleptics and other serotonin-affecting medications, though terminated at least two weeks before the PET study in this case, may have been a confound. However, when we grouped the subjects based on drug free period and dose of antipsychotics taken and analyzed the BP data in terms of these groups, we could find no significant relationship between drug free period or type and dose of antipsychotics and BP, using both parametric and non-parametric (rank-order) analyses.

Conclusions: This study showed significant reductions in BP with MDL in cingulate, orbito-frontal, pre-frontal and temporal cortices, matching findings from previous PET studies of 5HT2A binding with PET as well as post-mortem studies (Meltzer, et al, 1999; Adams, et al, 2004). In addition, we show additional cortical regions that were not observed in those studies, which focused only on the frontal and pre-frontal cortices. In summary MDL is a suitable approach and practical for not only investigating the pathophysiology of schizophrenia, but providing a convenient short-lived selective radiotracer for drug development studies.

COMPENSATORY INVOLVEMENT OF IPSI- AND CONTRALATERAL CORTEX AREAS IN APHASIA

Wolf-Dieter Heiss¹, Alexander Thiel², Lutz Winhuisen²

¹Max Planck Institute for Neurological Research, Cologne, Germany, ²Department of Neurology, University of Cologne, Cologne, Germany

CBF DURING CHANGES IN HEAD OF BED POSITION IN ACUTE, ISCHEMIC STROKE MONITORED AT THE BED-SIDE: HEMISPHERIC EFFECT OF INFARCT

Turgut Durduran^1,2, Chao Zhou², Brian L. Edlow³, Guoqiang Yu², Regine Choe², Brett L. Cucchiara³, Mary Putt⁴, Qaisar Shah³, Scott E. Kasner³, Arjun G. Yodh², Joel H. Greenberg³, John A. Detre³

¹Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA, ²Department of Physics and Astronomy, University of Pennsylvania, Philadelphia, PA, USA, ³Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA, ⁴Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, PA, USA

Introduction: In acute ischemic stroke, a flat head of bed (HOB) position is often used to improve perfusion to regions with presumably impaired autoregulation of CBF. Direct measurement of CBF at the microvascular level and at the bed-side is currently not feasible in intensive care units. We have developed an optical technique ? diffuse correlation spectroscopy ? that has enabled us to measure CBF changes in acute, ischemic stroke during changes in head of bed position (HOB). Hypothesis: CBF response to changes in HOB position on the infarcted hemisphere will be large compared to the “healthy” hemisphere providing evidence of impaired autoregulation in ischemic or penumbral cortex. Methods: Optical measurements of relative CBF during HOB manipulations were obtained from patients with acute hemispheric ischemic stroke (n=17) with probes placed on the forehead measuring CBF near the frontal poles. CBF was measured sequentially for 5 min each at HOB positions: 30, 15, 0, −5 and 0 degrees and was normalized to its value at 30 degrees. A linear mixed-effects model was used to analyze the statistical significance of relevant parameters. Results: A representative figure of a SF patient with unilateral ischemic stroke is shown. HOB position was found to be a significant factor (p<0.05) in both hemispheres, but was a stronger factor in the infarcted area (p<0.02) which also showed a larger variation. However, “time after stroke” was not a significant factor (p=0.45). Repeated measurements at flat HOB showed excellent repeatability at each study day (R>0.9). Conclusion: Diffuse optical correlation spectroscopy permits the measurement of relative CBF at the bedside. With further improvements in this technology, these measurements will provide valuable information permitting the manipulation of hemodynamic parameters so as to maximize the management of acute stroke patients.

IMAGING APOPTOSIS BY SPECT AFTER STROKE AND THE EFFECT OF MINOCYCLINE ON REDUCING APOPTOTIC CELL DEATH

Xiannan Tang^1,2, Francis Blankenberg³, Danye Cheng³, Midori Yenari¹

¹Department of Neurology, UCSF and SF VAMC, San Francisco, CA, USA, ²Department of Anesthesia, Stanford University, Palo Alto, CA, USA, ³Department of Radiology, Stanford University, Palo Alto, CA, USA

Introduction: Programmed cell death (apoptosis) is thought to play a role in brain injury following stroke. There are obvious advantages to being able to image apoptosis and the progression of stroke noninvasively. Annexin V is an endogenous protein with high affinity to phosphatidylserine (PS). Externalization of membrane PS is an early sign of apoptosis. We previously showed that radiolabeled annexin V can be used to noninvasively detect apoptosis in stroke models, and that annexin V binding occurs primarily in ischemic neurons. Minocycline is a common antibiotic that has strong anti-apoptotic and anti-inflammatory properties. Because of these properties minocycline may be of benefit in reducing neuronal apoptosis from ischemia and subsequent post-ischemic inflammation if administered soon after a stroke. In this study we explored whether 99mTc-annexin V imaging using SPECT (single photon emission computed tomography) could be used to monitor a potential treatment response to minocycline in experimental stroke. Methods: CB6/F1 adult male mice underwent unilateral distal middle cerebral artery occlusion (dMCAO) and were survived 1, 3, 7 and 30 days. Animals were given 45 mg/kg minocycline (or vehicle) i.p. 30 min and 12 hour after dMCAO, then 22.5 mg/kg twice daily up to the time of sacrifice. In each group, mice were injected with 5-10 mCi of 99mTc Annexin V 2 hours before undergoing SPECT. Sensorimotor function were studied using the cornerand ladder tests, then brains were collected for study of infarct volume and assessed for apoptosis using TUNEL staining and activated microglia using isolectin B4 (IB4) histochemistry. Double labeling was performed to identify neurons (NeuN), microglia (isolectin B4 or tomato lectin) and astrocytes (GFAP) that bound annexin V. Results: 99mTc-annexin V uptake in the injured (left) hemisphere was significantly increased by ischemia, and was decreased 2–3 fold by minocycline at all time points. Minocyline reduced infarct size as seen histologically and improved behavioral indices up to 30 d. The pattern of infarct volume histologically measured at different time points after stroke is highly consistent with that measured by SPECT. This was correlated to reduced numbers of TUNEL positive cells (P<0.05) & microglia (P<0.01) positive cells among treated mice. Double labelling studies demonstrated that annexin V bound mostly neurons at 1 and 3 d, and microglia at 7 and 30 d. Conclusions: Minocycline significantly reduces neuronal apoptosis and infarct size, and improves neurological outcome in mice following acute focal cortical ischemia. This correlated to patterns observed on annexin V SPECT imaging, and suggests that this may be a novel way to noninvasively monitor the progression of stroke and response to therapy.

CENTRAL ROLE OF ASTROCYTIC GLUTAMATE UPTAKE ACTIVITY IN ODOR-EVOKED FUNCTIONAL SIGNALS IN OLFACTORY GLOMERULI

Hirac Gurden^1,2,3, Barbara L'Heureux¹, Frederic Pain¹, Martine Guillermier², Laurent Pinot¹, Francoise Lefebvre¹, Albertine Dubois², Thierry Delzescaux², Roland Mastrippolito¹, Philippe Hantraye², Gilles Bonvento², Zach Mainen³, Philippe Laniece¹

¹Cnrs Umr8165, Imnc, Orsay, France, ²CNRS-CEA URA221, MIRCen, Fontenay-Aux-Roses, France, ³Cshl, New-York, NY, USA

Functional imaging signals are linked to metabolic and vascular activity, but how these processes are related to neural activity is not well understood. Recent data suggest that astrocytes play a key role in the mechanisms leading from neural transmission to functional signals. To provide direct insights into this issue, we studied the olfactory glomeruli (the functional modules allowing the first step of olfactory coding in the brain) where astrocytes express both GLT1 and GLAST glutamate transporters. To record odor-evoked metabolic and vascular changes in rat and mouse glomeruli, we used 1) imaging of intrinsic optical signals (IOS) 2) blood flow measurement by laser-Doppler flowmetry (LDF) 3) [14C]-2-Deoxy-D-Glucose (2DG) uptake.

First, we recorded IOS maps in both species in response to different odorants (phenol and benzaldehyde, 0.2 to 2% of saturated vapor) with changes in reflectance up to 1%. Second, using LDF, we measured more specifically blood flow in the local vascular network of the mouse glomeruli in response to benzaldehyde. LDF showed stimulus onset-locked and odor-specific blood flow increases from 20 to 30% in both species. Finally, we observed ~20% increase in glucose uptake among the subset of lateral and medial glomeruli activated bilaterally by benzaldehyde in awake mice.

We report that 1) blocking astrocytic glutamate transporters GLT1 and GLAST by TBOA blocks IOS (Gurden et al., 2006): this optical signal is reduced by about two-thirds, 36 ± 6% of control, P < 0.005 (t test, “control vs treated bulb”, n=5) but 2) 2DG uptake is not significantly different in the olfactory bulb of GLT1 knock-out mice compared to wild-types: respectively 17 ± 4% vs 20 ± 3% of relative 2DG uptake increase, n=6 in each group (the % represents relative changes calculated by the ratio of optical density values in activated / non-activated glomeruli). Thus, instead of GLT1, GLAST might have a central role in the neuro-energetic coupling in the olfactory glomeruli leading us to start the study of GLAST knock-out mice.

In conclusion, activation of olfactory glomeruli in rat and mouse is a suitable experimental system for the study of physiological neurovascular and neurometabolic couplings in vivo. According to our results, astrocytic transporter activity is crucial for the induction of functional optical signals. Finally, we believe that our current work on astrocytic transporter knock-out will help to determine whether glutamate uptake activity is the central signaling pathway leading from synaptic transmission to functional signals.

Supported by a Neuroscience grant from the ≪ Agence Nationale de la Recherche ≫ (Astroglo 2006).

IN VIVO DETECTION OF ADULT NEURAL STEM CELLS LABELLED IN SITU USING FERUMOXIDE-POLYCATION COMPLEX

Rachael Dobson^1,2, David L. Thomas³, David G. Gadian¹, Patrizia Ferretti², Mark F. Lythgoe¹

¹RCS Unit of Biophysics, UCL Institute of Child Health, University College London, London, UK, ²Developmental Biology Unit, UCL Institute of Child Health, University College London, London, UK, ³Wellcome Trust High Field Laboratory, Department of Medical Physics and Bioengineering, University College London, London, UK

Introduction: Neural stem cells (NSC) are located in the subventricular zone (SVZ) of the lateral ventricles in the adult brain. This population migrates towards the olfactory bulb along the rostral migratory stream (RMS), and has been shown to respond to ischaemic injury through increased neurogenesis and migration toward the infarct. Monitoring the timecourse of this behaviour in vivo has not been achieved with any imaging modality. Endogenous labelling with MRI contrast agents has been problematical due to diffusion of free agent outside the ventricles and low labelling efficiency. The polycation protamine sulphate enhances viral gene transfer through increased cell membrane contact and aggregation of virus. When complexed with Endorem, a ferumoxide contrast agent, protamine sulphate increases NSC labelling efficiency four-fold in vitr0. This complex, FePro, has been used to label stem cells in vitro. Here we demonstrate the utility of FePro to label adult NSC in situ in the SVZ, and we assess its labelling efficiency and regional distribution using MRI compared to Endorem alone.

Methods: Rats were injected with 2µl FePro or Endorem into the left lateral ventricle. FePro injected rats were sacrificed and perfused fixed with 4% paraformaldehyde at 24 hours, 7 days, and 28 days for histology, and scanned at 7 days and 28 days. Endorem-injected rats were scanned at 7 days and sacrificed. Animals were scanned using a 2.35T magnet interfaced to a SMIS console, with 2DFT coronal and sagittal spin echo sequences (TR 1500 ms; TE 120 ms; FOV 30 mm), and 3D gradient echo sequence (TR 80 ms; TE 40 ms; FOV 30 mm). Histological sections were stained with Prussian blue (PB) for iron detection and with Hematoxylin/Eosin, or for fluorescence immunohistochemistry with doublecortin and nestin.

Results: In Endorem-injected animals at 7 days, the injected contrast agent had diffused across the septal region into the right ventricle and along the RMS but did not localise with cells. In FePro-injected animals at 24hr, 7 days and 28 days, the contrast agent was localised to the left ventricle and some diffusion caudally into the left third and fourth ventricles occurred at later timepoints. At 28 days, signal was detected in the RMS. Immunohistochemistry confirms the presence of doublecortin-positive migrating cells in the SVZ and RMS. Histological sections show the FePro localising with cells in the SVZ, and with cells in the RMS.

Conclusions and discussion: Directional movement of free endorem along the RMS may be caused by the directional flow of CSF created by ciliary epithelium in the SVZ. The FePro complex prevents diffusion of endorem outside of the ventricles over a long period of time and enhances cell labelling. This may be due to restricted diffusion of the FePro complex through increased cell membrane interaction and increased internalisation of the complex. This labelling technique may provide an important methodology for monitoring the behaviour of endogenous stem cells in response to CNS injury using MRI.

CEREBRAL BLOOD FLOW AND METABOLISM IN PATIENTS WITH SILENT BRAIN INFARCTION

Ashkan Mowla¹, Reza Gharebaghi²

¹School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran, ²Shahed University, Tehran, Iran

OBJECTIVES: To evaluate CBF and metabolism in patients with Silent brain infarction (SBI) in relation to symptomatic stroke.

METHODS: The patients underwent PET and were separated into three groups; control group, with no lesions on CT (n=10, mean age 55), SBI group, with no neurological signs or history of stroke, but with ischaemic lesions on CT (n=10, mean age 61), and brain infarction group (BI group), with neurological deficits and compatible CT lesions in the area supplied by perforating arteries (n=20, mean age 56). Regional CBF, oxygen extraction fraction (OEF), cerebral metabolic rate for oxygen (CMRO2), and cerebral blood volume (CBV) were measured by PET.

RESULTS: Mean values for CBF to the cerebral cortex and deep grey matter were lower in the SBI group (32.5 (SD 5.8) and 33.3 (SD 6.9) ml/100 g/min, respectively) and in the BI group (30.8 (SD 5.2), 32.9 (SD 5.8), respectively) than in the Control group (36.0 (SD 6.6) and 43.5 (SD 9.5), respectively). Although mean CMRO2 of deep grey matter (2.36 (SD 0.52) ml/100 g/min) was significantly decreased in the SBI group compared with the control group (2.76 (SD 0.480), p<0.01), CMRO2 of the cortical area was as well preserved in the SBI patients (2.36 (SD 0.39)) as in the controls (2.48 (SD 0.32)) with a compensatory increase of mean OEF (0.45 (SD 0.06) and 0.41 (SD 0.05), respectively).

CONCLUSIONS: Patients with SBI showed decreased CBF and CMRO2 in deep grey matter. On the other hand, decreased CBF with milder increased OEF, resulting in preserved CMRO2 in the cerebral cortex indicates the presence of occult misery perfusion, suggesting that patients with SBI have reduced cerebral perfusional reserves.

MULTIVARIATE SYSTEM IDENTIFICATION OF CEREBRAL AUTOREGULATION

Tingying Peng¹, Alexander. B. Rowley^1,2, Stephen. J. Payne¹

¹Department of Engineering Science, University of Oxford, Oxford, UK, ²Computing Laboratory, University of Oxford, Oxford, UK

ACID-BASE MANAGEMENT AFFECTS ENDOTHELIUM-DEPENDENT CEREBRAL VASOREACTIVITY IN MODERATELY HYPOTHERMIC RATS

Izumi Yuzawa, Masaru Yamada, Kiyotaka Fujii

Department of Neurosurgery, Kitasato University School of Medicine Sagamihara, Kanagawa, Japan

Background and aims: The effect on the cerebral circulation of acid-base management under moderate therapeutic hypothermia is not fully understood. We examined cerebral vasoreactivity to topically administered vasodilators and hypercapnia under conditions of normothermia (37°C), induced hypothermia (30°C) with alpha- and pH-stat management, and rewarmed normothermia in rats.

Methods: Using laser-Doppler flowmetry and the closed cranial window method, we measured changes in regional cerebral blood flow (rCBF) and pial arteriole reactivity. rCBF changes during carbon dioxide (CO2) inhalation, and vasoreactivity to acetylcholine (Ach), an endothelium-dependent vasodilator, sodium nitroprusside (SNP), an endothelium-independent nitric oxide (NO) donor, and cilostazol, a phosphodiesterase III inhibitor were compared in normothermic rats, hypothermic rats subjected to alpha- and pH-stat management, and in rats rewarmed to normothermia.

Results: Ach-induced vasodilation was significantly attenuated under alpha-stat- but not pH-stat management and rewarming completely restored the normal vasodilatory response. CO2 reactivity was preserved under both acid-base management protocols.

Conclusion: At moderate hypothermia, endothelium-dependent vasodilation is reversibly impaired under alpha-stat- but not pH-stat conditions, indicating that acid-base management significantly affects endothelial cell function under moderate hypothermia.

LINEARIZABILITY OF NEURO-VASCULAR COUPLING RESULTING FROM LOW-FREQUENCY INTERMEDIATE-DURATION STIMULI

Tina Rasmussen^1,2,3,4, Martin Lauritzen^3,4, Ryuta Kawashima¹

¹Institute of Development Aging and Cancer, Tohoku University, Sendai, Japan, ²Department of Control Engineering, Institute of Electronic Systems, Aalborg University, Aalborg, Denmark, ³Department of Medical Physiology, Panum Institute, University of Copenhagen, Copenhagen, Denmark, ⁴Department of Clinical Neurophysiology, Glostrup University Hospital, Glostrup, Denmark

Previous studies [Ances et al. 2000, Martindale et al. 2003 and 2005] have investigated neuro-vascular coupling in rat somato-sensory cortex resulting from stimulation with a single frequency and a single duration. They have shown that this system can be approximated with a linear dynamic mathematical model when the stimulation evoking the system dynamics has a low frequency, that is, 5 Hz or less, and a short duration, that is, 5 s or less.

We investigate the limits of linearizability in rat cerebellar cortex. We present evidence that this system can be approximated with a linear dynamic mathematical model when the stimulation frequency is kept below 7 Hz and the stimulation duration is fixed at 15 s (an intermediate duration). Specifically, when the stimulation duration is fixed to 15 s, a single model is able to cover the system dynamics resulting from multiple frequencies in the range 0.5–7. Hz.

We furthermore investigate the coupling during application of stimulation with frequencies in the range 2-30 Hz and durations of 60 s and 600 s (long durations). Based on the characteristics of the deviations of the experimental data from a linear dynamic model, we hypothesize that the neuro-vascular system in rat cerebellar cortex can be linearized for durations up to 20 s and frequencies up to 7 Hz.

CNTF-ACTIVATED ASTROCYTES DISPLAY METABOLIC PLASTICITY AND INCREASED RESISTANCE TO METABOLIC INSULTS

Carole Escartin^1,4, Karin Pierre⁴, Emmanuel Brouillet¹, Thierry Delzescaux², Martine Guillermier¹, Marc Dhenain¹, Nicole Deglon^1,2,3, Philippe Hantraye^1,2,3, Luc Pellerin⁴, Gilles Bonvento¹

¹CEA, DSV, DRM, CNRS URA 2210, Service Hospitalier Frederic Joliot, Orsay, France, ²CEA, DSV, DRM, U2IBP, Service Hospitalier Frederic Joliot, Orsay, France, ³MIRCen, CEA, Fontenay-aux-Roses, France, ⁴Departement De Physiologie, Universite De Lausanne, Lausanne, Switzerland

Introduction: High energy demands of neurons make them vulnerable to adverse effects of energy impairment. Astrocytes have been shown to regulate the flux of energy substrates to neurons. In pathological situations, astrocytes are activated but the consequences on brain energy metabolism are still poorly characterized.

Materials and Methods: To characterize the metabolic profile of reactive astrocytes, we used our two previously described models of astrocyte activation through either lentiviral-gene transfer of the cytokine ciliary neurotrophic factor (CNTF, a cytokine known to activate astrocytes) in the rat striatum in vivo or exposure of primary neurone astrocyte mixed cultures to recombinant CNTF (1). We assessed 2-[18F]-deoxy-D-glucose (FDG) uptake in vivo using autoradiography and micro-PET, the level of expression of different proteins by immunoblots and fluorescent immunohistochemistry, the rate of oxidation of both [14C]-3-hydroxybutyrate (BHB) and [14C]-palmitate in vitro and finally the effects of toxic levels of extracellular fatty acids (0.2mM palmitate for 24h) and glycolysis (0.4 mM iodoacetate for 10 min) inhibition upon neuronal survival in vitro.

Results: We found that CNTF-induced astrocyte activation in the striatum was associated with a marked rearrangement of metabolic features in absence of any pathological process. Glycolysis was stably reduced, AMP-activated protein kinase (AMPK, a central regulator of energy metabolism) was activated both in vivo and in vitro and key proteins of ketolysis (monocarboxylate transporter 1 MCT1, succinyl-CoA:3-oxoacid CoA-transferase, SCOT and beta-hydroxybutyrate dehydrogenase, BDH) and β?oxidation (carnityl-palmityl transferase, CPT1, the rate-limiting enzyme for β-oxidation of fatty acids) were overexpressed by CNTF-activated astrocytes in vivo. In vitro, the oxidative rate of both fatty acids and ketone bodies was significantly enhanced by 100 and 30%, respectively. This metabolic plasticity of CNTF-activated astrocytes significantly improved their ability to resist to glycolysis inhibition, ultimately providing protection to neurons. Such an effect was partially mediated by the transport of monocarboxylates. We finally found that CNTF-activated astrocytes, which perform β-oxidation two times more efficiently than controls, were totally resistant to toxic palmitate levels in vitro.

Conclusion: These results show that CNTF-activated astrocytes, with their increased capacity to use alternative energetic substrates besides glucose, have a strong protective potential to face severe metabolic insults and confirm their crucial role as ‘metabolic nurses' for neurons.

NEUROPROTECTIVE EFFECT OF XG-102 ADMINISTERED ALONE OR IN COMBINATION WITH TISSUE PLASMINOGEN ACTIVATOR

Karine Wiegler¹, Christophe Bonny², Didier Coquoz³, Lorenz Hirt¹

¹Department of Neurology, CHUV, Lausanne, Switzerland, ²Department of Medical Genetics, CHUV, Lausanne, Switzerland, ³Xigen Pharmaceuticals, Lausanne, Switzerland

A NOVEL INHIBITOR OF ADVANCED GLYCATION AND ENDOPLASMIC RETICULUM STRESS REDUCES INFARCT VOLUME IN RAT TRANSIENT FOCAL ISCHEMIA

Shunya Takizawa¹, Yuko Izuhara², Yuko Morita¹, Tsuyoshi Uesugi¹, Shigeharu Takagi¹, Toshio Miyata²

¹Division of Neurology, Department of Internal Medicine, Tokai University School of Medicine, Isehara, Kanagawa, Japan, ²Institute of Medical Sciences 2, Tokai University School of Medicine, Isehara, Kanagawa, Japan

Background and aims: Angiotensin II receptor blockers inhibit formation of advanced glycation end products (AGEs) and reduce oxidative stress, and should therefore ameliorate ischemic cerebral damage, but their hypotensive effect limits their clinical use. We have discovered a novel non-toxic inhibitor of advanced glycation and oxidative stress, TM2002, by means of high-throughput screening of the anti-oxidative and AGEs-lowering activities of over 1300 compounds. We here demonstrate its in vivo effectiveness in both transient and permanent focal ischemia models in rats.

Methods: TM2002 (5.58 mg/kg twice a day) was administered intravenously in a rat transient focal ischemia model, and we evaluated its effect on infarct volume and neurological function. Brain tissues were examined immunohistochemically for heme oxygenase-1, advanced glycation end products (AGEs), oxygen regulated protein (ORP) 150, heat shock protein (HSP) 70i, HSP60, and TUNEL. Further, in a photothrombotic occlusion model, we evaluated the dose-dependence of the effect of TM2002 on infarct volume and neurological functions, conducted time-window studies of its neuroprotective efficacy, and compared its effect on infarct volume with that of the known neuroprotective agent NXY-059.

Results: There was no statistically significant difference in any physiological parameter, including mean arterial blood pressure, between the TM2002- and vehicle-treated groups throughout the experiment. TM2002 significantly reduced infarct volume compared with the vehicle (79.5 ± 52.8 vs. 183.3 ± 68.7 mm3, p < 0.01), and reduced the numbers of cells immunohistochemically positive for heme oxygenase-1, AGEs, ORP150, and TUNEL, especially in the penumbra, as compared with the vehicle-treated group. Further, in the rat photothrombotic occlusion model, TM2002 (2.79, 5.58, and 11.16 mg/kg twice a day) dose-dependently reduced the infarct volume (to 242.1±102.3, 201.3±47.7, and 171.3±48.0 mm3, respectively), and improved the neurological deficits. In this model, TM2002 reduced infarct volume even when administered 1 hour after occlusion, and the efficacy of TM2002 was equivalent to that of NXY-059.

Conclusions: Our novel synthetic compound significantly ameliorated ischemic cerebral damage through reduction of ER stress, advanced glycation, and oxidative stress, independently of blood-pressure lowering. Since the efficacy of TM2002 was equivalent to that of NXY-059 in the photothrombotic occlusion model, a clinical trial of TM2002 in stroke patients might be warranted.

ISCHEMIC TOLERANCE INDUCED BY REPEATED HYPERBARIC OXYGEN IN RAT BRAIN: TIME WINDOW AND GENOME-WIDE GENE AND PROTEIN EXPRESSION

Takao Hirata¹, Ying Jun Cui¹, Takeshi Funakoshi², Yoichi Mizukami², Yu-ichiro Ishikawa³, Futoshi Shibasaki⁴, Mishiya Matsumoto¹, Takefumi Sakabe¹

¹Department of Anesthesiology and Resuscitology, Yamaguchi University School of Medicine, Yamaguchi, Japan, ²Center for Gene Research, Yamaguchi University, Yamaguchi, Japan, ³Center for Microarray Analysis, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan, ⁴Department of Molecular Cell Physiology, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan

THE PROTECTIVE EFFECT OF RAPID ISCHEMIC PRECONDITIONING ON THE PI3K/AKT PATHWAY IN FOCAL ISCHEMIA IN RAT

Hanfeng Zhang¹, Xuwen Gao¹, Gary Steinberg^1,2, Heng Zhao^1,2

¹Department of Neurosurgery, Stanford University, Stanford, CA, USA, ²Stanford Stroke Center, Stanford University, Stanford, CA, USA

Background and Aims: Two protective time windows for preconditioning have been identified: rapid preconditioning, performed 30 min to one hour before onset of a subsequent severe ischemia, and delayed ischemic tolerance, performed one to several days before a severe ischemia. Only a few reports are available for rapid ischemic tolerance, and the protective mechanisms of rapid tolerance are largely unexplored. Here we investigate whether rapid preconditioning reduces infarction in a focal ischemic model generated by permanent distal MCA occlusion plus transient bilateral CCA occlusion and study its potential protective mechanisms. It is well documented that the Akt/PKB survival pathway is involved in the protective effect of delayed preconditioning in global ischemia. However, few studies have addressed its role in focal ischemia and none have studied its role in rapid preconditioning. Akt activity is regulated by phosphorylation via several molecules upstream of the Akt kinase. Akt is activated via phosphoryation by PDK1 (either directly or indirectly). Conversely, PTEN dephosphorylates and inactivates Akt. Upon activation, Akt phosphorylates GSK3β and downregulates its activity. Phosphorylation of β-catenin by GSK3β leads to degradation of β-catenin and apoptosis. Here we report that rapid preconditioning reduces infarction in a distal permanent MCA occlusion model and this protection correlates with activation of the Akt pathway.

Methods: Focal ischemia was generated by permanent distal MCA occlusion plus 30 min of transient bilateral CCA occlusion in male SD rats. Rapid ischemic postconditioning was performed by occluding the left distal MCA for 15 min. Test ischemia was induced 1h later. For Western blot, ischemic brains were harvested 1, 5 and 24 h after stroke. Whole cell homogenates from the ischemic core and penumbra were prepared for Western blots. Protein levels of phosphorylated-Akt (Ser473) (P-Akt), P-PDK1, P-PTEN, activated β-catenin and pan-β-catenin were measured. In vitro Akt kinase assays were also performed to evaluate Akt activity.

Results: Rapid ischemic preconditioning significantly reduced infarct size as measured 2 days post-stroke (P<0.05). P-Akt transiently increased from 1 to 5h in the ischemic penumbra (P<0.05, n=5-7, Two way NOVA) but not in the ischemic core. Preconditioning did not further enhance protein level of P-Akt at 1 and 5h, but attenuated its decreases at 24h in both the penumbra and core. Akt kinase assays showed that preconditioning attenuated decreases in Akt activity at 5 and 24h in the penumbra (P<0.05). Levels of P-PDK1 decreased from 5h after stroke, but there was no difference between animals treated with and without preconditioning. The levels of P-PTEN (inactive PTEN) decreased after stroke, and preconditioning attenuated its decreases at 24h in both the penumbra and core. Cleaved β-catenin appeared as early as 1h and the levels of cleavage products further increased at 24h in both the penumbra and core. Preconditioning attenuated β-catenin cleavage at 24h in both the penumbra and core.

Conclusion: Rapid preconditioning reduces infarction in a distal permanent MCA occlusion model, and this correlates with activation of the Akt survival pathway.

THE BK CHANNEL OPENER NS1619 INDUCES NEURONAL PRECONDITIONING IN VITRO

Tamas Gaspar¹, Ferenc Domoki^1,2, Keita Mayanagi¹, James A. Snipes¹, Ferenc Bari², David W. Busija¹

¹Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston Salem, NC, USA, ²Department of Physiology, Faculty of Medicine, University of Szeged, Szeged, Hungary

Background and aims. Targeting the large conductance Ca activated potassium (BK) channels is a novel approach to induce preconditioning (PC). NS1619 (NS) is a selective opener of the BK channel and was shown to protect the heart against an otherwise lethal ischemic insult. Recently, the BK channel has been reported to be present in the inner mitochondrial membrane of rat neurons, but its role in tolerance induction in the central nervous system has not yet been explored. The aim of our in vitro study was to examine whether NS can induce PC and protect rat cortical neuronal cultures against toxic stimuli. We also investigated its effect on reactive oxygen species (ROS) production, plasma membrane potential, and expression and activity of cytoprotective proteins.

Methods. Cortical neurons of 18-day old Sprague-Dawley rat fetuses were cultured in B27 supplemented Neurobasal medium. One week after plating, a 6-hour daily treatment with different doses of NS (10 ? 100 µM) for 3 days was introduced. Eighteen hours after the last treatment, neuronal cultures were exposed to 180 min of oxygen and glucose deprivation (OGD), 6 hour of exogenous hydrogen peroxide toxicity (80 µM), or 60 min of glutamate excitotoxicity (200 µM). Cell viability was evaluated with CellTiter 96 AQueous One Solution assay. ROS generation, plasma and mitochondrial membrane potentials were monitored using hydroethidine, di-8-ANEPPS, and tetramethylrhodamine ethyl ester, respectively. The expression of manganese dependent superoxide dismutase (MnSOD), cupper and zinc dependent superoxide dismutase (CuZnSOD), glutathione peroxidase (GPx), and catalase was examined using Western blotting. Enzyme activity of antioxidants was assessed with a microplate reader using commercially available kits.

Results. Application of NS increased ROS generation, depolarized mitochondria, and hyperpolarized the cell membrane of neurons. Three-day treatment with NS dose-dependently protected the cell cultures against OGD, H2O2, and glutamate, although higher concentrations of the compound were needed to elicit PC than previously reported in the heart. The highest tolerance was observed against H2O2 where a dose of 40 µM provided complete protection (viability: control, 100.0±1.97%; untreated, 50.1±2.81%*; NS 40 µM, 93.2±3.25%#; *p<0.05 vs. control; #p<0.05 vs. untreated), whereas even higher doses were required against OGD (viability: control, 100.0±1.61%; untreated, 45.9±0.58%*; NS 100 µM, 88.4±0.56%*#; *p<0.05 vs. control; #p<0.05 vs. untreated), and the protection was found to be incomplete against glutamate (viability: control, 100.0±1.21%; untreated, 48.2±0.42%*; NS 100 µM, 74.4±1.06%*#; *p<0.05 vs. control; #p<0.05 vs. untreated). Co-treatment with the BK channel antagonist Paxilline did not block ROS generation and did not antagonize the neuroprotective effect of NS. Despite the protection afforded by 40 µM NS against H2O2, this concentration of the compound did not change the activity and/or expression of ROS scavengers.

Conclusion. Our results demonstrate that NS1619 is a potent neuroprotective agent in vitro. However, the need of relatively high doses of NS1619, and the effects on the cell membrane potential indicate the involvement of non-mitochondrial mechanisms in the induction of neuroprotection.

NEUROPROTECTIVE EFFECTS OF GLNVA ON INFLAMMATION-INDUCED DOPAMINERGIC NEURONAL CELLS DEATH

Yi-Ching Lo, Yi-Chin Lin

Department and Graduate Institute of Pharmacology, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan-R.O.C.

Background and aims: Glyceryl nonivamide (GLNVA) has been reported to prevent subarachnoid hemorrhage-induced cerebral vasospasm. In this study, the anti-inflammatory and neuroprotective effects of GLNVA were investigated in mouse microglial BV-2 cells and human dopaminergic neuron SH-SY5Y cells.

Methods: we characterize the anti-inflammatory and neuroprotective effects of GLNVA on lipopolysaccharide (LPS)-induced neuronal cytotoxicity by measuring the ratio of dead cells, production of inflammatory cytokines? anti-inflammatory cytokine and the inflammatory protein expression.

Results: The results showed that pretreatment of BV-2 with GLNVA diminished LPS-induced nitric oxide production, overexpression of iNOS and gp91phox and intracellular ROS. GLNVA also reduced COX-2 expression, IκBα / IκBβ degradation and the overproduction of TNF-α, IL-1β and PGE2. However, GLNVA augmented anti-inflammatory cytokine IL-10 production on LPS-stimulated BV-2 cells. In co-culture conditions, we employed LPS stimulated BV-2 as a reactive microglial model and GLNVA decreased the reactive microglial neurotoxicity-induced cell death and overexpression of nNOS and gp91phox on SH-SY5Y cell.

Conclusions: GLNVA possesses neuroprotective effect by inhibiting oxidative stress and inflammation induced by LPS.

CYCLOSPORIN-A PREVENT MITOCHONDRIAL PERMEABILITY TRANSITION SIMILARLY IN BOTH BRAIN CORTEX AND SPINAL CORD

Saori Morota^1,2, Magnus Hansson J.³, Yoshihisa Kudo¹, Eskil Elmer³, Nagao Ishii², Hiroyuki Uchino²

¹Laboratory of Cellular Neurobiology, Tokyo University of Pharmacy and Life Science, Hachioji, Japan, ²Department of Anesthesiology, Tokyo Medical University Hachioji Medical Center, Tatemachi, Department of Anesthesiology, Tokyo Medical University Hachioji Medical Center, Hachioji, Japan, ³Laboratory for Experimental Brain Research, Department of Clinical Sciences, Lund University, Lund, Sweden

The mitochondrial permeability transition (mPT) is defined as sudden increase in permeability of the inner mitochondrial membrane and is primarily caused by calcium overload. mPT has been increasingly implicated as a potential pathogenetic mechanism in neurodegeneration. Varying sensitivity to mPT has been demonstrated for regions within CNS as well as between tissues. This variance has been thought that possibly contributes to different efficacy of mPT inhibitor Cyclosporin-A (CsA) on different region e.g. brain and spinal cord, and selective vulnerability. However, comparison of mitochondrial function and mPT sensitivity is possibly hampered by confounding influence due to different tissue handling. The objectives of present study were to stringently compare mitochondria from the brain cortex and spinal cord with respect to mPT sensitivity, calcium retention capacity (CRC) and respiratory function, and further, to re-evaluate the efficacy of CsA on mitochondria isolated from brain cortex and spinal cord. In present study, we induced whole-body hypothermia in animals before tissue extraction to counteract differential tissue degeneration during extraction. Overall, the features of mPT and cyclosporin-A sensitivity on administration of calcium bolus loads were similar. CsA also demonstrated equal relative improvement of CRC in these mitochondria although spinal cord mitochondria were able to retain less calcium than brain cortex mitochondria. Further we evaluated whole-body hypothermia of animal as a technique to minimize possible confounding influence by differential tissue handling and found it improve active phosphorylating rates of respiration. The present findings imply that the general features of mPT and the effect of CsA as mPT inhibitor are similar in brain cortex and spinal cord mitochondria.

A RCT OF MILD HYPOTHERMIA THERAPY IN SEVERE TRAUMATIC BRAIN INJURY IN JAPAN - A PRELIMINARY REPORT

Tsuyoshi Maekawa, Nariyuki Hayashi, Keiki Ogino, Jun Takezawa, Seigo Nagao, Yasuo Ohashi, Susumu Yamashita, Kotaro Kaneda, Ryosuke Tsuruta, Shunji Kasaoka

Hypothermia for Head Injury Study Group in Japan

NEUROPROTECTIVE EFFECTS OF INCREASED FOREBRAIN MINERALOCORTICOID RECEPTOR EXPRESSION

Malcolm Macleod¹, Karen Horsburgh², Sung-Eun Bae³, Roderick Carter³, Dirk Stenvers³, Jill Fowler², Joyce Yau³, Celso Gomez-Sanchez⁴, Megan Holmes³, Christopher Kenyon³, Jonathan

Seckl³, Maggie Lai³

¹Department of Clinical Neurosciences, University of Edinburgh, Edinburgh, Scotland, UK, ²Centre for Neuroscience Research, University of Edinburgh, Edinburgh, Scotland, UK, ³Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, Scotland, UK, ⁴Division of Endocrinology, University of Mississippi Medical Center, Jackson, MS, USA

Background and Aims: Glucocorticoid control of neuronal function requires actions mediated through both the mineralocorticoid (MR) and glucocorticoid receptor (GR). Chronic overactivation of GR is detrimental, targeting some neurons for apoptosis, disturbing neuroendocrine control and impairing cognition and behaviour. In contrast, the effects of increased MR signalling have yet to be explored in detail, perhaps because MR is largely occupied at basal steroid levels so modulation of signalling is more dependent on changes in receptor density than on ligand concentration. What data there are suggest a possible role in the promotion of neuronal survival and improved learning and memory. We have previously shown that neuronal MR induction occurs early in response to cellular injury in vivo and in vitro and is associated with enhanced survival. However, the causality of this association and any relevance for human health has yet to be established.

Methods: We therefore (i) generated a forebrain specific MR overexpressing transgenic mouse (MR-Tg) under the control of the Cam KII alpha promoter and measured the extent of neuronal death following transient global cerebral ischaemia; (ii) measured the expression of MR and GR in human hippocampal sections following initially non-fatal cardiac arrest. Given the localization of the transgene, MR-Tg mice also served as a useful model in assessing the role of MR in hypothalamic-pituitary-adrenal (HPA) axis regulation, in cognition and behaviour.

Results: Neuronal death measured 3 days after 20 minutes of bilateral common carotid artery occlusion was reduced by 33.4 ± 4.2% in the caudate nucleus, CA-1 and CA-2 regions of the hippocampus compared to wildtype littermate controls (F(1,68)=10.4, p=0.002). MR-Tg mice also demonstrated improved memory retention in the Morris water maze (p<0.02), spent significantly more time in the anxiogenic central zone of the open field test (p<0.05) and displayed decreased latency to enter (p<0.05) and spent more time in the light arena (p<0.05) in the dark-light exploration test compared to wildtype controls. Basal plasma corticosterone levels, GR expression, the glucocorticoid response to restraint stress, basal blood pressure and general locomoter activity was not affected by genotype. Patients dying 4 days to 8 weeks after initially non-fatal cardiac arrest had increased hippocampal expression of MR but not GR, suggesting modulation of MR expression observed in experiments in vitro and in vivo is also seen in humans.

Conclusions: Increased neuronal MR attenuates neuronal loss in cerebral ischaemia and has beneficial effects on memory and mood ?which are often compromised in humans post-stroke. These effects are not due to an effect of circulating glucocorticoid levels, which are unchanged, and identify increased MR expression as a therapeutic target not only in cerebral ischaemia but also in disorders of mood.

PROTECTIVE EFFECTS OF DIHYDROGINSENOSIDE RB1 ON ISCHEMIC BRAIN DAMAGE AND COMPRESSIVE SPINAL CORD INJURY THROUGH UPREGULATION OF VEGF AND BCL-XL

Ryuji Hata, Pengxiang Zhu, Fang Cao, Keiichi Samukawa, Masahiro Sakanaka

Department of Functional Histology, Ehime University Graduate School of Medicine, Toon, Ehime, Japan

NO FURTHER DECREASE OF 5-HT2A RECEPTORS IN PATIENTS WITH MILD COGNITIVE IMPAIRMENT: A TWO-YEAR FOLLOW-UP STUDY

Lisbeth Marner¹, Gitte Moos Knudsen¹, Karine Madsen¹, Steven Haugbol¹, Soren Holm³, Willian Baare⁴, Steen G. Hasselbalch^1,2

¹Neurobiology Research Unit, The Neuroscience Center, Rigshospitalet, Copenhagen, Denmark, ²Memory Disorders Research Unit, The Neuroscience Center, Rigshospitalet, Copenhagen, Denmark, ³Department of Clinical Physiology and Nuclear Medicine, Rigshospitalet, Copenhagen, Denmark, ⁴Danish Research Center for Magnetic Resonance, Hvidovre Hospital, Copenhagen, Denmark

Background and aims. Previous postmortem studies have revealed a profound reduction in the serotonin 2A (5-HT2A) receptor binding in patients with Alzheimer's disease (AD). In vivo studies of patients with AD with assessment of the 5-HT2A receptors using [18F]-Altanserin positron emission tomography (PET) have confirmed this finding. In patients with mild cognitive impairment of the amnestic type (MCI), we have previously demonstrated a 20–30% reduction in cortical 5-HT2A-receptor binding as compared to healthy subjects1. Here we present a two-year follow-up of both patients and healthy subjects.

Methods. At baseline (n= 16 patients, 17 healthy subjects) and at two-year follow-up (n= 14 patients, 12 healthy subjects), all individuals were investigated with magnetic resonance imaging (MRI) and [18F]-Altanserin PET using a bolus-infusion approach. Volumes of interest (VOI's) were delineated automatically on coregistered MRI images, and partial volume (PV) corrected and non-partial volume (nonPV) corrected VOI values were extracted from the PET images. The binding potential of specific binding (BP1) was calculated using cerebellum as a reference region devoid of 5-HT2A-receptors (CREF) and metabolite corrected plasma values (CPLASMA) according to BP1=(CVOI-CREF)/CPLASMA where CVOI is the radioactivity concentration in the VOI. As the temporal and parietal cortical regions often show hypometabolism in early AD and at the same time have a high 5-HT2A-receptor density, they were chosen for the present analysis.

Results. In the two-year follow-up period, 9 out of 14 patients with MCI had progressed to fulfill AD criteria. In both MCI patients and healthy subjects, no significant change in binding was found as compared to the investigation carried out two years earlier. In MCI patients, the average PV-corrected BP1 was 2.16 at baseline and 2.09 at two-year follow-up (95% C.I. for difference: [-0.48 0.19]) and in healthy subjects BP1 was 2.62 and 2.54 (95% C.I. for difference: [-0.35 0.20]). The average nonPV-corrected BP1 MCI patients was 0.70 at baseline and 0.62 at two-year follow-up (95% C.I. for difference: [-0.21 0.05]) and in healthy subjects 0.97 and 0.89 (95% C.I. for difference: [-0.18 0.04]). The patients that converted to AD during the follow-up period did not differ significantly from the patients that did not convert.

Conclusions. We conclude that the reduced levels of 5-HT2A-receptor binding in MCI patients are relatively stable, even in patients that convert to AD. Our finding suggests that the lower cortical 5-HT2A-receptor binding in MCI patients is due to brain changes that take place in the very early stages of AD before symptoms are present or perhaps even constitute a permanent trait marker for the risk of developing AD.

MALE-SPECIFIC ISOFLURANE PRECONDITIONING NEUROPROTECTION IN EXPERIMENTAL FOCAL STROKE IS TESTOSTERONE DOSE-SPECIFIC AND ANDROGEN RECEPTOR-DEPENDENT IN MICE

Lan Wang, P.D. Hurn, S.J. Murphy

Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA

Background and Aims: We have shown that neuroprotection due to isoflurane preconditioning (IsoPC) 24 hours before experimental stroke is male-specific, with the neuroprotective benefits of IsoPC being lost in females and estradiol-treated animals. We evaluated the testosterone (T) dose relationship and the role of the androgen receptor (AR) in the ischemic sensitivity of IsoPC male mouse brain.

Methods: Selected groups of male and castrated (CAST) mice were treated with 21-day release subcutaneous implants containing T (1.5 or 5 mg) or flutamide (FLUT, 5 mg), an AR antagonist, 1 week before preconditioning. Males ± FLUT, CAST ± FLUT, CAST + 1.5T ± FLUT, and CAST + 5T ± FLUT were preconditioned for 4 hours with air (sham preconditioning, Sham PC) or 1.0% IsoPC. Mice then underwent 2 hours of middle cerebral artery occlusion 24 hours after preconditioning. Laser-Doppler flowmetry (LDF) was used to monitor cortical perfusion. Brains were collected at 22 hours reperfusion. Cortical and striatal infarct volumes (% contralateral structure) were determined by digital image analysis of 2 mm thick coronal brain slices stained with 2,3,5-triphenyltetrazolium chloride.

Results: Relative LDF changes were equivalent among groups. In contrast to corresponding Sham PC groups, IsoPC decreased cortical and striatal infarct volumes in males and CAST + 1.5T, and striatal infarct volumes in CAST + 5T (Table). IsoPC had no effect on cortical and striatal infarct volumes in males + FLUT, CAST ± FLUT and T-treated CAST + FLUT mice, and on cortical infarct volume in CAST + 5T as compared to corresponding Sham PC groups (Table).

Conclusions: IsoPC neuroprotection in experimental stroke is testosterone-dependent and dose-specific, with IsoPC neuroprotection being lost in androgen-deficient CAST, males and CAST + 1.5T being completely protected, and CAST + 5T only showing reduced striatal injury. Testosterone's effects in IsoPC ischemic male brain are likely AR-mediated as FLUT blocked IsoPC neuroprotection in male and T-treated CAST ischemic brain. Further studies are needed to explore the role of other androgens like dihydrotestosterone and to confirm an AR-dependent mechanism in male-specific IsoPC neuroprotection.

NICOTINAMIDE REDUCES THE INTRACEREBRAL CELLULAR INFLAMMATORY RESPONSE FOLLOWING TRANSIENT FOCAL CEREBRAL ISCHEMIA IN RATS

Tsung-Ying Chen^1,2, Yu-Hsiang Kuan¹, Hung-Yi Chen^1,3, Ming-Yang Lee¹, E-Jian Lee¹

¹Neurophysiology Laboratory, Neurosurgical Service, Department of Surgery, National Cheng Kung University Medical Center and Medical School, Tainan, Taiwan-R.O.C., ²Department of Anesthesiology, Buddhist Tzu-Chi University and Buddhist Tzu Chi General Hospital, Hualien, Taiwan-R.O.C., ³Institute of Pharmacy, China Medical University, Taichung, Taiwan-R.O.C.

We have previously shown that nicotinamide protects against energy depletion and decreases infarction in a rat model of cerebral ischemia [1], [2]. It is known that postischemic cell-mediated neuroinflammation exaggerates the initial brain insult caused by ischemia. Herein, we explored whether nicotinamide mitigates the cellular inflammatory response after transient focal cerebral ischemia for 90 min in rats. Nicotinamide (NICO, 500 mg/kg) or vehicle was given intravenously at reperfusion onset. Immunohistochemistry and flow cytometric analysis were used to evaluate the cellular inflammatory response at 48 hr after reperfusion. Relative to controls, NICO-treated animals resulted in significantly decreased macrophages (4.57(x105)±3.00 vs. 2.83(x105)±1.42) and T lymphocytes (0.74(x105)±0.57 vs. 0.28(x105)±1.42) in the bloodstream (P>0.05). Treatment with NICO also significantly decreased the cellular inflammatory response in the ischemic hemisphere. Specifically, melatonin effectively decreased the extent of neutrophil emigration (Ly6G-positive/CD45-positive) and macrophage/activated microglial infiltration (CD11b-positive/CD45-positive) by 55% (P<0.01) and 66% (P<0.01), respectively. In addition, NICO significantly decreased the population composition of T lymphocyte (CD3-positive/CD45-positive; P<0.001) by 39%. This NICO-mediated decrease in the cellular inflammatory response was accompanied by both reduced brain infarction by 34% (P<0.05) and improved surviving neurons in penumbral cortex and caudoputamen by 112% (P<0.005) and 250% (P<0.001), respectively. Thus, intravenous administration of NICO upon reperfusion effectively decreased the emigration of circulatory neutrophils, T-lymphocytes and macrophages/monocytes into the injured brain and inhibited focal microglial activation following cerebral ischemia-reperfusion. The finding demonstrates nicotinamide's inhibitory ability against the cellular inflammatory response after cerebral ischemia-reperfusion, and further supports its pleuripotent neuroprotective actions suited as a therapy of candidate for further clinical trials in the context of ischemic stroke.

TETRACYCLINE DERIVATIVES INHIBIT CALPAINS AND PROTECT MOUSE NEURONS AGAINST CEREBRAL ISCHEMIA

Susan Jiang¹, Jittiwud Lertvorachon², Sheng Tao Hou¹, Yasuo Konishi², Jacqueline Webster¹, Geoff Mealing¹, Eric Brunette¹, Joseph Tauskela¹, Edward Preston¹

¹NRC Institute for Biological Sciences, National Research Council Canada, Ottawa, ON, Canada, ²NRC Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada

LYSINE 134 OF BASIC FGF IS A PREREQUISITE FOR ITS NEUROPROTECTIVE ACTIVITY

Susanne Klumpp, Karsten Rose, Dorothee Kriha, Stefanie Pallast, Vera Junker, Josef Krieglstein

Institut Fuer Pharmazeutische Und Medizinische Chemie, Westfaelische Wilhelms Universitaet Muenster, Muenster, Germany

Introduction: Basic fibroblast growth factor (bFGF) is involved in various cellular processes, e.g. cell proliferation, angiogenesis and neuroprotection. In previous work we could provide evidence that binding of ATP to or phosphorylation of bFGF is essential for its biological activity [1]. Therefore, site-directed mutagenesis was used to identify the amino acids of bFGF responsible for ATP binding or phosphorylation. By doing so we discovered that heparin binding and, correspondingly, the neuroprotective potency of bFGF was abolished in one of the mutants generated.

Methods: Escherichia coli TG1 cells carrying the plasmid pCytexp1-bFGF were used for overexpression of bFGF. Site-directed mutagenesis was performed according to the manual of the QuikChange mutagenesis kit (Stratagene). The neuroprotective effect of bFGF and mutants was tested in primary cultures of embryonic (E19) hippocampal neurons damaged by staurosporine (200 nM) and in a mouse model of focal cerebral ischemia.

Results: Replacing lysine 134 of bFGF by alanine resulted in a growth factor protein that could be labeled with [γ−32P]ATP to the same extent as the wildtype (wt) protein. However, this mutant bFGF(K134A) did not protect the cultured hippocampal neurons against damage whereas wt-bFGF and various other mutants did (Fig. 1). Furthermore, phosphorylation of Akt and ERK-1,2 was significantly diminished in cultured neurons treated with bFGF(K134A) indicating decreased cellular signalling compared to neurons treated with wt-bFGF. Notably, only wt-bFGF, but not bFGF(K134A), was capable of reducing the infarct area after occlusion of the middle cerebral artery of mice. A heterotrimeric complex consisting of one heparin and two bFGF molecules is known to interact with the bFGF receptor. Lysine 134 of bFGF is located within the heparin binding site. Indeed, binding of bFGF(K134A) to heparin acrylic beads was markedly reduced in comparison to that of wt-bFGF.

Conclusion: Our results demonstrate that lysine 134 plays a pivotal role for the neuroprotective activity of bFGF. Since labeling with [γ−32P]ATP is unaffected by this mutation but heparin binding is diminished we suggest that the single amino acid substitution K134A is detrimental to neuroprotection of bFGF because of the reduced heparin binding affinity.

Fig. 1: Replacement of lysine 134 in bFGF abolished the neuroprotective effect. Scheffe test: *p<0.001.

THROMBIN, FACTOR XA AND ACTIVATED PROTEIN C CAN PROTECT THE RAT CORTICAL AND HIPPOCAMPAL NEURONS FROM GLUTAMATE-INDUCED DEATH

Lubov Gorbacheva¹, Tatyana Storozhevykh², Vsevolod Pinelis³, Svetlana Strukova⁴

¹Lomonosov Moscow State University, Department of Human and Animal Physiology, Moscow, Russia, ²The Scientific Centre for Children Health, Russian Acad. Med. Sci., Moscow, Russia, ³The Scientific Centre for Children Health, Russian Acad. Med. Sci., Moscow, Russia, ⁴Lomonosov Moscow State University, Department of Human and Animal Physiology, Moscow, Russia

Background and aims. Glutamate-induced damage of calcium homeostasis in neurons is one of the mechanisms of brain injury (hemorrhagic insults, ischemia, neurodegenerative diseases and etc.) (Hossain MA., 2005). It is known, that serine proteinases can participate in regulatory process in CNS (Xi, G. et al, 2003; Kawabata, A. et al, 2005; Zlokovic, B. et al, 2005). We investigated the role of thrombin (Th), factor Xa (FXa) and activated protein C (APC) in cell survival and calcium homeostasis in cultured rat cortical and hippocampal neurons.

Methods. Studies were performed using 7-10-day primary culture of neurons from cortex or hippocampus obtained from brain of 1-3-day-old Wistar rats. 24 h after the cell exposure (30 min) to 100µM glutamate (Glu) or serine proteases and glutamate neuron death in culture was estimated. Cell death was determined using two approaches, biochemical (MTT-test and the measurement of lactate dehydrogenase activity in the medium) and morphological (Hoechst 33342, Ethidium bromide and Syto-13). Using fluorescent probes Fura-2/AM or Fura-2FF/AM intracellular calcium concentration ([Ca2+]i) was measured. Student,s t-test was used for statictical treatment.

Results. We observed that exposure neurons to Glu led to the increase of apoptotic cells (to 33%) at 24 h. The pretreatmant of culture with Th (0,1-10 nM), FXa (1-10 nM) and APC (Sigma) (0,05-10 nM) significantly (on average in two times) reduced the Glu-induced neuronal death. But the most effective protective concentrations of APC were different for cortical and hippocampal neurons (0,1nM and 0,05 nM, accordingly). High concentrations of serine proteases (100nM Th and 100nM APC) have not neuroprotective effect at Glu-induced neuron death and caused cell death in 32% of cells. Inactivated with PMSF Th, FXa and APC did not protect neurons from Glu-induced toxicity.

As known, the apoptotic effect of glutamate is based on the overloading of cells by calcium and disruption in the regulation of calcium homeostasis. In our research, Th- and FXa- (but not APC) induced decrease of the cell number in which the basal level of intracellular calcium was recovered after removal of glutamate, was revealed.

At the same time only Th elicited the rapid transient increase in intracellular calcium in hippocampal and cortical neurons. Using the blockers of IP3 and ryanodine receptors, 2-ABP and dantrolene, was shown the predominant role of the IP3 receptors in the calcium response of neurons to Th.

Conclusions. The present findings suggest that Th, FXa and APC at low concentrations (< 10 nM) has the protective effect on cortical and hippocampal neurons survival at Glu-induced toxicity. Hippocampal neurons sensitivity to APC was higher than one to cortical neurons. However, most likely these protective effects have not identical mechanisms.

FUNCTIONAL VARIATION IN THE SEROTONIN TRANSPORTER GENE DECREASES IN VIVO BRAIN SEROTONIN 2A RECEPTOR BINDING

Steven Haugbol¹, Lars H. Pinborg¹, Vibe G. Frokjaer¹, David Erritzoe¹, Lisbeth Marner¹, Steen G. Hasselbalch¹, Claus Svarer¹, Lis Hasholt², Soren Holm³, Olaf B. Paulson⁴, Gitte M. Knudsen¹

¹Neurobiology Research Unit, Rigshospitalet, Copenhagen University Hospital and Center for Integrated Molecular Brain Imaging, Copenhagen, Denmark, ²Section of Neurogenetics, Department of Medical Biochemistry and Genetic, University of Copenhagen, Copenhagen, Denmark, ³PET and Cyclotron Unit, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark, ⁴Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital, Copenhagen, Denmark

BACKGROUND AND AIMS: There is increasing evidence that the developing brain is under profound serotonergic control. Various polymorphisms of the serotonin transporter (5-HTT) gene are known to be associated with anxiety, depression, and vulnerability to affective disorders. The postsynaptic serotonin 2A receptor (5-HT2A) receptor is widely distributed in the brain with highest density in cortical brain regions. The receptor is down-regulated both in genetically modified mice lacking the 5-HTT and after chronic blockade of the serotonin reuptake site with SSRIs. These observations support that 5-HT2A receptor density is regulated by brain serotonin levels that in turn are adjusted by the 5-HTT. Clinically, the 5-HT2A receptor has been implicated in psychosis, depression and different behavioral traits. The 5-HT2A receptor agonists elicit hallucinations and antagonists are useful both as antipsychotics and as adjuvant treatment in depression. AIMS: Here we investigate the effects of two frequent functional polymorphisms in the 5-HTT gene (SLC6A4) on the 5-HT2A receptor binding. METHODS: 104 healthy volunteers (39 women and 65 men; mean: 41 years; age range: 18 ? 80 years) underwent [18F]altanserin PET scanning for assessment of the 5-HT2A receptor binding in the brain. The genotypes were assessed by polymerase chain reaction, oligonucleotide primers, and electrophoreses. One polymorphism is positioned in the promoter region (5-HTTLPR) and is fixed in a short allele and a long allele. The other is located in the second intron (STin2) and the alleles most frequently contain 10 and 12 tandem repeats. We applied multivariate repeated measures ANOVA (SAS statistical software package) to analyze the effect of the length polymorphisms. RESULTS: Genotype frequencies of 5-HTTLPR and STin2 all conformed to the Hardy-Weinberg equilibrium. In all examined brain regions, 5-HT2A receptor binding was lower in subjects with STin2 10/10 and 10/12 in comparison to subjects carrying 12/12. Multivariate repeated measures ANOVA of the group allelic effect of STin2 on 5-HT2A receptor binding were significant, F = 4.54, P = 0.03. The allelic effect of 5-HTTLPR itself showed no significant effect on 5-HT2A receptor binding, F = 2.56, P = 0.11. However, when adding both class variables (genotypes) to the repeated ANOVA model the effect of the STin2 polymorphism became more significant, F = 6.29, P = 0.01. CONCLUSION: The finding adds further evidence to the generally accepted view that differential expression of the 5-HTT governs serotonergic neurotransmission. Here, we suggest that the combination of two 5-HTT polymorphisms, the 10 and the s allele, leads to a lifelong increase in endogenous 5-HT tone and a subsequent desensitization of 5-HT2A receptors. This finding is consistent with the understanding of serotonin being profoundly involved in neurodevelopment and its possible implications in neuropsychiatric diseases via the 5-HT2A receptor.

SEROTONIN TRANSPORTER (SERT) AVAILABILITY MEASURED WITH [11C]DASB PET IS REDUCED IN OBSESSIVE-COMPULSIVE DISORDER INEDEPENTLY FROM THE SERT GENE PROMOTER VARIANT

Swen Hesse¹, Katarina Stengler-Wenzke², Ralf Regenthal³, Georg-Alexander Becker¹, Marianne Patt¹, Anita Seese¹, Daniel Teupser⁴, Henrik Roth⁵, Henryk Barthel¹, Kai Kendziorra¹, Wolfgang Heinke⁶, Ulrich Hegerl², Osama Sabri¹

¹Department of Nuclear Medicine, University Hospital, University of Leipzig, Leipzig, Germany, ²Department of Psychiatry, University Hospital, University of Leipzig, Leipzig, Germany, ³Institute of Clinical Pharmacology, University of Leipzig, Leipzig, Germany, ⁴Institute of Laboratory Medicine, Clinical Chemistry and Molecular Diagnostics, University of Leipzig, Leipzig, Germany, ⁵Department of Diagnostic and Interventional Radiology, University Hospital, University of Leipzig, Leipzig, Germany, ⁶Department of Anaesthesiology and Intensive Care Therapy, University Hospital, University of Leipzig, Leipzig, Germanty

Introduction: Derived from the clinical effectiveness of selective serotonin reuptake inhibitors (SSRIs) it is currently under debate whether the brain serotonin transporters (SERT) are altered in patients with obsessive-compulsive disorder (OCD). However, first in-vivo imaging studies on SERT employing SPECT/PET did not reveal consistent results (Hesse et al., 2004). The aim of the present study was therefore to measure SERT availability in OCD patients using the SERT-selective PET radiotracer [11C]DASB. Since a polymorphism of the SERT gene promoter might have an impact on the in vivo SERT availability in OCD, we furthermore compared the SERT PET measures of promoter variants (S, short allele, L, long allele).

Methods: Ten drug-naive OCD patients (ICD 10 F42.0–42..2; 2 females, age 39.1 ± 14.3 years; age of onset 24.5 ± 15.5 years; genotype S/L:L/L = 1:1.5) underwent dynamic PET for 2 hours after intravenous bolus injection of 370 MBq [11C]DASB. OCD severity was assessed using the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). Only patients with Y-BOCS >/= 20 points were included into the study. Depressive symptoms were rated using the Beck Depression Inventory (BDI). Parametric images of SERT binding potential (BP) were generated from the PET data by the multi-linear reference tissue model with 2 parameters and the cerebellum as the reference tissue (Ichise et al., 2003) after model validation with arterial input function determination and regional analyses of BP values was performed. The BP images were co-registered with 3D MRI data using the MultiModality software (HERMES Medical Solutions) for delineating the regions of interest (ROI, striatum and sub-regions, thalami, midbrain/brainstem raphe, and cortical areas). For considering probable effects of nicotine on SERT BP, subjects were classified as 0, non-smokers, to 4, heavy smokers.

Results: In OCD patients, the BPs were significantly reduced in the SERT-rich brain region, e.g. the (left) thalamus and the midbrain raphe (2.03 ± 0.27 and 1.88 ± 0.24 versus 2.81 ± 0.26 and 2.70 ± 0.10 in age-matched controls, p = 0.005 and p = 0.001) and in limbic areas (hippocampus) but not in cortical areas. The BP values did neither correlate with YBOCS, smoking status, age nor differ between S/L and L/L carriers. There was an overall trend toward a negative association between BDI and BP values, which became significant in the thalamus (e.g. right: R = -.687, p = 0.041), the caudate (right: R = -.743, p = 0.022), and frontal regions (e.g. the right medial frontal cortex: R = -.712, p = 0.031).

Conclusion: The SERT availability is reduced in OCD patients confirming own previous [123I]s-CIT SPECT studies (Hesse et al., 2004). This reduction seems to involve not only SERT-rich brain regions but also limbic areas (hippocampus) but does not depend on SERT gene promoter polymorphism. We assume that, in OCD; SSRIs act mainly via SERT-dependant anti-depressant effect. Verification of this hypothesis, however, will require comparative on-medication versus off-medication studies.

BOLD SIGNAL AMPLITUDE SHOWS AN INVERSE DEPENDENCE ON BASELINE CBF

Joy Liau¹, Thomas T. Liu²

¹Department of Bioengineering, University of California at San Diego, San Diego, CA, USA, ²Department of Radiology, University of California at San Diego, San Diego, CA, USA

BRAIN ACTIVATION OF ACTUAL BUT NOT SIMULATED CAR-DRIVING USING PET AND [18F]FDG

Myeonggi Jeong¹, Manabu Tashiro¹, Keiichiro Yamaguchi¹, Laxsmi Singh¹, Masayasu Miyake¹, Ren Iwata², Masatoshi Itoh¹

¹Division of Nuclear Medicine, Tohoku University Cyclotron and Radioisotope Center, Sendai, Miyagi, Japan, ²Division of Radiopharmaceutical Chemistry, Tohoku University Cyclotron and Radioisotope Center, Sendai, Miyagi, Japan

IMPAIRED SENSITIVITY TO PAIN IN MULTIPLE FUNCTIONAL DISORDER

Ruta Kuzminskyte^1,2, R. Kupers^2,3, J. Geday³, P. Videbech⁴, P. Fink¹, A. Gjedde^2,3

¹Research Clinic for Functional Disorders and Psychosomatics, Aarhus University Hospitals, Aarhus, Denmark, ²Center of Functionally Integrative Neuroscience, Aarhus University Hospitals, Aarhus, Denmark, ³Pathophysiology and Experimental Tomography Center, Aarhus University Hospitals, Aarhus, Denmark, ⁴Institute of Biological Psychiatry, Aarhus University Hospital, Aarhus, Denmark

Background and aim: Some patients complain of physical symptoms from various organ systems that are not attributable to any known conventionally defined disease and are not adequately supported by clinical or paraclinical findings. Such patients are suffering from Multiple Functional Disorder (MFD) and their symptoms are functional somatic symptoms. Functional somatic symptoms are painful, severe, and persistent and have a significant impact on patients' well-being and functioning.

Given the lack of peripheral pathology in MFD patients, research focuses on the central nervous system as the potential source of the problem. The aim of this study was to examine cerebral responses to painful heat stimulation in MFD patients and compare these with responses in healthy controls.

Methods: Fourteen MFD patients and 14 healthy controls participated in the study. All subjects underwent a psychiatric (SCAN) interview, a thorough somatic check-up and quantitative sensory testing, during which individual thermal pain thresholds were determined. We used water (H215O) uptake detected with PET to measure changes in regional cerebral blood flow (rCBF). Measurements were made in triplicate at baseline and during painful stimulation. Stimulus strengths were determined individually as each subject's pain threshold. After every tomography session, the subjects rated the intensity of the induced pain.

Results: At baseline, the rCBF were similar in MFD patients and healthy controls, and mean stimulus temperatures and mean ratings of pain during painful stimulation did not differ between the two groups. However, differences in rCBF (both increases and decreases) were observed between the two groups during pain stimulation. In pain, the MFD patients had significantly lower increase of rCBF contralaterally in inferior parietal lobe (BA 40), and bilaterally in inferior frontal gyri (BA 10) when compared to controls.

Conclusions: The MFD patients and the controls reported similar perception of painful stimuli induced by similar stimulus temperatures. Nevertheless, the MFD patients exhibited significantly lower increase in rCBF during painful stimulation contralaterally in secondary somatosensory cortex and bilaterally in prefrontal cortices when compared to controls, suggesting that pain perception is impaired in the MFD patients.

EFFECT OF BRAIN HYPOTHERMIA ON CEREBRAL GLUCOSE METABOLISM IN ACUTE BRAIN INJURY: EVALUATION WITH FLUORODEOXYGLUCOSE POSITRON EMISSION TOMOGRAPHY

Nobuyuki Kawai¹, Yoshihiro Nishiyama², Yuka Yamamoto², Takashi Tamiya¹, Seigo Nagao¹

¹Department of Neurological Surgery, Kagawa University School of Medicine, Kita-Gun, Kigawa, Japan, ²Department of Radiology, Kagawa University School of Medicine, Kita-Gun, Kagawa, Japan

Background and aims: Brain hypothermia (BH) has been used for brain protection in patients with acute brain damage. Suppression of the metabolism and subsequent lowered blood flow in the brain has been considered as a major protective mechanism of BH. To evaluate the effect of BH on glucose metabolism, we examined cerebral glucose metabolism during and after BH in patients with acute brain damage.

Methods: We investigated 4 patients (3 severe head injury and 1 cardiac arrest) who were treated with mild BH (34-34.5?C) for 3–5 days. All patients were sedated with midazolam and buprenorphine and paralyzed with vecuronium during BH. Cerebral metabolic rate for glucose (CMRglc) was determined quantitatively with 18F-fluoro- deoxyglucose positron emission tomography (FDG-PET) kinetic analysis during and after BH.

Results: Case 1 was 43 year-old female with diffuse brain injury (GCS 8). PDG-PET during BH showed diffuse glucose hypometabolism (CMRglc 25.5 µmol/100g/min) in the brain. Follow-up FDG-PET examined on the 28th hospital day showed markedly increased glucose metabolism (CMRglc 53.3 µmol/100g/min) except for the contusional brain. The patient had moderate disability at discharge. Case 2 was 69 year-old male with focal brain injury (GCS 5). The patient received craniotomy for delayed traumatic intracerebral hematoma. PDG-PET during BH showed diffuse glucose hypometabolism (CMRglc 34.9 µmol/100g/min) in the brain. Follow-up FDG-PET examined on the 30th hospital day showed mildly increased glucose metabolism (CMRglc 45.5 µmol/100g/min). The patient had severe disability at discharge. Case 3 was 57 year-old male with diffuse brain injury (GCS 6). PDG-PET during BH showed diffuse glucose hypometabolism (CMRglc 35.4 µmol/100g/min) in the brain. Follow-up FDG-PET examined on the 60th hospital day showed further decreased glucose metabolism (CMRglc 28.2 µmol/100g/min). The patient did not show any improvement in his consciousness. Case 4 was 45 year-old male with out-of-hospital cardiac arrest. After successful resuscitation, he was semicomatous with GCS of 8. PDG-PET during BH showed diffuse glucose hypometabolism (CMRglc 35.3 µmol/100g/min) in the brain. Follow-up FDG-PET examined on the 15th hospital day showed markedly increased glucose metabolism (CMRglc 58.8 µmol/100g/min). The patient fully recovered with no neurological deficits.

Conclusions: In two patients who showed favorable recovery (Case 1&4), BH suppressed the cerebral glucose metabolism approximately 40–50%. Considering the effect of sedative drugs on the metabolism using during BH, the suppressive effect of BH on the cerebral glucose metabolism is mild. In two patients who showed unfavorable recovery (Case 2&3), follow-up FDG-PET study showed decreased or mildly increased glucose metabolism in the brain. Temporal changes in the cerebral glucose metabolism using FDG-PET may be a useful indicator of neurological recovery in patients with severe acute brain damage.

CLINICAL APPLICATION OF REAL-TIME CEREBRAL PERFUSION IMAGING USING DIGITIZED DSA IMAGE DATA

Yoshikazu Nakajima¹, Masayuki Hirata², Amami Kato², Toshiyuki Fujinaka², Yoshifumi Teramoto¹, Mamoru Ito¹, Toshiki Yoshimine²

¹Department of Neurosurgery, Rinku General Medical Center, Izumisano, Osaka, Japan, ²Depatment of Neurosurgery, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

Introduction: We have previously reported the utility of a perfusion imaging (PI) method that displays color-coded images of global and local cerebral perfusion based on DSA image data (1)–(3). The previous system processed digitized data converted from an analog signal on videotape. The aim of this study was to evaluate the clinical utility of a new system that can directly analyze digitized data from serial DSA images in a DICOM format on a portable personal computer (PC).

Methods: We created the new system program using the Code Warrior C language image (Metrowerks, USA). The system was set up to operate on Windows XP (Microsoft Corp., USA) using Scion Image for Windows (Scion Corp, USA). It can process DSA image data in a DICOM format, provided by the workstation with DSA equipment (CRS-PC, Cardiac View Station; GE OEC Medical Systems, Inc.). We applied this system to assess changes in cerebral perfusion in the treatment of acute apoplexy.

Results: This system was able to visualize cerebral perfusion using 20 DSA images (512 × 512-pixels) in less than 1 second. Information regarding changes in flow velocity of the cerebral vessels and global perfusion status could be provided in semi-real time. Moreover, as compared to our previous system, the new system can detect smaller differences in signal density in the high-density range (Table 1). This is important for clinical use because the signal density of cerebral vessels on DSA is around this high range.

Conclusion: The system described here can provide indispensable information for objectively evaluating DSA image data in semi-real time. It could be especially beneficial for intra-operative monitoring of cerebral perfusion during endovascular surgery.

DOES BOLD FUNCTIONAL MRI HAVE THE CAPABILITY OF MAPPING SUB-MILLIMETER-SCALE FUNCTIONAL COLUMNS?

Chan-Hong Moon, Hiro Fukuda, Sung-Hong Park, Seong-Gi Kim

Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA

[Background and aims] The ability of blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) in mapping sub-millimeter-scale functional columns still remains debatable mainly due to the spatially non-specific large vessel contribution, poor sensitivity and reproducibility, and lack of independent evaluation. Furthermore, regions with the highest conventional BOLD signals may indicate neurally inactive domains rather than active columns if, for instance, the results from optical intrinsic signal (OIS) are directly applicable. To examine these issues, here we performed BOLD fMRI at an ultrahigh magnetic field to map orientation-selective columns of isoflurane-anesthetized cats.

[Methods] Gradient-echo (GE) BOLD fMRI (TE = 18 ms, TR = 500 ms) and spin-echo (SE) BOLD fMRI (TE = 40 ms, TR = 2 s) with 0.3×0.3×1.0 mm3 resolution were obtained in six anesthetized cats (10 hemispheres) at 9.4 Tesla. SE BOLD technique at high magnetic fields is mostly sensitive to microvessels including capillaries, and has been shown to improve spatial localization to the parenchyma. As a reference for the known neural interpretation, cerebral blood volume (CBV)-weighted fMRI (TE = 10 ms, TR = 1 s) was also obtained after the injection of monocrystalline iron oxide nanoparticles (MION). Visual responses were induced by 800-sec-long full-field moving gratings with eight orientations (10 s for each orientation, 10 repetitions) without inter-stimulation gap. Fourier analysis was applied to obtain maps of orientation-specific signal (1/80 Hz). For neural interpretation, BOLD iso-orientation maps were directly compared to MION-based CBV fMRI maps and correlation coefficients (R) were calculated. For further evaluation (n = 3 of 6 cats, 4 hemispheres), the BOLD maps were also compared to the isoorientation maps in the same cats from CBV-weighted OIS.

[Results] Temporally-encoded continuous cyclic orientation stimulation with Fourier analysis reduces orientation-non-selective signals such as draining artifacts and allows us to obtain conventional GE BOLD iso-orientation maps with high reproducibility (R = 0.74 ± 0.12, n = 10 hemispheres). Further reduction of large vessel contribution was achieved using, SE BOLD, but with overall decreased sensitivity; the ratio of SE BOLD contrast-to-noise ratio (CNR) to GE BOLD CNR was approximately 1 to 3. The both GE and SE BOLD iso-orientation maps were significantly correlated with maps determined by MION-based CBV fMRI (R = 0.78 ± 0.06 for GE and 0.76 ± 0.08 for SE). Furthermore, the average least-distance for local maxima of isoorientation domains between GE BOLD and OIS maps was 0.22 ± 0.02 mm (n = 4 hemispheres). Since these distances are shorter than the column width of 0.70 mm (i.e., a half of the measured inter-columnar distance), we consider that the BOLD iso-orientation maps match with OIS isoorientation maps. These results suggest that the highest BOLD signals are localized to the sites of increased neural activity when column-non-selective signals are suppressed.

[Conclusions] Thus, sub-millimeter-scale functional columns can be reliably mapped using conventional BOLD fMRI.

[Acknowledgements] We thank Michelle Tasker and Ping Wang for animal preparation. This study was supported by NIH (NS44589, EB003324) and the McKnight Foundation.

DECREASED SEROTONIN 5-HT1A RECEPTOR AVAILABILITY IN MEDICATION-NAIVE PATIENTS WITH MAJOR DEPRESSIVE DISORDER ? A PET STUDY USING [CARBONYL-11C]WAY-100635

Jussi Hirvonen^1,2, Hasse Karlsson^1,3, Jaana Kajander¹, Antti Lepola², Juha Markkula¹, Helena Rasi-Hakala¹, Kjell Nagren², Sargo Aalto⁴, Jouko Salminen⁵, Jarmo Hietala^1,2

¹Department of Psychiatry, University of Turku, Turku, Finland, ²Turku PET Centre, University of Turku and Turku University Central Hospital, Turku, Finland, ³Department of Psychiatry, University of Helsinki, Turku, Finland, ⁴Department of Psychology, Abo Akademi University, Turku, Finland, ⁵The Social Insurance Institution of Finland, Turku, Finland

TRANSIENT HYPERACTIVATION IN BILATERAL MOTOR RELATED AREAS DURING RECOVERY FROM POSTSTROKE MILD HEMIPARESIS-A STUDY WITH NIRS

Kotaro Takeda^1,2, Itsuki Imai³, Kimiko Maruyama¹, Yukihiro Gomi¹, Nobuaki Shimoda^1,2, Masao Hiwatari^1,2, Hiroyuki Kato^1,2

¹Center for Clinical Medicine and Research, International University of Health and Welfare, Nasushiobara, Japan, ²CREST, Japan Science and Technology Agency, Kawaguchi, Japan, ³Department of Rehabilitation, Nasu Neurosurgical Center, Nasushiobara, Japan

[Background and aims] Hemiparesis is the most common neurological deficit after stroke. It has been demonstrated that several motor related cortical areas in both hemispheres contribute to motor recovery from the poststroke dysfunction. Although recent fMRI and PET studies have suggested that a transient involvement of ipsilateral hemisphere plays an important role for the recovery, the mechanisms and the recovery process have been imcompletely understood. We used near-infrared spectroscopy (NIRS), which can noninvasively monitor the changes in concentrations of oxygenated hemoglobin (Oxy-Hb) and deoxygenated Hb (Deoxy-Hb), for investigating the change of cerebral activation areas in the patients with mild hemiparesis from acute to chronic phases after stroke. [Methods] NIRS recordings were performed using a 44-channel optical topographic system (ETG-4000, Hitachi Medical) during 5 cycles of 15 s hand grasping and 30 s resting periods while sitting on a chair. According to the international 10-20 system, 30 probes (22 channels on each hemisphere) were mounted on head shells and were placed on the scalp overlying the primary sensorimotor cortex (SM1) and the dorsolateral premotor cortex (PMd). The activation areas in 6 motor recovered patients (52-73 years old; 4 females and 2 males) with a cerebral infarction were examined longitudinally through acute to chronic phases after stroke. [Results] The 6 patients showed the same NIRS findings (Figure). During affected hand movements, the amounts of Oxy-Hb increased significantly in bilateral SM1 and PMd extensively at the acute phase. At the chronic phase, in contrast, significant activation was observed predominantly in the contralateral SM1, which pattern was similar to that seen in normal subjects. This change occurred sigmoidally about 30 days after stroke. [Conclusions] The activation pattern during affected hand grasping changed from the bilateral extensive activation pattern to the contralateral predominant pattern. In addition, the activation areas were focused on the contralateral SM1. The present findings suggest that motor recovery from mild hemiparesis is facilitated by an early, transient recruitment of bilateral hemispheres including SM1 and PMd, and that this period is critical for the therapeutic interventions that are aimed at functional recovery.

TRANSCRANIAL ULTRASONIC BRAIN TISSUE PERFUSION IMAGES BY POWER MODULATION IMAGING IN COMPARISON WITH SECOND HARMONIC IMAGING

Toshiyuki Shiogai¹, Kaori Ikeda¹, Masami Matsumoto¹, Aki Morisaka¹, Yoshinari Nagakane², Toshiki Mizuno², Masanori Nakagawa², Hiroshi Furuhata³

¹Departments of Clinical Neurosciences, Kyoto Takeda Hospital, Kyoto, Japan, ²Department of Neurology, Kyoto Prefectural University of Medicine, Kyoto, Japan, ³ME Laboratory, Tokyo Jikei University School of Medicine, Tokyo, Japan

Objective: In order to improve temporal resolution of ultrasonic tissue perfusion images, harmonic imaging with an increased reception frequency has frequently been utilized. For the introduction of transcranial brain tissue perfusion imaging, fundamental imaging without increased harmonic frequency reception and with transient response high transmitting acoustic power images, appears suitable while taking bone- and depth-dependant ultrasound attenuation into consideration. Power modulation imaging (PMI) uses a fundamental frequency and transmits pulses of alternating amplitude and is introduced into continuously transmitted real-time low power images by means of second-generation ultrasound contrast agents. The objective is to evaluate transcranial brain tissue perfusion images utilizing high power images transmitted by transient response, and to compare between PMI and conventional second harmonic imaging (SHI).

Methods: The subjects were 10 neurological patients (ages 34-92, mean 70; 8 male, 2 female). Utilizing a SONOS5500 S3 transducer (Philips) after a bolus intravenous LevovistR injection (400mg/ml, 2.5g), transient response images utilizing PMI (1.7MHz) and SHI (transmitting/receiving frequencies: 1.3/2.6MHz) were created every 2 seconds at the bilateral transtemporal diencephalic horizontal plain. The mechanical index of both images was 1.6. The investigation depth was 16cm and the focus was 8 cm. Five regions of interest (ROI) were placed in the bilateral basal ganglia and thalamus, and contra-lateral temporal lobe. Peak intensity (PI) and time to PI (TPI) based on time-intensity curves of the ROIs as analyzed by QLAB software (Philips), were compared between both images (n = 18).

Results: 1) Significant depth dependant PI attenuation was observed in both images. 2) PI in all PM ROIs tended to be higher than those of SHI, significantly in the contra-lateral side. 3) Significant depth dependant TPI prolongation was not observed in either image type.

Conclusions: Transcranial brain tissue perfusion transient response images utilizing PMI are superior to SHI in contra-lateral cerebral hemisphere visualization. This will contribute to shorter examination times by visualizing bilateral hemispheres via unilateral temporal windows. For quantitative evaluation of brain tissue perfusion, TPI is a more suitable parameter than PI.

INCOMPLETE BRAIN INFARCTION IN PATIENTS WITH MOYAMOYA DISEASE: STATISTICAL IMAGING ANALYSIS USING 123I-IMZ-SPECT

Jyoji Nakagawara, Kenji Kamiyama, Toshiaki Osato, Toshiiti Watanabe, Hirohiko Nakamura

Department of Neurosurgery, Nakamura Memorial Hospital, Sapporo, Hokkaido, Japan

COMPARISON OF C-11 ACETATE, C-11 METHIONINE AND F-18 FDG PET IN PATIENTS WITH BRAIN MALIGNANT AND BENIGN TUMORS

Yuka Yamamoto¹, Yoshihiro Nishiyama¹, Nobuyuki Kawai², Motoomi Ohkawa¹

¹Department of Radiology, Faculty of Medicine, Kagawa University, Kita-Gun, Kagawa, Japan, ²Department of Neurological Surgery, Faculty of Medicine, Kagawa University, Kita-Gun, Kagawa, Japan

Background and aims: F-18 FDG (FDG) PET has been used in brain tumor imaging. However, several studies have demonstrated diagnostic limitations of FDG PET for imaging brain tumors because of the high basal glucose metabolic rate of normal brain tissue. C-11 methionine (MET) PET has been reported to be useful for evaluating brain tumors. C-11 acetate (Acetate) is currently being investigated as a new tracer for imaging neoplasms, most notably prostate cancer. Acetate PET has promising potential for the detection of brain tumors, because its uptake in normal brain is low. We compared the uptake of Acetate, MET and FDG in patients with brain tumor.

Methods: Twenty-two lesions in 20 patients with known brain tumor were evaluated regarding the uptake of Acetate, MET and FDG. The diagnoses were 12 primary malignant tumors, 2 metastases, 8 primary benign tumors. All patients underwent PET with Acetate (15 min p.i.), MET (15 min p.i.) and FDG (40 min p.i.). PET results were evaluated by visual and semiquantitative analysis. For semi-quantitative analysis, the standardized uptake value (SUV) and tumor to contralateral normal brain tissue (T/N) ratio were determined by region-of-interest analysis.

Results: The mean value of SUV in the contralateral normal brain tissue of Acetate (1.5±0.3) was lower than that of MET and FDG (1.9±0.4 and 7.1±1.9, respectively) (P<0.001). The mean value of SUV in the contralateral normal brain tissue of MET was also lower than that of FDG (P<0.001). MET uptake was observed in all 14 malignant tumors, Acetate in 12 malignant tumors and FDG in 5 malignant tumors. The mean value of T/N ratio of MET (2.5±1.3) and Acetate (1.8±1.0) in malignant tumors was higher than that of FDG (1.3±1.0) (P<0.02). There was also a significant difference between T/N ratio of MET and that of Acetate in malignant tumors (P<0.02). Both MET and Acetate uptake was observed in 6 of 7 benign tumors and FDG in 1 benign tumor. The mean value of T/N ratio of MET (2.0±0.7) and Acetate (2.1±0.8) in benign tumors was also higher than that of FDG (0.7±0.1) (P<0.02). There were no significant differences between T/N ratio in malignant and benign tumors on MET, Acetate and FDG PET.

Conclusions: MET is more sensitive than Acetate or FDG for detecting brain tumors although Acetate may be a promising tracer for the detection of brain tumors.

GD-CHELATE MRI CONTRAST AGENTS DELIVERED TO BRAIN PARENCHYMA BY INTRA-CEREBRAL VENTRICULAR INFUSION ARE LOCALIZED TO THE EXTRACELLULAR SPACE

Paul Schornack, Ping Wang, Seong-Gi Kim

Department of Radiology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA

This study ventures to demonstrate that Gd-chelates are localized to extracellular space in brain parenchyma. Distribution characteristics of uncharged Gd-chelates, GdDTPA-BMA(Omniscan) and GdHPO3A(Prohance), are quantitatively compared during prolonged “incubation? in rat brain, in vivo, treating Prohance as a good extracellular marker. Since the blood brain barrier(BBB) precludes delivery by vascular routes, achieving mM agent concentrations throughout brain parenchyma is accomplished by infusing agents over several hours directly into lateral ventricle CSF(ICVinfusion). Millimolar agent concentrations occur throughout the brain for up to 20hr post-infusion, due to the limited clearance pathways of the BBB/CSF systems. With vascular infusion, agents are typically only present in tissues at mM concentrations for seconds to minutes and have essentially no contact with brain parenchyma cells due to the BBB. Conversely, the conditions of the ICV infusion experiment are very similar to those that exist in the studies of Cabella(2006) exploring intracellular labeling of cultured cells with different Gd-chelates by incubation(4hr) with 1-50mM agent. While both Gd-chelates used in the present study are uncharged and possess the same MW and size, their thermodynamic stabilities are markedly different; the macrocyclic GdHPCO3A logK=23.8 is 10^7 greater than the logK=16.5 of GdDTPA-BMA. The more thermodynamically stable GdHPCO3A was observed to not enter cells while GdDTPA-BMA appears to be taken up into cells (due to Gd3+ release from GdDTPA-BMA induced by phosphate).

Experiments used male Sprague-Dawley rats(390-440gm) with Isofluorane anesthesia. The ICV infusion cannula was positioned in the left lateral ventricle using a stereotaxic frame for delivery of identical quantities of either agent (30µL of 250mM at 6µL/hr over 5hr). T1 maps were obtained by inversion recovery gradient echo(Turboflash) MRI between 4 and 16hr after agent infusion using a 9.4T Varian system.

Agent clearance rate constants(kclearance) from ROIs of the R1(1/T1) maps throughout the brain were obtained. R1 histograms shown indicate similar distribution kinetics in the brain for both agents. Plots of kclearance with z-position show very similar clearance rates of these uncharged Gd-chelates throughout the entire depth of that ROI. The Gd-chelates in the present study clear out of the brain with time constants of 7-15hr with no agent apparent after 36hr; inconsistent with the observation that Gd-chelates taken up by pinocytosis in stem cells have been observed for up to 14 days in endosomes! Taking GdHPCO3A as a reliable extracellular marker under the conditions of this experiment, these results indicate that Gd-chelates remain predominantly in the extracellular space and do not enter cells during their prolonged presence in brain parenchyma.

EFFECTS OF PVH, ANTICLOTTING, ANTIPLATELET DRUGS, AND ANTIHYPERTENSIVES ON MICROBLEEDS

Yuichi Komaba¹, Kazumasa Arii², Noriko Tanaka², Yasuo Katayama¹

¹Department of Neurology, Nippon Medical School, Tokyo, Japan, ²Department of Neurology, Hakujikai Memorial Hospital, Tokyo, Japan

INITIAL EXPERIENCE OF NEW COREGISTRATION SOFTWARE BRAIN EASY ANALYSIS TOOL FOR 201TL SPECT IN BRAIN TUMORS

Yasushi Shibata, Akira Matsushita, Manabu Akimoto, Tetsuya Yamamoto, Shingo Takano, Akira Matsumura

Department of Neurosurgery, Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Ibaraki, Japan

IN VIVO IMAGING OF THE PENUMBRA IN MICE USING SMALL ANIMAL 18F-FDG PET

Benjamin Klaesner¹, Uta Mamrak², Marc Huisman¹, Markus Schwaiger¹, Alexander Drzezga¹,

Nikolaus Plesnila²

¹Department of Nuclear Medicine, Klinikum Rechts Der Isar, Technical University of Munich, Munich, Germany, ²Laboratory of Experimental Neurosurgery, Department of Neurosurgery & Institute for Surgical Research, University of Munich Medical Center, Grosshadern, Munich, Germany

OXYGEN METABOLISM CHANGES BEFORE AND AFTER SHUNTING OPERATION IN PATIENTS WITH IDIOPATHIC NORMAL PRESSURE HYDROCEPHALUS (I-NPH)

Junichi Miyamoto, Kazunori Tatsuzawa, Yasuo Inoue, Katsuyoshi Mineura

Department of Neurosurgery, Kyoto Prefectural University of Medicine, Kyoto, Japan

Objective - The aim of the present study is to evaluate the changes in cerebral oxygen metabolism before and after ventriculo-peritoneal shunt (VPS) in patients with idiopathic normal pressure hydrocephalus (i-NPH).

Methods - Eight patients with i-NPH underwent VPS and an 15O-positron emission tomography (15O-PET) study before and approximately 3 months after VPS. The relationship between the changes in oxygen metabolism before and after VPS was analyzed.

Results - In five patients, the symptoms improved based on the classification by Krauss (1) (good responders) after VPS. In three patients, the symptoms improved subjectively following VPS (poor responders). The postoperative regional metabolic rate of oxygen (rCMRO2) of the good responders increased significantly. The postoperative regional oxygen extraction fraction (rOEF) in reduced in the poor responders.

Conclusion - The changes in oxygen metabolism imply metabolic pathogenesis and predict responsiveness to shunting operation in iNPH.

Figure - The graphs show the results of the 15O-PET study in the frontal white matter (A), putamen (B), and thalamus (C). Graphs show the results of rCBF (a), rCBV (b), rOEF (c), and rCMRO2 (d). The continuous line shows the results of good responders and the broken line shows the poor responders. rCMRO2 increases apparently after VPS in the good responders. In the poor responders, the change of rOEF is apparent without any improvements of rCMRO2.

BRAIN HISTAMINE H1 RECEPTOR OCCUPANCY OF A NEW ANTIHISTAMINE, BEPOTASTINE, MEASURED BY PET AND [11C]DOXEPIN

Manabu Tashiro¹, Duan Xudong¹, Motohisa Kato², Masayasu Miyake², Shouichi Watanuki², Youichi Ishikawa³, Yoshihito Funaki³, Ren Iwata³, Masatoshi Itoh¹, Kazuhiko Yanai²

¹Division of Nuclear Medicine, Tohoku University Cyclotron and Radioisotope Center, Sendai, Miyagi, Japan, ²Department of Pharmacology, Graduate School of Medicine, Tohoku University, Sendai, Miyagi, Japan, ³Division of Radiopharmaceutical Chemistry, Tohoku University Cyclotron and Radioisotope Center, Sendai, Miyagi, Japan

A NEAR-INFRARED SPECTROSCOPY (NIRS) STUDY OF LEXICAL PITCH ACCENT PROCESSING IN JAPANESE SPEAKERS

Yutaka Sato¹, Reiko Mazuka^1,2, Yuko Sogabe¹

¹Laboratory for Language Development, BSI, RIKEN, Wako, Saitama, Japan, ²Department of Psychology and Neuroscience, Duke University, Durham, NC, USA

HEMODYNAMIC CHANGES AFTER SPECTACULAR SHRINKING DEFICIT IN PATIENTS WITH CEREBRAL INFARCTION IN COMPARISON TO MRI FINDINGS

Yoshiharu Taguchi, Shutaro Takashima, Kouzi Hirano, Nobuhiro Dougu, Shigeru Toyota, Kortaro Tanaka

The Department of Neurology, Toyama University Hospital, Toyama, Japan

Introduction: Spectacular shrinking deficit (SSD) refers to a rapid recovery from a major hemispheric syndrome within a few hours after onset of stroke (1). SSD is considered to result from rapid embolic lysis, fragmentation and migration of embolus before irreversible damage occurs (1). However, little is known about the cerebral hemodynamic changes after SSD (2). We evaluated changes in regional cerebral blood flow (rCBF) using 99mTc-ECD single-photon emission computed tomography (SPECT) in patients with SSD, and compared these findings with MRI images.

Methods: We examined rCBF by 99mTc-ECD SPECT in 15 patients after SSD. The SPECT study was performed within 8 days after SSD and the follow-up SPECT study was done if patients revealed hyperperfusion. The obtained SPECT data were analyzed semiquantitatively using a three-dimensional stereotaxic region-of-interest template (3DSRT) reported by Takeuchi et al. (3). Brain MRI or CT was examined in all patients to evaluate the size of lesion.

Results: The examined patients were 8 men and 7 women, aged 70.8±11.9 years (mean±SD). In regard to type of cerebral infarction, 9 patients were cardiogenic embolism, 3 patients were artery to artery embolism and 3 patients had unknown embolic source. The time interval between the onset of cerebral hemispheric syndrome and the subsequent rapid improvement was 2.25±3.73 hours (mean±SD). SPECT showed hyperperfusion (111.45±6.03%) in the reperfused territory compared to the contralateral side in 6 patients, hypoperfusion (93.83±3.55%) in 6 patients and normoperfusion (100.74±2.95%) in 3 patients. The follow-up SPECT showed hypoperfusion (87.74±3.55%) in 4 patients and normoperfusion (100.74±3.95%) in 2 patients out of the 6 patients who revealed hyperperfusion on the first SPECT study. The SPECT study was performed significantly earlier in the patients revealing hyperperfusion (3.2±1.2 days after SSD) than in those revealing hypoperfusion (6.0±1.8 days)(p<0.05). In patients without any abnormal areas in diffusion-weighted MRI, the above SPECT findings were very helpful to assess the size of initial ischemic area.

Conclusion: The present study indicated that the initial ischemic areas become hyperperfused during the first few days after SSD, followed by a gradual decline of CBF. Measurement of rCBF by SPECT is very helpful to assess the size of initial ischemic area, especially in patients with no abnormality in diffusion-weighted MRI.

BRAIN METHIONINE METABOLISM IN HUMANS:A CORRELATION WITH CEREBRAL OXYGEN UTILIZATION STUDIED WITH PET AND SPM

Ryo Ashida, Kunihiro Kawanishi, Nobuhiko Shiraki, Haruka Maeda, Taku Fujimoto, Yasuhiro Ohsaki, Ashik Bin Ansar, Katsufumi Kajimoto, Yasukazu Kimura, Hiroki Kato, Makiko Tanaka, Eku Shimosegawa, Jun Hatazawa

Department of Nuclear Medicine and Tracer Kinetics, Osaka Univesity Graduate School of Medicine, Suita, Osaka, Japan

A number of experimental studies indicated that methionine (MET) is metabolized by neurons, glial cells, and blood vessels. Although normal distribution of MET brain uptake was demonstrated in experimental animals (1), it is still unknown in humans. In this study, we measured local brain uptake of [11C]-MET in neurologically normal subjects using PET. By applying SPM99, we created a standardized map of MET uptake in humans. We compared a pattern of MET uptake with that of cerebral metabolic rate of oxygen (CMRO2).

COMPARISON OF HIPPOCAMPAL ATROPHY ASSESSED USING NEW MRI SOFTWARE (VSRAD) IN ALZHEIMER'S DISEASE, VASCULAR DEMENTIA, AND DEMENTIA WITH LEWY BODIES

Kazunori Okada¹, Nobuyuki Umegae¹, Takuya Yamaguchi¹, Shuhei Yamaguchi²

¹Department of Neurology, Oda Municipal Hospital, Oda City, Shimane Prefecture, Japan, ²Department of Internal Medicine III, Shimane University School of Medicine, Oda City, Shimane Prefecture, Japan

[Introduction] The early diagnosis of Alzheimer's disease (AD) is important for determining when to administer drugs such as cholinesterase inhibitors, and for estimating the prognosis. Matsuda et al. have recently developed new magnetic resonance imaging (MRI) analysis software, the Voxel-based Specific Regional Analysis System for Alzheimer's Disease (VSRAD), which analyzes brain volume by voxels and yields a hippocampal atrophy index using the Z score. To clarify the usefulness of VSRAD, we compared the severity of disease, intellectual function, and hippocampal atrophy in subjects with AD, vascular dementia (VD), and dementia with Lewy bodies (DLB).

[Subjects] All of the subjects were outpatients and they could walk without assistance. The subjects consisted of 30 probable AD patients (mean age 80.2 yr), 10 VD patients (76.3 yr), 3 DLB (71.0 yr), and 8 normal controls (NC; 66.4 yr). Each disease was diagnosed according to the DSM-IV and the guidelines for DLB.

[Methods] Neuropsychological examinations administered consisted of the Mini-Mental State Examination (MMSE) for all subjects and the Alzheimer's Disease Assessment Scale (ADAS-J cog.) for AD. The impairment of daily life was assessed using the Clinical Dementia Rating (CDR). Head MRI images were obtained in 1.5-mm-thick sagittal slices and converted into DICOM files. The images were analyzed on a personal computer using the VSRAD system.

[Results]

[Conclusions] The VSRAD system provides useful information about hippocampal atrophy that distinguishes AD from VD. In addition, it indicates the severity of AD patients.

The English in this document has been checked by at least two professional editors, both native speakers of English. For a certificate, see: http://www.textcheck.com/cgi-bin/certificate.cgi?id=PZz4gV

(1) We recommend that you include the above statement in the end of your document to inform reviewers that the English has been professionally checked. Many reviewers are not native speakers of English and this information may be valuable to them. (The text can be deleted later).

2) We strongly recommend that you do not make ANY changes to this document. Textcheck should be the last process except final formatting. No further editing should be necessary unless there is text that the editors have misunderstood.

If any text has been misunderstood:

Rewrite the sentence(s), SAVE, and upload the completed document to: http://www.textcheck.com/cgi-bin/login.cgi?revision=70110ok for a final check. (To upload, click on the link above, or copy it into the address bar in Internet Explorer, Netscape, or other web browser.)

(Please do not mark your changes in any way, do not send as a new document, and do not send by e-mail unless necessary.)

Requests for a final check should be made within 1 month.

[The above applies ONLY to requests for a final check. Any new request for editing (such as changes post review) should be uploaded as a new document via ‘Submit Document’ on our www site, with a note that the file is a revision of ‘70110ok’. Text with minor changes may qualify for a discount of 50%. This discount will not apply if the document is extensively rewritten.]

EXOGENOUS GROWTH FACTORS AMELIORATE BEHAVIORAL DEFICITS THROUGH ACTIVATION OF ENDOGENOUS NEURONAL REGENERATION IN YOUNG AND AGED RAT HUNTINGTON MODELS

Jun-ichi Tanaka^1,3, Gakushi Yoshikawa^1,3, Soichi Oya^1,3, Nobuhito Saito¹, Takaaki Kirino^2,3, Nobutaka Kawahara^1,3

¹Department of Neurosurgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan, ²Research Institute International Medical Center of Japan, Tokyo, Japan, ³Solution Oriented Research for Science and Technology, Japan and Science Technology Corporation, Saitama, Japan

Introduction: We previously reported that exogenously administrated growth factors (GFs) activated neural progenitors and recruited them to the hippocampal CA1 region in rat global ischemia, which was accompanied by improvement of impaired functions. Here rises a question whether these self-repair mechanisms also occur in other non-neurogenic sites not only in young but in aged adults. To address these issues, this study was conducted using a rat model of Huntington's disease. Methods: Young (2 months of age) and aged (more than 18 months of age) male Wistar rats were given quinolinic acid (100 nMol/1ul) or PBS directly into the right striatum. Two days later, the treated groups were given either a cocktail of EGF and FGF-2 or vehicle solution intraventricularly for 7 days, with simultaneous administration of BrdU (100mg/kg, ip.) twice daily for the last 3 days. Some of non-treated animals were sacrificed just after injuries. Brains were removed after 6 to 7 weeks, and immunohistochemical investigations were carried out on serial coronal sections. Histological quantification was conducted in 3 rectangular areas with sizes of 0.5 × 2 mm2, L: lateral adjacent to the lesion core, IM: intermediate and M: medial close to the lateral ventricle. To confirm newborn cells as neurons, double staining was performed using markers for NeuN, Hu, Dcx or GFAP together with BrdU. Fluorogold, a retrograde tracer, was injected for 3 GF-treated rats into the ipsilateral globus pallidus 3 days prior to sacrifice. Phenotypes of new neurons were classified by antibodies against choline-acetyltransferase (ChAT), parvalbumin, neuropeptide Y and DARPP-32. Behavioral study was carried out using the apomorphine (1.0 mg/kg)-induced rotation test. In the younger groups, the staircase test was also conducted. Difference with P<0.05 were considered statistically significant. Results: In the young adults, by cresyl violet staining, the number of neurons increased by 5.8% without treatment compared with that at 2 days, which was further augmented significantly by the GFs (42.8%) especially in the area L and IM as compared to vehicle control (3.2 %) (unpaired t-test, n=5-9/group). These cells were confirmed as mature neurons by double staining for BrdU and NeuN. Although the same trends were observed in aged animals, they did not reach a significant difference (P=0.08, n=4, each). Intriguingly, immunostaining revealed a relatively large increase of ChAT-positive newborn cells (from 12 to 39%). Double staining for Fluorogold and BrdU identified colabeled cells in the rostral striatum, indicating the integration of these newborn cells into the existing circuits. Rotometer test revealed the significant recovery in GFs-treated groups in both young and aged animals (paired t-test, n=7-11). Staircase test in young GF-treated animals showed good recovery of an impaired forelimb function equal to the unlesioned side; while vehicle-treated animals remained unskillful (n=16 and 9). Conclusions: Exogenously administered GFs augmented the endogenous regenerative capacity in the striatum, and improved behavioral functions in both young and aged adult Huntington's disease rats. This novel regenerative strategy without transplantation may have a therapeutic potential in various neuronal injuries in the striatum even in the aged adults.

ENGRAFTMENT AND DISPERSION OF HUMAN EMBRYONIC STEM CELL-DERIVED MULTIPOTENT NEURAL STEM CELLS IN EXPERIMENTAL STROKE MODELS

Marcel Daadi, Gary Steinberg

Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA

Cellular therapy for stroke has demonstrated promising functional recovery in experimental stroke models. However, the current sources of stem cells are not amenable for the needs of large-scale production and multi-centered clinical trials. Human embryonic stem cells (hESCs) have the potential to provide an indefinite and renewable source of a wide range of normal cell types. However, the isolation of homogenous and self-renewable sources of neural progenitors capable of differentiating into the principal neural lineages from hESCs remains largely unexplored.

We carried out a series of experiments to isolate stable, homogenous and expandable multipotent neural stem-progenitor cell lines (NSPCs) from hESCs. The hESCs were cultured in serum free medium containing epidermal growth factor, basic fibroblastic growth factor and leukemia inhibitory growth factor. After 3 days in vitro (DIV), the cells appeared aggregated in clusters or spheres. These primary spheres were collected and replated in fresh culture media supplemented with factors. Secondary spheres were then generated and expanded for 5 to 6 passages. To ascertain self-renewal ability under clonal conditions, a single cell suspension made from dissociated spheres were plated at clonal density (10 cell/µl). After 48 hrs, plated single cells underwent a first cell division and gave rise to progeny expressing nestin, a neurepithelial stem cell marker. The single cell-derived spheres continued to grow over the 7 DIV. At this stage, all the neurospheres generated were nestin+ and did not express the pluripotency markers SSEA4, OCT4 or Nanog. The NSPCs showed stable growth and a consistent 2.7±0.2 fold increase every 5 to 7 days. The population doubling at each passage averaged at 1.4±0.1. When induced to differentiate, 36.6±2.7 of the cells expressed nestin, 62.5±2.8% expressed the neuronal marker TuJ1, 1.9±0.3% were GFAP+ astrocytes and 7.1±0.4% were oligodendrocytes and expressed galactocerebroside.

The NSPCs were then evaluated in the rat middle cerebral artery occlusion (MCAO) model of stroke. Sprague Dawley rats (n=10) were subjected to 1.5 h MCAO. One week later, immunosuppressed rats were transplanted with NSPCs (2 × 105) into the ischemic boundary zone in the striatum. Animals were sacrificed 1 week later and the brains processed for histo-pathology and immunocytochemistry. Grafted NSPCs, identified with a human-specific nuclear marker, demonstrated a 14.2±3.1% survival rate and 61.2±4.7% migrated at least 200 µm away from the graft core and penetrated an average distance of 806.4±49.3 µm into the stroke-damaged tissue. They extended neurite outgrowth into the host parenchyma and did not show tumor formation. Characterization of the grafted cells demonstrated the absence of pluripotency markers and the presence of neural-specific markers. Transplanted cells expressed nestin and differentiated into TuJ1+ neurons, GFAP+ astrocytes and CNPase+ oligodendrocytes with 6.5±0.9% of the neurons expressing doublecortin.

We have demonstrated that under defined culture conditions, pluripotent hESCs can give rise to homogenous, expandable multipotent NSPCs. Transplantation of NCPSs into infarcted neural tissue resulted in good survival, migration and neural differentiation into the principal CNS cell types. These findings provide preliminary evidence of potential use of hES derivatives in cell therapy for stroke.

MICROGLIA PROMOTES ASTROGLIOGENESIS OF THE NEURAL STEM CELLS THROUGH ACTIVATING JAK/STAT SIGNALING

Pengxiang Zhu, Ryuji Hata, Fang Cao, Masahiro Sakanaka

Department of Functional Histology, Ehime University Graduate School of Medicine, Toon, Ehime, Japan

ANOXIA INCREASES THE NUMBER OF ADULT NEURAL STEM CELLS BY INHIBITION OF APOPTOSIS

Heinrich F. Burgers¹, Martin Henrik Maurer¹, Dominik Schelshorn¹, Armin Schneider², Wolfgang Kuschinsky¹

¹Department of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany, ²SYGNIS Pharma AG, Heidelberg, Germany

Introduction: Neurons are damaged by deprivation of oxygen and glucose. It is not known whether their precursors, i.e. adult neural stem cells are likewise damaged by oxygen and glucose deprivation. We have investigated the influence of anoxia, glucose deprivation and oxygen glucose deprivation on proliferation and survival of adult neural stem cells in vitro. Since erythropoietin is known as neuroprotective agent, we also tested its effect.

Methods: We cultured adult neural stem cells (NSC) from the subventricular zone of adult rat brains and exposed them to normoxia (controls) or anoxia (5% CO2 in N2) for 24h. Glucose deprivation was induced adding 5mM 2-deoxy-D-glucose. Cell viability was determined using a colorimetric assay (CellTiter96 AQueous One; Promega, Mannheim, Germany). Total caspases activity was recorded using a homogeneous caspases assay (Roche, Penzberg, Germany). Caspase-3/7 activity was determined using the Apo-ONE Homogeneous Caspase-3/7 Assay (Promega). Proliferative activity was determined by BrdU uptake and subsequent immunostaining of fixed cells on coverslides. Proliferation rates were determined by the proliferation dependent dilution of a fluorescent intracellular dye, CFSE (Sigma-Aldrich, Munchen, Germany) using FACS (BD Bioscience, Heidelberg, Germany). Erythropoietin (Janssen-Cilag, Neuss, Germany) concentrations ranged from 0.1 to 100 u/mL.

Results: Cell viability of NSC increased by more than 50% during anoxia compared to normoxic controls (n=240, p<0.01). This increased cell viability could be caused by either (1) reduced apoptosis or (2) by increased proliferation.

(1) The two caspases assays of apoptosis yielded different results: The general caspases activity was significantly reduced by about 50% (n=15, p<0.05) whereas the specific activity of caspases-3/7 of anoxic cells was not affected (n=18, n.s.), indicating a pathway of apoptosis different from caspases-3/7. (2) Proliferation activity was tested by BrdU incorporation (24h). About 99% of the normoxic cells had incorporated BrdU, whereas only 71% of the anoxic cells were found BrdU-positive, indicating decreased proliferation activity in anoxic cells. CFSE staining showed identical effects.

Under glucose deprivation cell viability was reduced to 44% (n=36, p<0.05) regardless of anoxia or normoxia (n.s.).

Erythropoietin did not affect cell viability in normoxia and anoxia, whereas under oxygen glucose deprivation, erythropoietin moderately increased cell viability by up to 6% (n=216, p<0.05). Again, total caspases activity was reduced in a dose dependent manner by 25% (n=24, p<0.05) in normoxia and anoxia whereas caspase-3/7 activity of anoxic cells was not affected.

Conclusions: The main finding of the present study is an increased viability of adult neural stem cells during anoxia. This effect, which is known from other types of stem cells, can be explained by reduced overall (homogeneous) caspases activity, but not by specific effects on caspases-3/7 or on cell proliferation. Glucose deprivation is detrimental to cell viability, even during anoxia.

The erythropoietin-induced inhibition of homogeneous caspases activity did not result in an increased viability during the short experimental period (24h). However, we cannot exclude a positive effect of erythropoietin during longer observation periods.

Support: German Ministry of Education and Research (National Genome Research Network, NGFN2), Dell GmbH, Estate of Friedrich Fischer.

ZINC CHELATION BY CLIOQUINOL OR VESICULAR ZINC DEPLETION BY TARGETED DELETION OF ZNT3 REDUCES HIPPOCAMPAL NEUROGENESIS FOLLOWING HYPOGLYCEMIC INSULTS

Sang Won Suh^1,3, Seok Joon Won^2,3, Aaron Hamby^1,3, Christian Sheline⁴, Raymond Swanson^1,3, Jialing Liu^2,3

¹Department of Neurology, UCSF, San Francisco, CA, USA, ²Department of Neurosurgery, UCSF, San Francisco, CA, USA, ³Department VAMC, San Francisco, San Francisco, CA, USA, ⁴Dpepartment of Neurology, Washington University, St. Louis, MO, USA

Neurogenesis after brain injury not only leads to the replacement of damaged cells but might also contribute to functional recovery, suggesting the possibility of endogenous neural repair. In the brain, neurogenesis occurs in the subgranular zone (SGZ) of the dentate gyrus (DG) where high amount of vesicular zinc is localized in the presynaptic terminals. The aim of this study is to determine whether the vesicular zinc plays a modulating role in neurogenesis either under normal or pathological conditions. In this study, we investigated the role of zinc on hippocampal neural regeneration in a rodent model of hypoglycemia using cell permeable zinc chelator, clioquinol (CQ) or targeted deletion of the zinc transporter 3 (ZnT3 KO). In the rodent model of hypoglycemia induced by insulin followed by 30-minute isoelectric electroencephalogram (EEG), extensive neuronal loss was observed in the hippocampus including area CA1 and the DG. At this time, a transient increase in progenitor cell proliferation in the DG SGZ was detected, leading to an increase of immature neuroblasts. Most of the surviving newborn cells assumed a neuronal phenotype within one week in the DG. Animals treated with CQ (30 mg/kg) twice per day for 1 week after hypoglycemia showed decreased progenitor cell proliferation in the SGZ compare to vehicle treated ones, with a parallel reduction in the number of neuroblasts. Using a mice model of hypoglycemia, wild type mice showed substantial increase of progenitor cell proliferation one week after hypoglycemia. However, ZnT3 KO mice showed less number of neuroblasts in the SGZ both under normal and hypoglycemic conditions. Our data suggest that synaptic zinc chelation or genetic depletion of vesicular zinc significantly reduces progenitor cell proliferation in SGZ. For the first time, our present study demonstrates that vesicular zinc in the hippocampus modulates neurogenesis in the adult brain not only under normal condition but also following insulin-induced hypoglycemia. The mechanisms involved in the Zn-mediated inhibition of hippocampal neurogenesis warrant further investigation.

Supported by Juvenile Diabetes Research Foundation (SWS, JDRF2-2006-113; JL NIHR21 NS051576)

IMPLANTATION OF A NEW POROUS GELATIN-SILOXANE HYBRID INTO A BRAIN LESION AS A POTENTIAL SCAFFOLD FOR TISSUE REGENERATION

Kentaro Deguchi¹, Takeshi Hayashi¹, Kanji Tsuru³, Mikiro Takaishi², Mitsuyuki Nagahara³, Shoko Nagotani¹, Yoshihide Sehara¹, HanZhe Zhang¹, Satoshi Hayakawa³, Tatsushi Kamiya¹, Masahiro Miyazaki², Akiyoshi Osaka³, Nam-ho Huh², Koji Abe¹

¹Departments of Neurology, Graduate School of Medicine, Dentistry and Pharmacy, Okayama University, Okayama, Japan, ²Departments of Cell Biology, Graduate School of Medicine, Dentistry and Pharmacy, Okayama University, Okayama, Japan, ³Biomaterial Laboratory, Faculty of Engineering, Okayama University, Okayama, Japan

Background: For brain tissue regeneration, a scaffold for migrated or transplanted cells with vessels is important if the tissue become pannecrotic. The present study was designed to determine whether the implantation of porous gelatin-siloxane hybrids can serve as a scaffold for brain tissue regeneration.

Methods: We used male Wistar rats of 12-weeks old. A new porous gelatin-siloxane hybrid derived from the integration of gelatin and 3-(glycidoxypropyl) trimethoxysilane (GPSM) was implanted into a defect of the cerebral cortex, which was located at 3.0 mm anterior to 2.0 mm posterior of the bregma (5 mm in length) and 2.0–4..0 mm (2 mm in width) lateral to the midline, and the depth of the lesion was approximately 2.0 mm below the skull surface. The block of GPSM was presoaked in basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF) or only vehicle, and the time of implantation was just after or at 30d of brain defect making. The animal was decapitated and the frozen sections were prepared. In order to identify the cell types in the newly formed tissue, immunohistochemical analysis was performed for N-acetylglucosamine oligomers (NAGO), an endothelial cell marker, GFAP, ionized calcium-binding adapter molecule-1 (Iba1), NeuN and bromodeoxyuridine (BrdU). Additionally, in order to determine whether BrdU-labeled cells differentiate into vascular endothelial, astroglial and microglial cells, double immunofluorescent studies were performed for BrdU and NAGO, GFAP or Iba1. Furthermore, we observed axonal growth into the scaffold by double staining for BrdU and MAP2.

Results: The control brain showed distortion of the cerebral cortex and enlargement of the ipsilateral lateral ventricle (Fig.1 A, C). On the other hand, the GPSM implanted into the lesion remained at the same site for 60 days, kept integrity of the brain shape, and attached well to the surrounding brain tissues (Fig.1 B, D). Marginal cavities of the scaffolds were occupied by newly formed tissue (Fig.1 E), where newly produced vascular endothelial, astroglial and microglial cells were found with BrdU positivity. The numbers of those cells were dose-dependently increased with bFGF and EGF. Extension of dendrites was also found from the surrounding cerebral cortex to the newly formed tissue, especially with addition of bFGF and EGF.

Conclusion: The present study showed that GPSM had biocompatibility. It can thus serve as a potential scaffold for cell migration, angiogenesis and dendrite elongation with dose-dependent effects of bFGF and EGF.

NEURAL STEM CELLS EXERT A NEUROPROTECTIVE ACTION BY CHANGING THE ISCHEMIC MICROENVIRONMENT

Principato Maria-Cristina¹, Carmen Capone¹, Simona Frigerio², Stefano Fumagalli¹, Maurizio Gelati², Claudio Storini¹, Mery Montinaro¹, Marco De Curtis³, Eugenio Parati², De Simoni Maria-Grazia¹

¹Laboratory of Inflammation and Nervous System Diseases, Mario Negri Institute, Milano, Italy, ²Laboratory of Neurobiology and Neuroregenerative Therapies, Carlo Besta Neurological Institute, Milano, Italy, ³Clinical Epileptology and Experimental Neurophysiology Unit, Carlo Besta Neurological Institute, Milano, Italy

Cell replacement strategies are particularly promising candidate therapies for stroke, a condition in which a widespread loss of different cell types occurs that in principle may benefit from exogenous stem cells replacement of lost or damaged cells. In the present study we have used a murine model of focal, transient ischemia to investigate if: 1) infused NSC are able to decrease ischemic injury and restore brain function; 2) these cells induce changes in the environment in which they are infused; 3) changes in brain environment consequent to transient ischemia are relevant for NSC action.

Neurosphere-derived stem cells were obtained from brains of C57BL/6 newborn mice. NSC were infused intracerebroventricularly 4h or 7d after 30min focal ischemia obtained by intraluminal MCAO. Behaviour was evaluated in an open field paradigm at day 7 or 14 after ischemia/reperfusion. The impairment in the numbers of rears and of contacts with objects, measures related to exploratory behavior and sensory-motor activity respectively, observed in ischemic mice, was significantly improved in ischemic mice receiving NSC. Neuronal loss, evaluated in striatum and in cortex, was significantly reduced at day 7 and day 14 in ischemic mice infused with NSC at 4h compared to controls and to ischemic mice receiving NSC at 7d. Infusion of adult fibroblasts in the same experimental conditions was not effective. Assessment of infused NSC distribution revealed that they migrated from the ventricle to the brain parenchyma, they were observable up to day 14, but they progressively decreased in number. In mice receiving NSC at day 7 and in sham-operated mice, few cells located near the injection site could be observed only at 24h, indicating that the survival of these cells in brain tissue relates to the ischemic environment. The mRNA expression of trophic factors such as IGF-1, VEGF-A, TGF-alpha1, BDNF and SDF-1beta increased 24h after NSC infusion in ischemic mice treated with NSC at 4h compared to sham-operated and to mice receiving NSC at 7d. Ischemic brains infused with NSC showed higher activation of microglia/macrophages compared to sham-operated mice receiving NSC or to ischemic brains receiving saline injection showing that microglia/macrophages activation is required for NSC action.

In conclusion, neurosphere-derived NSC are able to reduce functional impairment and neuronal damage after ischemia/reperfusion injury. The reciprocal interaction between NSC and the ischemic environment is crucial for NSC protective actions. Indeed NSC survival is enhanced by the ischemic environment and NSC can induce the activation of chemokine and neurotrophic factors. NSC protective effects also relates to the interaction with activated microglia/macrophage. Based on these observations we propose that NSC exert their protective effect by a bystander control of the ischemic environment.

THERAPEUTIC TIME WINDOW FOR INTRAVENOUS TRANSPLANTATION OF HUMAN NEURAL STEM CELLS AFTER CEREBRAL ISCHEMIA IN ADULT RATS

Miyeoun Song, Yong-Ju Kim, Yoon-Ha Kim, Jina Roh, Byung-Woo Yoon

Department of Neurology, Clinical Research Institute, Seoul National University Hospital, Seoul, South Korea

Introduction: It was reported that human neural stem cells (hNSCs) improve functional recovery in the adult rats with experimental focal ischemia when administered one day after stroke. However, the optimal time to administer these cells after cerebral ischemia has not been established. In this study, we intravenously delivered hNSCs at various time after induction of ischemia in the rat middle cerebral arterial occlusion (MCAo) model to determine specific therapeutic time for efficacy.

Methods: After MCAo, Sprague-Dawley male rats weighing 250-300 g were divided into ischemia-only and ischemia-hNSC groups. In ischemia-hNSC group, hNSCs (4×106 cells) were intravenously injected at different time points (3hr, 1d, 3d, 7d, 14d, and 28d after MCAo, respectively). After transplantation, rats were evaluated for 4 weeks (at 1 day before and after ischemia, and weekly thereafter) using modified neurological severity score (mNSS) and cylinder test. To measure the infarction size, Nissl staining was used.

Results: Histological analysis at 28 days after transplantation revealed reduced infarction volume comparing with the ischemia-only group. Moreover, infarction volume of 1d-ischemia-hNSC group (150±37 mm3, n=6) was significantly decreased (p=0.02) as compare with these of 28d-hNSCs-ischemia group (202±39 mm3, n=6). On mNSS, 3hr (4.5±2.2, n=8, p=0.028) and 1d-ischemia-hNSC groups (3.5±2.3, n=6, p=0.014) showed better performance from 4 weeks after the transplantation than control group (6.9±1.5, n=7). Conclusion: Our study demonstrates that intravenous administration of hNSCs 3hr to 1 day after stroke could reduce the size of infarction, and result in functional improvement.

Grant support: This study was supported partly by a grant from the Korea Health 21 R & D Project, Ministry of Health & Welfare (Project No.: A040004), and partly by a National R & D Program Grant of The Ministry of Science and Technology. (Project No. SC-3111), Republic of Korea.

CHARACTERIZATION OF A FOCAL EAE-MOG MODEL FOR THE STUDY OF REACTIVATION OF CNS LESIONS BY A SYSTEMIC INFLAMMATORY RESPONSE

Sebastien Serres¹, Yanyan Yiang², Sandra Amor³, Daniel C. Anthony², Nicola R. Sibson¹

Experimental Neuroimaging Group, Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK, Experimental Neuropathology Laboratory, Department of Pharmacology, University of Oxford, Oxford, UK, Department of Immunobiology, Biomedical Primate Research Centre, Rijswijk, The Netherlands

Multiple sclerosis (MS) is a chronic inflammatory disorder in which 70% of MS patients display a relapsing-remitting form of the disease. It is thought that bacterial and viral infections may trigger relapses and progression of MS. In this study, our aim was to characterise a focal EAE-MOG model targeted in the brain by intracerebral injection of cytokines, and to determine whether this lesion could be reactivated by a systemic inflammatory response. The lesions were monitored by Magnetic resonance imaging (MRI) and immnunostaining.

Male Lewis rats were immunised by subcutaneous injection of MOG peptide (35-55) or MOG protein. Three weeks later, a solution containing 1.45µg of TNFα + 1µg of IFNγ (in 2 · 1 sterile saline) was injected into the corpus callosum of the same animals. Subsequently, animals underwent MRI at days 0, 1, 7, 14, 21 and 28. T1- (pre- and post-gadolinium) and T2-weighted images, as well as apparent diffusion coefficient and regional cerebral blood volume (rCBV) maps were obtained using a 7T horizontal bore magnet. Four weeks after cytokine injection, the animals were challenged systemically by intraperitoneal injection of lipopolysaccharide (LPS; 100 µg/kg) and underwent further MRI examinations 6 and 24h later. Further animals were perfusion-fixed at the same timepoints as the MRI for immunohistochemical analysis.

At 1 day post injection, we observed an acute effect of the cytokines with breakdown of the BBB, increased rCBV and ADC changes in the injected hemisphere. We also observed a significant disruption of the signal on T2-weighted images in the corpus callosum at the injection site. The longitudinal study of T2-weighted images revealed the presence of a chronic hypointense signal in the striatum below the injection site and an increase in ventricle size with incomplete recovery of the signal from the corpus callosum [Figure 1a]. ED1 immunostaining correlated with this hypointense signal, suggesting involvement of activated macrophage in this effect. The disruption and remission of the corpus callosum structure is correlated to an inflammation process as revealed by ED1 staining. The study of the ongoing disease also revealed an increase in the left/right ratio of rCBV followed by recovery to normal values by 28 days [Figure 1b].

Induction of a systemic inflammatory response with LPS challenge, resulted in an increase in the left/right rCBV ratio at both 6 and 24h after LPS injection [Figure1b], together with recruitment of both macrophages as revealed by ED1 staining [Figure1c]. These findings show that the reactivation of inflammatory processes by a bacterial endotoxin within a previously quiescent CNS lesion can be obtained in a clinically-relevent model of MS.

EXPRESSION OF GLUCOSE TRANSPORTER MOLECULES IN ADULT NEURAL STEM CELLS: INFLUENCE OF HYPOXIA AND HYPERGLYCEMIA

Martin Henrik Maurer, H. Kenny Geomor, Heinrich F. Burgers, Dominik W. Schelshorn, Wolfgang Kuschinsky

Department of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany

Background and aims: Glucose constitutes the main substrate for the nervous tissue. It is transported by specialized glucose transporter molecules, the GLUT family. The facilitative glucose transport protein GLUT1 is expressed at the blood-brain barrier and in astrocytes, whereas the GLUT3 isoform is responsible for the neuronal glucose uptake. Almost nothing is known about the energy metabolism of neural stem cells in the adult brain with regard to functional similarities and differences compared to neurons and astrocytes. In the current study, we addressed the questions, whether neural stem cells express glucose transporter molecules and whether this expression can be regulated by typical pathophysiological stimuli such as hypoxia and exposure to high glucose concentrations.

Methods: Neural stem cells were isolated from the subventricular zone of adult rat brains and cultured using standard protocols. For hypoxia, neural stem cells were incubated in a hypoxia chamber for 24 h in 5% CO2, 95% nitrogen. For hyperglycemia, the glucose concentration in the cell culture media was doubled. GLUT1 and GLUT3 expression was compared by RT-PCR, Western blotting, and immunocytochemistry.

Results: Neural stem cells expressed mRNA for GLUT1 and GLUT3, but not for GLUT2 and GLUT4. We found co-expression of GLUT1 and GLUT3 in the same cells throughout, indicating that there are no subpopulations of neural stem cells expressing only one glucose transporter type. Both hypoxia and high glucose concentrations increased GLUT1 and GLUT3 expression on the mRNA and protein levels. Whereas GLUT1 mRNA concentrations increased moderately (about 1.4-fold), GLUT3 mRNA expression increased strongly (about 10-fold) in hypoxia, hyperglycemia, or the combination thereof. On the protein level, GLUT1 showed a high baseline expression which did not change under the experimental conditions, whereas GLUT3 increased about 2-fold.

Conclusions: The data show the existence of GLUT1 and GLUT3 in neural stem cells and a moderate degree of regulation in their expression. Since neural stem cells are capable of differentiating both into astrocytes (expressing GLUT1), and neurons (expressing GLUT3), we interpret the co-expression of GLUT1 and GLUT3 in the neural stem cells as sign of maintained multipotency.

Acknowledgements: Supported by the German Ministry of Education and Research (BMBF) within the National Genome Research Network NGFN-2 (01GS0496, MHM and WK), the German Research Foundation (DFG, MA 2492/2-2, MHM and WK) and the Estate of Friedrich Fischer (MHM).

NEURODEVELOPMENTAL STUDIES IN C57B/L6 MICE ASSESSED BY IN VIVO DTI

Halima Chahboune^1,2, Laura. R. Ment², William. B. Stewart³, Douglas. L. Rothman⁴, Fahmeed Hyder⁵

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA, ²Department Of Pediatrics, Yale University, New Haven, CT, USA, ³Department of Surgery, Yale University, New Haven, CT, USA

DTI is a noninvasive technique that allows studies of neurodevelopment in transgenics or mutants. Prior in vitro mouse brain DTI studies at high magnetic fields that have revealed tissue anisotropy variations were conducted in fixed brain with DTI signal acquisitions requiring several hours to a day. The purpose of this study was to assess the feasibility of in vivo DTI to quantify maturation.

VOXEL-WISE DETECTION OF AMYLOID DEPOSITION WITH PITTSBURGH COMPOUND B (PIB) IN NORMAL AGING: A PARTIAL VOLUME CORRECTION-BASED APPROACH

Hiroto Kuwabara¹, Anil Kumar¹, Jitka Sojkova^1,2, Olivier Rousset¹, Mohab Alexander¹, Weiguo Ye¹, Daniel Holt¹, Robert Dannals¹, Susan M. Resnick², Dean F. Wong^1,3

¹Department of Radiology, Johns Hopkins University, Baltimore, MD, USA, ²National Institute On Aging, Baltimore, MD, USA, ³Department of Psychiatry, Johns Hopkins University, Baltimore, MD, USA

Background and Aims: Identification of voxel-wise amyloid deposition in normal aging may be hampered by high inter-subject variability and the apparent greater retention of PIB in white matter than gray matter. We employed a new approach to compare observed binding potential (BP) maps to partial volume correction (PVC)-predicted BP maps that served as individuals' own control, to cope with inter-subject variability while accounting for gray-white matter composition at the voxel level.

Methods: Twenty-eight nondemented participants of Baltimore Longitudinal Study of Aging (age: 79.2 ± 8.0; 11 males and 17 females) underwent one 90-min dynamic PIB-PET scan. BP maps of PIB scans were constructed by the bolus-plus-infusion transformation (BPIT; Kuwabara et al., 2002). Gray matter, white matter, and cerebro-spinal fluid space segments were identified on MRI, transferred to PET space, and smoothed by PET camera specific Gaussian kernels to reproduce the partial volume effect. The resulting geometric transformation matrix (M; Rousset et al., 1998), is an n by 3 matrix where n is the number of voxels covering the three segments. PV-corrected BP values of the three segments (T, a 3 by 1 matrix) were obtained by left matrix division using Gaussian elimination. The PVC-predicted BP map was constructed by inserting the M*T product to the PVC voxels and −1 to the remaining voxels. Thus, PVC-predicted BP maps represent weighted mean BP values of the three segments at the voxel level assuming a homogenous BP value for each segment throughout the brain. Comparison of observed BP maps to PVC-predicted BP maps, using SPM2, allows for exploring regions that show significant deviation from the homogeneity assumption.

Results: PV-corrected BP values were 0.14 ± 0.15 (unit-less; mean ± SD) for gray matter, 0.54 ± 0.10 for white matter, and −0.73 ± 0.12 for cerebro-spinal space. Table lists four clusters (observed BP values > PVC-predicted BP values) identified with SPM (p<0.05, corrected; volume > 1 ml) across the 28 subjects.

Conclusions: The approach successfully identified four regions of relative increases in PIB binding in nondemented older adults with stringent statistical criteria. The results suggest the PVC-based approach can detect subtle increases in PIB binding in nondemented subjects with a low cortical BP value despite a several fold higher white matter BP value. Further studies on demented populations will be required to understand the clinical significance of PIB accumulation at this early stage of amyloid deposition.

Grant Support: Intramural Research Program, NIA, NIH; N01-AG-3-2124; K24 DA00412 (DFW)

CEREBRAL ACTIVATION AND DEACTIVATION DURING NAVIGATION IN A VIRTUAL REALITY ENVIRONMENT IN PATIENTS WITH MCI. AN O15-WATER PET ACTIVATION STUDY

Marc Daniel Wermke¹, Timo Grimmer², Isabell Stangier¹, Stephan Nekolla¹, Alexander Kurz², Markus Schwaiger¹, Alexander Drzezga¹

¹Nuclear Medicine Department, Technische Universitat, Munich, Germany, ²Psychiatry Department, Technische Universitat, Munich, Germany

Background and aims: In patients with Alzheimer's disease (AD), characteristic abnormalities of cerebral activation during cognitive performance have been previously observed. The aim of the current study was to assess the cerebral activation and deactivation patterns during navigation in a virtual reality (VR) environment in healthy volunteers and patients with mild cognitive impairment (MCI) and to correlate the cerebral activation with navigation speed.

Methods: A healthy control group (n=11, right handed, 3 female, age 67+/−10 years) and a matched group of 11 MCI-patients (4 female, 71+/−9 years) underwent a navigation task in a computer-based 3D VR-labyrinth, based solely on visual clues. During active navigation 8 O15-water PET-scans were performed (370 MBq i.v.-bolus). In each subject navigation speed (time from start to destination) was measured. In addition, an FDG-PET-scan at rest condition was performed in all subjects. Statistical analysis (correlation analysis of O15 water activation with navigation speed and group comparison of O15-water and F-18 FDG PET scans) was carried out, using SPM02 software (Wellcome Inst., London, UK), p<0.05, FDR-corrected.

Results: Analysis of FDG PET-data revealed a significantly lower metabolism exclusively in the posterior cingulate cortex in the MCI-group compared to controls. Controls showed a significantly better navigation performance (required time: 67±46 sec, MCI-group 139±53 sec). Analysis of O15 water PET showed a significantly different pattern of activation in MCI and controls. Stronger activations in controls were found in the left cerebellum, the occipital lobe (Brodman area [BA] 18,19) and the sensorimotor cortex (BA 3,4) and in MCI in the prefrontal cortex (BA 9), the SMA (BA 6) and the anterior cingulate cortex (BA 32). Relevant deactivation of auditory cortical regions (BA 41,42,22) ?representing cross-modal inhibition- was detected exclusively in controls. In addition different patterns of correlation between navigation speed and regional brain activation were found in the two groups. Navigation speed correlated with activation in left cerebellum, left premotor cortex (BA 6) and right thalamus in controls and in MCI with activation in left temporal cortex (BA 20), left posterior cingulate cortex (BA 31) and right cerebellum.

Conclusions: Several cortical key regions have already been associated with human navigation in healthy subjects. In our study MCI showed activations differing from the activations found in controls during navigation. Also the deactivation of task-irrelevant cerebral regions observed in controls, was absent in MCI. This suggests that MCI is characterized by a progressive impairment of cross-modal cerebral deactivation-functions, resulting in a decreased ability to direct attention primarily to the relevant cognitive modality. In addition, a correlation analysis revealed a significantly different network associated with successful navigation-performance in MCI compared to controls. These results might indicate compensatory changes or functional deficits in MCI due to ongoing neurodegeneration.

FAMILIAL ALS WITH A NOVEL L126S MUTATION IN THE SOD1 GENE SHOWING MILD CLINICAL FEATURES AND LOW PENETRANCE IN FEMALES

Yasushi Takehisa, Tetsuro Murakami, Kentaro Ohta, Tomoko Kurata, Nobutoshi Morimoto, Makiko Nagai, Tatsushi Kamiya, Koji Abe

Department of Neurology, Okayama University Graduate School of Medicine and Dentistry, Okayama, Okayama, Japan

Back ground: Mutations in the SOD1 gene are responsible for approximately 25% of familial amyotrophic lateral sclerosis (ALS) cases. However, the correlation between the clinical and pathological features and the various SOD1 gene mutations has not been well characterized.

Objective: To screen the SOD1 gene in search of potential mutations and to obtain clinical and pathological datas for three Japanese families with ALS.

Design: Clinical histories and neurological findings, gross and microscopic pathological features, and DNA analysis of the SOD1 gene.

CEREBRAL GLUCOSE METABOLISM BY PET IN MILD COGNITIVE IMPAIRMENT

Jyrki Kuikka^1,4, Anne Hamalainen², Tuija Kangasmaa¹, Minna Rusanen², Eini Niskanen², Ritva Vanninen³, Hilkka Soininen^2,4

¹Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital, Kuopio, Finland, ²Department of Neurology, Kuopio University Hospital, Kuopio, Finland, ³Department of Clinical Radiology, Kuopio University Hospital, Kuopio, Finland, ⁴Kuopio University Neuroscience Centre, Kuopio, Finland

STATINS PROTECT NEURONS AND ASTROCYTES AGAINST ISCHEMIA-LIKE INSULTS, BUT NOT MICROGLIA-INDUCED TOXICITY

Hualong Ma^1,2, Zhen Zheng^1,2, Maya Koike^1,2, Scott Zamvil¹, Midori Yenari^1,2

¹Department of Neurology, University of California San Francisco, San Francisco, CA, USA, ²Department of Neurology, Veteran Affairs Medical Center, San Francisco, CA, USA

It has been reported by several groups that statins, a class of cholesterol lowering drugs, are neuroprotective. The underlying mechanism for such protection is thought to be due, in part, to its ability to enhance cerebral blood flow by upregulating endothelial nitric oxide synthase (eNOS). However, recent work has also suggested an anti-inflammatory effect. We evaluated the effects of atorvastatin on neuronal and astrocyte cell death as well as on microglial activation in an in vitro ischemia model. Primary neuronal cultures were prepared from cortices of E16 Swiss Webster mice and used for experiments at 12 DIV. Primary microglial and astrocyte cultures were obtained from 1 day old mice. Co-cultures of astrocytes with neurons (AN) with or without microglia (NAM) were prepared by seeding 2×104 cells/ml (neuron), and 2×105 cells/ml (microglia) on top of astrocytes. Cocultures were used for experiments 48 hours (microglia) and 12 days (neuron) later. Cultures were subjected to oxygen and glucose deprivation (OGD) for 90 minutes (neuron containing cultures) and 6 hours (astrocytes alone) followed by reperfusion for 24 hours. Overall cell death was evaluated by using LDH assay and trypan blue staining. Neuronal death was estimated using a MAP-2 ELISA based assay. Following OGD, treatment of primary astrocytes (A) with 1µM atorvastatin led to less cell death by 90% (P=0.001) compared to control. Coculture with microglia (AM) showed increased cell death by 2 fold (P<0.01). However, atorvastatin failed to affect the extent of microglia-exacerbated cell death (NS). Atorvastatin treatment of primary neuron cultures (N) also showed reduced cell death by 23% (P<0.001). Interestingly, the addition of microglia to neuron cultures (NM) did not increase cell death, as was seen with astrocytes. Atorvastatin treatment of NM cultures failed to affect cell death compared to untreated NM and N cultures. In cocultures of neuron and astrocytes (AN), statin treatment decreased cell death by 38% (p < 0.01). Although addition of microglia to AN coculture (ANM) increased neuron cell death by 2 fold (P <0.005), statin treatment again showed no effect on microglia-potentiated neuronal cell death. In another experiment, we measured the expression of MHC-class II molecules on cultured BV2 cells (murine microglia) using FACS analysis. After stimulation with LPS (10 µg/ml) for 24 h, microglia showed increased activation by FACS analysis of MHC-class II antigen expression. Treatment of these cells with atorvastatin failed to inhibit MHC-II expression. In conclusion, our data suggest a mechanism of direct protection of neuron and astrocytes against ischemia-like insults in culture by atorvastatin. This protection appears to be independent of vascular changes/cerebral blood flow or microglial activation/toxicity. We further show that microglia alone are not toxic to neurons under conditions of OGD, but appear to require the presence of astrocytes to increase neuronal death.

TISSUE SODIUM CONTENT AND IRREVERSIBLE ISCHEMIC TISSUE DAMAGE DURING FOCAL BRAIN ISCHEMIA

George LaVerde¹, Edwin Nemoto¹, Charles Jungreis³, Alexander Kharlamov², Steven Jones², Costin Tanase¹, Fernando Boada¹

¹Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA, ²Department of Anesthesiology, Allegheny General Hospital, Pittsburgh, PA, USA, ³Department of Radiology, Temple University, Philadelphia, PA, USA

HMG COA REDUCTASES REDUCES ISCHEMIC BRAIN INJURY OF WISTAR RATS THROUGH DECREASING OXIDATIVE STRESS ON NEURONS

Shoko Nagotani, Takeshi Hayashi, Kentaro Deguchi, Hanzhe Zhang, Yoshihide Sehara, Atsushi Tsuchiya, Toru Yamashita, Violeta Lukic, Tatsushi Kamiya, Koji Abe

Department of Neurology, Okayama University Graduate School of Medicine and Dentistry, Okayama, Japan

Background: Statins possess many effects besides cholesterol lowering effect, which is called pleiotrophic effects. They show neuroprotective effect against ischemic damage, but how they protect neurons is not exactly made clear. As they have anti-oxidative property and oxidative stress is implicated in ischemic brain damage, we speculated that anti-oxidative property of statins plays am important role. We therefore investigated statins' influences on the oxidative neuronal damage in the brain after ischemia.

Methods: We used male Wistar rats (10 weeks old, 230 g). Atorvastatin (20 mg/kg), pitavastatin (10 mg/kg), simvastatin (20 mg/kg), or vehicle (methyl cellulose) was per-orally administered for 14 days, and then 90 min of middle cerebral artery occlusion was imposed on the rats. At 24 hr after CBF restoration, the animals were anesthetized and blood samples were collected. Triglyceride (TG), total-cholesterol (T-cho), high-density lipoprotein-cholesterol (HDL-cho), and low-density lipoprotein-cholesterol (LDL-cho) were measured using these samples. Just after blood collection, the animals were decapitated and the brains were removed. In order to evaluate oxidative stress, production of 4-hydroxynonenal (HNE) and 8-hydroxy-2′-deoxyguanosine (8-OHdG) were investigated by immunohistochemistry. Infarction volume was investigated using same sections by cresyl violet staining.

Results: In the vehicle group, massive infarction was confirmed and HNE and 8-OHdG are robustly produced. In the statins-treated group, the infarction was smaller and the HNE and 8-OHdG production was less prominent than the vehicle group (Figure). Among the statins investigated, simvastatin was most effective for reducing oxidative stress and infarction volume. On the other hand, statin-treatment did not decrease the cholesterol levels in the rats.

Conclusion: Statins possess anti-oxidative proterty, which is implicated in neuroprotection against ischemic injury. Drug's lipophilic property correlated with their neuroprotective property, suggesting that entering into brain is critical for exerting their effects.

PGE2-EP3 RECEPTOR STIMULATION LEADS TO EXACERBATION OF STROKE AND EXCITOTOXIC INJURY

Muzamil Ahmad¹, Abdullah Ahmad¹, Sylvain Dore^1,2

¹Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, SOM, Baltimore, MD, USA, ²Department of Neuroscience, Johns Hopkins University, SOM, Baltimore, MD, USA

Background: Inflammation has been shown to play a major role in the pathological response and outcome of stroke and other central nervous system disorders. Inflammation is mediated at least in part by prostaglandins (PGs), which are produced through the cyclooxygenase (COX) pathway. PGE2 mainly binds to a family of G-protein-coupled receptors known as EP receptors. The EP3 receptor mediates the activation of several signaling pathways, leading to changes in cAMP levels, calcium mobilization, and activation of phospholipase C. Our goal in this study was to determine the role of EP3 in brain injury induced by ischemia-reperfusion injury.

Method: Using 25-30g male C57BL/6 mice, we investigated the effects of a selective EP3 agonist on 90-min middle cerebral artery occlusion (MCAO)-induced infarct volume, relative cerebral blood flow (CBF), mean arterial blood pressure (MABP), and other physiological parameters. Furthermore, since excitotoxicity is involved in the resulting injury caused by ischemia/reperfusion, we went on to determine the effect of EP3 agonist on lesion size after NMDA-induced toxicity. All animal protocols were in accordance to NIH guidelines and approved by the JHU Animal Care and Use Committee.

Results: Intracerebroventricular injection of the EP3 agonist before MCAO significantly enhanced hemispheric infarct volumes. Infarct volume increased in a dose-dependent manner as compared with the vehicle-treated group (n = 9/dose). As estimated by laser-Doppler flowmetry, relative CBF decreased significantly in each group of mice after insertion of the nylon monofilament but the differences in CBF reduction among the treated groups were not significantly different. No statistically significant differences in pH, PaCO2, PaO2, or MABP were detected before ischemia, during ischemia, or during reperfusion among the groups of mice. On the other hand, mouse pretreatment with the EP3 agonist dose dependently aggravated the brain injury caused by NMDA injection.

Conclusion: Some conflicting findings have been reported in the literature in regard to the role of EP3 receptors in health and disease; some investigators have reported that EP3 affords protection, whereas others state that it aggravates toxicity. Considering that PGE2 is often called the “pro-inflammatory” prostaglandin, we have used the mouse models of cerebral ischemia and induced excitotoxicity to elucidate and better understand the EP3 function. We found that stimulation of the EP3 receptor with a selective agonist significantly exacerbated the lesion volume following NMDA injection and the infarct volume after transient ischemia, while none of the physiological parameters studied were significantly affected. To our knowledge, this is the first report to show that stimulation of the EP3 receptors exacerbates brain damage; thus deletion or pharmacologic blockade of this receptor might limit brain damage induced by pro-inflammatory COX metabolites, and these receptors could be used as therapeutic targets for the prevention or reduction of ischemia-reperfusion-associated brain injury. (This study was supported by an AHA postdoctoral fellowship and by NIH grants.)

BENEFITS OF PRE-TREATMENT WITH EDARAVONE IN TPA INTRAVENOUS THERAPY FOR ACUTE CEREBRAL INFARCTION

Yoshifumi Teramoto¹, Yoshikazu Nakajima², Mamoru Ito², Makoto Dehara², Takao Soda³, Tomonari Yabuuchi², Koji Tokiyoshi², Toshiki Yoshimine⁴, Mamoru Taneda¹

¹Department of Neurosurgery, Kinki Universiy School of Medicine, Osaka-Sayama, Japan, ²Department of Neurosurgery, Rinku General Medical Center, Izumisano, Japan, ³Department of Neurology, Rinku General Medical Center, Izumisano, Japan, ⁴Department of Neurosurgery, Osaka Universiy School of Medicine, Suita, Osaka, Japan

[Objective] In thrombolytic therapy, the administration of edaravone, a free radical scavenger, before thrombolysis can potentially remove the free radicals that are generated upon effective recanalization, inhibit hemorrhagic infarct and severe cerebral edema, and improve vital and functional prognosis.

[Cases and Methods] Of the 60 patients who have been treated for acute cerebral infarction at our clinic since November 2005, when tPA intravenous therapy first became available for this condition in Japan, and who ranged between 5 and 22 on the National Institute of Health Stroke Scale (NIHSS) at the time of first visit, 13 patients (seven men and six women aged between 56 and 79 years; mean 71.5 years) who underwent tPA therapy were examined in this study. Of these 13 patients, 54% had cardiogenic cerebral infarction, 38% had cerebral atherothrombosis, and 8% had lacunar infarction (according to TOAST classification). Towards evaluating the benefits of pre-treatment with edaravone, we used MRI and CT to assess the change in cerebral edema over time in patients who underwent tissue-type plasminogen activator (tPA) intravenous therapy with antecedent intravenous administration of edaravone.

[Results] The tPA intravenous therapy was started 48 to 147 (mean: 108) minutes after the onset of stroke symptoms. Recanalization was achieved in 38% of the patients, was not achieved in 31%, and the status of recanalization remained unknown in 31%. Symptoms improved in five patients, did not change in four patients and were aggravated in four patients (including one patient with the complication of hemorrhage). In all 11 patients who received edaravone (30 mg) intravenously before (5 to 45 minutes before; mean 20.7 minutes) tPA, cerebral edema was limited to the area surrounding the infarction lesion and there was no rapid exacerbation of consciousness or neurologic symptoms secondary to extensive brain swelling with midline-shift or severe cerebral edema secondary to hemorrhagic infarct. Cerebral hemorrhage at a site distinct from the infarction lesion was observed in one patient with midline-shift developed in one of the two patients who received edaravone after (5 and 25 minutes after) intravenous injection of tPA.

[Discussion] Evaluation of the clinical cases of acute cerebral infarction treated with tPA intravenous therapy suggests that pre-treatment with edaravone can inhibit white matter injury or cerebral edema due to recanalization. Future studies including randomized controlled trials on a large number of cases are desired.

CEREBROVASCULAR CO2 REACTIVITY EXAMINED AFTER CHRONIC PRETREATEMENT OF AN ANGIOTENSIN II TYPE 1 ANTAGONIST IN RATS WITH ACUTE ISCHEMIC STROKE

Shigenori Mizusawa¹, Yasushi Kondoh¹, Kazuhiro Nakamura¹, David Wright¹, Svend Strandgaard², Iwao Kanno^3,1

¹Akita Research Institute for Brain and Blood Vessels, Akita, Japan, ²Department of Nephrology, Herlev University Hospital, Copenhagen, Denmark, ³Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan

INTRODUCTION: Pretreatment of Angiotensin II type 1 (AT1) antagonist, candesartan, reduces the infarct size in middle cerebral artery occlusion (MCAO) model rats [1]. Candesartan is thought to affect larger cerebral resistance vessels and lower the limit of cerebral blood flow (CBF) autoregulation. The reactive site of CO2 is smaller resistance vessels, therefore treatment with candesartan should not affect cerebrovascular CO2 reactivity. We examined CO2 reactivity under chronic pretreatment of candesartan to test this relationship.

METHOD: Twelve Sprague-Dawley rats were administrated with candesartan (0.3mg/kg/day;n=6) or vehicle (0.1N Na2CO3;n=6) for 14 days prior to experimentation by using subcutaneously implanted osmotic minipumps. Ischemic regions were induced by occluding the left middle cerebral artery with embolic thread and an endotracheal tube was inserted for artificial ventilation and CO2 inhalation. After preparation, the animals were set in the bore of a 4.7T MR spectrometer and a CBF image was acquired using a continuous arterial spin labeling (CASL) method 3 hours after the MCAO [2]. Inspired CO2 was then adjusted to 0.8% and 1.6% of ventilated volume and CASL scans were acquired after CBF stabilized. The CO2 reactivity was calculated by exponential curve fitting using the least squares method and the ischemic area was defined using the apparent diffusion coefficient (ADC).

RESULTS: The figure shows mean CBF values measured at various PaCO2 levels. In the healthy hemisphere, the CO2 reactivity in animals treated with candesartan (2.1%/mmHg) was significantly lower than vehicle (3.3%/mmHg). CBF response to PaCO2 was well conserved in the ischemic penumbra (70 % < ADC < 90% of mean value in control hemisphere) especially in the candesartan group.

DISCUSSION: Vasodilatation limits further vessel extension resulting in decreased CO2 reactivity. The results suggest that candesartan induced persistent vasodilation of smaller resistance vessels hence decreased response to CO2. A smaller infarction observed in MCAO animals treated with candesartan may be due to persistent vasodilation of smaller resistance vessels in the ischemic penumbra.

MECHANISMS OF ANGIOTENSIN II-INDUCED ENDOTHELIAL DYSFUNCTION: ROLE OF INFLAMMATION, INTERLEUKIN-6 AND INTERLEUKIN-10

Sean P. Didion¹, Dale A. Kinzenbaw¹, Laura I. Schrader¹, Frank M. Faraci^1,2

¹Department of Internal Medicine, University of Iowa, Iowa City, IA, USA, ²Department of Pharmacology, University of Iowa, Iowa City, IA, USA

Background and Aims: Hypertension has an enormous negative impact on the carotid artery and cerebral circulation. Angiotensin II (Ang II) plays a major role in hypertension, in part by promoting vascular inflammation. The goal of this study was to examine the role of interleukin 6 (IL-6, a pro-inflammatory cytokine) and IL-10 (an anti-inflammatory cytokine) in vascular dysfunction produced by Ang II.

Methods: Responses of carotid arteries from wild-type, IL-6 and IL-10 deficient mice (IL-6 -/and IL-10 -/−, respectively) were examined following either overnight incubation with Ang II (1 or 10 nmol/L) or chronic infusion of Ang II in vivo (1.4 mg/kg per day for 10-14 days).

Results: Baseline blood pressure and Ang II-induced hypertension were similar in control and IL-10 -/− mice. Ang II-infusion for 10 days produced modest impairment of endothelial function in control mice but marked impairment in IL-10 -/− mice [e.g., 100 µmol/L acetylcholine (ACh), an endothelium-dependent agonist, produced 102±3 and 55±1% relaxation in vessels from vehicle-and Ang II-infused IL-10 -/− mice, respectively]. Overnight treatment with 1 nmol Ang II had no effect on responses to ACh in vessels from control mice but markedly reduced responses to ACh in IL-10 -/− (e.g., 100 µmol/L ACh produced 82±8 and 37±5% relaxation in vehicle- and Ang II-treated arteries, respectively). This effect of Ang II in IL-10 -/− mice was reversed by PEG-SOD, a scavenger of superoxide. Vascular superoxide levels increased following local treatment with Ang II in IL-10 -/− but not control mice. In studies of IL-6, baseline blood pressure and Ang II-induced hypertension were similar in control and IL-6 -/− mice. Ang II-infusion (14 days) produced impairment of responses to ACh in arteries from control, but not IL-6 -/− mice (e.g., 100 µmol/L ACh produced 51±7 and 80±5% relaxation in vessels from control and IL-6 -/− mice, respectively). IL-6 deficiency also protected against endothelial dysfunction produced by acute treatment with 10 nmol/L Ang II, a concentration that produced endothelial dysfunction in arteries from control mice. Endothelial dysfunction could be reproduced in vessels from IL-6 -/mice following overnight incubation with Ang II plus recombinant IL-6 (0.1 nmol/L). Responses to nitroprusside, an endothelium-independent agonist, were not affect by either genotype or treatment, suggesting that the effect of Ang II for endothelium was selective. Consistent with this functional data, Ang II-induced increases in vascular superoxide levels were attenuated in IL-6 -/mice.

Conclusions: Taken together, our findings provide the first evidence that IL-6 promotes and IL-10 limits Ang II-mediated oxidative stress and endothelial dysfunction. These effects were found to occur both in vivo and in vitro suggesting effects of both cytokines occur within the vessel wall independent of changes in arterial pressure. These findings have implications for mechanisms of hypertension-induced vascular dysfunction and carotid artery disease, a major risk factor for ischemic stroke.

OVEREXPRESSION OF COPPER-ZINC SUPEROXIDE DISMUTASE ATTENUATES CYTOSOLIC PHOSPHOLIPASE A2 ACTIVATION BY BLOCKING THE P38 MAPK PATHWAY AFTER FOCAL CEREBRAL ISCHEMIA

Chikako Nito, Deborah J. Myer, Kuniyasu Niizuma, Pak H. Chan

Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA

Background and purpose: Cytosolic phospholipase A2 (cPLA2) is a key enzyme that mediates arachidonic acid (AA) metabolism and thus causes cerebral ischemia-induced oxidative injury, blood-brain barrier (BBB) dysfunction and edema. Recent reports have shown that p38 mitogen-activated protein kinase (MAPK) is related to phosphorylation and activation of cPLA2 and AA release in a Ca2+/calmodulin-dependent kinase II. However, the involvement of the p38 MAPK pathway in cPLA2 activation, and the role of reactive oxygen species (ROS) in the expression of p38 MAPK/cPLA2 after ischemia-reperfusion (IR) injury in the brain remain unclear.

Methods: To investigate these issues, we used a transient middle cerebral artery occlusion (MCAO) model with rats that overexpress copper-zinc superoxide dismutase (SOD1) and with wild-type rats. After 90 min of MCAO, the cortex of the affected side was removed at 6 and 24 h and 3 and 7 days. We also administered SB203580, a potent p38 MAPK inhibitor, into the rat brains and examined phospho-p38 MAPK expression and cPLA2 levels measured by Western blot analysis and activity assays. Infarct or edema volume was evaluated by 2,3,5-triphenyltetrazolium chloride staining.

Results: Western blot analysis and activity assays demonstrated that expression of phospho-p38 MAPK was increased after reperfusion, maximally on day 1, and that phospho-cPLA2 significantly increased at 1 day (Figure 1; *p< 0.05,). The intraventricular administration of SB203580 provided a significant suppression of phosphorylation and activation of cPLA2 (p< 0.05). Moreover, SOD1 overexpression significantly diminished phosphorylation and activation of both p38 MAPK and cPLA2 (Figure 2; *p<0.05, ##p<0.05, respectively) and reduced cortical infarct or edema volume at 3 days (p < 0.05 and p< 0.01, respectively) after reperfusion.

Conclusions: These results suggest that ROS may induce cPLA2 activity through the p38 MAPK pathway and promote BBB disruption in the brain cortex after IR injury.

APPEARANCE OF INTRA-NUCLEAR MATRIX METALLOPROTEINASES IS INVOLVED IN THE EARLY STAGES OF NEURONAL DEATH FOLLOWING BRAIN ISCHEMIA

Anna Rosell, Eloy Cuadrado, Maria Borrell-Pages, Miriam Navarro, Mar Hernandez-Guillamon, Anna Penalba, Laura Ortega, Joan Montaner

Neurovascular Research Laboratory, Institut De Recerca, Hospital Vall D'Hebron, Barcelona, Spain

Background and aims: Matrix Metalloproteinases (MMPs) are known to play a critical role increasing brain damage after brain ischemia both in humans and in animal models. Although MMPs are best known as being either secreted or transmembrane proteins, recent publications have identified unexpected locations for different MMPs. A variant of MMP-11 has been described as an intracellular active protease and MMP-14/MT1-MMP has been identified in the centrosome. Moreover, nuclear MMP-2 has been identified in cardiac myocytes and nuclear location of MMP-3, and the fact that can induce apoptosis via its catalytic activity, has been recently described in human liver cells. In the present study we aimed to search for MMP expression and cellular location following brain ischemia.

Methods: Several samples from patients who had an ischemic stroke within the previous 5 days were included in the study. On autopsy, by an experienced neuropathologist, morphological features and last available neuroimages were used to guide brain tissue sampling from ischemic ipsilateral hemisphere or from contralateral hemisphere. Finally, 19 samples from 7 ischemic stroke patients (12 infarct samples and 7 contralaterals) were studied. Brain tissue samples from 2 patients that died due to a non-inflammatory pathology were used as controls.

To detect MMPs in human brain homogenates we used a protein array technology (Searchlight), which allowed us to measure MMP-1, MMP-2, MMP-3, MMP-8, MMP-9, MMP-10 and MMP-13 concentrations.

To study the temporal profile of MMP activation in cells, in vivo experiments were conducted in a rat permanent focal MCAO model, occlusion times: 30, 60, 90, 120 minutes and 48 hours (n=3 per time-point).

In situ zymography that detects gelatinase/collagenase activity was carried out on human and rat brain sections, including some controls using a metalloproteinase inhibitor (phenanthroline). Reaction products were examined with fluorescence microscope and using confocal microscopy.

Results: Brain tissue from ischemic patients showed higher levels of MMP-2, MMP-3, MMP-8, MMP9, MMP-10 and MMP-13 in the infarct area as compared with contralateral area (p=0.005, p=0.049, p=0.041, p=0.031, p=0.021 and p=0.001, respectively) while in situ zymography showed a clear collagenase/gelatinase activity at the nucleus of neuronal cells.

Regarding the temporal profile of MMP activation we observed an early nuclear gelatinase/collagense activity (at 30 minutes) that increased together with the occlusion time. This time-dependent response was perceived on the number of cells with nuclear activity and on the intensity. After 60 minutes of occlusion the gelatinase/collagenase activity also appeared in the cytoplasm of the cells following again a time-dependent response.

Finally our controls confirmed our findings since the sham rats did not show nuclear activity and gelatinase/collagenase activity was blocked after treatments with phenanthroline both in human and rat sections.

Conclusions: Several MMPs are up-regulated after stroke as well as a new location of gelatinase/collagenase activity is identified into the nucleus of brain cells at the early stages of brain hypoxia/ischemia. The unexpected location of this activity merits further research in order to identify the responsible matrix metalloproteinases, to understand their contribution to neuronal death and to explore new treatments for stroke.

ISCHEMIC PRECONDITIONING VIA EPSILON PKC INDUCES ERK1/2 AND STAT3 PATHWAYS IN MIXED CORTICAL NEURONS/ASTROCYTES

Eun Joo Kim¹, Jesus Purroy^1,2, Anna M. Planas², Miguel A. Perez-Pinzon¹

¹Department of Neurology and Neuroscience, Cerebral Vascular Disease Research Center, University of Miami Miller School of Medicine, Miami, FL, USA, ²Institut D Investigacions Biomediques De Barcelona, Consejo Superior De Investigaciones Cientificas-Institut D Investigacions Biomediques August Pi I Sunyer, Barcelona, Catalonia, Spain

Ischemic preconditioning (IPC) is a protective mechanism against ischemia/reperfusion injury in brain and heart. Our previous studies have shown that IPC via epsilon PKC activation induced the phosphorylation of extracelluar signal regulated kinase1/2 (ERK1/2) and cyclooxygenase-2 (COX-2) leading to the neuroprotection (1). However, transcriptional signal pathways by which IPC regulates COX-2 in brain remain undefined. The signal transducers and activators of transcription (STATs) upregulate COX-2 expression during cardioprotection (2). A previous study with epsilon PKC knock out mouse model suggests that serine-727-STAT3 phosphorylation was mediated by epsilon PKC activation (3). Therefore, in this study, we tested the hypothesis that IPC via epsilon PKC activates STATs transcriptional signaling pathways.

Methods: Mixed cortical neuron/astrocyte cell cultures are prepared from rat embryo/neonatal rat (18-19 days old/1-2 days old) respectively. Cell cultures were exposed to 1 h of IPC and 48 h of reperfusion later. Epsilon PKC specific activating peptide (100 nM) was administrated to cell culture for 1 h. Cell lysates were isolated from different intervals of reperfusion after IPC or epsilon PKC specific activating peptide treatment for immunoblotting with antibody against p-ERK1/2 and P- ser727-STAT3.

Results: Ischemic preconditioning induced the serine-727 phosphorylation of STAT3 from 15 min and was sustained for 2 h of reperfusion. STAT3 phosphorylation was accompanied by the phosphorylation of ERK1/2 (15 min-1 h of reperfusion). Similarly, epsilon PKC specific activating peptide (100 nM) also induced the phosphorylation of STAT3 and ERK1/2. To test whether IPC-induced STAT3 activation is mediated by ERK1/2 activation or by directly PKC activation, PD98059 (MAPK-K inhibitor,0.01 mM) was administrated to cell cultures during and after IPC. PD98059 prevented the IPC-induced phosphorylation of STAT3. Our previous studies showed that Inhibition of ERK1/2 prevented IPC-induced COX-2 expression (1). Based on our findings in future studies, we will further define whether STAT3 activation induces COX-2 expression after IPC.

Conclusion: Our findings suggest that IPC via epsilon PKC activation is mediated by ERK1/2->STAT3 pathway and the signaling pathways of epsilon PKC->ERK1/2->STAT3 might be involved COX-2 expression following IPC.

This study was supported by PHS grants NS34773, NS05820, NS045676 and AHA Florida& Puerto Rico Affiliate grant 0525331B.

PKC MEDIATES THE SERINE-727 PHOSPHORYLATION OF STAT1 AND STAT3 IN ASTROCYTES FOLLOWING OXIDATIVE INJURY

Jesus Purroy^1,2, Roser Gorina¹, Eun Joo Kim², Anna M. Planas¹, Miguel A. Perez-Pinzon²

¹Institut D'Investigacions Biomediques De Barcelona, CSIC-IDIBAPS, Barcelona, Catalonia, Spain, ²Cerebral Vascular Disease Research Center, Department of Neurology, University of Miami Miller School of Medicine, Miami, FL, USA

Signal transducers and activators of transcription (STATs) are transcription factors which reside in the cytoplasm until they are activated by a tyrosine phosphorylation. A second phosphorylation on Ser-727 is believed to induce maximal transcriptional activation of STAT1. Both STAT1 and STAT3 are quickly phosphorylated in serine-727 in astrocytes following oxidative stress with H202. We set out to test whether PKC was the kinase involved in the serine-727 phosphorylation of either STAT1 or STAT3.

Methods: Astrocytes purified from rat neonates were cultured in 60mm dishes for 12 days prior to treatment. Ten micromolar chelerythrine, 50 µM NSC-87877 or 30 µM PD08059 were added to the plates 30 minutes prior to treatment with 300 µM H202 for times ranging from 5 minutes to 90 minutes. Cell lysates were obtained either whole or fractioned (cytosolic/membrane). PVDF membranes were probed with antibodies against delta-PKC, epsilon-PKC, P-Ser727-STAT1, P-Ser727-STAT3, STAT1 and STAT3.

Results: Inhibition of PKC activity with chelerythrine prevented the serine-727 phosphorylation of both STAT1 and STAT3. Analysis of the translocation of two PKC isozymes showed that epsilon-PKC (ePKC) is deactivated immediately following oxidative stress, whereas delta-PKC is still in the membrane fraction 20 minutes after incubation of cultured astrocytes with 300 µM hydrogen peroxide. We then tested whether PKC is a direct kinase of STATs or whether it mediates this phosphorylation by either activating a downstream kinase or inhibiting a phosphatase. The serine-727 phosphorylation of STAT1 and STAT3 is not prevented by either the inhibition of phosphatase Shp2 with NSC-87877 or the ERK1/2 pathway with PD08059. We are currently testing whether other kinases or phosphatases are part of the phosphorylation pathway between one PKC isozyme and STAT1.

Conclusions: A PKC isozyme mediates the phosphorylation of both STAT1 and STAT3 in serine-727 following oxidative injury in astrocytes. Preliminary data suggest that delta-PKC may be the PKC isozyme involved, either directly or via a signaling cascade.

Grant support: Grant 0525393B from the American Heart Association and CICYT (SAF2005-05793-CO2).

EFFECT OF CILOSTAZOL ON CEREBRAL PRODUCTION OF NITRIC OXIDE DURING CEREBRAL ISCHEMIA AND REPERFUSION IN MICE

Harumitsu Nagoya, Nobuo Araki, Takeshi Ohkubo, Yoshio Asano, Daisuke Furuya, Kimihiko Hattori, Tomokazu Shimazu, Yasuo Ito, Yuji Kato, Kunio Shimazu

Department of Neurology, Saitama Medical University, Iruma-Gun, Saitama, Japan

Background and aims: It is suggested that cilostazol prevents a recurrence of cerebrovascular disorders, especially lacunar stroke. Recently, it is reported that cilostazol reduces cerebral infarction volume by scavenging free radical and inhibiting apoptosis. We investigated the effect of cilostazol on the cerebral nitric oxide production in mice in steady state and during ischemia and reperfusion.

Methods: Thirty-one male C57BL/6 mice were used: control group (n=12), and cilostazol group (n=19). Rectal temperature was kept between 37.0 and 37.5 °C under halothane inhalation anesthesia. A catheter was inserted into a femoral artery for measuring blood pressure (BP). Ringer's solution was perfused at a constant rate 2 µL/min through a microdialysis probe stereotaxically inplanted into the left striatum. We collected perfusate every 10 minutes. A laser Doppler probe was fixed on the surface of contralateral skull and measured cerebral blood flow (CBF) continuously. Forebrain cerebral ishemia was made by clipping both common carotid arteries, using Zen clips for 10 minutes and afterward clips were removed and reperfused. Levels of nitric oxide (NO) metabolites, nitrite and nitrate, in the dialysate were determined using the Griess reaction. Results: [I] Effect of Cilostazol treatment on the baseline of BP, CBF, and NO metabolites; Mean BP decreased from the baseline level (65mmHg) to 62mmHg 20 minutes after the administration of cilostazol (p<0.05). CBF also decreased from the baseline level to 93.5% and 91.0% in 20 minutes and 30 minutes after the administration of cilostazol, respectively (p<0.05). The level of nitrite decreased from 2.1 µmol/L to 1.8 µmol/L in 30 minutes after the administration of cilostazol (p<0.01). The level of nitrate was not changed after the administration of cilostazol. [II] Effect of cilostazol on the change in BP, CBF, and NO metabolites during cerebral ischemia and reperfusion; (1) BP: Mean BP in the cilostazol group after ischemia/reperfusion was significantly lower than that of control group (Fig. 1). (2) CBF: During ischemia, CBF decreased to 7.8%, and 4.8% of the baseline level in the cilostazol group and the control group, respectively. There was no significant difference between the two groups. (3)%change of nitrite : The level of nitrite in the cilostazol group after 50, 60, and 70 minutes from the start of reperfusion were lower than those of the control group (p<0.05) (Fig. 2). (3)%change of nitrate : No significant difference was seen between the two groups. Conclusion: These data suggested that cilostazol reduces the nitric oxide production in the reperfusion phase and may protect ischemic neuronal injury.

NADPH OXIDASE IS ACTIVATED, BUT NOT DIRECTLY INVOLVED IN NEURONAL DEATH INDUCED BY HYPOGLYCEMIA-GLUCOSE REPERFUSION

Kazuki Nagasawa, Taichi Kakuda, Sadaki Fujimoto

Department of Environmental Biochemistry, Kyoto Pharmaceutical University, Kyoto, Japan

[Background and aim] In diabetic patients receiving insulin therapy, hypoglycemia followed by glucose supplementation induces significant neuronal death in the central nervous system, and its consequences are severe clinical problems. Although reactive oxygen species (ROS) was reported to be involved in this type of neuronal death, detail mechanism for the ROS generation has not been understood well so far. The aim of this study was to clarify whether activation of NADPH oxidase, a well known ROS generating enzyme in phagocytic cells, was involved in neuronal death induced by the treatment of glucose deprivation followed by glucose reperfusion (GD/R) using primary cultured rat cortical neurons.

[Methods] Neurons were obtained from E18 rat embryo, cultured in B-27-supplemented EMEM, and used at 8-10DIV. GD/R treatment was performed as follows: ‘glucose deprivation’ was initiated by twice washes of neurons with glucose-free BSS, and they were incubated in glucose-free BSS for the designated times, and then cultured for an appropriate time interval in BSS (for the short time incubation) or B-27 minus-AO-supplemented EMEM (for long time incubation) as ‘glucose reperfusion’. Expression of proteins and mRNAs were determined by western blot and real-time PCR, respectively. Activation of NADPH oxidase was assessed by translocation of p47phox, its cytosolic component under resting conditions, to membrane fraction with an immunocytochemical analysis. ROS generation was monitored by staining with 5-(and 6-)carboxy-2′,7′-dichlorodihydrofluorescein diacetate (DCF) using a fluorescence microscopy, and cell viability was determined by an MTT asssay and/or a trypan blue exclusion test.

[Results] The GD/R treatment decreased the intracellular NAD+ concentration, and induced the accumulation of poly(ADP-ribose) (PAR) and nitrotyrosine, and decrease of neuron viability, which was accompanied with translocation of apoptosis inducing factor, but no cleavage of poly(ADP-ribose) polymerase-1, and was recovered by coadministration of N-acethyl-L-cycteine, a nonspecific ROS scavenger, and 3-aminobenzamide, a PARP inhibitor, indicating it was confirmed to be mediated by ROS generation and extensive PARP-1 activation. A nonspecific NMDA receptor antagonist, MK801, had no effect on the GD/R-induced decrease of neuron viability, its contribution being negligible. The neurons expressed functional NADPH oxidase and it was activated by the GD/R treatment, but did not generate detectable ROS with DCF staining, and its pharmacological inhibition by apocynin deteriorated the GD/R-induced decrease of neuron viability. On the other hand, a chelator for extracellular zinc, CaEDTA, almost completely abolished the GD/R-induced ROS generation and PAR accumulation, and improved the neuron viability.

[Conclusion] These findings suggested that NADPH oxidase is not directly involved in the GD/R-induced neuronal death, and its activation of neuronal NADPH oxidase might contribute to maintain the intracellular redox balance in oxidative stress-loaded neurons, and furthermore demonstrated that extracellularly chelatable zinc plays a central role in GD/R-induced neuronal events.

HYPEROXIDIZED PEROXIREDOXINS AND GLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE IMMUNOREACTIVITY AND PROTEIN LEVELS ARE CHANGED IN THE GERBIL HIPPOCAMPAL CA1 REGION AFTER ISCHEMIC INSULT

Ki-Yeon Yoo¹, In Koo Hwang¹, Dae Won Kim², Jung Hoon Choi³, Seoung Uk Lee¹, Ok Kyu Park¹, In Se Lee³, Soo Young Choi², Moo Ho Won¹

¹Department of Anatomy and Neurobiology, College of Medicine, Hallym University, Chuncheon, South Korea, ²Department of Biomedical Science and Research Institute for Bioscience and Biotechnology, Hallym University, Chuncheon, South Korea, ³Department of Anatomy and Cell Biology, College of Veterinary Medicine and BK21 Program for Veterinary Science, Seoul National University, Seoul, South Korea

Oxidative stress is a major pathogenic event occurring in several brain disorders and is a major cause of brain damage due to ischemia/reperfusion. Thiol proteins are easily oxidized in cells exposed to reactive oxygen species (ROS). In the present study, we investigated transient ischemia-induced chronological changes in hyperoxidized peroxiredoxins (Prx-SO3) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH-SO3) immunoreactivity and protein levels in the gerbil hippocampus induced by 5 min of transient forebrain ischemia. Weak Prx-SO3 immunoreactivity is detected in the hippocampal CA1 region of the sham-operated group. Prx-SO3 immunoreactivity was significantly increased 12 h and 1 day after ischemia/reperfusion, and the immunoreactivity was decreased to the level of the sham-operated group 2 days after ischemia/reperfusion. Prx-SO3 immunoreactivity in the 4 days post-ischemia group was increased again, and the immunoreactivity was expressed in glial components for 5 days after ischemia/reperfusion. GAPDH-SO3 immunoreactivity was highest in the CA1 region 1 day after ischemia/reperfusion, the immunoreactivity was decreased 2 days after ischemia/reperfusion. Four days after ischemia/reperfusion, GAPDH-SO3 immunoreactivity increased again, and the immunoreactivity began to be expressed in glial components from 5 days after ischemia/reperfusion. The patterns of changes in Prx-SO3 and GAPDH-SO3 protein levels in the ischemic CA1 region were similar to the patterns of the immunohistochemical changes in the ischemic CA1 region. In conclusion, Prx-SO3 and GAPDH-SO3 immunoreactivity and protein levels in the gerbil hippocampal CA1 region are significantly increased 12 h-24 h after ischemia/reperfusion and their immunoreactivity begins to be expressed in glial components from 4 or 5 days after ischemia/reperfusion.

CAVEOLIN EXPRESSION IS ESSENTIAL FOR NMDA-MEDIATED ERK1/2 PHOSPHORYLATION IN CULTURED NEURONS

Brian Head, Hemal Patel, Yasuo Tsutsumi, John Drummond, David Roth, Piyush Patel

Department of Anesthesia, University of California, VA San Diego, San Diego, CA, USA

Ischemic preconditioning (PC) is a phenomenon whereby various stimuli significantly reduce the vulnerability of the brain to subsequent lethal ischemia. Although the endogenous signaling mechanisms remains unknown, previous work suggests that NMDA receptor (NMDAR) mediated signaling plays a critical role in neuronal PC. NMDAR activation leads to downstream activation of pro-survival kinases Src and ERK1/2. Caveolin (Cav), a marker of caveolae, scaffolds signaling complexes in discrete microdomains in most cells, including the Src tyrosine kinase and ERK1/2 and recent work from our laboratory has shown that the NR2B subunit of the NMDAR is scaffolded by caveolin-1. Src has been demonstrated to be activated by NMDA and by sub-lethal ischemia, both of which induce PC. It is therefore conceivable that transient activation of Src and ERK1/2 is necessary to achieve PC and that Cav-1 expression and scaffolding of Src and ERK1/2 is essential for maintaining the fidelity of this signaling cascade during neuronal preconditioning. The present study was conducted to test this hypothesis. Neurons were subjected to sublethal oxygen glucose deprivation (OGDSL; 25 min) or exposed to NMDA (10 µM; 25 min) 24 hr prior to lethal OGD (OGDL; 60 min) and neuronal cell death was assessed after 24 hr with trypan blue. NMDA or OGDSL blocked OGDL-mediated neuronal cell death (one-way ANOVA P<0.0001). The NMDAR antagonist, MK801, and the selective Src kinase inhibitor, PP2, attenuated both NMDA and OGDSL-mediated neuronal preconditioning in vitro. Immunoblot analysis of OGDSL-treated or NMDA-stimulated neurons revealed a significant enhancement in phosphorylated Cav-1 (P-Cav-1), P-Src, P-ERK1/2, and P-p38 (P<0.05). Treatment of neurons with Cav-1-siRNA (72 hr) reduced Cav-1 protein expression; in these neurons, Src and ERK1/2 phosphorylation in response to NMDA exposure was not observed by immunoblot and fluorescence microscopy. In neurons harvested from Cav-1 null mice, NMDA or OGDSL failed to enhance P-Src or P-ERK1/2. Collectively, these data indicate that Src activation plays an important role in PC. The data are consistent with the hypothesis that Cav-1 expression is required for NMDA and OGDSL-mediated activation of Src and ERK1/2, two pro-survival kinases involved in neuronal preconditioning.

MECHANISMS OF OXIDATIVE STRESS-INDUCED CALCIUM SIGNALING IN HUMAN CNS PERICYTES

Masahiro Kamouchi, Takanari Kitazono, Junya Kuroda, Kuniyuki Nakamura, Yuji Shono, Noriko Hagiwara, Hiroaki Ooboshi, Setsuro Ibayashi, Mitsuo Iida

Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

TETRAHYDROBIOPTERIN DEFICIENCY RESULTS IN OXIDATIVE STREAA AND ENDOTHELIAL DYSFUNCTION IN CAROTID ARTERIES OF SALT-SENSITIVE HYPERTENSION

Hui-Wen Yu^1,2, Alex Chen¹

¹Departments of Pharmacology and Neurology, Neuroscience Program and Cell and Molecular Biology Program, Michigan State University, East Lansing, MI, USA, ²Department of Medicine, Guangzhou Red Cross Hospital, Guangzhou, China

Introduction: Tetrahydrobiopterin (BH4) is an essential cofactor of endothelial nitric oxide (NO) synthase (eNOS). When BH4 levels are decreased, eNOS becomes uncoupled to produce superoxide anion (O2-) instead of NO, contributing to endothelial dysfunction. GTP cyclohydrolase I (GTPCH I) is the rate-limiting enzyme for de novo BH4 synthesis. The hyperphenylalaninemic mouse mutant (hph-1 mice) displays 90% deficiency of the endothelial-specific GTPCH-?. This study tested the hypothesis that BH4 deficiency leads to oxidative stress and endothelial dysfunction in carotid arteries of deoxycorticosterone acetate (DOCA)-salt hypertension, a salt-sensitive model characterized by suppressed plasma renin level due to sodium retention, but manifests in eNOS uncoupling.

Methods: DOCA-salt hypertension was created in 8-10 week-old male hph-1 and wild-type mice (both of C57BL/6 background). Systolic blood pressure was measured by tail-cuff method. Carotid arteries were obtained three weeks after surgery. Endothelium-dependent relaxations in carotid arteries in response to acetylcholine (ACh, 1 nM −10 µM) were assessed by isometric tension recordings. Vascular superoxide levels were assayed by lucigenin-enhanced chemiluminescence.

Results: There was no significantly difference in systolic blood pressure between wild-type and hph-1 mice (104.50±2.58 vs. 104.67±2.62 mmHg, n=6-8, p>0.05). However, systolic blood pressure was significantly increased in wild-type DOCA-salt mice when compared to Sham-operated control mice (136.25±3.71 vs. 104.50±2.58 mmHg, n=8, p<0.001), which was further elevated in DOCA-salt treated hph-1 mice (136.25±3.71 vs. 156.17±3.72 mmHg, n=6-8, p<0.01). Consistently, vascular superoxide levels were significantly increased in DOCA-salt treated hph-1 mice when compared to the Sham controls (1.91±0.25 vs.0.56±0.10 nmol/min/mg tissue, n=6, p<0.01). Endothelium-dependent relaxations in carotid arteries in response to ACh (1 nM-10 µM) were significantly decreased in DOCA-salt treated hph-1 mice as compared to Sham controls in the presence of PEG-catalase (250U/mL, n=4-8, p<0.05), a membrane-permeable scavenger of hydrogen peroxide (n=5-8, p<0.05). The relaxation responses were abolished by NOS inhibitor NG-nitro-L-arginine methyl ester (L-NAME, 0.3 mM, n=4-8, p<0.001).

Conclusions: These findings demonstrate that deficiency of eNOS cofactor BH4 leads to oxidative stress-induced endothelial dysfunction of carotid arteries, parallel with enhanced blood pressure level in DOCA-salt hypertensive mice.

EFFECTS OF FENOFIBRATE, A PEROXISOME PROLIFERATOR-ACTIVATED RECEPTOR ALPHA AGONISIT, ON NITRIC OXIDE PRODUCTION DURING CEREBRAL ISCHMIA AND REPERFUSION IN MICE

Takeshi Ohkubo, Nobuo Araki, Yoshio Asano, Daisuke Furuya, Tomokazu Shimazu, Harumitsu Nagoya, Yasuo Ito, Yuji Kato, Mikiko Ninomiya, Kunio Shimazu

Division of Neurology, Department of Internal Medicine, Saitama Medical University, Iruma-Gun, Saitama, Japan

Background and aims: Nitric oxide (NO) plays an important role in the pathogenesis of neuronal injury during cerebral ischemia. Fenofibrate, used clinically as a lipid-lowering agent, activates peroxisome proliferator-activated receptor α (PPARα). It has been reported that fenofibrate reduces cerebral infarct size, but its direct action for NO production has not been clarified. We investigated the effects of fenofibrate on NO production during cerebral ischemia and reperfusion.

Methods: Twenty one C57BL/6 mice were used in the study. Fenofibrate (50mg/kg/day) was orally administered for 4 days before each brain ischemia in 9 mice (fenofibrate group), and the drug was not administered in the remaining 12 mice (control group). The animals were anesthetized with 2% halothane and maintained with 0.5–1% halothane. NO production was continuously monitored by in vivo microdialysis. A microdialysis probe was inserted into the left striatum and perfused with Ringer's solution at a constant rate of 2µl/min. After 2 hours equilibrium period, fractions were collected every 10 minutes. A laser Doppler probe was placed on the right skull surface. Global ischemia was produced by clipping both common carotid arteries using Zen clips for 10 minutes. The levels of nitrite (NO2-) and nitrate (NO3-) in the dialysate samples were measured by the Griess reaction.

Results: (1) Blood Pressure: No significant differences were observed between fenofibrate group and control group. (2) Cerebral Blood Flow (CBF): CBF decreased to 4.6±0.9% (mean±SE) in fenofibrate group and 4.9±0.7 in control group during ischemia. After reperfusion, CBF transiently returned to baseline values and then gradually decreased significantly in both groups. There were no significant differences between the two groups. (3) NO Metabolites (Fig1, Fig2, Fig3): In fenofibrate group, NO2- levels at baseline (2.12±0.26*µmol/L) and at from 30 (2.23±0.26*) to 60 minutes (2.10±0.27*) after reperfusion were significantly higher than those in control group (1.50±0.16, 1.43±0.16, 1.32±0.14, respectively). The levels of NO3- during ischemia (1.34±0.20*) was significantly higher than that in control group (0.89±0.091), but no significant differences were observed between the two groups after reperfusion. Total NO (NO2-+NO3-) levels at baseline (3.99.0±0.28*) and at 30 (4.48±0.27*) and 40 minutes (4.97±0.24**) in fenofibrate group were significantly higher than those in control group (2.93±0.25, 3.56±0.28, 3.92±0.23, respectively). [*: p<0.05, **: p<0.01]

Conclusions: The above data indicate that fenofibrate affects the whole NO production in baseline and in early phase after reperfusion. It has been reported that PPARα activators upregulates eNOS expression, mainly through mechanisms of stabilization eNOS mRNA. These data suggest fenofibrate may exert the brain protective effect through the influence on NO metabolism.

NEUROPROTECTIVE ROLE OF PRAS IN APOPTOTIC NEURONAL CELL DEATH AFTER STROKE

Atsushi Saito¹, Pak H. Chan²

¹Department of Neurosurgery, Aomori Prefectural Central Hospital, Aomori, Japan, ²Department of Neurosurgery, Stanford University, Palo Alto, CA, USA

The Akt signaling pathway contributes to regulation of apoptosis after a variety of cell death stimuli. A novel proline-rich Akt substrate (PRAS) was recently detected and found to be involved in apoptosis. However, the role of the PRAS pathway in apoptotic neuronal cell death after ischemia remains unknown. Phosphorylated PRAS (pPRAS) and the binding of pPRAS/phosphorylated Akt (pPRAS/pAkt) to 14-3-3 (pPRAS/14-3-3) were detected, and their expression transiently decreased in mouse brains after transient focal cerebral ischemia (tFCI). Liposome-mediated pPRAS cDNA transfection induced overexpression of pPRAS, promoted pPRAS/14-3-3, and inhibited apoptotic neuronal cell death after tFCI. These results suggest that PRAS phosphorylation and its interaction with pAkt and 14-3-3 might play an important role in neuroprotection mediated by NGF in apoptotic neuronal cell death after tFCI.

CORTICAL COLD LESIONS AND BRAIN PRECONDITIONING. DISTINCT SIZE THRESHOLDS FOR PROTECTION AND EFFECTS ON RESTING CBF

Liang Zhao, Thaddeus S. Nowak Jr.

Department of Neurology, University of Tennessee Health Science Center, Memphis, TN, USA

Background and Aims: Recent studies indicate that preconditioning (PC) effects of brief middle cerebral artery (MCA) occlusion may be mediated by early CBF recovery in penumbral regions during subsequent occlusions (1). CBF improvement was also noted after PC by NaCl-induced cortical lesions (2). Both studies identified variable effects of such procedures on resting CBF in the hemisphere subjected to manipulations, suggesting that injury-associated reduction in metabolism could also contribute to the protected state. This study compared the impact of lesion size on resting CBF and protection after cortical cold injury.

Methods: Spontaneously Hypertensive Rats under halothane anesthesia were subjected to cortical cold lesions by brief application of a liquid nitrogen-chilled metal rod to thinned skull overlying frontal cortex. Rod size and exposure times were varied to produce lesions of different sizes. In one group of rats CBF was evaluated the following day by a quantitative autoradiographic method using [14C]iodoantipyrine as the diffusible tracer. The same series of serial frozen sections was later stained with hematoxylin/eosin to assess cold lesion volume. Another group was subjected to permanent tandem occlusion of the right MCA and common carotid artery, with infarct and cold lesion volumes evaluated 24 h later.

Results: Mean cortical CBF relative to the contralateral hemisphere was not impacted until cold lesion volume exceeded 5 cubic mm (Figure). In contrast, reductions in infarct volume were evident in association with the smallest cold lesions detected. Sham surgical procedures had no effect (not shown).

Conclusions: These results indicate that even small cortical lesions that have no impact on resting CBF have robust preconditioning effects against subsequent focal ischemia. Prior reductions in brain CBF and metabolism are therefore not required for such protection. Further experiments are necessary to fully assess the efficacy of cold lesion PC, but the reductions in infarct volume observed here approach the maximum seen after ischemic PC (1). This suggests that cold lesions provide an alternative, easily titrated PC method that permits independent assessment of the insult to frontal cortex, avoiding potential confounds of variable MCA territory injuries that arise after ischemic PC.

Supported by USPHS grant NS42267 (TSN)

PROSTANOIDS, NOT REACTIVE OXYGEN SPECIES, MEDIATE COX-2-DEPENDENT NEUROTOXICITY

Yasuhiro Manabe¹, Josef Anrather², Takayuki Kawano², Kiyoshi Niwa², Ping Zhou², Elizabeth Ross², Alexander Kunz², Koji Abe³, Costantino Iadecola²

¹Okayama Medical Center, Okayama, Okayama, Japan, ²Division of Neurobiology, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, New York, NY, USA, ³Department of Neurology, Graduate School of Medicine and Dentistry, Okayama University, Okayama, Japan

Background and aims: The prostaglandin synthesizing enzyme cyclooxygenase-2 (COX-2) has emerged as a critical pathogenic factor in brain diseases associated with activation of N-methyl-D-aspartate (NMDA) receptors. However, the COX-2 reaction products responsible for these deleterious effects have not been identified. Methods and results: We found that the brain damage produced by direct injection of NMDA into the somatosensory cortex is attenuated by the COX-2 inhibitor or in COX-2-null mice, but that the associated production of free radicals is not. Conclusions: This study directly implicates COX-2-derived prostanoids, rather then radicals, in the COX-2-dependent component of the damage mediated by NMDA receptors and strengthen the rationale for using COX-2 inhibitors in the treatment of neurological diseases associated with glutamate neurotoxicity.

INDUCTION OF UBIQUITIN, UCH-L1 AND PARKIN AND SELECTIVE VULNERABILITY OF MOTOR NEURON IN SPINAL CORD AFTER TRANSIENT ISCHEMIA

Takashi Yamauchi¹, Masahiro Sakurai², Koji Abe³, Yoshiki Sawa⁴

¹Department of Cardiovascular Surgery, Sakurabashi Watanabe Hospital, Osaka, Osaka, Japan, ²Department of Cardiovascular Surgery, National Hospital Organization Sendai Medical Center, Sendai, Miyagi, Japan, ³Department of Neurology, Okayama University Graduate School of Medicine, Okayama, Okayama, Japan, ⁴Department of Cardiovascular Surgery, Osaka Graduate School of Medicine, Suita, Osaka, Japan

Background and aims: Vulnerability of motor neuron in spinal cord against ischemia is considered to play in the development of delayed paraplegia after operation of thoracic aorta. However, the reasons of such vulnerability are not fully understood. Recently, ubiquitin system has been reported to participate in neuronal cell death. In the present study, we investigated the expression of ubiquitin system molecules and discussed the relationship between the vulnerability and ubiquitin system after transient ischemia in spinal cord.

Methods: Fifteen minutes spinal cord ischemia of rabbits was applied with use of a balloon catheter. In this model, spinal motor neuron shows selectively delayed neuronal death whereas other spinal neuron such as inter neurons survive. Immunohistochemical analysis and western blotting for ubiquitin system molecules, ubiquitin, deubiquitylating enzyme ubiquitin (carboxy-terminal hydrolase 1;UCH-L1) and E3 ubiquitin ligase parkin, were examined.

Results: In cytoplasm, ubiquitin and UCH-LI were induced strongly both in inter neuron and motor neuron at early stage of reperfusion, but the prolonged expression was observed in motor neuron. Parkin was induced strongly at 3 hr after reperfusion, but the immunoreactivity returned to almost the same level of sham group at 6 hr in both neurons. In neuclei, ubiquitin, UCH-L1 and parkin were strongly induced in inter neuron whereas no upregulation of these proteins were observed in motor neuron.

Conclusions: These results indicate that the vulnerability of motor neuron of spinal cord might be partially attributed to the different response in ubiquitin and its related molecule response after transient ischemia.

MECHANISMS UNDERLYING THE BENEFIT OF REPEATED PRECONDITIONING INSULTS IN A RAT 4-VESSEL OCCLUSION MODEL OF GLOBAL ISCHEMIA

Masayuki Ueda, Thaddeus S. Nowak Jr.

Department of Neurology, University of Tennessee Health Science Center, Memphis, TN, USA

Introduction: Repeated priming insults can increase the efficacy of ischemic preconditioning (1), but underlying mechanisms are unknown. This study compared hippocampal CA1 neuron protection after single and repeated priming insults monitored by DC potential measurement in a rat 4-vessel occlusion (4-VO) model (2).

Methods: All procedures were carried out under halothane anesthesia. Vertebral arteries were first cauterized, with recovery overnight. Common carotid arteries were occluded for 2 min and the duration of ischemic depolarization was recorded in each hippocampus (n=15), and this was repeated 2 days later. A comparison group (n=18) received single priming insults yielding depolarizations of varied duration. Test occlusions were produced 2 days after the final priming insult, and hippocampi experiencing standard insults of 7.5–9..5 min depolarization were included in subsequent analyses (14 of 15). Rats were perfusion-fixed 7 days after the test insult and CA1 neurons were counted in paraffin sections stained with hematoxylin/eosin.

Results: The first priming occlusions resulted in depolarizations of 50-104 sec. The second occlusion produced depolarizations of comparable duration (54-106 sec), most within 20 sec of that produced during the first occlusion (6 longer and 4 shorter), but 4 hippocampi failed to depolarize. The lag between occlusion and depolarization was 88 / 15 and 98 / 16 sec after first and second occlusions, respectively (excluding depolarization failures), but this delay increased to 133 / 35 seconds for the test insult. The magnitude of tolerance induction after repeated priming occlusions was well correlated with the duration of the longer of the two depolarization produced during priming insults, following the relationship defined for single priming insults in previous studies (Figure), although none reached the optimal preconditioning range (2-3 min).

Conclusion: There was no obvious cumulative effect of repeated insults that each produced short, suboptimal depolarizations. Efficacy may have been somewhat improved by increasing the probability of achieving a longer priming insult. There was no other advantage to repeated priming in this model, in which depolarization defined test insult severity. However, the delay in ischemic depolarization would reduce the severity of insults defined by occlusion duration. Since such depolarization delay appears to be more prominent in rats than in gerbils (2),(3) there may be species differences in this effect.

Supported by USPHS grant NS32344 (TSN)

ESTROGEN REPLACEMENT THERAPY POTENTIATES POST-ISCHEMIC CEREBRAL NEUTROPHIL ACTIVITY IN DIABETIC OVARIECTOMIZED RATS VIA RAGE

Hao-Liang Xu, Hae-Kyung Lee, Dale Pelligrino

Neuroanesthesia Research Department, University of Illinois at Chicago, Chicago, IL, USA

BACKGROUND: Although neuroprotective benefits of estrogen replacement therapy (ERT) have been established in multiple experimental ischemia models, the results of large-scale clinical studies have not supported the animal findings, and even hinted that ERT may be detrimental. This implies that conditions may exist whereby ERT may exacerbate rather than limit ischemic neuropathology. We recently reported that, in the diabetic, ovariectomized (OVX) rat, ERT promotes neutrophil adhesion and infiltration in the post-ischemic brain (rather than the reductions seen in non-diabetics), and exacerbates ischemic brain damage (1). In that study, a linkage between enhanced neutrophil infiltration and increased neuropathology was established. In the present study, we hypothesized that ERT-related exacerbation of post-ischemic inflammation in diabetics is linked to an increase in the presence of advanced glycation end-products (AGE) and the AGE receptor (RAGE) (2).

METHODS: Two principal strategies were used to test this hypothesis: a) RAGE inhibition (using a soluble RAGE ectodomain decoy inhibitor [sRAGE]; 425 ng in aCSF, injected icv 24h prior to study); or b) AGE disruption (4-6 wk treatment with the anti-AGE drug, ALT-711, via drinking water). To that end, we monitored post-ischemic neutrophil (rhodamine-tagged) adhesion/infiltration (intravital microscopy/closed cranial window system) in diabetic (streptozotocin [STZ]-injected; studied at 4–6 weeks post-STZ) OVX female rats, with or without chronic ERT (17B-estradiol [E2], 0.1 mg/kg/day for 1 week). All rats were subjected to 20 min of transient forebrain ischemia (carotid occlusion + hypotension) and 10 h reperfusion.

RESULTS: The data summarized in fig. 1 (sRAGE treatment) and fig. 2 (ALT-711 treatment) is taken from OVX + E2 rats and is expressed as the total area of rhodamine reactivity (intra- plus extravascular neutrophils) divided by the viewed pial venular area (means +/− SE). Extravasation of neutrophils in the vehicle-treated rats was seen starting at 8h reperfusion, while no infiltration was seen in the inhibitor-treated rats. In OVX rats, no infiltration was observed. Moreover, no effects of sRAGE or ALT-711 (vs vehicle) were seen (not shown). In ERT females, sRAGE and ALT-711 treatments were associated with substantial reductions in post-ischemic pial venular leukocyte adhesion and infiltration.

CONCLUSIONS: These results suggest that the exacerbation of post-ischemic leukocyte adhesion by chronic ERT in diabetic OVX females involves a RAGE-related process.

REGULATORY CD4FOXP3+ T-LYMPHOCYTES PREVENT DELAYED INFARCT GROWTH IN MURINE FOCAL ISCHEMIA

Roland Veltkamp¹, Elisabeth Suri-Payer², Clemens Sommer³, Claudia veltkamp⁴, Henrike Doerr³, Thomas Giese⁵, Arthur Liesz¹

¹Department of Neurology, University of Heidelberg, Heidelberg, Germany, ²Department of Immunogenetics, DKFZ, Heidelberg, Germany, ³Department of Neuropathology, University of Mainz, Mainz, Germany, ⁴Department of Internal Medicine IV, University of Heidelberg, Heidelberg, Germany, ⁵Department of Immunology, University of Heidelberg, Heidelberg, Germany

Background and Aims: Inflammatory cascades contribute substantially to secondary ischemic brain damage. Regulatory CD4+CD25+Foxp3+ T lymphocytes (Treg) are important anti-inflammatory modulators in various inflammatory diseases. We studied the role of Tregs in the progression of brain damage after ischemic stroke.

Methods: Permanent focal cerebral ischemia was induced by transtemporal middle cerebral artery occlusion (MCAO). Tregs were eliminated either by preischemic depletion with a mAb (clone PC61) in C57Bl/6 mice or by adoptive transfer of CD4+CD25- into rag2-/− mice. Infarct volume and cerebral cytokine expression (RT-PCR) were measured at various time points after MCAO. Tcell invasion was shown by flow cytometry and immunohistochemistry, respectively.

Results: Depletion of Tregs had no effect on infarct size 24 h after MCAO. Treg-depleted mice tended to have greater infarcts 72 h, and had significantly larger infarct volumes 7 d after MCAO (control: 7.4mm3; antibody-treated mice: 12.1mm3). Thus, Tregs prevented delayed infarct growth. Correspondingly, transfer of MACS sorted CD4+25- T cells into lymphocyte-deficient rag2-/− mice resulted in larger infarcts (12.9mm3) than transfer of CD4+ cells (6.8 mm3; p<0.05). In mice lacking Tregs, RNA levels of proinflammatory cytokines were significantly more elevated in the ischemic hemisphere compared to control (6 h after MCAO: TNF?: 4fold and IL-1? 2fold; 72 h after MCAO: IFN?: 5 fold). Treg invasion into the ischemic brain became first detectable 72 h after MCAO by immunohistochemistry and FACS.

Conclusions: Tregs are master anti-inflammatory modulators in ischemic stroke which substantially reduce secondary infarct progression by downregulating proinflammatory cytokine ?induced cell death. Based on our findings, this effect is probably mediated by early humoral signalling and by delayed Treg invasion. Better understanding of this postischemic crosstalk between immune cells and the damaged CNS may lead to new therapeutic strategies for ischemic stroke.

EFFECTS OF 24-HOUR EXPOSURE TO HIGH GLUCOSE CONCENTRATIONS ON OXIDATIVE METABOLISM IN RAT ASTROGLIAL CULTURES

Shinichi Takahashi, T. Abe, N. Suzuki

Department of Neurology, Keio University School of Medicine, Shinjuku-Ku, Tokyo, Japan

CEREBRAL BLOOD FLOW IN WILD & ADENOSINE A2A KNOCK-OUT MICE DURING ACUTE HYPOXIA: EFFECTS OF A2A ANTAGONIST, ZM 241385

Grzegorz J. Miekisiak, Alexander Beylin, David K. Kung, JiSuk K. Yoo, Adam L. Sandler, H. Richard Winn

Department of Neurosurgery, Mount Sinai School of Medicine, New York, NY, USA

CAUDATE PHOTOTHROMBOTIC INFARCTION OF RODENT: NEUROPATHOLOGICAL AND BEHAVIORAL STUDY

Toshihiko Kuroiwa, Guohua Xi, Masanobu Okauchi, Ya Hua, Timothy Schallert, Julian T. Hoff, Richard F. Keep

Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA

Basal ganglia infarction is typically caused by small end-artery occlusion, which is different from cortical infarction typically caused by large artery occlusion. We have developed a model of caudate infarction in rodent induced by photothrombotic occlusion of parenchymal small vessels using a stereotaxically placed optic fiber, and analyzed the time course of behavioral and pathological changes.

[Methods] Male SD rats (n=20) and male Mongolian gerbils (n=24) were used in this study. Optic fiber 0.5 mm or 0.75 mm in diameter coated with fluorocarbon resin was placed stereotaxically to the left caudate nucleus under isoflurane anesthesia, Rose Bengal dye (20mg/kg) was intravenously injected and caudate nucleus was exposed to cold white light (lighting intensity, 900 ? 2400 Lux) via the optic fiber for 3, 5 or 10 min. A battery of neurological tests was performed repeatedly during 7 days after light exposure. The brain was removed after transcardiac perfusion with fixative at different post-exposure time-intervals, and prepared for light-microscopical and electron- microscopical examination. Vascular perfusion was examined in the brain removed shortly after intravenous injection of FITC-dextran. BBB opening was detected by Evans Blue extravasation, and ischemic brain edema was quantified by tissue gravimetry. [Results] Round infarction with thrombosed parenchymal small vessels was produced in the caudate nucleus. The infarct size increased in parallel the light exposure time (1.5mm ? 3.5mm in diameter). The infarcted tissue was surrounded by a thin layer (0.12mm ? 0.14mm) of selective neuronal death with interstitial microvacuolation (ischemic edema) with BBB opening to serum proteins. The density of vessels with FITC-dextran was decreased in the peri-infarct tissue where dilated vessels were scattered. Electron microscopy revealed platelet thrombi occluding parenchymal small vessels, astrocytic swelling, dark neuronal change and hydropic oligodendrocyte with pyknosis. Neurological test revealed circling toward left, left corner turn, impaired right-hindlimb placing, which was milder and transient (Foot-fault placing score= 14.2 ± 9.0) in 3–5 min exposure group and more severe (22.5 ± 10.0, p<0.05) and lasted longer in 10 min exposure group. [Conclusion] Small infarction localized in the caudate nucleus was produced by photothrombotic occlusion of the parenchymal small vessels using a stereotaxically placed optic fiber in rodents. The infarction was surrounded by a thin layer of selective neuronal death and hypoperfusion. Severity and duration of neurological deficit parallels the infarction size. We have established a rodent model of basal ganglia infarction induced by photothrombotic occlusion of the parenchymal small vessels.

Figure. Photothrombotic infarction of caudate nucleus in rat (left) and parenchymal small vessels occluded with platelet thrombus

IMAGING OF EXPERIMENTAL EPILEPSY SURGERIES: CALLOSOTOMY, MULTIPLE SUBPIAL TRANSECTION AND DEEP BRAIN STIMULATION

Tatsuya Tanaka, Kiyotaka Hashizume, Akira Hodozuka, Seiji Takebayashi, Masayuki Kunimoto

Department of Neurosurgery, Asahikawa Medical College, Asahikawa, Japan

Bakground and Aim: In various epilepsy surgery, seizure-free rate of each surgery is still controversial. And cost-effectiveness of surgery is still discussing. In the present study, experimental epilepsy surgeries were performed in cats and in rats. Electrophysiological and metabolic approaches were made.

Methods: Under intraperitoneal pentobarbital anesthesia, electrodes or chemitrodes were placed and chronic preparation was made. (1) After seizure induction, deep brain stimulation(DBS) was performed. (2) For multiple subpial transaction(MST) or callosotomy, craniotomy was made in all animals and the dura was opened unilaterally. Bipolar electrodes were implanted and EEG recording was made. Two micrograms of kainate was injected into unilateral sensori-motor cortex and focal cortical seizures were elicited. MST or callosal section was performed using stereotactic coordinates. EEG analysis was made. (3) Under intraperitoneal pentobarbital anesthesia, femoral artery and femoral veins were canulated. Craniotomy was made in all rats and the dura was opened unilaterally. Two micrograms of kainite was injected into unilateral sensori-motor cortex and focal cortical seizures were elicited. MST or callosal section was performed using stereotactic coordinates. Intravenous injection of 14C-deoxyglucose was made and the rats were processed for autoradiography.

Result: All cats and rats exhibited focal cortical seizures after kainite injection. DBS suppressed seizure propagation. Seizure activity between two MSTs was suppressed and seizure propagation to contralateral cortex was clearly suppressed after callosotomy. But, kainite-induced seizures remained in the focus and seizure propagation was observed to the subcortical structures in both cases. (3) After MST or callosotomy, seizure propagation to contralateral cortex(hypermetabolic area) was suppressed. Hypermetabolic area in the focus, ipsilateral basal ganglia and thalamus, and contralateral basal ganglia remained even after MST or callosotomy.

Conclusion: MST or callosotomy may be a palliative surgical option only to reduce seizure activities in the cortex. But seizure activities of the focus remained even after the procedures. DBS may be the promising procedure when epileptic focus locate in the eloquent cortex.

ADAPTIVE GLIAL STRESS RESPONSES IN CHEMICAL PRECONDITIONING: FOCUS ON ASTROCYTIC GLUTAMINE SYNTHETASE BUFFERING GLUTAMATE-MEDIATED NEUROTOXICITY

Akihiko Hoshi, Toshiki Nakahara, Keiko Shimizu, Kazuhiro Endo, Teiji Yamamoto

Department of Neurology, Fukushima Medical University, Fukushima, Japan

Astrocytic control of glutamatergic activity through glutamate transporters plays a pivotal role in neuroprotection. In astrocytes, approximately two-thirds of glutamate is converted by glutamine synthetase (GS) into glutamine (glutamate-glutamine cycle), that is, astrocytes serve as a sink to take up the released glutamate from neurons and utilize it to synthesize glutamine. 3-Nitropropionic acid (3-NPA) is an irreversible inhibitor of succinate dehydrogenase in the tricarboxylic-acid cycle, and provides ischemic tolerance to the brain. So far, there have been no reports on the relationship of astrocytic GS and ischemic tolerance by chemical preconditioning. We have investigated the temporal profile of GS expression parallel with glial fibrillary acidic protein and heat-shock protein 70 (HSP70) in order to test the hypothesis that astrocytes have a major role in 3-NPA-induced chemical preconditioning. In our rat model of permanent focal ischemia, preconditioning with 3-NPA singnificantly reduced the subsequent neurological deficits and infarct volume within 24-72 hours after treatment. Immunohistochemically, protoplasmic astrocytes in the cortex and striatum were activated in terms of upregulation of GS and more abundant protoplasmic processes with 3-NPA preconditioning, however, HSP70 expression could not be induced. Thus, the activation of astrocytes and upregulation of GS play an important role in 3-NPA-induced preconditioning. In view of glutamate being imposed on the cerebral ischemic damage, the astrocytic GS may contribute to 3-NPA-induced ischemic tolerance.

EFFECTS OF MAGNESIUM TREATMENT IN A MODEL OF INTERNAL CAPSULE LESION IN SPONTANEOUSLY HYPERTENSIVE RATS

Clotilde Lecrux¹, Chris McCabe², Chris J. Weir³, Lindsay Gallagher², Jim Mullin⁴, Omar Touzani¹, Keith W. Muir⁵, Kennedy R. Lees⁶, I. Mhairi Macrae²

¹University of Caen, Centre CYCERON, CNRS UMR 6185, Caen, France, ²7TMRI Facility and Wellcome Surgical Institute, Division of Clinical Neuroscience, University of Glasgow, Glasgow, Scotland, UK, ³Robertson Centre for Biostatistics, University of Glasgow, Glasgow, Scotland, UK, ⁴Department of Clinical Physics and Bioengineering, Southern General Hospital, Glasgow, Scotland, UK, ⁵Division of Clinical Neurosciences, University of Glasgow, Institute of Neurological Sciences, Southern General Hospital, Glasgow, Scotland, UK, ⁶University Department of Medicine and Therapeutics, Gardiner Institute, Western Infirmary, Glasgow, Scotland, UK

Background and aims - Previous studies have demonstrated that magnesium is neuroprotective in several models of cerebral ischemia in rats. The Intravenous Magnesium Efficacy in Stroke (IMAGES) trial found no significant effect of magnesium sulphate overall but subgroup analysis detected benefit in patients with lacunar stroke syndromes and/or with higher than median blood pressure. We therefore aimed to assess the effects of magnesium sulphate administration following white matter damage in vivo in spontaneously hypertensive rats (SHR).

Methods - We developed a model involving a lesion of the left internal capsule by a local injection of endothelin-1 (ET-1) (200 pmol) in adult, male SHR. We evaluated the effect of subcutaneous injections of magnesium sulphate (1×300 mg/kg 30 mins before + 4×200 mg/kg each hour after) on functional recovery, 3 and 10 days after ET-1 injection. Group sizes were n=15 vehicle, n=13 magnesium. We assessed lesion size after 2 days using MRI T2-weighted images and after 11 days using histological staining specific for white matter (luxol fast blue). Infarct volumes were statistically analysed using Student's unpaired t-test and behavioural data using ANCOVA, ANOVA and PLSD Fisher tests.

Results ? ET-1?induced lesion in the internal capsule produced significant motor deficits on both the forelimb and the hindlimb for 10 days in the vehicle group (figure), which were significantly ameliorated by the magnesium treatment in both the cylinder (left forelimb use, p<0.01 and both forelimb use, p<0.03 versus vehicle) and the walking-ladder (right hindlimb score, p<0.02 versus vehicle) tests. Infarct volumes (expressed as percentage of ipsilateral hemisphere) were not significantly different between animals receiving vehicle or magnesium sulphate treatments: 1.2 ±0.1% and 2.0±0.5%, respectively using MRI and 0.3±0.1% and 0.5±0.1%, respectively from histology at 11 days post-stroke.

Conclusions - We have developed a new model of white matter ischaemia in SHR in which we were able to assess specific behavioural deficits persistent for 10 days. Furthermore, we demonstrate that magnesium treatment can provide an improvement of motor abilities, suggesting that magnesium remains an appealing protective substance in the case of white matter ischaemic damage in the rat.

REDUCTION IN OXIDATIVE STRESS BY SOD1 OVEREXPRESSION ATTENUATES ACUTE BRAIN INJURY AFTER SUBARACHNOID HEMORRHAGE VIA ACTIVATION OF AKT/GSK3B SURVIVAL SIGNALING

Hidenori Endo^1,2, Kuniyasu Niizuma¹, Miki Fujimura², Teiji Tominaga², Pak H. Chan¹

¹Department of Neurosurgery, Department of Neurology and Neurological Sciences, and Program in Neurosciences, Stanford University School of Medicine, Stanford, CA, USA, ²Department of Neurosurgery, Tohoku University Graduate School of Medicine, Sendai, Japan

Introduction: Subarachnoid hemorrhage (SAH) results in a high mortality rate, despite sophisticated medical management and neurosurgical techniques. Recent studies have emphasized the importance of acute brain injury after SAH. Oxidative stress plays important roles in the pathogenesis of acute brain injury after SAH. We have reported that copper/zinc-superoxide dismutase (SOD1) is a crucial endogenous enzyme responsible for eliminating superoxide, and that overexpression of SOD1 reduces cell injury after SAH. In cerebral ischemia, SOD1 overexpression decreased ischemic neuronal death through activation of the Akt/GSK3b survival pathway. In this study, we investigate the relationship between SOD1 and the Akt/GSK3b pathway in acute brain injury after SAH. We address this issue by examining apoptotic cell death, superoxide production, and phosphorylation of Akt/GSK3b after SAH, using both wild-type (Wt) and SOD1 transgenic (Tg) rats.

Materials and methods: To examine the relationship between SOD1 and the Akt/GSK3b pathway following SAH, we used a perforation SAH model in both Wt and SOD1 Tg rats. One, 6 and 24 hours after SAH, samples were taken from the cerebral cortex, and used for Western blot analysis and immunohistochemical analysis. To investigate apoptotic cell death in acute brain injury after SAH, DNA fragmentation was analyzed with a commercial enzyme immunoassay. Early production of superoxide anions (O2?) during SAH was investigated with the use of hydroethidine (HEt).

Results: The mortality rate at 24 hours for the SOD1 Tg rats (7.69%) was significantly decreased compared with the Wt rats (31.4%). DNA fragmentation in the SOD1 Tg rats was significantly decreased at 24 hours compared with the Wt rats at the same time point. O2? production shown by oxidized HEt signals was observed in the cerebral cortex at 1 hour in the Wt rats, and was markedly decreased in the SOD1 Tg rats. Western blot analysis showed that immunoreactivity of phospho-Akt and phospho-GSK3b was significantly increased at 6 hours compared with the control brain samples in the Wt rats. In the SOD1 Tg rats, the increase in phospho-Akt and phospho-GSK3b levels was more prominent and persistent. In a comparison between the two groups, phospho-Akt and phospho-GSK3b were significantly increased at 24 hours in the SOD1 Tg rats. An immunohistochemistry study showed slight immunoreactivity of phospho-Akt and phospho-GSK3b in the cerebral cortex of the control brains. This immunoreactivity became stronger at 24 hours in both groups. In the SOD1 Tg rats, phospho-Akt and phospho-GSK3b expression at 24 hours was more prominent than in the Wt rats. A double immunofluorescent study demonstrated that phospho-Akt-positive cells colocalized with phospho-GSK3b-positive cells in the cerebral cortex 6 hours after SAH. Moreover, phospho-GSK3b-positive cells colocalized with neurons at the same time point.

Conclusions: This study suggests that oxidative stress plays a significant role in acute brain injury after SAH, and that the neuroprotection of SOD1 is partly mediated by activation of the Akt/GSK3b survival pathway. Reducing oxidative stress, thereby activating the survival pathway, could be a therapeutic target for acute brain injury after SAH in a clinical situation.

TEMPORAL AND REGIONAL EVOLUTION OF AQUAPORIN 4 EXPRESSION AND MAGNETIC RESONANCE IMAGING IN A RAT PUP MODEL OF NEONATAL STROKE

Jerome Badaut¹, Stephen Ashwal², Beatriz Tone², Amber Wang³, Luca Regli¹, Hui R. Tian², Andre Obenaus³

¹Neurosurgery Research Group, CHUV, UNIL, Lausanne, Switzerland, ²Department of Pediatrics, Loma Linda University School of Medicine, Loma Linda, CA, USA, ³Department of Radiation Medicine, Loma Linda University School of Medicine, Loma Linda, CA, USA

Aim : Cerebral edema contributes to the morbidity and mortality and can readily be observed using magnetic resonance imaging (MRI). Recent studies show that aquaporin 4 (AQP4, water channels) is induced early after stroke and potentially participates in the development of brain edema(1). We studied whether induction of AQP4 correlated with edema development in the 10-day old rat pup undergoing transient middle cerebral artery occlusion (tfMCAO) using high field (11.7 Tesla) MRI with T2WI and DWI followed by immunohistochemical investigation of AQP4 protein expression.

Methods : Focal ischemia in the rat pup was accomplished using a filament model of temporary unilateral occlusion of the left MCA(2). MRI was performed at 24-hours, 72-hours, and 28-days after tfMCAO. AQP4 immunolabeling was evaluated on slices corresponding to the MRI acquisitions using an infrared-scanner (LiCor) for quantification and epifluorescence microscopy.

Results : At 24-hours, we observed increased T2 values and decreased ADC within injured cortical and striatal regions that reflected the onset of edema formation. Coincident with these MR changes were significant increases in AQP4 expression in the border regions of injured tissues. Increased AQP4 expression was observed on astrocyte end-feet in contact with blood vessels. Striatal imaging findings were still present at 72-hours with a slow normalization of AQP4 expression in the peri-infarct regions. At 28-days AQP4 expression normalized throughout the brain except striatal T2 and ADC values which remained increased. Our findings show that induction of AQP4 is increased during the period of active edema formation in the peri-lesional region but at 28 days does not correlate with brain water content as observed with MRI.

Conclusion : Induction of AQP4 on astrocyte end-feet at 24 and 72 hours could participate in the clearance of water and solutes from brain parenchyma after ischemia in the immature rat brain.

Figure : A) Macroscopic comparison of MRI data and AQP4 immunohistochemistry at 24 hrs after tfMCAO occlusion : A1) A representative T2 image and A2) ADC-image show respectively an increase and decreased signal intensities within the ischemic region which are suggestive of edema formation. A3) Illustration of the AQP4-IR shows that the intensity of AQP4 is increased in border of the red line.

B) The intensity of the AQP4 staining was significantly increased around blood vessels (arrowheads) in cortex tissues in vicinity the regions where ADC was decreased and T2WI-values were increased. C) In frontal cortex of the contralateral hemisphere the intensity of AQP4-IR around blood vessels (arrow-heads, D) was higher than in the contralateral parietal cortex (arrow-heads, D). Bar : B-D=40 micrometer.

PREFERRED BLOOD-BRAIN TRANSFER OF O-(2-[18F]FLUOROETHYL)-D-TYROSINE (D-FET) INTO THE PORCINE BRAIN

Peter Brust¹, Victoria Makrides², Rainer Hinz⁵, Wolfgang Weber⁴, Hans-Jurgen Wester⁴, Steffen Fischer¹, Katja Huggel², Vadivel Ganapathy⁶, Francois Verrey², Reinhard Bauer³

¹Institute of Interdisciplinary Isotope Research, Leipzig, Germany, ²Institute of Physiology, University of Zurich, Zurich, Switzerland, ³Institute of Molecular Cell Biology, University Hospital Jena, Jena, Germany, ⁴Department of Nuclear Medicine, Technische Universitat Munchen, Munchen, Germany, ⁵Wolfson Molecular Imaging Centre, The University of Manchester, Manchester, UK, ⁶Department of Biochemistry and Molecular Biology, Medical College of Georgia, Augusta, GA, USA

Large neutral amino acids (LNAA) are transported by a variety of transporters, among which the SLC7A5/LAT1 transporter is regarded as the predominant LNAA transporter at the blood-brain barrier (BBB). LAT1 has broad substrate specificity, accepting even the D-isomers of amino acids as high affinity substrates. Also, LAT1 is frequently overexpressed in malignant tumors providing the rationale for the use of radiolabeled LNAA such as L-FET for imaging intracerebral tumors with PET. Recently, experimental evidence was obtained from various authors that D-enantiomers of some LNAA are superior to L-enantiomers with regard to the imaging of certain peripheral tumors. Therefore, in this study the transport of D-FET across the blood-brain barrier was studied with PET in anaesthetized piglets and patients after subtotal resection of brain tumors and compared with L-FET.

Ten piglets of mixed German domestic breed were anesthetized with 0.25% isoflurane in 65% nitrous oxide and 35% oxygen. L-[18F[FET (98 ± 11 MBq, n=4) or D-[18F]FET (128 ± 34 MBq, n=6) in 10 ml saline was intravenously infused within 2 min. Compartmental modeling of PET data was used to estimate the rate constants for the blood-brain (K1) and the brain-blood (k2) transfer of both radiotracers. Six patients were investigated with L-FET-PET and D-FET-PET after subtotal resection of brain tumors. Standardized uptake values (SUV) were calculated in brain cortex and lesions. Additionally, two different approaches were used to assess the interaction of L-FET and D-FET with transporters known to have affinity for LNAA. First, we studied the hLAT1 transporter in HRPE cells (human retinal pigment epithelial cells). Second, hLAT1, LAT2, XPCT and PAT1 were studied in Xenopus laevis oocytes.

The normalized brain uptake in piglets of D-FET 5 min after injection is about 140% higher than that of L-FET. The rate constants for the blood-brain and brain-blood transfer of D-FET in piglets are significantly higher (by factor 3-4) than those for L-FET. Both findings indicate that in piglets the BBB transport of D-FET is significantly faster than that of L-FET. This observation is in contrast with previous studies demonstrating high and stereoselective uptake of L-FET in mouse brain, and low uptake of D-FET into mouse brain (Wester et al., 1999). Also, the studies in humans revealed an initial uptake of D-FET which was similar to L-FET but followed by a rapid washout from brain resulting in a significantly higher uptake of L-FET at later time points. Thus, the transport kinetics for D-amino acids seems to differ among species. The differences in brain uptake of L-FET and D-FET can only be explained by differences in their affinity to specific transport systems. L-FET showed at least 20-fold higher in vitro affinity than D-FET towards LAT1 and LAT2 whereas both enantiomers have similar affinities to XPCT and PAT1 transporters.

In conclusion, we have demonstrated specific transport of L-FET and D-FET across the BBB which cannot be fully explained by known transport systems. The study provides further evidence for the presence of an unknown, D- amino acid preferring transporter / transport system at the blood-brain barrier.

DIFFERENCES IN POSTOPERATIVE GEOMETRICAL AND HEMODYNAMIC CONDITIONS OF CAROTID ARTERIES BETWEEN CAROTID ARTERY STENTING AND CAROTID ENDARTERECTOMY WITH PATCH ANGIOPLASTY

Koji Tokunaga, Kenji Sugiu, Ayumi Nishida, Shigeki Ono, Keisuke Onoda, Isao Date

Department of Neurological Surgery, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan

Background and aims: Carotid endarterectomy with a patch graft (Patch CEA), allowing marked increase in vessel diameters and reduction of the short- and long-term risk for ischemic stroke, has been our treatment of choice for patients with carotid artery stenosis. Recently, carotid artery stenting (CAS) has emerged as an alternative to CEA, particularly for patients at high surgical risk. However, postoperative geometrical and hemodynamic differences between these two treatment modalities have been rarely investigated. The purpose of this study was to evaluate the postoperative changes in configurations and flow velocities of carotid arteries after Patch CEA compared to CAS. Methods: Thirty-eight consecutive patients with carotid artery stenosis treated by CAS or Patch CEA were included. Pre- and postoperative configurations of carotid arteries were evaluated by conventional digital subtraction angiography and/or 3D-CT angiography. Ultrasonographic examinations were performed using a color duplex scanner with a 7.5(7.2–8..6)-MHz linear probe or a 5-MHz sector probe (Aplio, Toshiba Medical Systems Corp., Tokyo). Doppler waveforms were recorded in the middle portion of the common carotid artery (CCA) and the internal carotid artery (ICA) bulb to measure parameters of flow velocities including peak flow velocities (PFV), end diastolic velocities (EDV) and mean velocities (Vmean). Results: Twenty-five patients were treated by CAS, and Patch CEA was performed for 13 patients. There were no differences in preoperative degrees of stenosis (83% in CAS group vs 79% in CEA group) and flow velocities of CCA and ICA between two groups. The postoperative mean degree of stenosis was 12% and 0% in CAS group and Patch CEA group, respectively. Mild stenoses tended to be left at the severely calcified plaque after CAS. After Patch CEA, the configurations of the ICA bulbs were long and widely restored. The averages of diameters of the postoperative ICA bulbs measured by ultrasonography were 4.5 mm in CAS group and 7.0 mm in Patch CEA group (P<0.01). There were no differences in the averages of postoperative PFV, EDV and Vmean of CCA between two groups, whereas the averages of PFV, EDV and Vmean of ICA in CAS group were 80 cm/s, 25 cm/s and 42 cm/s, respectively, and significantly greater than those (53 cm/s, 16 cm/s and 28 cm/s, respectively) in Patch CEA group (P<0.01). Conclusions: There are significant differences in postoperative geometrical and hemodynamic conditions between CAS group and Patch CEA group. Suboptimal resolution of carotid stenosis may lead to acute/subacute occlusion or restenosis in the chronic stage. On the other hand, excessive increase in vessel diameter may cause recurrent atheromatous plaque because of atherogenic low and disturbed shear stress. Further studies including long-term clinical follow-up and computer flow dynamics are needed to clarify the impact of geometrical and hemodynamic differences between CAS and Patch CEA on clinical courses of patients with carotid stenosis.

GENDER-RELATED TEMPOROSPATIAL DIFFERENCES IN BLOOD-BRAIN BARRIER PERMEABILITY FOLLOWING EXCITOTOXIC LESIONS IN THE RAT

Karim Khallout¹, Jerome Toutain¹, Nicolas Delcroix², Eric T. MacKenzie¹

¹Centre Cyceron UMR 6185, Caen, Lower Normandy, France, ²Centre Cyceron UMR 6194, Caen, Lower Normandy, France

GENDER-RELATED TEMPOROSPATIAL DIFFERENCES IN BLOOD-BRAIN BARRIER PERMEABILITY FOLLOWING EXCITOTOXIC LESIONS IN RATS

Khallout K.1., Toutain J.1, Delcroix N.2, MacKenzie E.T.1

University of Caen, CNRS UMR 6185 1, UMR 6194 2 Centre CYCERON Bd H. Becquerel 14074 BP 5229 Caen, FRANCE, khallout@cyceron.fr

Introduction: The object of these investigations was to study brain permeability in male and female rats following excitotoxics lesions as a function of time, so as to better understand the effects of gender-related hormones on the blood-brain barrier (BBB). We examined, through the use of Evans blue albumin (EBA) as a tracer, the differences in extravasation between the two sexes.

Methods: All striatal lesions were provoked by the intrastriatal injection of 75 nmoles NMDA in 3 µL. EBA extravasation was quantified by fluorescent microscopy over serial coronal sections as was the volume tissue necrosis (thionine staining cf. fig. 2).

Results (cf. fig. 1): The insult provoked a volume of EBA extravasation, which correlated with the tissue necrosis; the lesions volumes were also highly sex and time-dependent (p<0.001). Statistically, the data were analyzed by polynomial two degree equations followed by ANOVA.

Conclusions: While the disruption of BBB during ischaemia has been frequently studied, there are few time-based publications. Excitotoxic lesions have been much less examined. We clearly demonstrate a quadratique time-dependent dysfunction of the BBB following intrastriatal lesions and differences in time-related (p<0,001) volumes between the two genders (males : R2 = 0,4532; p<0,02 and females : R2 = 0,7008 p<0.001). The quantification of the extravasation between the two sexes has been analyzed on a basis of the consolidated lesions. This study merits to be continued with the use of ovariectomized rats in order to examine the reasons for the exacerbation and delayed resolution between the two sexes. Such approaches are underway.

Acknowledgements: The authors thank many other members of the group that have contributed to this study.

C1-INHIBITOR ATTENUATES NEUROBEHAVIORAL DEFICITS FOLLOWING CONTROLLED CORTICAL IMPACT BRAIN INJURY IN MICE

Luca Longhi^1,2, Carlo Perego¹, Elisa Zanier^1,2, Fabrizio Ortolano^1,2, Paolo Bianchi¹, Luigi Bergamaschini³, Nino Stocchetti¹, Maria Grazia De Simoni²

¹Department of Anesthesia and Critical Care Medicine, Milan University, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli E Regina Elena, Milano, Italy, ²Mario Negri Institute, Milano, Italy, ³Department of Internal Medicine, Milan University, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli E Regina Elena, Milano, Italy

Background and aims: The goal of the study was to evaluate the neurobehavioral effects of the C1-inhibitor (C1-INH), an endogenous inhibitor of complement and contact-kinin pathways, following controlled cortical impact (CCI) brain injury in mice.

Methods: Mice (n = 36) were anesthetized with intraperitoneal administration of sodium pentobarbital (65 mg/kg) and subjected to CCI brain injury. At 10 minutes postinjury, animals randomly received an intravenous infusion of either C1-INH (15 U) or saline (equal volume, 150 µl). A second group of mice received identical anesthesia, surgery, and saline to serve as uninjured controls. Neurobehavioral motor outcome was evaluated weekly from 1 to 4 weeks postinjury by performing Neuroscore. Cognitive function was evaluated at 4 weeks postinjury using the Morris water maze (MWM) test of learning ability.

Results: Consistently, brain-injured mice receiving C1-INH showed attenuated neurological motor deficits during the 4-week period compared with injured mice receiving saline (Figure).

At 4 weeks postinjury all mice were able to swim in the MWM without notable physical impairment and demonstrated an ability to learn the visuo-spatial task, as reflected by decreasing latencies to find the platform over the three-day period. Brain-injured mice receiving C1-INH showed an average latency of 30.7 ± 1.9 seconds which was significantly faster (better performance) than the average latency of brain-injured animals receiving saline (42.9 ± 3.1 seconds, p < 0.01).

Conclusion: Post-traumatic administration of the endogenous complement inhibitor C1-INH significantly attenuates neurobehavioral motor and cognitive deficits associated with traumatic brain injury. C1-INH confirms to possess clear neuroprotective actions as previously shown in ischemia/reperfusion brain injury (1).

AGMATINE ATTENUATE BRAIN EDEMA THROUGH LESSENING THE EXPRESSION OF AQUAPORINS AFTER TRAUMATIC BRAIN INJURY

Yong Woo Lee^1,2, Jae Hwan Kim^1,2, Jin Woo Chang^1,3, Kyung Ah Park², Won Taek Lee², Jong Eun Lee^1,2

¹Brain Korea 21 Project for Medical Sciences, Yonsei University College of Medicine, Seoul, South Korea, ²Department of Anatomy, Yonsei University College of Medicine, Seoul, South Korea, ³Department of Neurosurgery, Yonsei UniversityCollege of Medicine, Seoul, South Korea

Background and aims: Brain edema is a pathophysiological condition of increased water content due to a variety of brain injuries, including trauma, ischemia and progression results in major focal brain dysfunction. The cold injury model is one of the established models for the study of vasogenic brain edema development and secondary rain injury[1]. The aquaporins are a family of water-selective membrane channels. Aquaporin-1, −4 and −9 have been identified as the three major water channels in the brain[2]. Agmatine is a primary amine formed by the arginine decarboxylase and is known as an antagonist at N-methyl-D-aspartate (NMDA) receptors and an inhibitor for NOS. Previously we reported the neuroprotective effect of agmatine on infarct volume and brain edema following cerebral ischemia[3]. In this study, we investigate that administration of agmatine regulates aquaporin expression and attenuates brain edema after TBI.

Materials and methods: For the cold traumatic injury model, the skull of the rat was mounted in a stereotactic device for trephination (diameter for 4mm) above the right parietal cortex leaving dura mater intact. Cortical lesion was induced by attachment of metal probe (diameter 3mm) cooled by liquid nitrogen onto the brain surface 5 times for 30 seconds. Agmatine was treated 30min after traumatic cold injury (100mg/kg, IP). Water content was analyzed in 24hours after TBI. The expression of aquaporins was examined with immunohistochemistry and immunoblotting.

Results: The physiological parameters before cold traumatic injury were monitored and maintained throughout the experiment. Trauma-induced loss of brain volume and the number of eosinophilic neurons were reduced and improved neurological deficits in agmatine treatment group. The water content in cold traumatic injured brains at 24hours after insults was measured and decreased in agmatine treatment group. We examined the expression of aquaporin-1,-4 and −9 using immunohistochemistry and immunoblotting in cerebral cortex and hippocampus. In all time point, aquaporin-1 was decreased in agmatine treatment group compared to experimental control after TBI. Aquaporin −4, −9 were similar expressed in agmatine treatment group and experimental control 1, 2 day after TBI. But 1week after TBI, the expressions of aquaporin −4,-9 were decreased in agmatine treatment group.

Conclusion: Agmatine could attenuate brain edema through lessening water content by suppression of the expression of aquaporins. These results may suggest agmatine as a novel therapeutic material for vasogenic and cytosolic edema including traumatic brain injury and brain tumor.

CEREBRAL VASOREACTIVITIES IN PATIENTS WITH CAROTID STENOSIS HELP TO PREDICT HYPERPERFUSION AFTER CAROTID INTERVENTION

Ting-Yu Chang¹, Ho-Ling Liu^2,3, Tai-Cheng Wu¹, Chien-Hung Chang¹, Hsiu-Chuan Wu¹, Yeu-Jhy Chang¹, Tsong-Hai Lee¹

¹Stroke Section, Department of Neurology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan-R.O.C., ²Department of Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan-R.O.C., ³MRI Center, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan-R.O.C.

Background: Carotid arterial stenting andendarterectomy are both common carotid interventions in managing patients with critical or symptomatic stenosis. Among various complications of carotid intervention, hyperperfusion syndrome is rare but the most devastating one leading to poor outcome. It is important to predict and to discover hyperperfusion early before and after such interventional procedures. The aim of this study is to investigate whether cerebral vasoreactivity (CVR) is a useful parameter to identify patients at risk for hyperperfusion after receiving carotid intervention. We also compare the influences on CVR in patients with either unilateral or bilateral carotid stenosis.

Method: Six patients with carotid stenosis were enrolled and divided into two groups: one is unilateral ICA tight stenosis (stenosis>= 70%), and the other is bilateral ICA tight stenosis (including ICA occlusion). Before intervention, baseline CVRs of patients were evaluated under breath-holding condition by functional MRI. Evaluations of cerebral perfusion were done by perfusion MRI before and 3–5 days after carotid intervention. Cerebral blood flow (CBF), cerebral blood volume (CBV), mean transit time (MTT), and time to peak (TTP) were generated following deconvolution of an arterial input function as parameters. Clinical hyperperfusion was defined as patients with one of following symptoms or signs after procedures: dizziness without hypotension, headache, seizures, and new onset cerebral hemorrhage.

Result: In unilateral ICA tight stenosis group, patients with reduced CVR (CVR<=10%) showed remarkable changes on post-procedural MR perfusion with increased CBF and decreased TTP. Clinical hyperperfusion syndrome was identified in one patient, who had reduced CVR. In bilateral ICA tight stenosis group, CVR was reduced over the hemisphere ipsilateral to the artery with higher grade of stenosis. The CVR did not show correlation with the MR perfusion change after carotid intervention.

Conclusion: Functional MRI is a useful and reproducible tool in evaluating CVR. Perfusion MRI is sensitive in detecting changes of cerebral perfusion. In patients with unilateral tight stenosis of carotid artery, CVR may help to identify subjects at risk for hyperfusion after carotid intervention.

CPP, BLOOD GLUCOSE, AND BRAIN EXTRACELLULAR GLUTAMATE AND GLUCOSE IN SEVERE TRAUMATIV BRAIN INJURY

Yasuhiro Kuroda¹, Susumu Yamashita¹, Kazumi Kumagai¹, Yuko Abe¹, Ken-Ya Kawakita¹, Shiro Yamashita¹, Takehiro Nakamura¹, Nobuyuki Kawai², Nobuhiro Maekawa¹, Seigo Nagao²

¹Department of Emergency and Critical Care Medicine, Kagawa University Hospital, Miki, Kagawa, Japan, ²Department of Neurological Surgery, University of Kagawa, Faculty of Medicine, Miki, Kagawa, Japan

Objective; Using brain microdialysis in severe traumatic brain injury, we investigated the time course of brain extracellular glutamate and glucose concentrations to find the relationship between Microdialysate (MD) concentrations and physiologic variables (cerebral perfusion pressure: CPP, blood glucose).

Material/Method; The study was approved by the ethic committee of Kagawa University Hospital and informed consent was obtained by the relatives. Six patients (21-69 years, 5 male. 1 female) with severe traumatic brain injury were enrolled. Neuromonitoring was performed by the consists of cerebral microdialysis (CMA60) and intracranial pressure (ICP) measurements (Codman). These probes were placed in a penumbra area (in a case with craniotomy) or right frontal lobe (in a case with burr holing) as confirmed by following-up cranial computed tomography. MD concentrations of glutamate and glucose, and value of ICP were measured every one hours.

Results; MD glutamate concentration was maximal (47 uM) immediately after probe insertion and then decreased to under 10 uM, 24 hours after brain injury in five cases with preserved CPP. MD glucose concentrations (normal value: 1.7 mM) were variable with blood glucose level (70 ? 160 mg/dl) in these patients. In one patient with no CPP, MD glutamate was sustained to increase over 300 uM and MD glucose decreased to zero mM irrespective with the value of blood glucose level.

Discussion; Initial increase of MD glutamate may reflect the degree of primary brain damage if CPP was preserved. MD glucose may change due to the blood glucose if CPP was preserved. Decrease of MD glucose in bad outcome patient could be related with the hypoxic/ischemic episode (lack of delivery of glucose and/or anaerobic glycolysis).

ACUTE SYSTEMIC ERYTHROPOIETIN THERAPY REDUCES CEREBRAL VASOSPASM AND DELAYED ISCHEMIC DEFICITS AFTER ANEURYSMAL SUBARACHNOID HEMORRHAGE: A DOUBLE-BLIND, RANDOMIZED-CONTROLLED TRIAL

Ming-Yuan Tseng, Hugh Richards, Marek Czosnyka, Peter J. Hutchinson, John D. Pickard, Peter J. Kirkpatrick

Department of Neurosurgery, Addenbrooke's Hospital, University of Cambridge, Cambridge, UK

Background: In vivo studies have demonstrated neuroprotective properties of erythropoietin (EPO), similar to ischemic preconditioning. Furthermore, acute therapy with recombinant human EPO appears to be both safe and potentially beneficial after ischemic stroke. In this phase-II randomized controlled trial, we wished to investigate whether high-dose EPO would protect against vasospasm-related delayed ischemic deficits (DIDs) after aneurysmal subarachnoid hemorrhage (aSAH).

Methods: Eighty aSAH patients (age mean 58.6, range 24-82 years) were randomized to receive iv EPO (30,000 IU Epoetin-beta, NeoRecormon, Roche, Welwyn Garden City, UK) or 0.9% saline on days 0, 2, 4 following admission starting within 72 hours of the bleed. Primary endpoints were the incidence, severity, and duration of cerebral vasospasm, and the duration of impaired autoregulation estimated from transcranial Doppler. Secondary endpoints were the incidence of vasospasm-related DIDs, and the modified Rankin scale (MRS) and National Institutes of Health Stroke Scale (NIHSS) scores at discharge.

Results: No trial-related complications were observed. Pre- and post-randomization characteristics were well balanced between the trial arms. The incidence and duration of vasospasm were significantly reduced in the EPO arm from 65% to 40% (p=0.025, Figure 1) and from 4.0 to 1.6 days (p=0.007) respectively. Impaired autoregulation was shortened from 6.6 to 3.5 days (p<0.001) for ipsilateral recordings, and vasospasm-related DIDs were reduced from 40 to 7.5% (p=0.001, Figure 2) in the EPO arm. Patients treated with EPO also demonstrated a more favorable MRS (MRS 1& 2, 40.0 vs. 67.5%, p=0.014) and lower NIHSS scores (11.7 vs. 5.1, p=0.005).

Conclusions: Acute systemic treatment with high-dose EPO appears to be safe and reduces the physiological and clinical manifestation of cerebral vasospasm after aSAH. Long-term follow-up is required to confirm that the EPO therapy improves early outcome.

NEWLY DEVELOPED FOCAL CEREBRAL ISCHEMIC AND REPERFUSION MODEL USING ENDOVASCULAR TECHNIQUE IN PRIMATES

Mitsuhito Mase¹, Fumiko Ono², Naoyuki Saito², Yuko Katakai², Keiji Terao³, Yasuhiro Yoshikawa⁴, Kazuo Yamada¹

¹Department of Neurosurgery and Restorative Neuroscience, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, ²The Corporation for Production and Research of Laboratory Primates, Tsukuba, Japan, ³Tsukuba Primate Center for Medical Science, National Institute Biomedical Innovation, Tsukuba, Japan, ⁴Department of Biomedical Science, Graduate School of Agricultural and Life Science, The University of Tokyo, Tokyo Tokyo, Japan

Background and Purpose: Control of focal cerebral ischemia and reperfusion injury is very important clinical issue. There has been many studies using rodent models, but frequently, the results could not be directly applied to human due to difference of species. The purpose of this study is to establish an easy and reproducible focal cerebral ischemia and reperfusion model in primate with following condition; 1) less-invasive procedures, 2) estimation of the severity and extent of ischemia using MRI during and after the ischemia, 3) long term survival, 4) monitoring higher cognitive function.

Materials and methods: Macaca fascicularis (n=9, BW: 3-5kg) were used. Under general anesthesia, an infusion catheter (4Fr) was introduced into the right internal carotid artery through the right femoral artery. A micro-catheter (0.021inch) was inserted into the right meddle cerebral artery (MCA) using micro-guide wire (0.010inch). The tip of the micro-catheter was at the MCA bifurcation to occlude the MCA completely. After the three-hour ischemia, a reperfusion was made by a withdrawal of the micro-catheter. Serial MR images (T1, T2, diffusion, and perfusion weighted images) were obtained using 3.0T MRI (Allegra, SIEMENS) during and after the ischemia. Regional cerebral blood flow (CBF) and volume (CBV) were calculated from the perfusion data. Neurological examinations (motor, sensory) were monitored daily, and performance in a food retrieval task (1) were assessed twice a week for 5 weeks after the ischemia. A food retrieval task is a test to estimate the capacity of spatial memory. Each animal has already been trained to complete the task for two months before the stroke.

Results: At first, preliminary experiments were performed using 5 animals to decide the endovascular occlusion method, the duration of ischemia, anesthetic methods, and suitable acquisition parameters of MR images. Afterwards, MCA occlusion and reperfusion model was made using 4 animals, which could survive until sacrifice at 5 weeks. The severity and extent of ischemia can be visualized using diffusion and perfusion weighted MR images during ischemia. Ischemic core was seen in the basal ganglia, which was extended to the frontal and temporal cortices to various extents. Severe brain swelling (edema) was shown in the right hemisphere at day 3 in all cases. At that time CBV and CBF also significantly increased, which were normalized until two weeks. All animals had left mild hemiparesis, improving quickly. Performance in a food retrieval task was severely disturbed until 2 or 3 weeks after the stroke, which was not correlated to the severity of hemiparesis, and gradually improved. Histological examination confirmed that permanent infraction existed mainly in the caudate nucleus, putamen, and temporal cortex.

Conclusion: We have developed MCA occlusion and reperfusion model in primates. This model is made by less-invasive endovascular techniques, and can allow long-term survival with estimating higher cortical function. Difference of severity and extent of ischemia of each animal can be shown on MRI. This model could be very useful to evaluate new therapeutic drugs for human cerebral ischemia.

EXACERBATED EDEMA, INFLAMMATION AND BRAIN DAMAGE IN TYPE-2 DIABETIC MICE SUBJECTED TO FOCAL ISCHEMIA

Raghu Vemuganti¹, Kellie K. Bowen¹, Jin Liang¹, Douglas L. Feinstein²

¹Department of Neurological Surgery, University of Wisconsin, Madison, WI, USA, ²Department of Anesthesiology, University of Illinois-Chicago, Chicago, IL, USA

One of the limiting factors in stroke therapeutic development is the use of healthy rodents to understand the mechanisms of ischemic neuronal death. In humans, diabetes increases the risk of stroke incidence as well as stroke-induced mortality. The db/db mouse (C57BLKS/J-m+/+Leprdb/db) is a model of type-2 diabetes with 4 times higher blood sugar than its normoglycemic genetic control (db/+ mouse). To understand the mechanisms that render diabetics to increased brain damage, we evaluated the effects of transient middle cerebral artery occlusion (MCAO) in adult db/db mice. Following a 2h transient MCAO, the mortality rate within the first day was 7-fold higher in the db/db mice over db/+ mice (n = 11/group; p<0.05). Following a 45 min transient MCAO and 3 days of reperfusion, there was a significant increase in the size of infarct (by 52±8%; p<0.05), cerebral edema (by 52±8%; p<0.05) and neurological dysfunction in db/db over db/+ mice (n = 9/group). The db/db mice also showed significantly higher post-ischemic inflammatory markers (ICAM1+ capillaries, extravasated macrophages/neutrophils and expression of inflammatory genes IL-1β, IL-6, ICAM1, MCP1 and P-selectin) compared to db/+ mice. When db/db mice were fed with a chow enriched with rosiglitazone (a PPARγ agonist known to control blood glucose levels and inflammation in diabetics), transient MCAO-induced infarction, edema, neurological dysfunction and inflammatory gene expression decreased significantly indicating the potential of PPARγ as a future stroke therapeutic target in diabetics. Funded in part by US NIH and AHA.

THE USE OF CEREBROLYSIN IN THERAPY OF STROKE-INDUCED HERPETIC INFECTION

Andrew E. Kulchikov¹, Alexander N. Makarenko², Irina G. Vasileva³

¹Orlov Regional Clinical Hospital, Orel, Russia, ²Orlov State University, Orel, Russia, ³Institute of Pharmacology and Toxicology of AMS of Ukraine, Kiev, Ukraine

Background and aims: the study of possibility of reactivation of latent herpetic infection in patients with acute ischemic stroke (AIS) and its correction by Cerebrolysin (“Ebewe”, Austria).

Methods: the patients of both sexes in the age from 55-85 years with primary AIS, in which the DNA of herpes simplex virus (HSV) was marked by means of polymerase chain reaction (PCR), were included in the study. These patients were divided in 2 groups: basic (receiving standard therapy and Cerebrolysin) and control (receiving standard therapy). Neurologic deficit was assessed by means of NIH-NINDS, modified Rankin scale and Barthel index. Infectious complications were defined by SIRS. HSV in blood was assessed by PCR method and enzyme multiplied immunoassay (EMIA). The immune status was assessed, CT/MRI of brain and laboratory tests were made.

Results: PCR on HSV in blood was positive in 98 (27,7%) from 353 patients with acute ischemic stroke. Exactly these 98 patients were included in the study and divided in 2 groups: basic and control. Cerebrolysin was injected in the dose of 10 ml intravenously during 10 days, in the first 12 hours from the onset of the disease. To the 10th day of AIS in the basic group the accelerated restoration of neurologic dysfunctions was observed (according to NIH-NINDS scale, modified Rankin scale and Barthel index), the reduction of infectious syndrome level by SIRS scale, the reduction of quantity of positive PCR and EMIA on HSV, normalization of the disturbed indexes of immune system (p<0,01) were marked.

Conclusions: in AIS in patients in terms of immunodepression the reactivation of latent herpetic infection is marked in 27,7 %, Cerebrolysin is the effective drug in therapy of this complication of stroke due to neuroimmunoresolving action.

A PROTECTIVE INVOLVEMENT OF BIP, AN ER-RESIDENT PROTEIN, ON NEURONAL DEATH INDUCED BY TRANSIENT FOREBRAIN ISCHEMIA IN GERBIL

Yasuhisa Oida¹, Masamitsu Shimazawa¹, Atsushi Oyagi¹, Takashi Kudo², Kazunori Imaizumi³, Hideaki Hara¹

¹Department of Biofunctional Molecules, Gifu Pharmaceutical University, Gifu, Japan, ²Department of Clinical Neuroscience, Psychiatry, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan, ³Division of Molecular and Cellular Biology, Department of Anatomy, Faculty of Medicine, University of Miyazaki, Kiyotake, Miyazaki, Japan

Background and aims: Endoplasmic reticulum (ER)-stress is caused by the accumulation of unfolded proteins in the ER lumen, and is associated with stroke and neurodegenerative diseases such as Parkinson's and Alzheimer diseases. The purpose in the present study was to investigate the involvement of ER-stress in the execution of cell death in transient forebrain ischemia model of gerbil. In addition, we evaluated the effects of a selective inducer of BiP to examine whether induction of BiP inhibits the neuronal cell death induced by the ischemia.

Methods: We assessed the expression patterns of immunoglobulin heavy chain binding protein (BiP)/glucose-regulated protein (GRP) 78, an ER-resident molecular chaperone whose expression serves as a good marker of ER-stress, activating transcription factor (ATF)-4, and C/EBP homology protein (CHOP) by immunohistochemistry and/or Western blotting after transient forebrain ischemia in gerbils. Double-fluorescent staining of CHOP immunohistochemistry and the terminal deoxynucleotidyl transferase-mediated DNA nick-end labeling (TUNEL) method were performed to clarify the involvement of CHOP in cell death.

Results: Immunohistochemical and Western blot analyses of the CA1 showed that BiP expression increased at 12 hr, reached a peak at 3 days, and gradually decreased thereafter as compared with that of control group. A transient increase was shown in CA3 at 1 day after ischemia, but unchanged in dentate gyrus and cortex. Furthermore, the signal for ATF-4 and CHOP was increased at 1 day and 3 days in CA1, and 12 hr in CA3. Co-localization of CHOP immunoreactivity and DNA fragmentation were detected by the TUNEL method 3 days after ischemia in CA1, but not in CA3. Moreover, we evaluated the effect of a selective inducer of BiP (BiP inducer X; BIX) against global transient forebrain ischemia. BIX significantly induced the expression of BiP mRNA in in vivo. Furthermore, BIX protected the cell death and the increase of TUNEL-positive cells induced by the ischemia.

Conclusions: These findings indicate that ER stress may play a pivotal role in the post-ischemic neuronal death in the hippocampal CA1 subfield and that the induction of BiP could prevent the neuronal damage.

SELECTIVE OCCLUSION OF THE MIDDLE CEREBRAL ARTERY IN MICE BY AN INTRALUMINAL APPROACH

Jianya Ma, Thaddeus S. Nowak Jr.

Department of Neurology, University of Tennessee Health Science Center, Memphis, TN, USA

Background and Aims: A significant limitation in experimental stroke research is the heterogeneity of models employed, which complicates comparison of results among laboratories. This is a particular concern with intraluminal filament methods, both because filament design and fabrication vary and because the internal carotid artery (ICA) obstruction and diffuse ischemia they produce limit their application to short occlusion intervals, often in strains with appreciable collateral perfusion. Technical issues and strain variation concerns are amplified in mice. This report describes implementation of an occluding device for mice consisting of a silicone cylinder molded onto nylon filament that can traverse the ICA and enter the middle cerebral artery (MCA) to produce selective, reversible occlusion, extending recently published methodology for rats (1).

Methods: Silicone cylinders of either 150 or 200 micron diameter and averaging 500 microns in length were molded onto 8–0 nylon filaments 10 mm in length. These were introduced into the cranial circulations of halothane-anesthetized adult (22-25 g) male mice of BALB/c, C57BL/6 and DBA/2J strains via the left external carotid artery as routinely described for intraluminal occlusions. Filaments were heparinized and mice received 100 units/kg subcutaneous heparin to reduce risk of secondary occlusion due to clot formation along the filament. After 24 hours brains were removed to assess filament position and circle of Willis anatomy, and to measure infarct volumes after triphenyltetrazolium staining.

Results: Successful MCA entry was achieved in 70% (7/10) of BALB/c and 81% (47/58) of C57BL/6 using 200 micron devices, and 88% (7/8) of DBA/2J after diameter reduction to 150 microns. Failures resulted from acute branch angles at the origin of the MCA. Cortical infarct volumes were 86 +/− 14 and 39 +/− 4 cubic mm (mean +/− SD) in BALB/c and C57BL/6 mice, respectively, with no infarcts seen in DBA/2J mice, in agreement with previous data on relative vulnerability to focal ischemia in these strains (2),(3). No mortality was observed.

Conclusions: These results demonstrate the application of a refined intraluminal occlusion method in mice, using filaments manufactured to close tolerances. Dimensions can be modified to accommodate variations in vascular caliber among strains. Selective occlusion of the MCA permits the study of long durations of focal ischemia while maintaining the advantages of the widely used intraluminal approach.

Supported by USPHS grant NS42267 (TSN)

EARLY CHANGES OF GENE EXPRESSION IN THE PERIFOCAL BRAIN TISSUE AFTER TRAUMATIC INJURY

Marc Schuerings¹, Markus Haak², Li Li², Felix Engel³, Norbert Gretz², Lothar Schilling¹

¹Division of Neurosurgical Research, Medical Faculty Mannheim, University of Heidelberg, Heidelberg, Germany, ²Medical Research Center, Medical Faculty Mannheim, University of Heidelberg, Heidelberg, Germany, ³Department of Molecular Genetics, German Cancer Research Center (DKFZ), Heidelberg, Germany

DIFFERENTIAL REGULATION OF ANGIOTENSIN RECEPTOR EXPRESSION AFTER EXPERIMENTAL BRAIN TRAUMA

Serge Thal¹, Sebastian Wyschkon¹, Raimund Trabold², Nikolaus Plesnila², Kristin Engelhard¹, Christian Werner¹

¹Department of Anaesthesiology, Johannes Gutenberg University, Mainz, Germany, ²Laboratory of Experimental Neurosurgery, Department of Neurosurgery and Institute for Surgical Research, University of Munich Medical Center, Grosshadern, Munich, Germany

*Introduction: *Several studies have investigated the influence of the renin-angiotensin system on secondary brain damage after cerebral ischemia. Angiotensin II receptor type 1 (AT1) seems to increase cerebral damage while angiotensin II receptor type 2 (AT2) may offer protection, preventing secondary brain damage. No studies have investigated the modulation of AT1 and AT2 after brain trauma. We, therefore, measured the expression of AT1 and AT2 after brain trauma in mice to evaluate the potential time window of a therapeutic intervention.

*Methods:* Male C57Bl/6 mice were anaesthetized with isoflurane (0.8 %) and a pneumatic brain trauma was induced on the right parietal cortex (controlled cortical impact (CCI), 8 m/s, 1 mm depth). Brain tissue was removed 15 min, 3, 6, 12 and 24 hr (n=6 each) after CCI and from naive animals (n=7). The absolute amount of AT1 and AT2 mRNA was determined by real-time RT-PCR (Lightcycler^R) in the traumatised and the contralateral hemisphere. The results were normalized against beta-2-microglobin as control gene. Statistics: ANOVA on RANKS, p<0.05.

*Results:* The expression of AT1 was significantly reduced at the site of contusion (15 min: −10 %, 3 hr: −10 %, 6 hr: −20%, 12 hr: −30 %*, 24

hr: −60 %* vs. naive, p<0.05). In the contralateral, non-traumatised tissue the AT1 expression was reduced 15 min after trauma, returning to baseline 3-12 hr after trauma, and finally dropping again at 24 hr (15 min: −20 %, 3 hr: 0 %, 6 hr: 0 %, 12 hr: −10 %, 24 hr: −20 % vs. naive).

The expression of AT2 in the contusion area was increased in comparison to healthy animals 15 min until 12 hr after CCI (+150 to 200 %), finally dropping to −20 % vs. naive 24 hr after CCI. In the contralateral hemisphere the AT2 expression was also increased by 200 %, dropping to 150 % 24 hr after CCI.

*Conclusion:* Brain trauma results in an immediate up-regulation of AT2 and down-regulation of AT1 in the injured hemisphere. The AT1/2 expression profile is similar to data in brain ischemia. The beneficial regulation of AT1 and AT2 in injured tissue might represent a cellular rescue mechanism. This suggests that a therapeutic intervention at AT2 within a window of opportunity (3-12 hr) after CCI would encounter increased expression of AT2 with a potential reduction of secondary brain damage.

REGULATION OF CALCIUM IN CEREBROVASCULAR ENDOTHELIUM BY TRANSIENT RECEPTOR POTENTIAL (TRP) CHANNELS

Sean Marrelli^1,2, R. Bryan^1,2

¹Department of Anesthesiology, Baylor College of Medicine, Houston, TX, USA, ²Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, TX, USA

IN VIVO SIMULTANEOUS OBSERVATION OF CEREBRAL MICROVESSELS DURING REPERFUSION AFTER BRAIN ISHCHEMIA IN RATS USING SYNCHROTRON RADIATION

Masanori Morita¹, Motohisa Ohkawa¹, Syuhei Miyazaki¹, Keiji Umetani², Kouichirou Suzuki¹

¹Department of Emergency Medicine, Kawasaki Medical School and Kawasaki Medical School Hospital, Kurashiki, Okayama, Japan, ²Japan Synchrotron Radiation Research Institute, SPring-8, Sayo-Gun, Hyogo, Japan

BACKGROUND AND PURPOSE: To clarify cerebral circulation disorders after ischemia and reperfusion, evaluation in vivo by visualization of intracerebral microvessels is appropriate. However, usual method is difficult to examine the blood vessels in the entire cerebrum including the striate arteries distributed in the cerebrum. A new angiography system, which consists of monochromatic synchrotron radiation (MSR) as an X-ray source and a high-resolution camera with a video system as a detector, has been developed in Hyogo, Japan. This angiography system has nearly parallel X-rays, and consequently can visualize microvessels of 20-30 µm in diameter. We aimed at in vivo simultaneous observation of superficial cortical and intracerebral microvessels during reperfusion after transient forebrain ischemia.

METHODS: Transient brain ischemia was induced by the method of ten-minute four-vessel occlusion in rats under general anesthesia. Angiographic images were then sequentially obtained at 3 frames/second. Changes in the cerebral microvessels were observed before and 1, 5, 10, 15, 20, and 30 min after transient cerebral ischemia using the MSR angiography system.

RESULTS: Diameters of ICA, MCA, and striate artery significantly increased 1 minute after reperfusion, while the pial arteriole caliber significantly decreased (80% of baseline). Thereafter, MCA, pial arteriole and striate artery significantly dilated 5 and 10 minutes after reperfusion. Although the ICA caliber of 30-minute reperfusion significantly decreased as compared with the basal value, the other 3 vessels' calibers maintained above basal value throughout the experiment. Early venous filling was observed at 5 and 10 minutes after reperfusion.

CONCLUSION: We herein describe and discuss the use of an in vivo experimental model in which changes in transient cerebral ischemia in the rat cerebral microcirculation were clearly depicted with an MSR angiography system. The MSR angiography system is useful for investigating morphological changes in both cortical and central branches of cerebral vessels in rats during reperfusion after cerebral ischemia. The pathophysiologic basis of early venous filling is incompletely understood. Clear detection of early venous filling in rats with the MRS angiography system could help us clarify it.

ENDOTHELIN RECEPTOR ANTAGONISM BLOCKS CALPONIN PHOSPORYLATION FOLLOWING BRAIN TRAUMA

Christian Kreipke, Randy Morgan, Jose Rafols

Department of Anatomy and Cell Biology, Wayne State University, School of Medicine, Detroit, MI, USA

INCREASE IN PO2 ON THE RAT BRAIN SURFACE BY PHTOACTIVATION

Satoshi Kimura, Katuyoshi Mineura

Department of Neurosurgery, Kyoto Prefectural University of Medicine, Kyoto, Japan

Introduction: The phenomenon that photostimulation induced vascular relaxation was reported previously (Ref. 1). In preliminary experiments, we found that it induced a rapid arteriolar vasodilation on rat brain surface (Fig.1). We developed the novel technique that we could measure the PO2 on the brain surface, using the oxygen-sensitive fluorescent membrane (Brain 05). We modified this technique to measure the PO2 on the brain and to photostimulate it simultaneously. The aim of this study was to investigate that the effect of photostimulation influences the oxygenation of rat brain.

Methods: An oxygen-quenching fluorescence dye, tris (1,10-phenanthroline) ruthenium (II) chloride hexahydrate, was embedded in a highly gas-permeable, opaque silicone-rubber film formed on a microscope coverslip. In this study, we modified the membrane to allow the one-tenth excitation light to reach brain surface. This modification enabled us to evaluate the PO2 on the brain surface and to photostimulate it simultaneously. Under anesthesia, the rat brain was subjected to 1 minute photostimulation after 1 minute shading. The trends of the PO2 on the brain were measured and analysed.

Results: We obtained 13% elevation of PO2 after stimulation, but then the PO2 of control period was unchanged. The difference of oxygenation was statistically significant.

Conclusion: We detected the increasing oxygen tension of the brain surface by photoactivation.

EXTRAORDINARY HIGH FLOW CAPILLARIES IN THE CEREBRAL MICROVASCULAR NETWORK

Miyuki Unekawa¹, Minoru Tomita¹, Yutaka Tomita², Takashi Osada¹, Haruki Toriumi¹, Norihiro Suzuki¹

¹Department of Neurology, Keio University School of Medicine, Tokyo, Japan, ²Department of Neurology, Institute of Brain and Blood Vessels, Mihara Memorial Hospital, Gunma, Japan

Capillary RBC velocity in the brain is temporally and spatially heterogeneous. Recent advances in technology with a high-speed laser-scanning confocal fluorescence microscope and Matlab domain home-made software (KEIO-IS2) developed by us have made it possible to examine the vast number of RBC appearing in the cortical microvasculature quantitatively. In this paper, we attempted to examine the frequency distribution function of RBC velocities in intraparenchymal tissue capillaries of the rat cerebral cortex. The method has been reported elsewhere (Tomita at al., JCBFM 25: S164, 2005). Briefly, we used 37 urethane-anesthetized Wistar rats with a closed cranial window in the temporo-parietal region. After FITC-labeled RBCs were injected into the circulating blood, the labeled RBCs observed in the ROI at 50 micrometer in depth below the brain surface were tracked at 500 frames/s and their velocities calculated with KEIO-IS2. Capillaries were identified one by one on FITC-dextran stained small vessels with a criteria of a diameter less than 10 micrometer. We obtained 4252 RBC velocities of single capillaries from approximately 100 samples each in 37 rats. The velocity ranged from 0.15 to 9.40 mm/s with a mean of 2.05±1.59 mm/s. The RBC number (N) was plotted in units 0.5 mm/s apart (Fig. 1). As shown in the frequency distribution function, capillary velocities were clustered at around 2.0 mm/s, smearing at higher velocities. Although RBC velocity increases were continuous, we tested the outlying value X statistically based on (x(mean)?X)/s, where s the standard deviation (Grubbs FE, Ann Math Stat 21: 27, 1950). We found that the capillaries having RBC velocity higher than 6.5 mm/s can be excluded statistically significantly (p<0.01). They must belong to vessels in a different category. The amount of oxygen delivered by arterial blood (D) is estimated by c×n×f, where c is the arterial concentration, n the capillary number (N in Fig.1), and f the flow linearly related to the velocity. Oxygen extracted at the place capillaries enter into the tissue is expressed as E = 1 ? e(exp)-B/f, where B is a constant related to the capillary surface area and length. The equation implies that gas in fast flowing blood is less extracted, with little gas exchange in the tissue and a large amount of gas reaching the vein. When E was subtracted from D, the amount of shunting blood can be obtained. Broad calculation of the shunting blood over the total arterial blood yielded 22% of the blood appearing in the vein without gas exchange. This might result from what have been called thoroughfare channels or non-nutritional capillaries carrying 22 % of the blood in reserve.

INTERINDIVIDUAL VARIATION OF BASELINE CBF: RELATION WITH CEREBRAL VASCULAR TONE ESTIMATED BY PET

Hiroshi Ito¹, Iwao Kanno¹, Masanobu Ibaraki², Kazuhiro Takahashi¹, Tetsuya Suhara¹, Shuichi Miura²

¹Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan, ²Department of Radiology and Nuclear Medicine, Akita Research Institute of Brain and Blood Vessels, Akita, Japan

Introduction: Inter-individual variations in normal human CBF at rest condition have been reported. Inter-individual variation of cerebral vascular tone it considered to contribute to this, and several determinants of cerebral vascular tone have been proposed. In the present study, the relation of inter-individual variation between CBF and cerebral vascular tone at rest condition was investigated by PET.

Methods: CBF and CBV at rest condition were measured by PET with H215O and 11CO, respectively, in 20 normal subjects. CBF at rest condition and cerebral vascular response to changes in PaCO2, which was calculated as the percentage changes in CBF per absolute change in PaCO2 (%/mm Hg), were measured by PET with H215O in another 20 normal subjects. Regions-of-interest were defined for the neocortical regions.

Results: The relation between CBF and CBV at rest condition was CBV = 1.00CBF^0.30. A significant negative correlation between baseline CBF and the vascular response to hypocapnia was observed (P<0.05). A tendency of negative correlation, but not significant, between baseline CBF and the vascular response to hypercapnia was also observed.

Conclusions: Because CBV is an indicator of cerebral vessel diameter if vessel length does not change, the relation between CBF and CBV, which is in good agreement with Poiseuille's law, indicates that inter-individual differences in CBF are related to those in cerebral vessel diameter corresponding to cerebral vascular tone. The correlations between baseline CBF and vascular responses support the assumption that cerebral vascular tone might incline toward vasoconstriction and vasodilatation when CBF is low and high between individuals, respectively. Although several determinants of cerebral vascular tone have been proposed, the mechanism of such inter-individual differences in cerebral vascular tone is unknown.

DILATION OF THE MIDDLE CEREBRAL ARTERY OF THE RAT DURING ACUTE HYPONATREMIA IS ENDOTHELIUM-DEPENDENT

Katarzyna Romaniuk, Ewa Kozniewska

Department of Neurosurgery, Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland

BACKGROUND AND AIMS. Hyponatremia is defined as an electrolyte disorder occurring when serum sodium level falls below 135 mmol/l. It is frequently observed after brain trauma or subarachnoid bleeding and is typically associated with increased plasma level of vasopressin (AVP). Although hyponatremia has been reported to impair the recovery of neurosurgical patients its effect on brain vasculature is practically unknown. Therefore, present study was carried out to determine the effect of low sodium and increased AVP concentration in the extracellular fluid on the calibre of the middle cerebral artery (MCA) of the rat.

METHODS. MCAs (n=38) harvested from the brains of the adult, male Wistar rats were used for this study. The vessels were mounted in the small organ chamber filled with 3-(N-morpholino) propanesulfonic acid (MOPS) ? buffered saline solution and pressurized. The chamber was placed on an inverted microscope equipped with videocamera. Changes in the diameter of MCA were observed on the monitor and analyzed with a help of a customer made software.

RESULTS. Decrease of the extracellular concentration of Na+ from 140 to 120 mM resulted in the dilation of MCA by 17% (from 143±10 to 167±10 micrometers, p<0.01). Dilation was abolished when endothelial cells were mechanically removed or NO synthesis was inhibited. Addition of 15 pg/ml of AVP to the extraluminal fluid during normonatremia prior to the administration of low sodium buffer did not affect the diameter of MCA. However, when low sodium buffer was given together with AVP, constriction of MCA by about 16% (p<0.02) instead of dilation was observed.

CONCLUSIONS. These results demonstrate for the first time that during acute hyponatremia healthy cerebral blood vessels dilate through endothelium- and NO-dependent mechanism. However, when endothelial control is abolished or impaired as following head trauma or subarachnoid hemorrhage, hyponatremia may result in vasoconstriction, particularly when associated with elevated plasma vasopressin levels. This finding may explain adverse vascular effects of AVP-associated hyponatremia observed in vivo.

THE HEMODYNAMIC RESPONSE DURING FUNCTIONAL ACTIVATION REFLECTS COUPLING RATHER THAN UNCOUPLING

Olaf B. Paulson^1,2, Egill Rostrup²

¹Neurobiology Research Unit, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark, ²MR Department, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark

Cerebral blood flow and cerebral metabolism is normally coupled, i.e., an increase in metabolic demand in the tissue and will lead to an increase of flow. However, during functional activation flow and glucose metabolism remain coupled as they increase proportionally, whereas oxygen metabolism only increases to a minor degree ? the so-called uncoupling of flow and oxidative metabolism. This physiological phenomenon forms the basis for studies of cerebral activation using PET or fMRI. Several studies have dealt with these aspects and theories have been forwarded regarding the mechanisms underlying the changes. Special attention and speculations and have been directed towards a potentially deficient oxygen supply to the tissue most distant from the capillaries.

In the present presentation we forward the theory that the essential mechanism responsible for the flow increase during functional activation is a tight coupling between flow and glucose metabolism. The uncoupling of flow and oxidative metabolism then remains a simple consequence of a less pronounced increase in oxygen consumption. Based in the literature on the cerebral microcirculation it is emphasized that physiological capillary heterogeneity and physiological recruitment of capillaries is mandatory in order to accomplish the increased demand for glucose supply to the cerebral tissue. These basic aspects of the cerebral microcirculation will result in an increased physiological density of capillaries in the cerebral tissue and thereby also to surplus in oxygen supply. The basic literature underlying these concepts will be discussed and related to the theories advanced in the field.

ESTIMATION OF OEF USING SPECT AND PERFUSION CT

Yoshikatsu Kawata

Department of Neurosurgery, Rumoi City Hospital, Rumoi, Hokkaido, Japan

[Objectives]The oxygen extraction fraction (OEF) is a very useful indicator of cerebral circulatory insufficiency. At present, PET is the only method to measure OEF, and the measurement of OEF is not available in general institutions without PET facilities. Therefore, we attempted to estimate OEF values from CBF and mean transit time (MTT) measured by SPECT and perfusion CT, which are available in general institutions. [Subjects and methods] We examined 444 areas (222 pairs), for which CBF and MTT were measured by SPECT and Perfusion CT at our institution, and 132 areas (66 pairs), for which CMRO2, OEF, CBF, and MTT values were available in the literature. The areas without cerebral infarctions on brain CT, and those whose cerebral circulatory dynamics were impaired unilaterally were subjected to the measurement at our institution. The results were then evaluated by a ratio of the measurements to those on the unaffected side. [Theoretical background]Assuming that CMRO2 values are normal (i.e. CMRO2 ratio = 1, ratio: affected side/unaffected side), then the OEF ratio equals the (1/CBF) ratio. Therefore, the theoretical OEF ratio can be estimated. However, when CMRO2 values are not normal, the OEF ratio cannot be estimated without further information. In this study, we determined the relationship among the CBF values, the MTT values, and the theoretical OEF values in the areas that are supposed to have normal CMRO2 values measured at our institution, and examined the deviation of the measured CBF and OEF values in the areas with abnormal CMRO2 values described in the literature from the above relationship. We considered that, if some relationship can be found between them, the OEF ratio may be estimated without measuring OEF. [Results] In the study of subjects measured at our institution (n=222 pairs), the CBF ratio showed a tendency to decrease linearly as the MTT ratio increased. The relationship can be expressed as the following formula: CBF ratio = −0.321(MTT ratio)+1.267 (R^2 = 0831, n = 222)(Formula 1). Assuming that the CMRO2 values are normal, then the OEF ratio = (MTT ratio)/{1.267(MTT ratio)-0.321} (Formula 2). The relationship between the CBF ratio and the OEF ratio was examined including the areas with abnormal CMRO2 in the literature. The ratio of the measured CBF ratio to the CBF ratio obtained from Formula 1 after measuring the MTT ratio (CBF’), and the ratio of the measured OEF ratio to the OEF ratio obtained from Formula 2 (OEF’) showed highly significant negative correlation {OEF'= −682CBF'+ 1.58 (R^2 = 0.658)}. When this regression line is used as a calibration curve, the OEF ratio can be estimated if the CBF ratio and the MTT ratio are known. Based on this, we calculated the theoretical OEF ratios of the subjects in the literature only from the CBF ratios and the MTT ratios, and they coincided with the measured OEF ratios very well. [Conclusion] It was considered that the OEF values can be estimated accurately from CBF and MTT.

ASSOCIATION OF POSTOPERATIVE CEREBRAL HYPERPERFUSIION WITH PREOPERATIVE HEMODYNAMIC IMPAIRMENT AND INTRAOPERATIVE ISCHEMIA

Nobukazu Komoribayashi¹, Kuniaki Ogasawara¹, Masakazu Kobayashi¹, Takeshi Fukuda¹, Hideo Saitoh¹, Kazunori Terasaki², Takashi Inoue¹, Akira Ogawa¹

¹Deparment of Neurosurgery, Iwate Medical University, Morioka, Japan, ²Cyclotron Research Center, Iwate Medical University, Morioka, Japan

EPSILON PKC INTERACTS WITH MITOCHONDRIAL ATP-DEPENDENT POTASSIUM CHANNEL DURING ISCHEMIC PRECONDITIONING IN ORGANOTYPIC SLICE CULTURES

Miguel A. Perez-Pinzon, Ami Raval, Kunjan R. Dave

Department of Neurology and Neuroscience, Cerebral Vascular Disease Research Center, University of Miami School of Medicine, Miami, FL, USA

Background: Mitochondrial ATP-dependent potassium (mtK+ATP) channel activation has been reported to mediate cardiac ischemic preconditioning (IPC) via generating reactive oxygen species (ROS). The role of the mtK+ATP channel in IPC induction in the brain is still not well defined. Previously we demonstrated that IPC-induced neuronal survival in the hippocampal CA1 required epsilon protein kinase C (εPKC) activation (1), (2). Thus, the goal of the present study was to define whether mtK+ATP channel activation is required during the triggering phase of IPC. Further, we hypothesize that εPKC-mediated neuroprotection requires mtK+ATP channel phosphorylation.

Methods: Hippocampal slices were obtained from 9-11 days old Sprague Dawley rats and cultured for 14-15 days. Slices exposed to oxygen/glucose deprivation (OGD) for 40 min ('test’ ischemia) were used as the ischemia group. Slices exposed to OGD for 15 min (or a pharmacological agent), 48 h prior to ‘test’ ischemia were used as the IPC/pharmacological preconditioned (PPC) groups. Propidium Iodide (PI) fluorescence images were obtained using a SPOT CCD camera and were digitized using SPOT advanced software. Percentage of relative optical intensity was used as an index of cell death. We confirmed the presence of mtK+ATP channel specific subunits (Kir 6.1 and Kir 6.2) in hippocampal mitochondria by Western blotting. Immunoprecipitation was carried out to study phosphorylation of Kir 6.2 in different experimental conditions. Results are expressed, as mean ± SEM. Statistical significance was determined with an ANOVA test followed by a Bonferroni's post-hoc test.

Results: Opening of mtK+ATP channel (diazoxide, 50 µM) 48 h prior to ischemia protected the hippocampal CA1 and mimicked neuroprotection mediated by IPC/PPC with the εPKC agonist. PI fluorescence level of ischemic-, IPC-, and mtK+ATP channel agonist (diazoxide) treated groups were 50.6 ± 5.2 (n = 6); 28.5 ± 4.9 (n = 4); 16.8 ± 7.6 % (50 mM; n = 8; p < 0.001, compared to ischemia) respectively. Blockade of the mtK+ATP channel using 5-hydroxydecanoic acid (HD) abolished IPC/εPKC mediated neuroprotection. PI fluorescence values of HD-treated and IPC groups were 58.13 ± 15.9 % (n =12) and 29.7 ± 15.7 % (n =8), respectively (p < 0.001). Similarly, PI fluorescence values of εPKC agonist and εPKC agonist plus HD-treated groups were 33.5 ± 6.7 % (n = 5), and 54.3 ± 3.3 % (n = 5), respectively. Finally, we determined if ischemic and εPKC agonist-mediated PPC leads to phosphorylation of mtK+ATP channel specific subunits Kir 6.2. The levels of Kir 6.2 phosphorylation were higher by 79 (n = 3, p<0.01) and 104% (n = 3, p<0.01) in IPC and PPC groups, respectively, when compared with mitochondria from sham treated slices (n = 3). Inhibition of εPKC activation after IPC prevented Kir 6.2 phosphorylation, suggesting that εPKC-mediated Kir 6.2 phosphorylation following IPC.

Conclusion: The signal transduction pathway that ensues following IPC/εPKC agonist-mediated PPC requires opening of the mtK+ATP channel.

Grant support: Supported by PHS grants NS34773, NS045676, NS054147

SEX-SPECIFIC ANTI-OXIDANT CAPACITIES IN ASTROCYTES

Mingyue Liu, Nabil J. Alkayed, Esteban Oyarzabal, Patricia D. Hurn

Department of Anesthesiology and Peri-Operative Medicne, Oregon Health and Science University School of Medicine, Portland, OR, USA

Introduction: Neuronal cell death after exposure to various cytotoxins is sex-specific; male cells being more sensitive to nitrosative stress and excitotoxity than female neurons (1). We previously demonstrated that astrocytes cultured from female brain (XX) are less susceptible to cell death induced by oxygen-glucose deprivation (OGD) and hydrogen peroxide (H2O2) than male (XY) astrocytes (2). In the current study, we tested the hypothesis that male and female astrocytes respond differently to oxidative stress, in part due to differences in antioxidant capacity between male and female astrocytes.

Methods: Primary cultured cortical astrocytes were prepared from 1-3-day old male and female rat pups separately and grown to confluency in steroid-free medium. Male and female rat pups were distinguished by a larger genital papilla and longer ano-genital distance in male vs. female pups. Sex was confirmed by inspecting internal organs and gonads after laporatomy, e.g. uterine horns in females and ductus deferens in males, and by multiplex polymerase chain reaction (PCR) using male-specific marker SRY (sex determination region on the Y chromosome, responsible for testes formation) and the universal marker myogenin (Myog), which is expressed in both males and females. At day 14 in vitro (div), cultures were homogenized for protein and mRNA isolation. The protein levels of three antioxidant enzymes: glutathione peroxidase (GPx), superoxide dismutase (SOD) and catalase (CAT), were determined by Western blotting in male and female astrocytes. Furthermore, GPx mRNA level was determined by RT-PCR, and GPx activity was measured in male and female astrocytes using a glutathione peroxidase activity assay kit measures GPx activity indirectly by a coupled reaction with glutathione reductase.

Results: The protein levels of SOD and CAT were not different between male and female astrocytes. However, the level of GPx protein was 3-fold higher in female compared to male astrocytes (3±0.6 fold, n=4, p<0.05). Similarly, GPx mRNA and activity were significantly higher in female compared to male astrocytes in female compared to male astrocytes (2±0.2 fold, n=3, p<0.05 for the mRNA and 2.4±0.1 fold, n=3, p<0.05 for the activity).

Conclusions: We conclude that female astrocytes express higher levels of GPx and exhibit higher GPx activity than male astrocytes. The sex difference in GPx expression and activity may afford female astrocytes with higher anti-oxidant capacity, and may underlie the difference in ischemic cell death between male and female astrocytes.

ESTRADIOL IS PROTECTIVE AGAINST ISCHEMIC BRAIN DAMAGE THROUGH SIGNAL TRASDUCER AND ACTIVATOR OF TRANSCRIPTION 3 (STAT3)

Suzan Dziennis¹, Taiping Jia², Oline K. Ronnekleiv², Patricia D. Hurn^1,2, Nabil J. Alkayed^1,2

¹Department of Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA, ²Department of Physiology and Pharmacology, Oregon Health and Science University, Portland, OR, USA

Background and Aims: Estradiol is neuroprotective in animal models of cerebral ischemia (1). Although the precise mechanisms are unknown, estrogen's beneficial properties are in part linked to rapid activation of signal transduction pathways that lead to the transcription of neuroprotective genes (2). Signal transducer and activator of transcription 3 (STAT3) is a transcription factor that is activated in response to both estrogen (3) and ischemia (4). Upon activation, STAT3 translocates to the nucleus where it upregulates neuroprotective genes, such as bcl-2 (5). We wished to determine whether estradiol replacement enhances STAT3 activation in female rat brain after focal cerebral ischemia and if STAT3 activation plays a protective role against ischemic brain injury. Methods: Adult female rats were ovariectomized (OVX) at 10-12 weeks and implanted with 25 µg 17s-estradiol (EST) 7 days prior to middle cerebral artery occlusion (MCAO). A 2-hour MCAO was induced by the intraluminal filament technique. Laser-Doppler cerebrocortical perfusion, mean arterial pressure, PO2, PCO2, pH, glucose and terminal plasma estrogen were measured. A single intraperintoneal injection of the STAT3 inhibitor JSI-124 (1mg/kg, was given 15 minutes after MCAO. Brains from separate groups were removed and processed for one of the following analyses: 1) brains were subdissected and rapidly frozen to prepare cortical nuclear and cytosolic extracts for phospho-STAT3 (P-STAT3) and total-STAT3 (T-STAT3) immunoblot analysis, 2) brains were sectioned and stained with 2,3,5-triphenyltetrazolium chloride (TTC) for infarct size measurement, or 3) brains were fixed in 4% paraformaldehyde for immunohistochemistry (IHC) with P-STAT3, Bcl-2 and MAP2 antibodies. Phospho-STAT3 was measured at 3 and 22 hours post reperfusion. IHC and infarct volume were measured at 22 hours post reperfusion. Results: Physiological variables were not different among groups. Estradiol increased P-STAT3 in the cytosolic and nuclear fractions at 3 hours following MCAO. In estradiol replaced animals, P-STAT3 remained elevated in nuclear fractions at 22 hours in ischemic cortex. Furthermore, the nuclear P-STAT3 co-localized with cells expressing neuronal and survival markers MAP-2 and Bcl-2. The P-STAT3 inhibitor JSI-124 abolished the effect of estradiol to reduce infarct size after MCAO in ovariectomized rats. Conclusion: Estradiol enhances STAT3 phosphorylation in ischemic brain, and activated STAT3 contributes to the hormone's neuroprotective property.

Grant support: Supported by F32 NS053065 (SD), NS 33668 (PDH) PO1 NS049210 (OKR, PDH and NJA)

DELAYED NEURONAL PRECONDITIONING WITH BMS-191095 IS MEDIATED BY CATALASE

Tamas Gaspar¹, Anna Busija¹, Bela Kis¹, Keita Mayanagi¹, Christina Tulbert¹, James A. Snipes¹, Ferenc Bari², David W. Busija¹

¹Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston Salem, NC, USA, ²Department of Physiology, Faculty of Medicine, University of Szeged, Szeged, Hungary

Background and aims. Selective opening of the mitochondrial ATP sensitive potassium (mitoKATP) channels with BMS-191095 (BMS) is an effective way to induce tolerance against toxic stimuli. Previous works from our laboratory demonstrated that stress-induced free radical production was completely abolished in BMS-preconditioned neurons, suggesting that improved free radical scavenging capacity plays an important role in the neuroprotective effect of BMS. The aim of our present in vitro study was to identify the mechanisms of delayed preconditioning induced by BMS.

Methods. Cortical neurons of 18-day old Sprague-Dawley rat fetuses were cultured in B27 supplemented Neurobasal medium. To induce preconditioning, the neuronal cultures were treated with BMS (30 µM) once a day for 1, 2, or 3 days. Twenty-four hours after the last treatment, neuronal cultures were exposed to glutamate (200 µM) for 60 min. In some experiments, different concentrations of the catalase inhibitor 3-aminotriazole (3AT; 1, 10, 100 mM) were applied to the cultures during and for 3 hours after glutamate excitotoxicity. Cell viability was evaluated with CellTiter 96 AQueous One Solution assay on the next day. ATP levels in cortical neurons were measured using CellTiter-Glo luminescent assay. The expression of manganese dependent superoxide dismutase (MnSOD), cupper-zinc dependent superoxide dismutase (CuZnSOD), catalase, Bcl-2, and heat shock protein 70 (HsP-70) was analyzed using Western blotting. The enzymatic activity of SODs and catalase was measured with a microplate reader using commercially available kits.

Results. BMS treatment resulted in higher ATP levels in resting cells (in 1E-11 mol/well: untreated, 10.65 ± 0.21; BMS 30 µM, 11.84 ± 0.22*; *p<0.05 vs. untreated). Neurons preconditioned with BMS could preserve ATP during exposure to glutamate (in 1E-11 mol/well: untreated, 4.4 ± 0.04; BMS 30 µM, 7.01 ± 0.10*; *p<0.05 vs. untreated). Treatment with BMS resulted in a dose-dependent elevation of catalase expression in both transcriptional and translational levels whereas the protein levels of MnSOD, CuZnSOD, Bcl-2, and HsP-70 did not change. A parallel increase in the enzymatic activity of catalase was observed while the activity of SODs did not change. Catalase overexpression could not be blocked with the putatively selective mitoKATP antagonist 5-hydroxydecanoate. The neuroprotective effect of BMS, however, could be partially and dose-dependently antagonized using 3-AT (cell viability after exposure to glutamate: untreated, 54.5 ± 1.32%; BMS 30 µM, 90.6 ± 3.12%*; BMS 30 µM + 3-AT 1 mM, 81.3 ± 1.32%*#; BMS 30 µM + 3-AT 10 mM, 69.9 ± 1.70%*#; BMS 30 µM + 3-AT 100 mM, 67.3 ± 2.01%*#; *p<0.05 vs. untreated; #p<0.05 vs. BMS-treated group).

Conclusion. Our results indicate that improved energy homeostasis and increased catalase expression, and activity play important roles in the neuroprotective effect of the mitoKATP opener BMS-191095.

MICROPLASMIN REDUCES DOSE-DEPENTLY MICROVASCULAR BASAL LAMINA DAMAGE IN EXPERIMENTAL FOCAL CEREBRAL ISCHEMIA

Dorothe Burggraf¹, Milan Vosko¹, Marion Schubert¹, Martin Dichgans¹, Gerhard Hamann²

¹Department of Neurology, Ludwig-Maximilians University Munich, Klinikum Grosshadern, Munich, Germany, ²Department of Neurology, HSK Dr. Horst Schmidt Klinik, Wiesbaden, Germany

Background and aims: Until now the only approved pharmacological therapy for acute stroke patients remains thrombolysis, by using recombinant tissue-type plasminogen activator (rt-PA, Alteplase). Thrombolytic therapy of ischemic stroke with rt-PA improves clinical outcome, but rt-PA also causes secondary intracerebral bleeding. Novel agents that achieve higher recanalization rates, lower hemorrhage rates, or both, might be able to improve the delicate balance of benefit and risk. An alternative substance could be microplasmin, a derivative of plasmin lacking the 5 “kringle” domain. We examined the effect of different microplasmin doses on microvascular damage, blood brain barrier disruption and the alternation from MMP-2 and MMP-9 and their endogenous inhibititors after focal cerebral ischemia in rats.

Methods: The suture model was used to induce a constant period of ischemia (3h) followed by reperfusion for 24h in rats. Microplasmin was injected with the chosen dose (0.5 mg, 1 mg, 2.5 mg, 5 mg or 10 mg; each N=6) at the end of ischemia. A control group was treated with saline. Another group was sham-operated.

Structural injury was detected by Nissl staining. The loss of collagen type IV, a major component of the basal lamina, was quantified by collagen western blotting. The amount of both gelatinases MMP-2 and MMP-9 was quantified by zymography. Their endogenous inhibitors (tissue inhibitor of MMPs: TIMPs) were analyzed by reverse zymography.

Results: We detected a significant reduction (p=0.01) of infarct volume in 10 mg microplasmin group (51.0 ± 22.6mm3) compared with the control group (167.3 ± 13.1mm3). A significant reduction (p<0.05) of basal lamina damage in basal ganglia region could be determined in 10 mg microplasmin animals in comparison to control group: 87±4% vs. 75±2%. With decreasing doses of microplasmin the infarction volume and the microvascular damage is decreasing (p<0.05; ANOVA). The extravasation of hemoglobin in basal ganglia was articulately reduced in 10 mg microplasmin (154±24%) treated rats compared with control group (442±124%). The MMP-activity (MMP-2 and MMP-9) was decreasing with rising microplasmin doses (p<0.05). However, the amount of the corresponding TIMPs remained unchanged by different doses microplasmin. Thus the balance between the proteases (MMP) and their inhibitors was shifted towards the inhibitory side.

Conclusions: Microplasmin showed a dose-dependent protective effect on microvascular basal lamina and blood brain barrier. With the high microplasmin dose (10 mg) the best neurological and microvascular protection could be detected. These results suggested that microplasmin might carry less risk for hemorrhage than tPA treatment of ischemic stroke. Thus, microplasmin may be safer than tPA in terms of damage of the microvasculature associated with thrombolytic therapy of ischemic stroke.

ROLE OF SPECIFIC MITOCHONDRIAL ATP-DEPENDANT POTASSIUM CHANNEL IN ISCHEMIC TOLERANCE

Megumi Watanabe^1,2, Ken-ichiro Katsura¹, Genki Mizukoshi¹, Ikuroh Ohsawa², Sadamitsu Asoh², Shigeo Ohta², Yasuo Katayama¹

¹Division of Neurology, Nephrology and Rheumatology, Department of Internal Medicine, Nippon Medical School, Tokyo, Japan, ²Department of Biochemistry and Cell Biology, Institute of Development and Aging Sciences, Graduate School of Medicine, Nippon Medical School, Kanagawa, Japan

Background and aims: When brain cells are subjected to brief ischemia, they become resistant to a subsequent lethal ischemic insult. We have reported roles of mitochondrial ATP dependent potassium (mito-KATP) channels in ischemic tolerance phenomenon for 2 different in vivo ischemic model. Damages in CA1 neuronal cells of gerbil global ischemia model, or infarct volumes of rat transient MCA-occlusion model, were attenuated because of ischemic tolerance phenomenon after preceding brief ischemic insult. The mito-KATP channel inhibitor, 5-hydroxydecanoate (5HD), canceled the tolerance, and mito-KATP channel opener, diazoxide (DZX), enhanced the tolerance. To clarify the mechanism of ischemic tolerance, we studied roles of mito-KATP channels on the ischemic tolerance phenomenon by using brain primary culture.

Methods: To mimic a in vivo ischemic condition, cells were exposed to oxygen-glucose deprivation (OGD). Cells were placed in OGD medium (without glucose, gassed with 95% N2/5% CO2), and incubated for the desired time with a chamber gassed with 95% N2/5% CO2. Survival neuron was distinguished with its morphological changes. At first, Cells were subjected to 10 min of OGD as a sublethal ischemia (PC), and 24 hours later, subjected to 80 min of OGD as lethal ischemia. 5HD and DZX were administrated 30 min before the lethal ischemia in the medium. We also measured mitochondrial membrane potential with tetra-methyl-rhodamine methyl ester (TMRM) under confocal micrography.

Results: In preconditioned groups cells were significantly more survived than in control group. In 5HD administrated group, the number of surviving cells were significantly reduced. DZX administration did not reduced the number of surviving neurons. Preconditioned cells were relatively depolarized, and 5HD administration inhibited the depolarization induced by the precondition.

Conclusions: Our results may suggest that mito-KATP channels play a key role in neuroprotection in ischemic tolerance phenomenon through mitochondrial membrane potential. Mitochodrial depolarization generated by ischemic tolerance phenomenon, may lead to inhibiting mitochondrial calcium overload upon lethal ischemia, and thereby protecting neurons.

CINNAMOPHILIN ATTENUATES GRAY AND WHITE MATTER DAMAGE AND IMPROVES NEUROBEHAVIORAL AND ELECTROPHYSIOLOGICAL RECOVERIES FOLLOWING TRANSIENT FOCAL CEREBRAL ISCHEMIA

Hung-Yi Chen^1,2, Tsung-Ying Chen^1,3, Chieh-Chih Huang¹, Yu-Hsiang Kuan¹, E-Jian Lee¹, Tian-Shung Wu^1,4

¹Neurophysiology Laboratory, Neurosurgical Service, Department of Surgery, National Cheng Kung University Medical Center and Medical School, Tainan, -R.O.C., ²Institute of Pharmacy, China Medical University, Taichung, Taiwan-R.O.C., ³Department of Anesthesiology, Buddhist Tzu-Chi University and Buddhist Tzu Chi General Hospital, Hualien, Taiwan-R.O.C., ⁴Department of Chemistry, National Cheng Kung University, Tainan, Taiwan & National Research Institute of Chinese Medicine, Taipei, Taiwan-R.O.C.

Introduction: Optimal functional integration requires not only the preservation of the cortex and the subcortical cytoarchitectures, but also an integrity of the anatomical, electrophysiological and metabolic links between gray and white matter [1]–[2]. Cinnamophilin (CINN; 20-80 mg/kg) dose-dependently protects against ischemic lesion via decreasing postischemic oxidative damage in mice [2]. We herein evaluate whether CINN could decrease gray and white matter damage and, therefore, enhance the restoration of neurobehavioral and electrophysiological outcomes following transient focal cerebral ischemia.

Methods: Male Sprque-Dawley rats, weighing 240-280 gm, were subjected to intraluminal suture occlusion for 60 minutes. CINN (80 mg/kg) or vehicle was given intravenously at 1 hour after reperfusion. Physiological parameters, including core temperature, local cortical perfusion and arterial blood gases, were measured before, during and after cerebral ischemia-reperfusion, whereas the somatosensory potentials evoked from each forepaw and hindpaw region were recorded prior to the ischemic insult and at 7 days after reperfusion. Postmortem gray matter damage was determined by quantitative image analysis of Nissl-stained sections. White matter damage was evaluated immunohistochemically by SMI-31 and myelin basic protein (MPB).

Results: CINN (80 mg/kg, i.v., n=9) given 2 h after the ischemic insult significantly reduced infarct volumes (27.3±7.3% vs. 39.9±5.8%, P<0.005) and improved neurobehavioral outcome by 47.4% (P<0.001), compared with those in control animals (n=9). CINN treatment was also noted to result in more surviving neurons in cortical (582±203 vs. 207±100, per mm2, P<0.001) and subcortical (201±67 vs. 646±153, per mm2, P<0.001) regions, compared to controls. In addition, CINN not only significantly improved somatosensory responses evoked from the contralateral, non-affected fore- and hindpaw by 33.2% and 26.9% (P<0.001, respectively), but also significantly improved contralateral transcallosal diaschisis evoked from the affected forepaw by 8% (P<0.05). Furthermore, we have observed that CINN significantly preserved more phosphorylated component-H of neurofilaments by 33.1% (52.6±12.4% vs. 19.6±12.7%, P<0.005) and improved the integrity of myelin by 28.2% (61.4±7.4% vs. 44.1±16.2, P<0.05) in the ischemic hemisphere.

Conclusion: Delayed treatment with CINN not only protects against gray and white matter damage, but also offers the advantage of enhancing neurohavioral and electrophysiological outcomes following transient focal cerebral ischemia in rats. These findings support that the agent may be worthwhile for further investigation for its clinical implications in the treatment of ischemic stroke.

THE SELECTIVE ESTROGEN RECEPTOR MODULATOR LY362361 IS NOT NEUROPROTECTIVE IN A RAT MODEL OF TRANSIENT FOCAL ISCHEMIA

Tracy Deanne Farr, Hilary V.O. Carswell, I. Mhairi Macrae

Wellcome Surgical Institute, University of Glasgow, Glasgow, Scotland, UK

The results from the Women's Health Initiative trial indicated that women receiving estrogen were 39% more likely to suffer a stroke than women receiving placebo. This has raised concern regarding the health risks of chronic estrogen prescription, and surprised the pre-clinical research community, since estrogen has consistently been shown to provide neuroprotection from stroke. However, there are some reports that reveal estrogen's capacity to exacerbate ischaemic injury (Bingham, et al., 2005. JCBFM. 25: 414; Theodorsson and Theodorsson, 2005. Peptides. 26: 2257; Harukuni, et al., 2001. Brain Res. 900: 127). In light of this, emphasis has been placed on selective estrogen receptor modulators (SERMs), which may offer improved alternatives to estrogen as neuroprotectants. The present study investigated the novel, brain penetrable SERM, LY362321, in a rat model of transient middle cerebral artery occlusion (MCAO). Female Sprague Dawley rats were ovariectomized and randomized to receive daily subcutaneous injections of 1mg/kg (n=11), 10mg/kg (n=10) of LY362321, or vehicle (n=11); the experimenter was blinded to the treatments. The left MCA was temporarily occluded (90minutes), with cortical blood flow monitoring, at 12 days post ovariectomy. Sensorimotor function was assessed using a neurological score prior to and daily following MCAO. Infarct size was assessed at 3 days post-MCAO on 1% 2, 3, 5-triphenyltetra-zolium chloride (TTC) stained sections. The results indicated that there were no significant differences amongst groups in cortical blood flow during the MCAO. Furthermore, there was no significant difference in infarct size (one way ANOVA (F(2,32) = 0.542, p = 0.587) amongst vehicle, 1, and 10mg/kg treated animals (22.9 · 16.4, 16.7 · 14.6, and 21.1 · 12.9, respectively, mean ± SD of % total brain area) (Figure 1). The MCAO induced a significant decline in neurological score in the vehicle group (from 14 down to 7 at 24hrs post-MCAO) but this was not affected by LY362321 at either dose. In conclusion, pretreatment with a low or high dose of the novel SERM, LY362321, did not significantly influence cerebral blood flow, infarct volume, or sensorimotor function in rats exposed to transient MCAO.

CEREBRAL METABOLIC RATE OF OXYGEN IS A POTENTIAL MARKER OF INSULT SEVERITY FOLLOWING HYPOXIA-ISCHEMIA IN NEWBORNS

Kenneth M. Tichauer¹, Jennifer Hadway¹, Daisy Y.L. Wong², R. Jane Rylett², Ting-Yim Lee^1,3, Keith St. Lawrence¹

¹Imaging Program, Lawson Health Research Institute, London, ON, Canada, ²Cell Biology Research Group, Robarts Research Institute, London, ON, Canada, ³Imaging Research Laboratories, Robarts Research Institute, London, ON, Canada

HEMODYNAMIC CHANGES AND OXYGEN METABOLISM DURING EVENT RELATED AND PROLONGED ACTIVATION OF A DISTINCT CORTICAL COLUMN IN RAT SOMATOSENSORY CORTEX

Georg Royl¹, Christoph Leithner¹, Heike Kaasch¹, Jan Philipp Mueller¹, Dirk Megow¹, Jens Steinbrink¹, Matthias Kohl-Bareis², Ulrich Dirnagl¹, Ute Lindauer¹

¹Department of Experimental Neurology, Charite - Universitaetsmedizin Berlin, Berlin, Germany, ²University of Applied Sciences Koblenz, RheinAhrCampus Remagen, Remagen, Germany

2D optical imaging of hemoglobin concentration and oxygenation was combined with cerebral blood flow (CBF) imaging with Laser Speckle Contrast Analysis to investigate the neurovascular response of the somatosensory cortex to mechanical deflection of a single whisker hair in the anesthetized rat. After optical imaging, a multi-electrode array was inserted into the area of activation and the experiment was repeated while measuring multiunit activity to accurately localize the site of neuronal activation (the activated whisker barrel). Two different stimulation paradigms were employed. One paradigm (single event) consisted of a single whisker hair deflection, randomized with a minimal interstimulus time of 15s. For the other paradigm (prolonged activation) the whisker hair was deflected periodically (4Hz) for 16s, with a minimal interstimulus time of 60s. Within the activated whisker barrel, this prolonged whisker stimulation led to a transient increase in CBF (5.0 ± 1%), total Hemoglobin (HbT, 0.7± 0.2%) and cerebral metabolic rate of oxygen (CMRO2, 3.7% ± 1%) and a decrease in deoxygenated Hemoglobin (Deoxy-Hb, −0.6 ± 0.3%), relative to baseline values. However, a single whisker deflection did not lead to a significant change in HbT and Deoxy-Hb while still causing an increase in CBF and CMRO2 by ~1.8% ± 0.7%. This dissociation of flow and hemoglobin related signals can be attributed to a short velocity related CBF increase without a significant change of cerebral blood volume or oxygenation. In conclusion, oxygenation related functional imaging like BOLD fMRI might be inferior to CBF based techniques when investigating fast single event related neuronal activation. Implications from this study considering the temporal and spatial profile of neurovascular and neurometabolic coupling will be discussed.

VASCULAR TERRITORIES MEASUREMENT USING ARTERIAL SPIN LABELING WITH A LOCALIZED LABELING RF COIL

Fernando F. Paiva^1,2, George C. Nascimento¹, Alberto Tannus², S. Lalith Talagala¹, Afonso C. Silva¹

¹National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA, ²Instituto De Fisica De Sao Carlos, Universidade De Sao Paulo, Sao Carlos, SP, Brazil

PREDICTION OF HYPERPERFUSION SYNDROME AFTER CAROTID STENTING BY PERFUSION CT

Chien-Hung Chang¹, Chi-Jen Chen³, Shy-Chyi Chin², Yeu-Jhy Chang¹, Ting-Yu Chang¹, Tai-Cheng Wu¹, Shan-Jin Ryu¹, Tsong-Hai Lee¹

¹Department of Neurology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan-R.O.C., ²Department of Diagnostic Radiology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan-R.O.C., ³Department of Diagnostic Radiology, E-Da Hospital and I-Shou University, Kaohsiung, Taiwan-R.O.C.

INTERLEUKIN-6 STIMULATES CIRCULATING HUMAN ENDOTHELIAL PROGENITOR CELL (EPC) ANGIOGENESIS IN VITRO

Y. Fan^1,2, J. Ye⁵, F. Shen^1,2, Y. Zhu^1,2, Y. Yeghiazarians⁴, W. Zhu^1,2, Y. Chen^1,2, M. Lawton³, W. Young^1,2,3,4, Guo-Yuan Yang^1,2,3

¹Center for Cerebrovascular Research, ²Anesthesia and Perioperative Care, ³Department of Neurological Surgery, ⁴Department of Neurology, ⁵Department of Internal Medicine

Introduction: Endothelial progenitor cells (EPCs) are defined as cells in bone marrow, circulating blood or adjacent tissue, which are positive for both stem cell markers CD34 and CD133, and endothelial cell markers such as VWF and KDR. EPCs contribute to postnatal vasculogenesis during wound healing, ischemic injury or tumor growth. EPCs respond to many growth factors and cytokines; however, it is still unclear if interleukin-6 (IL-6), a multifunctional cytokine that plays an important role in angiogenesis and vascular remodeling, has effects on EPCs. Here, we provide evidence that IL-6 stimulates EPC angiogenic changes in vitro.

Methods: For EPC culture, mononuclear cells were isolated from circulating blood using gradient centrifuge with Histopaq 1077 and seeded in fibronectin-coated plates. Dil-acLDL uptake, FITC-UEA binding assay and immunofluorescence of CD34, VWF, KDR and VE-cadherin were used to characterize the cultured EPCs. IL-6 receptor CD126 was detected on EPCs using immunofluorescence, and further identified by phosphorylation of ERK1/2 and STAT3 in response to IL-6 stimulation using Western blot. Finally, to determine if IL-6 stimulation causes EPC angiogenic changes, we performed EPC proliferation assay using BrdU incorporation assay, EPC migration assay using transwell system, and EPC tube formation using MetraGel. All the results were analyzed using a one-way ANOVA with Fisher's comparison tests. A random p< 0.05 is considered statistically significant.

Results: EPC colonies formed after 2 weeks in culture. The cultured EPCs grew like cobble stones and were able to uptake Dil-acLDL, bind FITC-UEA functionally and staining positive for cell markers CD34, KDR, VWF, and VE-cadherin. CD126 was expressed on EPCs. After stimulation with IL-6, two IL-6 downstream signal molecules ERK1/2 and STAT3 were phosphorylated in a dose-dependent manner and were inhibited by antibodies against IL-6 and IL-6 receptor. IL-6 induced EPC proliferation with a maximal effect at 25 ng/ml. Enhanced EPC migration had an optimal effect at 50 ng/ml compared with non-treated control groups (p<0.01), and showed a dose-dependent manner. Furthermore, EPC tube formation on MetraGel increased after different concentrations of IL-6 stimulation (p<0.01), with an optimal dose of 50 ng/ml. These effects were blocked significantly after using both IL-6 neutralizing antibody and IL-6 receptor antibody (p<0.01). The results demonstrated that IL-6 stimulation induced EPC angiogenic changes in vitro.

Conclusion: We successfully cultured EPCs form human circulating blood. We provide, for the first time, the evidence that EPCs express functional IL-6 receptor and that IL-6 can stimulate EPC proliferation, migration and tube formation upon binding of its receptor. These results provide a novel clue that EPCs may be involved in IL-6-related diseases.

CORRELATIONS BETWEEN FUNCTIONAL AND STRUCTURAL OBSERVATIONS AFTER TRANSPLANTATION OF HUMAN EMBRYONIC STEM CELL-DERIVED PROGENITORS INTO THE ISCHEMIC RAT BRAIN

Christine Seminatore¹, Jerome Polentes¹, Valerie Itier³, Laetita Aubry², Aurore Bugi², Thierry Delzescaux⁴, Marc Dhenain⁴, Philippe Hantraye⁴, Anselme Perrier², Brigitte Onteniente¹

¹Inserm Umr 549, Paris, France, ²Inserm Umr 861, Istem, Evry, France, ³Inserm Emi 0011, Creteil, France, ⁴Cea, Cnrs 2210, Orsay, France

Cell therapy with stem cell-derived neural progenitors has become a challenging medical perspective. Neural progenitors from a variety of stem cell sources can differentiate into neural cell types after transplantation into the ischemic brain. In some instances graft survival and cell differentiation into the required neural cell types are correlated with improvements of functional deficits. However, little is known on the possibilities to evaluate the efficacy of cell therapy in stroke in settings closer to the clinic, i.e. over a long period of time without sacrifice of the animals.

The aim of this study was to validate non invasive investigation techniques for embryonic stem cells (ESC) therapy in stroke using human ESC (hESC). We analyzed the differentiation and survival properties of early neural progenitors derived from the SA-001 hESC line (Cellartis AB, Sweden) over 4 months after transplantation into the striatum of adult male Sprague-Dawley rats 7 days after a 2 hours occlusion of the middle cerebral artery (tMCAO). Rats were analyzed with T2 magnetic resonance imaging (MRI, 3 Tesla) and two behavioral tests, the apomorphin-induced rotation test (Apo) for motor impairments, and the adhesive removal test (ADR) for sensory-motor impairments. Animals received an immunosuppressive tri-therapy containing cyclosporine A, azathioprin and methyl-prednisolone for 5 weeks.

MRI allowed identification of the graft as an hypo dense signal. 3D reconstitution was used to measure infarct and graft volumes. No correlations were found between infarction severity and graft volumes, nor between infarction severity and body weight modifications. Infarction severity correlated with Apo and ADR deficits. Transplantation into the peri-lesional parenchyma resulted in good survival of cells with neural (Pax6, GFAP, nestin) phenotype, although some non-neural cells (e.g. cytokeratin-14 positive) were also found. Grafted cells were surrounded by DARPP-32 fibers specific for the medium-spiny GABAergic host striatal neurons. When transplanted into the lesion cavity, hESC-derived progenitors gave rise to few neural cells and mainly to teratoma-like structures. No teratoma were formed when progenitors were transplanted into the unlesioned striatum, but the number of cytokeratin-14-labeled cells increased significantly. Teratoma-like structures could not be differentiated from neural grafts on 3T T2 MRI, indicating the limits of the technique. MRI also revealed a stabilization of the lesion at 1 month. Evolution of the lesion until 1 month was confirmed post-mortem by the presence of conspicuous inflammatory reaction at the lesion border in the absence of graft.

Partial (up to 40%) recovery of motor and sensory-motor functions were observed at 4 months in animals transplanted into the parenchyma. Motor deficits displayed a bi-phasic curve, with a partial recovery at 1 month followed by degradation of Apo test scores. This was not observed for the ADR test, suggesting an increased D2-receptors activity of GABAergic cells remaining on the lesioned side.

SC and PJ were equally involved in the work

HYPOFRONTALITY INDUCED BY VIOLENT VIDEO GAME PLAYING IN YOUNG ADULTHOOD

Yuan-Hwa Chou¹, Tung-Ping Su¹, Bang-Hung Yang², Shyh-Jen Wang²

¹Department of Psychiatry, Taipei Veterans General Hospital, Taipei, Taiwan-R.O.C., ²Department of Nuclear Medicine, Taipei Veterans General Hospital, Taipei, Taiwan-R.O.C.

Background and Aims: Violent video game playing is popular worldwide. It has been reported that brief exposure to a violent video game increased aggressive behavior. However, the effect of violent video game playing on human brain function is still unclear. The purpose of the study was to determine the effect of violent video game playing on cerebral blood flow (CBF) in young adulthood.

Methods: Twelve healthy subjects (averaged age was 22±3.2 years) who received three times single photon emission computed tomography (SPECT) examination with radiotracer 99mTc ECD for measuring CBF were recruited. The first SPECT measurement was measured as baseline, second and third SPECT measurements as study tasks. Each task measurement was performed after 30 min violent and mentality video game playing. There was two-week interval between SPECT measurements. Statistic Parametric Mapping (SPM2) was used for the mapping of CBF.

Results: The results showed that CBF in left frontal lobe was decreased (green area) significantly, whereas CBF was increased (red area) in occipital cortex after violent video game playing (Fig 1). However, the effect can not be found in CBF after mentality video game playing. To our knowledge this is the first data demonstrated hypofrontality induced by violent video game playing in young adulthood. The finding of hypofrontality is similar to those previous studies which showed hypofrontality in patient with major depression or schizophrenia.

Conclusion: As frontal lobe function is crucial in our cognitive function, executive function and emotion, we suggest that clinical study regarding relationship of violent video game playing and mental disorders has to be conducted in adolescence or young adulthood urgently.

THE FEASIBILITY OF 3D TENSOR TRACTOGRAPHY AND 3DAC AS A MODALITY TO ASSESS MOTOR FUNCTION IN PATIENTS WITH CEREBRAL INFARCTION

Keiji Igase¹, Yoshiaki Kumon¹, Shigeyuki Nagato¹, Kazuhiko Sadamoto², Takanori Ohnishi¹

¹Department of Neurosurgery, Ehime University, Toon, Ehime, Japan, ²Department of Neurosurgery, Washo-Kai Sadamoto Hospital, Ehime, Japan

Background and aims: In recent years 3D tensor tractography (DTT) has been applied for many CNS diseases to depict neuronal fibers through measuring a fractional anisotropy, meanwhile 3 dimensional anisotropic contrast (3DAC) has been recognized as an imaging method for delineating true neuronal fibers just as they were. However, there seems to be only a few reports that aim to evaluate the feasibility of both DTT and 3DAC to predict a motor recovery in patients suffering from cerebrovascular diseases. Thus we have embarked on investigating the possibility of these 2 methods focusing on a prediction in motor outcome in patients with cerebral infarction.

Methods: Nineteen patients with a brand-new lacunar infarct, who underwent DTT at least twice in each acute (mean 3.8 days) and subacute phase (mean 19.5 days), were enrolled, out of which only 4 patients underwent 3DAC study. Patients were separated to 2 groups, recovery and nonrecovery. Patients categorized to former group had almost complete recovery in motor function 3 month later, while those to latter had a remaining paresis even 3 month after the onset. DTT was implemented with 3 tesra MRI (SignaExcite; GE) and analyzed with dTV.IISR produced by the department of Radiology, Tokyo University. Pyramidal tract was delineated by setting each ROI, the cerebral peduncle for seed point and the motor cortex for target point. The number of pyramidal fibers was recognized as drawn lines obtained from a result display in this software, the ratio of numbers in the injured side to those in the intact side was measured. Meanwhile 3DAC image was produced with an imaging software through assigning 3 directions (anterior-posterior: A-P, right-left: R-L, superior-inferior: S-I) to RGB color.

Results: In acute phase mean ratio of both recovery and non-recovery groups were 70.8±21.6% and 63.5±23.4% respectively, resulting in no significant difference between two groups, meanwhile in subacute phase the ratio of recovery group was 100.5±28.3%, which revealed a significant difference compared with that of non-recovery group 46.8±10.3% (p<0.05). Furthermore, in a recovery group there was a significant difference between mean ratio in acute and that in subacute phases (p<0.001). 3DAC image obtained in acute phase disclosed that fibers in S-I direction appeared to be predominantly injured in non-recovery group.

Conclusions: The feasibility of DTT in predicting a recovery of motor function in patients with cerebral infarction has been shown, and the possibility that 3DAC image can estimate a motor outcome even in faster phase than DTT study has been suggested as well. Collecting cases might make an apparent answer to these questions.

ROLES OF BMPS IN THE DEVELOPMENT OF GLIOVASCULAR UNITS

Tetsuya Imura, Shinji Fushiki

Department of Pathology and Applied Neurobiology, Kyoto Prefectural University of Medicine, Kyoto, Japan

SURVIVAL AND DIFFERENTIATION INTO NEURONS OF TRANSPLANTED ES CELL-DERIVED PRECURSORS WITHIN STROKE LESION

Klaus G. Reymann¹, Claudia Buehnemann¹, Christian Bernreuther², Marcel Dihne², Holger Braun¹, Christoph Y. Malik³, Andreas Scholz³, Melitta Schachner²

¹Leibniz Institute for Neurobiology, Project Group Neuropharmacology, Magdeburg, Germany, ²Centre for Molecular Neurobiology, Universitaetskrankenhaus Hamburg-Eppendorf, Hamburg, Germany, ³Institute for Physiology, Giesen, Germany

NEUROGENESIS FOLLOWING CEREBRAL ISCHEMIA IN CEREBRAL CORTEX AND WHITE MATTER

Yasuaki Kokubo, Yasuhiko Matsumori, Atsushi Kuge, Sunao Takemura, Shinya Sato, Takamasa Kayama

Department of Neurosurgery, Yamagata University Faculty of Medicine, Yamagata, Japan

Backgrounds: Although cerebral cortex and white matter play an important role in brain function, there are few reports on neurogenesis in cerebral cortex and white matter following cerebral ischemia. Therefore we investigated cortical neurogenesis following focal cerebral ischemia (study1) and the dynamics of oligodendrocyte in corpus callosum with correlation of cognitive and motor function following chronic cerebral ischemia (study2) in rats.

Materials and Methods: (study 1) Male Sprague-Dawley (SD) rats (270-300 g) were subjected to 60-min middle cerebral artery occlusion. Proliferating cells were labeled by the cumulative administration of 5-bromo-2′-deoxyuridine-5′-monophosphate (BrdU) at 1, 2, 3, 4, 6, and 8 weeks after ischemia. We evaluated BrdU, NeuN positive cells in hippocampal dentate gyrus (DG) and the subventricular zone of the lateral ventricle (SVZ), and cerebral cortex by immunohistochemistry. (study 2) Male SD rats (210-250g) were subjected to permanent occlusion of bilateral common carotid arteries. We evaluated NG2 (maker of premature olidogendrocyte) and GST-π(maker of mature olidogendrocyte)positive cells in corpus callosum by immunohistochemistry after 2, 4, 6, 8 weeks of ischemia. For evaluation of cognitive and motor function, Morris's water-maze test and rotarod test were performed, respectively.

Results: (study1) BrdU-positive cells co-expressed NeuN in peri-infarct zone of cerebral cortex at 3 to 8 weeks after cerebral ischemia. BrdU-positive cells proliferated in the DG and SVZ. Their proliferation peaked at 3 weeks in the SVZ and at 4 weeks in the DG. (study 2) NG2 positive cells were increased at 4weeks of ischemia and GST-π positive cells were decreased at 2weeks of ischemia and then increased in corpus callosum, significantly. Deterioration of cognitive function was observed after 2weeks of ischemia. There was no motor dysfunction after ischemia. Although regeneration of oligodendrocyte was observed after 4 weeks of ischemia, deterioration of cognitive function was still remained.

Conclusions: We found no obvious migration of neural progenitor cells in cerebral cortex after ischemia. Therefore we postulate that endogenous dormant neural stem cells residing in the cortex are activated by ischemic insult to induce the proliferation of neural progenitors and differentiation into mature neurons. On the other hand, GST-π positive cells were increased after 4weeks of ischemia following decrease at 2 weeks. This result suggests that regeneration of oligodendrocyte occurs following chronic cerebral ischemia. However, we need further investigation regarding correlation with cognitive function. Although the issue is still controversial, progress of research on neurogenesis in cerebral cortex and white matter is required to develop the treatment for functional recovery of stroke patients.

GLUCOSE ACCELERATES DEATH AT REOXYGENATION IN A MODEL OF IN VITRO ISCHEMIA

Anna Serra-Perez, Anna M. Planas, Tomas Santalucia

Department of Pharmacology and Toxicology, IIBB-CSIC/IDIBAPS, Barcelona, Catalonia, Spain

Background and aims: Reduction of glucose and oxygen supply to the brain parenchyma can lead to neuronal cell death. Despite the immediate deleterious effects of ischemia, part of the subsequent tissue damage might be due to reperfusion. We have investigated the role of glucose in the induction of death after oxygen and glucose deprivation (OGD) followed by reoxygenation in an in vitro model.

Methods: Our model consisted of human neuroblastoma-derived SH-SY5Y cells differentiated with retinoic acid, which were subjected to 15h of OGD followed by a reoxygenation period of variable length, in which cells were incubated in control glucose concentrations.

Results: Counting of propidium iodide-positive nuclei in photomicrographs showed a significantly higher proportion of dead cells at 15h of OGD compared to cells that had been kept in the presence of glucose during anoxia for the same length of time. The difference widened in the hours following restoration of the oxygen and glucose supply. Nevertheless, incubating the cells with the glucose transport inhibitor cytochalasin B during reoxygenation or omitting glucose from the reoxygenation medium delayed cell death. This indicated that the utilization of glucose during reoxygenation impaired cell survival. The effects of glucose do not seem to depend on the availability of ATP, since analysis of the cellular ATP content showed that, although it was severely depleted at the end of the 15h-OGD, levels were eventually restored to almost normal after reoxygenation, regardless of the presence of glucose. We observed that by the end of the 15h-OGD, both cytochrome C and AIF were released from the mitochondrial fraction into the cytosol, which hinted at possible mechanisms mediating cell death after OGD and reoxygenation and that could involve the activation of apoptotic pathways. Nonetheless, we found no activation of caspase 3 during either the OGD or the reoxygenation periods, but an increase in the activity of calpain (as deduced from the analysis of the cleavage fragments of α-spectrin), which is often associated with necrotic death. Although nuclear import of AIF released from mitochondria has been associated with the development of caspase-independent apoptosis, we found no evidence for such import in our model.

Conclusions: In all, our data suggest that under these conditions OGD commits cells to a delayed death that is executed during reoxygenation in a way that requires the utilization of glucose by an as yet unknown mechanism.

TIME COURSE OF HUMAN SALIVARY STRESS PEPTIDES (AMYLASE, IGA, CHROMOGRANIN A) AFTRE PSYCHOLOGICAL STRESSOR EXAMINATION, UCHIDA-KRAEPELIN TEST

Koreaki Sugimoto¹, Aya Kanai¹, Noriaki Shoji²

¹Department of Psychosomatic Medicine, Tohoku Fukushi University, Sendai, Japan, ²Department of Oral Diagnosis, Tohoku University Graduate School of Dentistry, Sendai, Japan

[BACKGROUND AND AIM]Salivary peptide such as salivary amylase, IgA, and chromogranin A are known to be related to stress response. These peptides are secreted with noradrenalin exocytosis in the salivary, and show up-regulation under stressful conditions. So these are considered stress markers. Salivary is easy to collect without invasion. But the time course of stress response of these peptides is not clear. In this study we measured the human salivary α-amylase, IgA, and chromogranin A from the same sample which are taken from volunteers who were stressed by Uchida-Kraepelin test (U-K test). [METHODS] Volunteers (n=8) are all healthy women, aged 20-30 years old, and have had no medications, pregnant, oral disease and habituation of alcohol drunk. They didn't have food and drink for 30 min before the test. U-K test, psychological stressor as well as personality examination, was performed to these volunteers for 30 min at room temperature in routine way. Their salivary were collected before the test and 5, 10, 15, 20, 25, 30, 40, 50, 60 min after the test. Salivary α-amylase was examined by chemiluminescence's method (dry chemistry system), and IgA and chromogranin A were examined by enzyme immuno assay. [RESULTS AND CONCLUSIONS] Salivary α-amylase did not increased after U-K test, however, salivary IgA and chromogranin A were increased. A previous report showed that salivary α-amylase did not increase after U-K test, but increased and responded quickly while U-K test performance. Salivary IgA and chromogranin A are good stress markers after psychological stress conditions. Moreover we analyzed the relationship between salivary stress response and person's character obtained by U-K test.

This work was carried out in the Kansei Fukushi Research Center of Tohoku Fukushi University, Sendai, Japan, supported by “Academic Frontiers” Project for Private Universities: matching fund subsidy from the Ministry of Education, Culture, Sports, Science and Technology(2004~2008).

EXPRESSION OF GNRH BY ELECTROACUPUNCTURE IN RATS IN DIFFERENT ESTROUS CYCLE

Shaojun Wang, Bing Zhu, Zhigao Jin

Institute of Acupuncture & Moxibustion, China Academy of Chinese Medical Sciences Beijing,100700, China

PACAP has been found neuroprotective in numerous models of cerebral ischemia. PACAP-induced neuroprotection in vivo is partially based on its vasorelaxant effect. Cerebral ischemia occurring during the perinatal period often result in neurological sequelae. One possible contributor to the neurological damage is the cerebrovascular dysfunction, e.g. hypercapnia-induced arteriolar vasodilation is severely attenuated after ischemia reperfusion (I/R). The mechanism of PACAP induced cerebrovascular changes in newborns is unexplored.

We sought to characterize the vascular reactivity to PACAP-27 and −38 (naturally existing forms), as well as to shorter PACAP sequences in a piglet model that correlates well with the vascular physiology of newborn babies. Using the non selective cyclooxygenase (COX) inhibitor indomethacin (5 mg/kg, iv), the selective COX-1 inhibitor SC 560 (1 mg/kg, iv) and the selective COX-2 inhibitor NS 398 (1 mg/kg, iv), we investigated if COX-derived metabolites play a role in the cerebrovascular effects of PACAP. We also determined the effect of the nitric oxide synthase (NOS) inhibition by N-nitro-l-arginine methyl ester (L NAME, 15 mg/kg, iv). We examined the influence of short PACAP fragments (10-5 M topically) on PACAP-27 and-38 induced vasodilation. Further, we tested whether application of PACAP preserves hypercapnia-induced cerebrovascular dilator responses after I/R. Anesthetized (Na thiopenthal 40 mg/kg, ip, followed by α chloralose 40 mg/kg, iv), ventilated piglets (1 day old, 1 2 kg, n=104) were equipped with closed cranial windows. Pial arteriolar diameters were determined via intravital microscopy.

Topical PACAP-27 and −38 elicited similar, repeatable, dose dependent pial arteriolar dilations. Percent changes in diameters to 10-8, 10-7, and 10-6 M PACAP-38 and-27 were 6±1, 16±2, and 40±4, and 9±2, 19±3, 36±4 (mean±SEM), respectively. In contrast, the shorter segments of the peptides (6-27, 6-38, 6-15, 20-31, and 1-15) did not display any vasoactivity. Arteriolar dilations to PACAP-38 were abolished 20 min after indomethacin and SC-560 (to 10-6 M 35±4% vs. −1±1%*, and 36±6% vs. 1±1*, *p<0.05), but the responses were unaffected by NS-398 (to 10-6 M 39±4% and 45±8%, before and after the treatment, respectively). L NAME was also ineffective (34±5% vs. 26±5%). Co-application with PACAP 6-27 and 6-38 (putative PACAP receptor antagonists) reduced the PACAP induced caliber changes efficiently (e.g. changes to 10-6 M PACAP-38 were 46±10% vs. 7±3%* without and with PACAP 6-27), however other short segments had no impact on this response. Graded hypercapnia resulted in large, concentration-dependent, reversible increases in pial arteriolar diameters. Pretreatment with PACAP (10-8 M, topically for 30 minutes) preserved the hypercapnia-induced vasodilation after I/R (62±5% and 56±3% to 10% CO2 before and after I/R).

In summary, PACAP is a potent vasodilator in the neonatal cerebral circulation, and this vascular reaction depends critically on COX-1 derived prostanoids. The role of NO seems to be ancillary. Preservation of arteriolar dilator responsiveness to hypercapnia by PACAP may contribute to neuroprotection.

Supported by the Hungarian Research grants: OTKA-T046531, ETT-2006 and NKTH.

FRACTIONATED MANGANESE-ENHANCED MAGNETIC RESONANCE IMAGING

Nicholas A. Bock, J.B. Mackel, A.C. Silva

Cerebral Microcirculation Unit, Laborartory of Functional and Molecular Imaging, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA

Introduction: In manganese-enhanced MRI (MEMRI), manganese is administered systemically to an animal to improve contrast in the neuroarchitecture and to probe neuronal function. To maximize contrast, a large dose is administered at the risk of eliciting acute toxic side-effects. In the present work, we investigate administering manganese in small, fractionated doses over time to 1 lessen the acute toxicity of the Mn2+ for susceptible organs like the liver and heart, and 2 provide multiple periods of Mn2+ exposure in an animal to perform functional stimulations or pharmacological challenges.

Methods: Twenty male Sprague-Dawley rats (222-286 g) were used under the guidelines of the NINDS/NEI, NIH ACUC. For single-dose studies, animals received a tail-vein injection of 180 mg/kg MnCl2?4H2O as a 100mM solution pH-buffered in bicine and infused at 1.25 ml/h. The other rats received fractionated intraperitoneal doses of 30 mg/kg MnCl2?4H2O as a 25 mM solution every 48 hours to allow the Mn2+ to clear the heart and liver before the next dose. Animals were imaged live at 48 hours post-injection on a 7T Bruker animal MRI with a T1-weighted 3D sequence (FSE, rare factor = 2, TE = 12.3 ms, TR=150 ms, isotropic resolution = 167 µm). B1 inhomogeneity in the surface coil images was corrected using a transmit coil reference image, and the signals from various regions in the brain were measured and normalized to the signal from muscle in the head for inter-animal comparisons.

Results: The signal in the majority of brain regions saturated around a total cumulative dose of 180 mg/kg in rats receiving fractionated doses up to 270 mg/kg. The figure shows representative images at various fractions (top) and for the single 180 mg/kg dose (bottom). Both the contrast and SNR are better in the single-dose image and fine details of the neuroarchitecture, such as cortical lamination, are better visible. Regions that significantly enhanced at the p=0.05 level in the single-dose rats included the thalamus, CA1 and CA3 of the hippocampus, the hypothalamus, and the striatum.

Conclusion: Fractionating the Mn2+ dose greatly lessened the toxic burden; all fractionated rats appeared healthy after 18 days of injections, while the single-dose rats were euthanized after imaging because of necrosis in their tails and lethargic behaviour. The low signal change and faster saturation in the cortex suggests it receives Mn2+ via anterograde transport from other regions closer to the ventricles. Finally, the lower final signal in the fractionated rat images suggests that the cumulative manganese uptake is affected by a lower driving gradient following the 30 mg/kg versus the 180 mg/kg dose. Overall, fractionated MEMRI produced good images with a healthier outcome for the animals.

REGIONALITY AND RATE OF SODIUM ACCUMULATION IN RAT FOCAL CEREBRAL ISCHEMIA BY SODIUM MRI

Victor E. Yushmanov¹, Alexander Kharlamov¹, Boris Yanovski¹, George LaVerde³, Fernando E. Boada³, Stephen C. Jones^1,2

¹Department of Anesthesiology, Allegheny Singer Research Institute, Pittsburgh, PA, USA, ²Department of Neurology, Allegheny Singer Research Institute, Pittsburgh, PA, USA, ³Department of Radiology, MR Research Center, University of Pittsburgh, Pittsburgh, PA, USA

SEGMENTATION OF MR PERFUSION IMAGES FOR PATIENTS WITH CAROTID STENOSIS USING INDEPENDENT COMPONENT ANALYSIS

Yi-Hsuan Kao¹, Michael Mu-Huo Teng^2,3, Yu-Ching Lin¹, Io-Pong Lam¹, Yi-Hui Hsiao¹

¹Department of Biomedical Imaging and Radiological Sciences, National Yang Ming University, Taipei, Taiwan-R.O.C., ²School of Medicine, National Yang Ming University, Taipei, Taiwan-R.O.C., ³Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan-R.O.C.

Background and aims: Hemodynamic parametric images calculated from magnetic resonance (MR) perfusion images are used to evaluate cerebrovascular diseases. However, it is difficult to assimilate all available information on the parametric images. We report the segmentation of MR perfusion images for carotid stenosis patients using independent component (IC) analysis (1),(2).

Methods: MR perfusion images acquired from carotid stenosis patients were retrospectively segmented using the FastICA algorithm (1). For quantitative analysis, regions of interest for different tissue types were selected from the segmentation results and various parametric images were calculated.

Results: Figures A - F illustrate the IC segmentation results for a patient with a 95% right internal carotid stenosis and bad collateral blood flow. Figure (A) displays the normalized signaltime curves for the five output IC images shown in (B)-(F). The arterial-phase IC1 image (B) demonstrates a lack of the right-side middle and anterior cerebral arteries. The delay-supplied arteries, immediate gray matter, and collateral normal gray matter appear on IC2 image (C). Severely delay-perfused brain parenchyma at the stenotic side appears on the venous-phase IC3 image (D). Choroid plexus is separately displayed on IC4 image (E). Cerebrospinal fluid and white matter are shown on IC5 image (F). The relative cerebral blood volume image (G) demonstrates a slight hyper intensity area at the lesion site and this area corresponds to the delay-perfused area seen on the time to peak (H), Tmax (I), and mean transit time (J) images. Quantitative measurements on time to peak of the concentration-time curves for arteries at normal and stenotic sides were found to be related to the severity of stenosis observed on angiograms.

Conclusion: The IC analysis technique can simultaneous provides spatial and temporal information on local blood perfusions for carotid stenosis patients. The information correlates well with parametric images as well as angiograms.

DIFFERENCES IN HIPPOCAMPAL MR FINDINGS WITH WHITE MATTER SUPPRESSION SEQUENCE IN MCI AND ALZHEIMER DISEASE USING 3 TESLA MRI

Junko Takahashi¹, Satoshi Takahashi¹, Hisashi Yonezawa¹, Masako Kudo¹, Satoko Obara¹, Toshihide Shibata¹, Yasuo Terayama¹, Makoto Sasaki², Takashi Inoue³

¹Department of Neurology, Iwate Medical University, Morioka, Iwate, Japan, ²Department of Radiology, Iwate Medical University, Morioka, Iwate, Japan, ³Department of Neurosurgery, Iwate Medical University, Morioka, Iwate, Japan

Background and aims: Patients with Alzheimer disease(AD)decline their cognitive ability in almost same manner as the disease progresses. The pathology of AD have been throughly investigated, while neuroimaging techniques have been limited its possibility mainly due to their resolution. Recently the advent of 3 Tesla (3T) MRI provides fine anatomical details with high spacial and contrast resolutions to visualize regional cell loss in minute structures. To visualize and compare regional gray matter cell density of hippocampus, entorhinal area and substantia innominata using 3T MRI in patients with AD or mild cognitive impairment (MCI), inversion recovery (IR) sequence was employed. Methods: Subjects were 43 patients with AD (16 males and 27 females), 10 patients with MCI (3 males and 7 females; 74.3 ± 6.7 years old), and 12 controls (6 males and 6 females, 72.5 ± 13.8 years old) without neurological deficits whose MRI were normal. Functional Assessment Staging (FAST) among AD patient was IV for 27 patients (9 males and 18 females; 73.4 ± 8.8 years old), V for 9 patients (3 males and 6 females; 74.1 ± 7.2 years old), VI for 7 patients (4 males and 3 females; 68.3 ± 10.1 years old). All subjects underwent Mini-Mental State Examination (MMSE), Wechsler Adult Intelligence Scale-Revised (WAIS-R), Wechsler Memory Scale-Revised (WMS-R), Modified Wisconsin Card Sorting Test and neurological examination. All scans were performed using a Signa VH/I 3.0 Tesla MR imaging system (GE Medical Systems, Milwaukee, WI) and a standard head coil. We employed IR sequence (TR 4000ms, TE 20ms, TI 250 ms, matrix 256×256, FOV 220×220 mm, 3 mm thickness and 1 mm gap) to suppress white matter signals, words for emphasizing gray matter signals. Using gray matter emphasized image, ROIs were positioned on CA1, subiculum, entorhinal area and substantia innominata and their intensity were measured. Prior to IR sequence, all cases underwent conventional spin echo T2 weighted sequence (TR 3800ms, TE 80 ms, matrix 512×256, FOV 220×220 mm, 5 mm thickness and 1.5 mm gap). Results: The intensity in subiculum and entorhinal area were significantly decreased compared with CA1 (p<0.05) from MCI stage of each AD patients. On the other hand, intensity of substantia innominata, which was unchanged during MCI stage, declined after FAST IV stage of AD (p<0.05). Conclusions: By using 3T MRI, we can visualize detailed morphological changes of hippocampus, entorhinal area and substantia innominata, and furthermore observe longitudinal changes of those structures from MCI to AD patients. This method is a powerful tool to elucidate AD etiology and to evaluate the efficacy of anti-AD drugs including acetylcholinesterase inhibitor.

EVIDENCE FOR A VASCULAR CORRELATE OF THE BEREITSCHAFTPOTENTIAL USING NEAR-INFRARED SPECTROSCOPY

Christoph Drenckhahn^1,2,3,4, Hellmuth Obrig^2,4, Jens Steinbrink^2,4, Hannes Duemmler¹, Matthias Kohl-Bareis^4,5, Arno Villringer^3,4, Dreier Jens^1,2,4

¹Department of Experimental Neurology, Charite, Berlin, Germany, ²Department of Neurology CCM, Charite, Berlin, Germany, ³Department of Neurology CBF, Charite, Berlin, Germany, ⁴Berlin Neuroimaging Center, Berlin, Germany, ⁵RheinAhr Campus, University of Applied Sciences Koblenz, Remagen, Germany

Aim: Using near-infrared spectroscopy (NIRS) in human healthy subjects we investigated alterations in deoxy-hemoglobin (deoxy-Hb) concentrations as an indirect parameter for a hemodynamic response in the primary motor cortex (M1) and the frontal medial wall areas (FMWA). The FMWA comprises higher-order motor areas including the supplementary motor area, which are involved in preparation and initiating of voluntary movement. M1 and FMWA were measured simultaneously in order to identify a correlate of the Bereitschaftspotential following voluntary movement.

Method: 16 strongly right-handed volunteers underwent different finger tapping tasks which are known to be associated with increase in cerebral blood flow in M1 and FMWA. The tasks were designed as a block with repetitive finger tapping for 20s or as single trials with finger tapping either self-paced or externally triggered. 8 light sources and 8 detectors were positioned over the left M1 and the FMWA covering an area of 12.5×7.5 cm. Time course and amplitude were analysed for the concentration of deoxy-Hb. For spatial analysis, the recording points with significant decrease in deoxy-Hb were determined.

Results: We detected a circumscribed decrease in deoxy-Hb at two different locations corresponding to FMWA and M1 during functional activation. The FMWA showed significantly earlier decrease of deoxy-Hb compared with the M1 (self paced: ?†t=0.78 s, p<0.01 / cued: ?†t=0.54 s, p<0.01) demonstrating a vascular correlate of the Bereitschaftspotential. Onset in SMA started significantly earlier following self initiated movements than externally triggered movements (?†t=0.57 s, p<0.01) whereas the M1 failed to reach significance (?†t=0.32 s, p=0.12). Significant changes in deoxy-Hb were located within an area of ~10 cm2 in projection on M1 and ~15 cm2 in projection on SMA, respectively.

Conclusion: We established a non-invasive bedside method to measure changes of deoxy-Hb reflecting alterations in cerebral blood flow in primary and secondary motor areas with a high temporal resolution during functional activation in healthy subjects. Simultaneous investigation of M1 and SMA indicated a change of deoxy-Hb related to the Bereitschaftspotential.

SIGNIFICANCE OF HYPOINTENSE LEPTOMENINGEAL VESSELS ON T2*-WEIGHTED GRADIENT ECHO IMAGING FOR PREDICTION OF ACUTE ISCHEMIC LESION GROWTH

Kyung-Hee Cho, Hye J. Kim, A-Hyun Cho, Sun U. Kwon, Jong S. Kim, Dong-Wha Kang

Department of Neurology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea

STRIATAL AND EXTRASTRIATAL DOPAMINE D2 RECEPTOR OCCUPANCY BY PALIPERIDONE EXTENDED-RELEASE TABLETS: A PET STUDY

Ryosuke Arakawa^1,2, Hiroshi Ito¹, Akihiro Takano¹, Hidehiko Takahashi¹, Takuya Morimoto¹, Takeshi Sassa³, Katsuya ota⁴, Motoichiro Kato⁵, Yoshiro Okubo², Tetsuya Suhara¹

¹Molecular Neuroimaging Group, Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan, ²Department of Neuropsychiatry, Nippon Medical School, Tokyo, Japan, ³Asai Hospital, Chiba, Japan, ⁴Onda-Daini Hospital, Chiba, Japan, ⁵Department of Neuropsychiatry, Keio University School of Medicine, Tokyo, Japan

Background and aims: Paliperidone is a novel antipsychotic drug in development for the treatment of schizophrenia. It shows almost the same pharmacological profile as risperidone of which it is the active metabolite with high affinity for dopamine D2 receptor and serotonin 5-HT2 receptor. Paliperidone ER is the extended-release (ER) formulation of paliperidone, which enables low peak to trough fluctuations and it was approved by FDA (USA) for the treatment of schizophrenia in mid December 2006. In this study, we evaluated the therapeutic dose range of paliperidone ER in the patients with schizophrenia using positron emission tomography (PET). The difference in dopamine D2 receptor occupancy by paliperidone ER between the striatal and extrastriatal regions was also investigated.

Methods: Thirteen male patients (age range 22-40 yr; mean ± S.D., 29.4 ± 5.4 yr) who were diagnosed with schizophrenia according to the DSM-IV criteria participated in this study. They received no more than one oral antipsychotic drug prior to the entry. This study was conducted as a part of an open-label phase IIa trial of paliperidone ER in Japan (Sponsored by Janssen Pharmaceutical K.K.). Before the study initiation, a written informed consent (IC) was obtained from all patients participated. The paliperidone ER doses were 3 mg/day in 6 patients, 9 mg/day in 4 patients and 15 mg/day in 3 patients given once daily. The clinical symptoms were assessed with the positive and negative symptom scale (PANSS) at baseline (pre-dose) and at endpoint. The extrapyramidal symptoms were assessed by clinical observations. Two PET scans per patient were performed at least 2 weeks after the initiation of treatment with paliperidone ER on a same day with [11C]raclopride and [11C]FLB 457 for measurement of striatal and extrastriatal dopamine D2 receptor occupancy, respectively. This study was approved by the Ethics and Radiation Safety Committee of National Institute of Radiological Sciences, Chiba, Japan.

Results: The dopamine D2 receptor occupancy in the striatum measured with [11C]raclopride was 54.2% to 85.5% and that in the temporal cortex measured with [11C]FLB 457 was 34.5% to 87.3%. ED50 in the striatum was 2.38 mg/day and 6.65 ng/ml and that in the temporal cortex was 2.84 mg/day and 7.73 ng/ml. There was no significant difference of dopamine D2 receptor occupancy between the striatum and the temporal cortex at any doses. Average PANSS scores of all patients were 62.9 ± 16.5 at baseline and 58.5 ± 16.8 at endpoint. Three patients, 2 patients in 15mg and 1 patient in 9mg, showed extrapyramidal symptoms during the course of the study.

Conclusions: The data from this study suggests that about 5-10 mg of paliperidone ER has a comprehensive benefit-risk ratio. No significant difference of dopamine D2 receptor occupancy was observed between striatal and extrastriatal regions, indicating no regional selectivity of paliperidone.

THE EFFECT OF CAROTID ANGIOPLASTY AND STENTING (CAS) ON CEREBRAL ACTIVATION DURING FINGER MOVEMENTS: A FUNCTIONAL MAGNETIC RESONANCE IMAGING STUDY

Tai-Cheng Wu¹, June Hung¹, Jiun-Jie Wang², Yeu-Jhy Chang¹, Chien-Hung Chang¹, Ting-Yu Chang¹, Hsiu-Chuan Wu¹, Tsong-Hai Lee¹

¹Department of Neurology, Chang Gung Memorial Hospital, LinKou Medical Center, Taoyuan, Taiwan-R.O.C., ²Department of Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan-R.O.C.

Background Previous studies in patients who underwent carotid angioplasty and stenting (CAS) due to severe carotid stenosis have observed a significant hemodynamic increase in peak systolic and mean flow velocity by transcranial Doppler imaging associated with changes in cerebrovascular reactivity at 2 days and at 2 to 4 months postoperatively. However, it remains unclear whether task-related brain activation could be altered soon after CAS. Here we used functional magnetic resonance imaging (fMRI) to examine the effect of CAS on neural correlates of a motor task.

Methods Three male patients (median age 68 years, range 56-84) who underwent unilateral CAS participated this study. Each patient attended two separate fMRI sessions, on the day prior to CAS and on the day no later than 5 days after CAS respectively. During each scanning session, blood oxygenation level dependent (BOLD) sensitive images were acquired whilst patients were performing a finger tapping task. Image data were analyzed with statistical parametric mapping 2 (SPM2).

Results Prior to CAS, unilateral finger tapping activated contralateral and/or ipsilateral motor association areas, suggesting incomplete motor lateralization for some individuals. After CAS, all patients showed variable degrees of increase in BOLD signals in the motor association areas on the same side of CAS, specifically in response to movements of contralateral fingers only (cases no. 1 and no. 3) or ipsilateral fingers only (case no.2).

Conclusion Our study suggests that unilateral CAS may rapidly alter the pattern of cerebral activations, in response to the movement of one specific hand only. These effects on task-related cerebral activation may reflect the improvement of vasomotor reactivity following CAS.

EVOLUTION OF CMRO2 DURING STROKE RECOVERY IN RAT MODELS

Young R. Kim¹, Emiri Tejima², Bruce R. Rosen¹

¹MGH, MIT, HMS Athinoula A. Martino's Center for Biomedical Imaging, Harvard School of Medicine, Charlestown, MA, USA, ²Neuroprotection Laboratory, MGH, Harvard School of Medicine, Charlestown, MA, USA

Stroke impairs neuro-vascular functions at the cellular level and causes sensorimotor impairments. Some degree of sensorimotor recovery subsequently occurs, yet the changes in basal level function that are directly linked to this recovery are not well understood. Moreover, understanding and monitoring stroke recovery in humans using non-invasive brain imaging of Blood Oxygenation Level Dependent (BOLD) functional magnetic resonance imaging (fMRI) is hampered by uncertainties surrounding the relationship between restoration and remodelling of all of the functional components of the neurovascular unit (NVU) during the recovery process. Our data, based on the fMRI using electrical forelimb stimulation in a rat stroke model suggest a differential evolution in hemodynamic signals and metabolic measures during the period of sensorimotor recovery. In particular, early after stroke (~1day), when sensorimotor impairment is most severe, the ipsilesional sensorimotor cortex exhibits proportionately greater loss in CMRO2, than CBF and CBV, resulting in a smaller loss in BOLD (Figure 1). Thereafter, as sensorimotor recovery occurs post-stroke, the ipsilesional sensorimotor cortex exhibits early and progressive increases in CMRO2, with delayed increases in CBV and CBF responses. As a result, it is suggested that the level of sensorimotor function throughout the recovery process is strongly correlated with CMRO2 in the ipsilesional sensorimotor cortex, and less correlated with BOLD (Figure 1).

Figure 1. Representative functional activation maps of rat brains 1 day and 14 days (left and right two columns, respectively) after the 2 hour-occlusion of MCAO. First row: T2-wighted images and T2 maps. Second-fifth row: BOLD, CBF, CBV, and CMRO2 activation maps in response to electrical stimulation of unaffected and affected sides. Note decrease in BOLD, but increase in CMRO2, between 1 and 14 day time points.

AMYLOID IMAGING IN ALZHEIMER'S DISEASE AND MCI BY [11C]-PIB PET

Hitoshi Shimada^1,2, Hitoshi Shinoto¹, Shigeki Hirano¹, Kouichi Sato¹, Noriko Tanaka¹, Tsuyoshi Ota¹, Kiyoshi Fukushi¹, Tetsuya Suhara¹, Takamichi Hattori², Toshiaki Irie¹

¹Molecular Probe Group, Molecular Imaging Center, National Institute of Radiological Sciences, Chiba-Shi, Chiba, Japan, ²Department of Neurology, Chiba University Faculty of Medicine, Chiba-Shi, Chiba, Japan

Background and aims: Deposits of beta-amyloid plaques in the brain are one of the most characteristic neuropathological changes of Alzheimer's disease (AD). Pittsburgh Compound-B (PIB) is an amyloid-binding radiotracer for Positron emissipn tomography (PET), and using it can estimate the beta-amyloid deposition in vivo. In this study, we investigated the pattern of beta-amyloid deposition in patients with AD and mild cognitive impairment (MCI) by PET. Methods: Participants were 12 patients with AD (mean age: 69+/−8 years; male/female: 3/9; MMSE: 18+/−4), 8 MCI (mean age: 75+/−4 years; male/female: 5/3; MMSE: 27+/−2) and age-matched 13 healthy controls (mean age: 64+−9 years; male/female: 4/9; MMSE: 28+−2). A dose (about 10mCi) of [11C]-PIB was i.v. injected and sequential PET scans were performed for 90min PET scans with arterial blood sampling. Data were analyzed using Logan plot graphical analysis with the cerebellum as a reference region. Beta-amyloid deposition were compared between AD group and healthy control group using t-test in statistical parametric mapping 5 (SPM). Same statistical analysis was performed between MCI group and healthy control group. RESULTS: The SPM showed that amyloid deposition was significantly higher in parietal association area, cingulate, frontal association area and lateral temporal cortex in the AD group (uncorrected p<0.0001, extent>50), though no significant deposition were shown in sensorimotor, medial temporal and occipital cortexes. Five of eight patients with MCI showed the AD-like pattern and the rest showed healthy control-like pattern. Conclusion: The characteristic brain amyloid deposition was shown in AD by PET. The pattern of the amyloid deposition in MCI group was devided into two group, AD-like group and healthy control-like group. It may suggest that the former (AD-like group) will convert to AD and the latter (healthy control-like group) will remain the amnestic MCI. Follow-up study is needed.

RECOVERY OF CEREBRAL OXYGEN METABOLISM AFTER COGNITIVE REHABILITATION IN PATIENTS WITH HIGHER CORTICAL DYSFUNCTION DUE TO TRAUMATIC BRAIN INJURY

Mitsuhito Mase¹, Kazuo Yamada¹, Hidehiro Kabasawa², Tetuso Ogawa², Junko Abe², Yuri Nagano², Akihiko Iida³

¹Department of Neurosurgery and Restorative Neuroscience, Nagoya City University Graduate School of Medical Science, Nagoya, Japan, ²Departments of Rehabiliation, Nagoya City Rehabilitation and Sports Center, Nagoya, Japan, ³Departments Radiology, Nagoya City Rehabilitation and Sports Center, Nagoya, Japan

Background and Purpose: Neuropsychological impairment is one of the most crucial sequelae after mild traumatic brain injury (TBI), disturbing resumption of one's works or previous daily activity. We have already shown that the severity of this higher cortical dysfunction correlated to a decrease of cerebral metabolic rate of oxygen (CMRO2) and the severity of diffuse axonal injury [1]. After cognitive rehabilitation, these symptoms could improve to a certain extent by one or two years. However, the mechanism and time-course of the recovery have not been clearly shown. In order to clarify this, we studied cerebral blood flow (CBF) and metabolism using positron emission computed tomography (PET) in patients with these sequelae before and after cognitive rehabilitation.

Methods: Four patients (male 3, female 1, mean 26.0 year old) with neuropsychological impairments such as disturbances of memory, attention, and / or information processing after TBI due to traffic accidents were included in this study. The mean interval between the injury and the estimation was 6–9 (mean: 7.25) months. CBF, CMRO2, oxygen extraction fraction (OEF), cerebral metabolic rate of glucose (CMRGlu), and distribution of benzodiazepine receptors of brain (BZR) were measured before and after the 6-month cognitive rehabilitation. Higher cortical dysfunction was evaluated using Wechsler Adult Intelligence Scale-Revised (WAIS-R), Miyake's memory test (Miyake MT), Rey-Osterrieth complex figure test (Rey's Figure), Rivermead Behavioral Memory Test (RBMT), and paced auditory serial addition task (PASAT).

Results: After cognitive rehabilitation, scores of WAIS-R and Miyake's MT significantly improved in all cases. Rey's Figure, RBMT, and PASAT also improved, but the extent of recovery was varied in each patient. In all patients, CBF and OEF were within normal ranges, however, CMRO2 slightly decreased (luxury perfusion state) in almost all area before rehabilitation. The area where CMRO2 decreased was much wider than the area where cerebral contusion existed on MRI. CMRO2 improved in the whole brain in all patients after rehabilitation. CMRGlu also improved slightly after recovery of symptoms in 3 of the 4 patients, but one patient with decreased CMRO2 showed normal CMRGlu. BZR activity decreased in two cases before, which did not change after the rehabilitation.

Conclusions: Relative luxury perfusion state (decrease of CMRO2 with normal CBF and OEF) was observed in patients with cognitive impairments in almost 7 months after mild TBI. These symptoms improved after cognitive rehabilitation, and CMRO2 improved simultaneously. Changes of CMRGlu showed the similar tendency to changes of CMRO2. However, the sensitivity of the change was not good. BZR activity is not suitable to estimate cognitive impairment. We conclude that CMRO2 using PET can become a useful and objective marker for estimate severity and recovery of cognitive impairment due to diffuse axonal injury.

REACTIVITY OF CEREBRAL BLOOD VESSEL IN CADASIL

Yoshinari Nagakane¹, Mizuno Toshiki¹, Hideaki Tomimoto², Makaki Kondo¹, Noriko Ishigami¹, Aiko Tamura¹, Takahiko Tokuda¹, Yo Ushijima³, Tuneo Nishimura³, Masanori Nakagawa¹

¹Department of Molecular Neurology, Kyoto Prefectural University of Medicine, Kyoto, Japan, ²Department of Neurology, Kyoto University, Kyoto, Japan, ³Department of Radiology, Kyoto Prefectural University of Medicine, Kyoto, Japan

IMPROVING THE VISUALIZATION OF SPATIOTEMPORAL CHANGES IN CEREBRAL BLOOD FLOW THROUGH INTACT RAT/MOUSE SKULL BY LASER SPECKLE TEMPORAL CONTRAST ANALYSIS

Pengcheng Li, Jianjun Qiu, Shaoqun Zeng, Qingming Luo

Wuhan National Laboratory for Optoelectronics, Britton Chance Center for Biomedical Photonics, Huazhong University of Science and Technology, Wuhan, China

Laser speckle flowmetry based on the laser speckle contrast analysis has been used to assess two-dimensional (2-D) cerebral blood flow (CBF) with high spatial and temporal resolution. However, it can only be applied to the subjects whose skull were removed or thinned to transparency, or for the subjects whose skull were thin and transparent, such as mouse, but fails to reveal the CBF through the thick scattering skull. Here we describe a method that is able to reduce the influence of the static superficial scattering tissue on imaging the CBF and thus improve the visualization of CBF through the intact rat skull. It is based on the temporal statistics analysis of sequences of laser speckle images. Both physical model experiments and animal experiments were performed to demonstrate the feasibility of this method, and also its limitation. In the animal experiments, the spatiotemporal patterns of the propagation of changes in CBF during cortical spreading depression induced by KCl were investigated with the adult male rat and mouse, which were anesthetized and fixed in a stereotactic frame. The skin over the skull was removed and kept the skull intact. The experimental results show that comparing with the conventional laser speckle contrast analysis technique, our method significantly improve the resistance to the static superficial structure when the image area contains static scattering. Even for the case imaging CBF through mouse skull which is relative transparent, the results obtained by laser speckle temporal contrast analysis also show better image quality than conventional laser speckle contrast analysis.

Keywords: Cerebral blood flow, Laser speckle imaging, Intact skull

LARMINAR-DEPENDENT CBF RESPONSES INDUCED BY VISUAL AND HYPERCAPNIC STIMULATION

Tae Kim, Ping Wang, Seong-Gi Kim

Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA

QUANTITATIVE CEREBRAL BLOOD FLOW AND PERFUSION TERRITORIES MEASUREMENT IN NORMOTENSIVE AND HYPERTENSIVE RATS USING ARTERIAL SPIN LABELING

Fernando F. Paiva^1,2, Erica C. Henning¹, Alberto Tannus², Afonso C. Silva¹

¹National Institutes of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA, ²Instituto De Fisica De Sao Carlos, Universidade De Sao Paulo, Sao Carlos, SP, Brazil

DIFFERENCIAL BRAIN ACTIVATION DUE TO EXPERIENCE OF VISCERAL STIMULI IN HUMANS: A PET STUDY

Toyohiro Hamaguchi^1,2, Michiko Kano², Motoyori Kanazawa², Hisashi Rikimaru³, Satoshi Watanabe², Masatoshi Ito³, Kazuhiko Yanai⁴, Shin Fukudo²

¹Department of Occupational Therapy, Niigata University of Health and Welfare, Niigita, Japan, ²Department of Behavioral Medicine, Tohoku University Graduate School of Medicine, Sendai, Japan, ³Division of Nuclear Medicine, Tohoku University, Cyclotron Radio Isotope Center, Sendai, Japan, ⁴Department of Pharmacology, Tohoku University Graduate School of Medicine, Sendai, Japan

Irritable bowel syndrome is considered to be a disorder of brain-gut interactions, particularly of visceral hypersensitivity. However, the brain area related to initial sensitization provoked by the programming of visceral stimulation is still unknown. The hypothesis of this study is that brain activation due to the colonic distention is different between with initial intense stimulation and with intense stimulation after the mild stimulation, and its also hypothesized that brain activation due to sham colonic distention is different between before intense stimulation and after intense stimulation.

Forty-five normal volunteers participated in the study. Using a colonoscope, a bag was inserted into the descending colon and the colon was distended with barostat. Orders of stimuli was set with six different patterns. Regional cerebral blood flow was measured by PET. Brain image was analyzed with SPM 2.

The main effect of the sham stimulation (0 mm Hg) after 40 mm Hg distention denoted activation of the right cingulate gyrus (Brodmann area; BA 24, Z = 4.16 uncorrected p < 0.001), right insula (BA 13, Z = 4.08), and right medial frontal gyrus (BA46, Z = 3.92).

These findings suggest that the brain may behave differently in response to the different order of visceral stimulation. The determinants of the modulation of visceral perception and emotion may include the combined effects of experience of the stimuli.

123I-IOMAZENIL SPECT PERFORMED DURING THE ACUTE STAGE FOR PATIENTS WITH TRAUMATIC BRAIN INJURY: A NEW APPROACH TO EVALUATING CORTICAL DAMAGE

Hiroyasu Koizumi, Hirosuke Fujisawa, Tetsu Kurokawa, Shoichi Kato, Koji Kajiwara, Tatsuo Akimura, Sadahiro Nomura, Masami Fujii, Michiyasu Suzuki

Department of Neurosurgery, Yamaguchi University, School of Medicine, Ube, Yamaguchi, Japan

<Background> 123I-iomazenil were developed as radioligands of central benzodiazepine receptors distributed mainly in the cerebral cortex neurons. Because of the specific distribution of benzodiazepine receptors, these radioligands have been successfully applied to detect neuronal integrity in acute brain strokes.

Also, in the case of traumatic brain injury, neurons were irreversibly damaged in the cerebral cortex following cortical contusion injury. In addition to the morphological changes detected by CT/MRI scans, neuronal damage to the cortex post-contusion was also functionally visualized on 123I-iomazenil (IMZ) SPECT as a region of defects for RI accumulation.

<Methods> In this presentation, IMZ SPECT was performed for the patients with traumatic brain injury during the acute stage to functionally evaluate the cortical damage in comparison with morphological images detected by CT/MRI. In twelve patients, IMZ bindings were studied by SPECT during the acute stage following traumatic brain injury. Written informed consent was received from all patients beforehand. Three of the twelve patients were diagnosed with diffuse axonal injuries, while eight suffered focal brain injuries, such as cerebral contusions and acute subdural hematomas. One patient was suffered from hypoxic ischemic encephalopathy following an acute subdural hematoma.

<Results> In the case of hypoxic ischemic encephalopathy, IMZ SPECT revealed diffuse cortical disorders expressed in the form of a broad defect with IMZ accumulation. The cerebral cortex of intact with IMZ accumulation emerged in three patients; diagnosed as diffuse axonal injury. In eight patients diagnosed with cerebral contusion meanwhile, areas with markedly decreased IMZ binding, corresponding to regions on the CT/MRI scans, were detected on SPECT.

Surprisingly, in four of the eight patients diagnosed with cerebral contusion, areas with decreased IMZ bindings, expected to represent irreversible changes, had markedly reduced in size one month after the induction of the traumatic brain injury. However, in the remaining four patients, the size of areas with decreased IMZ binding remained unchanged at the chronic stage.

The reduction in the areas of defect for RI accumulation suggests that benzodiazepine receptors were additionally detected with IMZ binding at the chronic stage in the cortical lesion, where receptors were not detected at the acute stage. The mechanisms of this phenomenon are unclear and moreover, further investigations to add cases should be necessary.

MELODY PERCEPTION IN NONMUSICIANS - A FUNCTIONAL MRI STUDY

Miho Yamauchi, Takuya Hayashi, Akihide Yamamoto, Hiroshi Sato, Hidehiro Iida

Advanced Medical Engineering Center, Research Institute, National Cardiovascular Center, Suita, Osaka, Japan

DIRECT COMPARISON OF 201TL AND 99TC-MIBI SPECT OF GLIOMA BY A RECEIVER OPERATING CHARACTERISTIC ANALYSIS

Yasushi Shibata, Wataru Katayama, Tetsuya Yamamoto, Shingo Takano, Akira Matsumura

Department of Neurosurgery, Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Ibaraki, Japan

CALCIFICATION OF INTRACRANIAL ARTERIES-DETECTION AND EVALUATION USING MULTI-DETECTOR ROW CT (MDCT)

Masato Kin¹, Hideki Ohba¹, Masanori Mizuno¹, Yasuhiro Ishibashi¹, Makoto Sasaki², Yasuo Terayama¹

¹Department of Neurology, Iwate Medical University, Morioka, Iwate, Japan, ²Department of Radiology, Iwate Medical University, Morioka, Iwate, Japan

Background and purpose— The importance of calcification associated with atherosclerotic plaque in intracranial arteries has been debated. The purpose of the present study is to clarify the relation between calcification of intracranial arteries detection and evaluation using MDCT and risk factors for cerebral infarction.

Method— One hundred and thirteen (113) subjects (M:F=71:42, 67.0+10.7 years) including 88 with cerebral infarction and 25 without cerebral infarction were included. Detection and evaluation of calcification was performed using 16 lines of MDCT made by Toshiba Corporation. Analysis of the arterial image was performed using DICOM viewer software “ExaVision LITE” (ZIOSOFT, Inc.). Calcification of 10 different region of intracranial arteries were evaluated by calcification score (CS). Statistical analysis was performed using Dra??s SPSSa!.

Result— CS was significantly correlate with aging (p<0.001) and hypertension (p<0.05). CS of anterior circulation of intracranial arteries was significantly correlate with aging (p<0.001). CS of posterior circulation of intracranial arteries was significantly correlate with aging (p<0.01) and hypertension (p<0.05).

Conclusions— Aging and hypertension contribute to the calcification of intracranial arteries including both anterior and posterior circulation. Hypertension does not contribute to the calcification of anterior circulation. On the other hand, it contributes to the calcification of posterior circulation in the same way as coronary arterial system. The above data suggests that a factor contributing to calcification varies in intracranial arterial system.

TIMING OF CELL TRANSPLANTATION AFTER STROKE SIGNIFICANTLY INFLUENCES THE CELL-HOST INTERACTION

Hideo Shichinohe, Marcel Daadi, Anne-Lise Maag, Nobutaka Horie, Theo Palmer, Tonya Bliss, Gary K. Steinberg

Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA

Background and aims: Growing evidence suggests that cell transplantation holds great potential as a therapy for stroke. Several groups have shown that transplanted cells can survive in the stroke brain, migrate towards the injured tissue and improve neurological function. However, many questions must be answered before moving this therapy to the clinic. One fundamental variable that needs to be defined is the optimal time after stroke for transplantation. This study looks at two time points of cell delivery and describes the difference that the time of transplantation makes to the host response and graft biology.

Methods: Primary medial ganglionic eminence (MGE) neural precursors were isolated from E15 rat embryos carrying the transgene for enhanced green fluorescent protein. Stroke was induced in rats by distal middle cerebral artery occlusion (dMCAo). A suspension of MGE cells was transplanted into the rat cortex at day 2 (n=2) or day 14 (n=3) after stroke. Animals were sacrificed at day 42 post-dMCAo and brain sections stained for neural, monocytic and blood vessel markers.

Results: Cell survival differed drastically between the two transplant groups. Cells transplanted at day 2 post-stroke showed very little survival at day 42; the cells in the core appeared to be dead but a thin layer of cells survived around the periphery of the graft. In contrast, when examined at day 42, the cells transplanted at d14 exhibited much more robust survival throughout the graft and formed a much more elongated graft. There was a significant difference in transplanted cell survival between the two transplant groups ('day-2′ group; 0.7 ± 0.2 ×10^5 cells/mm3, ‘day-14′ group; 2.0 ± 0.3 ×10^5 cells/mm3, P=0.02). No difference in blood vessel density in and around the grafts was found between the groups. Therefore, differential vasculature did not account for the difference in cell survival. There was, however, a significant difference in the inflammatory response between the two transplant groups. The day-2 group showed a robust inflammatory response at 6 weeks after transplantation, with Iba1-positive (monocytic) cells both within and surrounding the graft. In the day-14 group at six weeks post transplantation, there were very few Iba1-positve cells in the graft and a dramatic lack of Iba1 cells immediately surrounding the graft, much fewer than in the rest of the hemisphere. However, in this Iba1-sparse area there were many host astrocytes which appeared to form a barrier around the graft precluding the monocytes.

Conclusions: Fetal neural precursors derived from the MGE survive poorly when transplanted two days post-dMCAo, possibly due to a prolonged host immune response. However, cells transplanted at 14 days post-dMCAo survived to a greater degree; they appeared to be protected from the monocytes by a wall of host astrocytes. Thus, the timing of transplantation after stroke has a long-term effect on the host response to the graft. This response can dramatically affect the graft's microenvironment and ultimately determine the success of cell-based therapy.

IDENTIFICATION OF NEURAL OUTGROWTH CELLS FROM PERIPHERAL BLOOD OF STROKE PATIENTS AND THEIR CONTRIBUTIONS TO ISCHEMIC BRAIN REGENERATION

Kon Chu, Keun-Hwa Jung, Soon-Tae Lee, Eun-Cheol Song, Dong-In Sinn, Jeong-Min Kim, Hee-Kwon Park, Manho Kim, Jae-Kyu Roh

Stroke and Neural Stem Cell Laboratory, Department of Neurology, Stem Cell Research Center, Seoul National University Hospital, Seoul, South Korea

Background: Bone marrow harbors tissue-committed stem cells that may migrate to the circulation and exert beneficial effects on regeneration. Recent studies have identified a subset of outgrowth cell population with endothelial phenotype in longterm cultures of peripheral blood mononuclear cells. The concept that peripheral blood-derived cells participate in neural regeneration remains highly controversial and no specific cell type has been identified. In this study, we undertook to characterize outgrowth cells (OCs) in peripheral blood culture from stroke patients and from healthy subjects, and to demonstrate their therapeutic efficacy against experimental ischemia.

Methods: Patients undergoing acute stroke was included. EPCs were isolated from the peripheral venous blood of 25 patients and 15 age-matched healthy volunteers. The yield of outgrowth endothelial cells was measured during the 3 months of culture. We transplanted the OCs in the ischemic cortex of adult rats and analyzed the functional recovery and engraftment pattern.

Results: OCs were isolated efficiently from stroke patients, and demonstrated heterogenous populations with different morphologies, e.g., cobblestone, palisading, or branching features. Two different types of OCs were identified: endothelial OCs; neuronal OCs, according to their morphology and differentiation patterns toward endothelial or neural lineages. The transplanted neuronal OCs survived in the ischemic brains of a rat model and improved functional recovery over 5 weeks post-transplantation. The majority of migrating, transplanted cells in ischemic rats had neuronal or endothelial phenotypes.

Conclusion: The feasibility of extracting and culturing neural progenitor cells in large numbers suggests that such autologous progenitor cells will be useful for applications ranging from basic research to cell-based therapy.

VEGFR2 ANTAGONISM ATTENUATES BEHAVIORAL IMPROVEMENTS ON A RADIAL ARM MAZE FOLLOWING TRAUMATIC BRAIN INJURY (TBI)

Randy Morgan, Christian Kreipke, Jose Rafols

Department of Anatomy and Cell Biology, Wayne State University, School of Medicine, Detroit, MI, USA

ISOFLURANE PRECONDITIONING ENHANCES POST-ISCHEMIC NEUROGENESIS IN ADULT MICE

Wenri Zhang, Lan Wang, Stephanie J. Murphy, Patricia D. Hurn, Ansgar M. Brambrink

Department of Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA

Background and Aims: Exposure to isoflurane prior to brain ischemia (“anesthetic preconditioning”) improves functional outcome and results in higher numbers of viable neurons after the insult, which has been attributed to neuroprotective effects of the volatile anesthetic. We tested the hypothesis that isoflurane preconditioning (isoPC) induces an up-regulation of post-stroke neurogenesis in adult male mice.

Methods: Adult C57BL/6 mice were preconditioned for 4 hours with air (sham preconditioning, shamPC) or 1.0% isoflurane (isoPC). Twenty-four hours after preconditioning, mice underwent 60 min of reversible middle cerebral artery occlusion (suture model). Animals were survived for 3 or 21 days, and received twice daily intraperitoneal injections of 5-bromo-2′-deoxyuridine-5′-monophosphate (BrdU, 50 mg/kg). At designated times, brains were processed (intravital perfusion-fixation, 4% PFA, free-floating 40 µm-cryosections) and stained for BrdU (proliferated cells), doublecortin (neuroblasts), and NeuN (mature neurons). Confocal microscopy in coronal sections was used to quantify double-labeled cells in dentate gyrus (DG), subventricular zone (SVZ), striatum, and peri-ischemic neocortex (overlay images; cells per tissue volume). Representative sections are stained for morphologic evaluation (MAP-2, Fluoro-Jade).

Results: We observed an increase in postischemic neurogenesis after isoPC (fig. 1). Maturation and survival of newborn cells were also enhanced by isoPC in ipsilateral dentate gyrus at 21 days after stroke (9959±423 vs. 2657±0 BrdU+ neuroblasts; 4310±1358 vs. 2030±0 BrdU+ mature neurons, n=1-4/group). Moreover, enhanced migration of newborn cells into damaged areas after ischemia was found in isoPC vs. shamPC animals.

Conclusions: The extent of functional impairment resulting from cerebral ischemia is generally considered to reflect the volume of lost brain tissue. However, newborn neurons may contribute profoundly to the amazing capacity for rehabilitation after cerebral ischemia, which apparently is enhanced after anesthetic preconditioning.

ENFORCED PHYSICAL TRAINING PROMOTES NEUROGENESIS IN THE SUBGRANULAR ZONE AFTER FOCAL CEREBRAL ISCHEMIA

Seung-Hoon Lee, Yun-Hee Kim, Young-Ju Kim, Miyeoun Song, Byung-Woo Yoon

Department of Neurology, Seoul National University Hospital, Seoul, South Korea

Background and Purpose: Cerebral ischemia increases neurogenesis in the subventricular zone (SVZ) and the subgranular zone (SGZ) of the dentate gyrus, which might be altered by enriched environment including voluntary physical activity. We tested whether enforced physical training (EPT) influences differentially on the neurogenesis of the SVZ and the SGZ after cerebral ischemia. Methods: Adult male Sprague-Dawley rats were subjected to focal cerebral ischemia for 2 hours, and were divided into EPT and non-EPT group. All rats in the EPT group were trained using rota-rod for 14 days. 5-bromo-2′-deoxyuridine (BrdU) was injected to determine levels of cell proliferation. Functional recovery was assessed by a set of behavioral test batteries. Extent of endogenous neurogenesis in the SVZ and the SGZ was quantified using immunofluorescence staining. Results: Although final infarction volume was not significantly different between the groups, functional recovery was better in the EPT group at 10 and 17 days after ischemia. In the SVZ, BrdU-labeling and double-labeling of BrdU/Dcx or BrdU/NeuN were not significantly between the groups. In the SGZ, EPT enhanced significantly BrdU-positive cells (EPT vs. non-EPT: 159.1±19.9 vs. 101.8±7.8, p=0.04), and the number of BrdU/Dcx doublelabeled cells was also increased in the EPT group (130.6±16.9 vs. 73.6±7.2, p=0.01). Conclusion: Our results indicated that EPT promotes neurogenesis in the SGZ of the dentate gyrus, but not in the SVZ after ischemia. Biochemical mechanism of the differential effects of EPT remains to be clarified.

IGF-1 AND FGF2 SYNERGISTICALLY PRMOTE ADULT NEURAL PROGENITOR PROLIFERATION UNDER NORMAL AND ISCHEMIC CONDITIONS

Robert Dempsey¹, Haviryaji Kalluri¹, Yi-ping Yan¹, Kurt Sailor², Raghu Vemuganti¹

¹Department of Neurological Surgery, University of Wisconsin, Madison, Madison, WI, USA, ²Department of Neurosurgery, Johns Hopkins Medical School, Baltimore, MD, USA

Transient focal ischemia (stroke) is a potent stimulator of neurogenesis in adult rodent brain. We previously observed that following transient middle cerebral artery occlusion (MCAO) in adult rats, the expression of IGF-1 is significantly elevated in the ischemic hemisphere. The neural progenitors in the subventricular zone (SVZ) and dentate gyrus (DG) expressed IGF-1 receptor. Inhibiting the IGF-1 activity by infusion of IGF-1 antibody into the lateral ventricle significantly prevented the ischemia-induced increase of BrdU positive cells in the both SVZ and DG. On the contrary, infusion of IGF-1 resulted in a significant elevation of BrdU positive cells in the both SVZ and DG of the post-ischemic brain. These results indicate that IGF-1 upregulation in the ischemic brain may increase neural progenitor proliferation or enhance their survival. Therefore, we used an in vitro culture system to dissect the mechanisms of IGF-1 effects on adult neural progenitor cells. IGF-I activated PI-3kinase/Akt pathway but not ERK pathway in the adult neural progenitor cells in vitro. Furthermore, LY294002 (PI-3 kinase inhibitor), but not U0126 (ERK inhibitor) inhibited the IGF-I induced survival of cells, while FGF2 (ERK activator) enhanced the IGF-I mediated survival of the cells as measured by cell proliferation assay. In addition, stimulation of ERK pathway by FGF2 induced the expression of cyclin D1, which is essential for the entry of cells into cell cycle, while IGF-I in the presence of FGF2 upregulated the expression of cyclin D1. Likewise, IGF-I in the absence or presence of FGF2 increased the phosphorylation of GSK supporting its role in the survival of NPCs. To further confirm the role of ERK activation in the proliferation, we cultured cells in FGF2 + IGF-I containing medium in the presence and absence of U0126 (ERK inhibitor) and show the inhibition of nestin expression in U0126 treated cells. The decrease in the cyclin D1 content in conjunction with the inhibition of nestin expression by ERK inhibitor, confirms the role of ERK in the proliferation of cells. Thus, these studies show that following focal ischemia, IGF-I might enhance the survival of the neural progenitor cells so that other mitogens like FGF2 might enhance their proliferation.

These studies were funded by NIH and AHA

INTRA-ARTERIAL TRANSPLANTATION OF BONE MARROW MONONUCLEAR CELLS IMMEDIATELY AFTER REPERFUSION AMELIORATES BRAIN INJURY FOLLOWING TRANSIENT FOCAL ISCHEMIA IN RATS

Nobuo Kamiya¹, Hironaka Igarashi², Yasuhiro Nishiyama¹, Masayuki Ueda¹, Yasuo Katayama¹

¹Division of Neurology, Nephrology and Rheumatology, Department of Internal Medicine, Nippon Medical School, Tokyo, Japan, ²Center for Integrated Human Brain Science, Brain Research Institute, University of Niigata, Niigata, Japan

Introduction: Transplantation of bone marrow cells on experimental cerebral ischemia has been reported to possess neuroprotective and regenerative effects (1),(2), however neuroprotective effect of the cells administered immediately after reperfusion has been rarely investigated. The present study was designed to examine whether ischemic cerebral injury can be ameliorated by bone marrow mononuclear cells (MMCs) administered immediately after reperfusion on transient focal ischemia in rat, and if so, to determine whether difference of the effectiveness is seen dependent on route of administration (intra-arterial and intravenous MMC administrations).

Methods: Male Sprague?Dawley rats, weighing 250?300g, were used in the study. Bone marrow was obtained autologously from the right femur of each animal, and MMCs were isolated prior to ischemia. Rats were subjected to transient focal ischemia for 90 min using the intraluminal suture technique, followed by MMC administration (1.0>10E7 cells) through the ipsilateral carotid artery (IA group) or the femoral vein (IV group) immediately after reperfusion. Several animals were received same procedure except for intravenous vehicle injection instead of MMC transplantation (control group). Animals were deeply anesthetized and decapitated 24 hrs after reperfusion. The brains were carefully removed and dissected into 2 mm-thick coronal sections, and the slices were stained with 2,3,5-triphenyltetrazolium chloride (TTC). The infarct volume was calculated by summing the infarct area of the TTC-stained sections from each animal.

Results: Averaged 1.13±0.25>10E7 MMCs were harvested from each rat. Total infarct volume was reduced in IA group (78.3±13.3 mm3), but not in IV group (119.0±35.7 mm3), compared with that in control group (140.6±47.8 mm3), with a statistical significance (p<0.05). The cortical infarct volume was also significantly decreased in IA group compared to control group, although striatal infarct volume was not statistically different among the groups.

Conclusion: The present data confirm significant neuroprotection following transient focal ischemia by autologous MMCs administered intra-arterially, but not intravenously, immediately after reperfusion in rats.

IN VIVO IMAGING OF NEUROKININ 1 RECEPTOR DISTRIBUTION WITH 18F-L829165 POSITRON EMISSION TOMOGRAPHY IN PARKINSON'S DISEASE AND LEVODOPA-INDUCED DYSKINESIAS

Yasuyuki Kimura^1,2, Alan Whone¹, Nicola Pavese¹, Richard Hargreaves³, Don Burn³, David J. Brooks¹

¹MRC Clinical Sciences Centre and Division of Neuroscience, Faculty of Medicine, Imperial College, Hammersmith Hospital, London, UK, ²Department of Nuclear Medicine and Tracer Kinetics, Osaka University Graduate School of Medicine, Suita, Osaka, Japan, ³Merck & Co., Inc.

Background: The majority of Parkinson's disease (PD) patients receving long-term replacement therapy with levodopa develop adverse events of motor fluctuations and dyskinesias. Alterations of substance P (SP) levels and neurokinin 1 (NK1) receptor availability have been reported to be associated with levodopa-induced dyskinesias (LID) in animal lesion and postmortem studies.

Methods: To determine whether changes in NK1 receptor binding are involved in the pathophysiology of human PD and LID, we measured in vivo NK1 receptor availability in normal subjects and dyskinetic and nondyskinetic PD patients using positron emission tomography (PET) and the NK1 receptor ligand 18F-L829165. Patlak linear graphical analysis was used to create parametric images of 18F-L829165 influx constants (Ki). Regional 18F-L829165 Ki's were calculated using region of interest (ROI) analysis. We also used statistical parametric mapping (SPM) to localise changes in 18F-L829165 Ki at a voxel level.

Results: In normal subjects, the highest 18F-L829165 Ki's were observed in the caudate nucleus and putamen. The globus pallidum and neocortex showed lower 18F-L829165 Ki's. Using ROI analysis, we found significantly lower 18F-L829165 Ki's in the caudate nucleus, ventral striatum, globus pallidus lateral segment, locus coeruleus, and parietal, temporal, and occipital lobes of PD patients compared with normal subjects. SPM confirmed the presence of decreased bilateral striatum, neocortex, and dorsal midbrain 18F-L829165 Ki in the PD patients. Although the ROI analysis did not reveal any significant difference in 18F-L829165 Ki between the non-dyskinetic and dyskinetic PD patients, SPM revealed relatively decreased right putamen and dorsal midbrain 18F-L829165 Ki's in the dyskinetic PD patients.

Conclusions: 18F-L829165 PET reveals in vivo the distribution of the cerebral NK1 receptor availability in normal subjects. The NK1 receptor binding is reduced in the striatum and cortex of PD patients and the presence of dyskinesias is associated with further striatal and midbrain reductions. The SP and NK1 receptor systems could, therefore, play a role in the pathophysiology of PD and levodopa-induced dyskinesias.

AGE-DEPENDENT CEREBROVASCULAR DYSFUNCTION IN TG2576 TRANSGENIC MOUSE MODEL OF HUMAN CEREBRAL AMYLOID ANGIOPATHY

Hwa Kyoung Shin¹, Monica Garcia-Alloza², Phillip B. Jones², Laura Borrelli², Brian J. Bacskai², Matthew P. Frosch², Bradley T. Hyman², Michael A. Moskowitz¹, Cenk Ayata^1,3

¹Stroke and Neurovascular Regulation Laboratory, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ²Alzheimers Disease Research Laboratory, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, ³Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA

Background and aims: The Tg2576 transgenic mouse model of human cerebral amyloid angiopathy (CAA) is characterized by age-dependent cerebrovascular deposition of amyloid-beta starting from 9 months of age and progressively worsening to involve most pial arterioles by 18 months. Here, we tested whether vascular amyloid-beta deposition impairs physiological CBF responses to 5% inhaled CO2, whisker stimulation or cortical spreading depression (CSD) in aged Tg2576 mice using laser speckle flowmetry non-invasively through intact skull.

Methods: We tested 8 month-old Tg2576 mice with elevated soluble amyloid-beta but no vascular amyloid-beta deposition, and 19 month-old Tg2576 mice with widespread and severe vascular amyloid-beta deposits, and compared these to wild type littermates.

Results: Hypercapnic hyperemia (5% CO2) was attenuated by 39% in 19, but not 8 month-old Tg2576 mice (p<0.05). The slope of the CBF increase was significantly attenuated in 19 month-old Tg2576 mice only (18±3 vs. 9±1%/min in wild type and Tg2576, respectively; p<0.05). The response to hypercapnia was also attenuated in aged wild type mice (p<0.05) but this did not explain the difference between the mutant strains. The CBF response to vibrissal stimulation was attenuated by 48% in 19, but not 8 month-old Tg2576 compared to wild type littermates (p<0.05). Aging alone also tended to attenuate the functional hyperemic response to vibrissal stimulation in 19 month-old wild type mice but the data were not significant (p=0.1). Both the transient hypoperfusion during CSD and the subsequent post-CSD oligemia were diminished in 19 but not 8 month-old Tg2576 compared to wild type littermates, suggesting that vasoconstrictor responses were also impaired in 19 month-old Tg2576 mice (p<0.05). Aging alone did not attenuate the vasoconstrictive CBF response to CSD in wild type mice. Anesthetic-induced changes in blood flow (i.e., Isoflurane-induced increase and alpha-chloralose-induced decrease in resting CBF) were both attenuated in 19 but not 8 month-old Tg2576 compared to wild type littermates, further suggesting that CAA impacts both vasodilator and vasoconstrictor responses.

Conclusion: Our data suggest that vascular amyloid-beta deposition rather than elevated soluble amyloid-beta is associated with cerebrovascular dysfunction when imaged non-invasively in these physiological models. Furthermore, advancing age is independently associated with impaired physiological CBF responses to hypercapnia and functional activation.

This work was supported by the American Heart Association (0335519N, Ayata), NIH (P50 NS10828, Moskowitz; AG08487, Hyman)

EFFECT OF AGE ON (R)-[11C]PK11195 BINDING IN HEALTHY SUBJECTS

Alie Schuitemaker^1,2, Bart N.M. van Berckel², Ronald Boellaard², Marc A. Kropholler², Philip Scheltens¹, Adriaan A. Lammertsma²

¹Department of Neurology, VU Medical Center, Amsterdam, The Netherlands, ²Department of Nuclear Medicine and PET Research, VU Medical Center, Amsterdam, The Netherlands

DENDRITIC ARBORIZATION AND ANCHORINGS OF NMDA AND AMPA RECEPTORS IN PRIMARY HIPPOCAMPAL NEURONS ARE CONTROLLED BY SIGMA-1 RECEPTORS AT ER

Shang-Yi Tsai, Teruo Hayashi, Tsung-Ping Su

Cellular Pathobiology Unit, CNRB, IRP, National Institute On Drug Abuse, NIH, DHHS, Baltimore, MD, USA

The endoplasmic reticular (ER) protein sigma-1 receptor (Sig-1R) has been implicated in many CNS diseases including depression, amnesia, pain, as well as in neuroplasticity associated with cocaine or methamphetamine abuse. Functionally, Sig1-Rs have been shown by us to affect differentiation in neuronal cell lines. This study, by employing the siRNA technique, examined the involvement of Sig-1Rs in the development of dendritic arborization and the associated anchorings of NMDA and AMPA receptors in rat hippocampal primary neurons. Hippocampal neurons treated with the siRNA against Sig-1Rs form fewer dendritic spines whereas possess more filopodia. The MAP-2B staining, which marks neuronal dendrites, shows that the siRNA treatment also impairs the growth and development of dendrites. Morphometric measurements indicate that siRNA-transfected neurons have shorter dendrites and possess fewer branches on the dendrite. The knockdown of Sig-1Rs by the siRNA not only reduces the level of marker proteins for synapse, including PSD-95 and synaptophysin, but also changes the localization of NMDA and AMPA receptors in hippocampal neurons. In siRNA-transfected neurons, NR-1 and GluR2/3 respectively form clusters and localize mainly at the dendritic shafts and not at spine heads where they normally exist. siRNAs also retard the extension of axons in hippocampal neurons. These results suggest that Sig-1Rs are related to a common mechanism critical at multiple steps in the development and maturation of dendritic arbors and, as a consequence, affect the anchoring of receptors or ion channels to the postsynaptic region. The exact mode of action of Sig-1Rs in this regard is under investigation.

SEROTONIN TRANSPOTER BINDING IN PATIENTS WITH OBSESSIVE COMPULSIVE DISORDERS: A PET WITH [11C] DASB

Ryohei Matsumoto^1,2, Hiroshi Ito¹, Yota Fujimura¹, Yoko Ikoma¹, Chie Seki¹, Jun Kosaka¹, Ryosuke Arakawa¹, Tomomichi Ando¹, Akihiro Takano¹, Hidehiko Takahashi¹, Kenji Fukui², Tetsuya Suhara¹

¹Department of Molecular Neuroimaging, Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan, ²Department of Psychiatry, Graduate School of Kyoto Prefectural University of Medicine, Kyoto, Japan

BENEFICIAL EFFECTS OF INTRATHECAL IGF-1 ADOMINISTRATION IN PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS

Nobutoshi Morimoto, Isao Nagano, Tetsuro Murakami, Makiko Nagai, Mito SHiote, Yasuyuki Ota, Kazunori Miyazaki, Tomoko Kurata, Mizuki Ishikawa, Yasushi Takehisa, Tatsusi Kamiya, Kouji Abe

Department of Neurology Graduate School of Medicine and Dentistry, Okayama University, Okayama, Japan

Background: Amyotrophic lateral sclerosis (ALS) is a progressive paralytic disorder caused by degeneration of the upper and lower motor neurons in the brain and spinal cord resulting in progressive paralysis of voluntary muscles. However, there is currently no effective pharmacological treatment for amyotrophic lateral sclerosis. In a transgenic mouse model of ALS, intrathecal infusion of insulin-like growth factor-1 (IGF-1) showed a promising increase in survival.

Objectives: To examine whether IGF-1 is effective for the prevention of disease progression, we performed a double-blind clinical trial to assess the effect of intrathecal administration of IGF-1 in patients with ALS.

Methods: Nine patients with ALS were randomly assigned to receive either a high dose (3 g/kg of body weight) or low dose (0.5 g/kg of body weight) of IGF-1 every 2 weeks for 40 weeks. The outcome measurements were the rate of decline of bulbar and limb functions (Norris scales) and forced vital capacity.

Results: The high-dose treatment slowed a decline of motor functions of the ALS patients in total Norris and limb Norris scales, but not in bulbar Norris or vital capacity. The intrathecal administration of IGF-1 had a modest but significant beneficial effect in ALS patients without any serious adverse effects.

Discussion: Intrathecal IGF-1 treatment could provide an effective choice for ALS although further studies in more patients are needed to confirm its efficacy and optimize dosages of IGF-1.

BRAIN ELECTRIC SIGNATURE OF DE NOVO PARKINSON'S DISEASE WITH LOW RESOLUTION BRAIN ELECTROMAGNETIC TOMOGRAPHY

Hideaki Tanaka¹, Mio Arai², Yuya Hoshino¹, Tomohiro Ogawa¹, Mikio Okayasu¹, Yuka Watanabe¹, Hidehiro Takekawa¹, Akinori Hozumi¹, Koichi Hirata¹

¹Department of Neurology, Dokkyo Medical University, Mibu, Tochigi, Japan, ²Department of Neurology, Tokyo Medical University, Kasumigaura Hospital, Kasumigaura, Ibaraki, Japan

Background and aims: The present study aimed to search for the brain electric signature of drug naive de-novo Parkinson's disease patients (PDPs) without dementia compared with normal controls (NCs).

Methods: We studied 16 drug naive de-novo PDPs and 16 age-, sex- and education- matched NC subjects. We recorded 20-channel spontaneous eyes closed resting EEGs (BioLogic Brain Atlas, 0.53–30.Hz band pass) according to the international 10/20 system. After recording, 20 artifact-free EEG epochs consisting of 2-second were selected for each subject. Global field power (GFP) for EEG frequency were determined by means of a Fast Fourier Transform (FFT) and compared between the two groups. Low Resolution Electromagnetic Tomography (LORETA) was used to compute the three-dimensional intracerebral distribution of electric activity for 7 EEG bands: delta (1.5–6.Hz), theta (6.5–8.Hz), alpha1 (8.5–10.Hz), alpha2 (10.5–12.Hz), beta1 (12.5–18.Hz), beta2 (18.5–21.Hz), beta3 (21.5–30.Hz). Images of two groups were statistically compared using non-parametric randomization tests with correction for multiple testing. To assess the cognitive functions, the Japanese version of the Mini-Mental State Examination (MMSE) and the New Modified Wisconsin Card Sorting test (WCST) were recorded.

Results: MMSE score revealed no significant differences between two groups, however PDPs showed significant reduction of achieved categories and significant increase of preservative errors in WCST. The EEGs of patients showed significant increase in slow activity (delta and theta) and significant reduction in fast activity (alpha2 and beta1). And alpha peak frequency of PDPs decreased. LORETA analysis indicated significant increase of slow activity (delta and theta) in the bilateral occipital lobe and significant decrease of fast activity (alpha2 and beta1) in the limbic lobe in PDPs compared with NCs. indicated occipital and limbic hypoactivities perfusion in denovo PDPs without dementia. This findings confirmed SPECT data which demonstrated occipital hypoperfusion corresponding to impairment of visual cognition.

CORTICAL NEURONAL AND GLIAL PATHOLOGY IN TGTAUP301L TRANSGENIC MICE: NEURONAL DEGENERATION, MEMORY DISTURBANCE AND PHENOTYPIC VARIATION

Tetsuro Murakami¹, T. Kawarabayashi², M. Ikeda³, E. Matsubara⁴, M. Nagai¹, T. Kurata¹, N. Morimoto¹, Y. Ohta¹, Y. Takehisa¹, K. Abe¹, D. Westaway⁵, P. St. George-Hyslop⁵, M. Shoji²

¹Department of Neurology, Okayama University Medical School, Okayama, Japan, ²Department of Neurology, Hirosaki University School of Medicine, Hirosaki, Japan, ³Department of Neurology, Gunma University Medical Shool, Gunma, Japan, ⁴Department of Alzheimer's Disease Research, National Institute of Longevity Sciences, National Center for Geriatrics and Gerontology, Aichi, Japan, ⁵Departments of Medicine (Neurology) and Laboratory Medicine and Pathobiology, Centre for Research in Neurodegenerative Diseases, University of Toronto, Toronto, ON, Canada

Recapitulation of tau pathologies in an animal model has been a long-standing goal in neurodegenerative disease research. We generated transgenic (TgTauP301L) mice expressing a frontotemporal dementia with parkinsonism linked to chromosome 17 (FTPD-17) mutation within the longest form of tau (“2N, 4R”). TgTauP301L mice developed florid pathology including neuronal pretangles, numerous Gallyas-Braak-positive neurofibrillary tangles and glial fibrillary tangles in the frontotemporal areas of the cerebrum, in the brainstem, and to a lesser extent in the spinal cord. These features were accompanied by gliosis, neuronal loss and cerebral atrophy. Accumulated tau was hyper-phosphorylated, conformationally changed, ubiquitinated and sarkosyl-insoluble, with electron microscopy demonstrating wavy filaments. Aged TgTauP301L mice exhibited impairment in hippocampally-dependent and -independent behavioral paradigms, with impairments closely related to the presence of tau pathologies and levels of insoluble tau protein. We conclude that TgTauP301L mice recreate the substantial phenotypic variation and spectrum of pathologies seen in FTDP-17 patients. Identification of genetic and/or environmental factors modifying the tau phenotype in these mice may shed light on factors modulating human tauopathies. These Tg mice may aid therapeutic development for FTDP-17 and other diseases featuring accumulations of four-repeat tau, such as Alzheimer's disease, corticobasal degeneration and progressive supranuclear palsy.

AGE-DEPENDENT PLAQUE DEPOSITION CORRELATES WITH A MORE EXPLORATORY PHENOTYPE IN ELDERLY TRANSGENIC APPSWE/PS1DE9 MICE WHEN ADMINISTERED WITH DOI

Ruben Christensen¹, Gitta Wortwein², Anders Brandt Jorgensen³, Tine Bryan Stensbol³, Gitte

Moos Knudsen¹, Susana Aznar¹

¹Neurobiology Research Unit, Copenhagen University Hospital, Copenhagen, Denmark, ²The Laboratory of Neuropsychiatry, Copenhagen University Hospital, Copenhagen, Denmark, ³H. Lundbeck A/S, Copenhagen Valby, Denmark

The transgenic APPswe/PS1dE9 animal model is regarded as a solid model of Alzheimer's disease (AD). These transgenic mice display an age-dependent increase in brain beta-amyloid (Abeta) plaque deposition, which occurs early (4-6 months) and is accelerated during their lifespan. Several PET studies have shown that cerebral 5-HT2A receptor binding is decreased in Alzheimer patients, correlating with the cognitive decline and neuropsychiatric symptoms, i.e. mood disturbances and depression. Furthermore, newer studies show that the 5-HT2A receptor is involved in anxiety modulation. We made use of these animals to study the association between plaque deposition and anxiolytic and changes in exploratory behaviour.

Different groups of APPswe/PS1dE9 transgenic and wild type mice, corresponding to age 4, 7 and 10 months, defined as pre-, mid- and late plaque stages respectively, were tested in the elevated plus maze and open field test. In the later test, animals were injected with a single injection of the 5-HT2A serotonin receptor agonist, DOI, to see whether the exploratory behaviour of the transgenic mice would differ by pharmacological activation of this receptor.

When tested in the elevated plus maze test, the transgenic APPswe/PS1dE9 mice did not show higher anxiety than the age-matched wild type mice. Both the transgenic and wild-type mice showed a decrease in time spending on the open arm with aging. In the open field test, transgenic animals showed a significant effect of DOI in the 7 months group. When stimulated with DOI the transgenic mice displayed a more active exploratory behaviour than their age-matched wild-type. In the 10 month old group no significant difference between the transgenic and wild-type was observed, probably due to the heavy plaque burden.

These preliminary results indicate an alteration in the responsiveness to the 5-HT2A agonist in the 7 months transgenic APPswe/PS1dE9, suggesting changes in the levels of this receptor in animals with beta-amyloid plaque deposition. Furthermore, these results indicate the involvement of the 5-HT2A in exploratory behaviour.

DELAYED ADMINISTRATION OF CYCLIC G-2ALLYLP REDUCED THE LONG-TERM FORELIMB AKINESIA INDEPENDENT STRIATAL DOPAMINE DEPLETION IN RATS

Jian Guan, Sam Mathai, Helena Kim, Rong Zhang, Rita Krishnamurthi

Liggins Institute, University of Auckland, Auckland, New Zealand

Background: Parkinson's disease (PD) is a neurodegenerative disorder due to progressive substantia nigra-striatum dopamine depletion. PD patients only experience bradykinesia, akinesia and loss of motor coordination after the loss of majority dopamine neurons in substantia nigra. During the pre-symptomatic period the brain appeared to be able to compensate and delay impending motor deficits, probably through endogenous reparative mechanisms including neurogenesis and the plasticity within cortical - basal ganglia circuit in both dopamine and non-dopamine neurotransmission. Cyclic glycol-2allyl-proline (cG-2allylP), derived from an endogenous DKP protects neurons from ischemic brain injury. Current study investigated the possible treatment effect of cG-2allylP in preventing 6-OHDA induced Parkinson motor deficit and dopamine depletion in rats.

Methods: Unilateral dopamine depletion was induced in adult rats. Two microinjections of 6-OHDA (7µg × 2 sites) were administered into right side striatum stereotaxically. 6-OHDA induced forelimb akinesia was examined using adjusting step test two weeks after the lesion. To have a comparable baseline crossed all groups, the rats were allocated into four treatment groups based on the number of adjusting steps made. Two weeks daily treatment of cG-2allylP (0 or 0.2 or 1 or or 5mg/kg/day, s.c) were initiated 14 days after the lesion. Parkinsonian deficit was examined at 6, 8, 10 and 12 weeks after the lesion using adjusting step tests. The nigral-striatal dopamine depletion was examined using tyrosine hydroxylase (TH) immunihistochemical staining 12 weeks after the lesion.

Results: 6-OHDA injection resulted to an approximately 90% loss of TH positive neurons in the SNc and around 80% loss of TH terminal staining in the striatum 12 weeks after the injury. A loss of ChAT positive neurons was also seen in the medial-ventral striatum, where 6-OHDA was injected. The number of adjusting steps was significantly reduced by approximately 80% 14 day after 6-OHDA lesion in all the groups. While the deficit remained in the vehicle treated group throughout the tests, the treatment of cG-2allylP significant increased the number of adjusting steps by 50%, suggesting a functional recovery. The treatment effect was not dose dependent with three doses tested showing similar effects. A trend towards increase in number of TH positive neurons was found in the SNc, however there was no improvement in TH terminal staining in the striatum. Cyclic G-2allylP treatment increased the number of doublecortin positive neuroblasts in the sub-ventricular zone compared to the vehicle treated group and without altering cell proliferation. A partial restoration in striatal ChAT neurons was also seen in the group treated with cG-2allylP.

Conclusion: Delayed treatments with cG-2allylP after the onset of motor deficit improved the long-term motor function following 6-OHDA induced striatum dopamine depletion. The functional recovery of cG-2allylP was independent on dopamine restoration in the striatum. The improved neurogenesis and plasticity in non-dopamine neurotransmission may have a role in cG-2allylP associated functional recovery, which need to be further examined.

IN VIVO DETECTION OF AMYLOID DEPOSITS IN ALZHEIMER'S DISEASE PATIENTS USING [11C]BF-227 PET

Nobuyuki Okamura¹, Yukitsuka Kudo², Shozo Furumoto², Manabu Tashiro³, Katsutoshi Furukawa⁴, Yoshihito Funaki³, Motohisa Kato¹, Satoshi Ito¹, Yoichi Ishikawa³, Masahiro Maruyama⁴, Hiroyuki Arai⁴, Hiroshi Fukuda⁵, Ren Iwata³, Masatoshi Itoh³, Kazuhiko Yanai¹

¹Department of Pharmacology, Tohoku University School of Medicine, Sendai, Japan, ²Tohoku University Biomedical Engineering Research Organization, Sendai, Japan, ³Cyclotron and Radioisotope Center, Tohoku University, Sendai, Japan, ⁴Department of Geriatrics and Gerontology, Tohoku University School of Medicine, Sendai, Japan, ⁵Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan

Background and aims: Progressive deposition of amyloid plaques is a critical event for the pathogenesis of Alzheimer's disease (AD). In vivo detection of amyloid deposits using positron emission tomography (PET) would thus prove useful for early diagnosis of AD and tracking disease progression. The purpose of this study is to evaluate the diagnostic utility of a novel PET imaging probe, 2-[2-(2-Dimethylaminothiazol-5-yl)ethenyl]-6-[2-(fluoro)ethoxy]benzoxazole (BF-227), in AD patients. Methods: Binding affinity of BF-227 to amyloid-β (Aβ) fibrils was calculated. Binding property of BF-227 to amyloid plaques was evaluated by neuropathological staining of AD brain sections. For clinical evaluation of [11C]BF-227, 10 normal subjects, 10 subjects with mild cognitive impairment (MCI) and 10 patients with AD participated in this study. Dynamic PET images were obtained for 60 min after administration of [11C]BF-227. Regional standardized uptake value (SUV) and the ratio of regional to cerebellar SUV were calculated as an index of [11C]BF-227 retention. Regional tracer distribution of AD patients was statistically compared with normal subjects on a voxel-by-voxel basis. Results: BF-227 displays high binding affinity to Aβ fibrils (Kd = 0.73 nM). BF-227 specifically binds amyloid plaques in AD brain sections. The clinical PET study using [11C]BF-227 demonstrated the retention of this tracer in cerebral cortices of AD patients, but not in those of normal subjects. All AD patients were clearly distinguishable from normal individuals using temporal SUV ratio. Cortical BF-227 retention in MCI subjects appeared either AD-like or control-like. Patient with frontotemporal dementia (FTD) showed no cortical retention of BF-227. Voxel-by-voxel analysis of PET images revealed that cortical BF-227 retention in AD patients is mainly distributed to posterior association area of the brain and corresponded well with the predilected site for amyloid plaque depositions containing dense Aβ fibrils. Conclusions: These findings suggest that [11C]BF-227 is a promising PET probe for in vivo imaging dense amyloid deposits in AD patients.

AGE-RELATED WHITE MATTER CHANGES, CEREBRAL MICROBLEEDS AND THEIR CORRELATION WITH COGNITIVE FUNCTION AND DEPRESSION

Monzurul Hasan Chowdhury¹, Atsushi Nagai², Kazuo Takahashi¹, Hirokazu Bokura¹, Shuhei Yamaguchi¹, Shotai Kobayashi¹

¹Deparment of Internal Medicine III, Shimane University School of Medicine, Izumo, Shimane, Japan, ²Deparment of Laboratory Medicine, Shimane University School of Medicine, Izumo, Shimane, Japan

Background and aims: Aging and hypertension are the two major contributory risk factors for the white matter changes. Hypertension also plays a leading role for the occurrence of cerebral microbleeds. In healthy people, without any cerebrovascular disease (CVD) risk factors, aging and cerebral amyloid angiopathy sometimes progress simultaneously and silently. In our study, the relationship of their symptoms, cognitive functions, and MRI findings was evaluated in normal population.

Methods: We searched the database of 1120 subjects without hypertension, diabetes mellitus and hyperlipidemia (as defined by the standard regulatory body), visited the Institute of Shimane Health Science for health screening of the brain over 11 years from 1995 to 2006. The mean age of our study population was 60.3 ± 10.0 years and sex ratio M/F was 608/512. We examined all the physical parameters and blood levels available in the database and also examined both the conventional and T2* weighted MRI to define the cerebral microbleeds, cerebral amyloid angiopathy and age related brain changes in our study population. The diagnosis of periventricular hyperintensity was performed on MRI and was graded from 0-4, where grade 0–2 were defined as mild PVH and grade 3–4 were defined as marked PVH. After grading of PVH, we performed the logistic regression analysis to evaluate the outcome of these 2 groups in respect of their age, blood pressure, blood values, cognitive functions with word fluency, Okabe's intelligence evaluation test (Okabe's test: a shortened version of WAIS), Kohs' block design test (Kohs' test) and Zung's self rating depression scale (SDS).

Results: In our study population, 38 subjects (3.4%) had PVH score 3-4. Logistic regression analysis showed that age and memory were significantly related with PVH (p < 0.0001, p < 0.001). T2* weighted MRI revealed 9 subjects had cerebral cortical microbleeds and 9 subjects had deep gray matter microbleeds. Comparisons of these two groups revealed that the cerebral cortical microbleeds group had higher age, larger deep white matter hyperintensity and hippocampal atrophy, associated with less memory score (Okabe's test) as compared with the deep gray matter microbleeds group.

Conclusion: We demonstrated that PVH is related with aging and memory impairment in healthy elderly individuals. It is suggested that when cerebral cortical microbleeds occur in elderly individuals without any known CVD risk factors, cerebral amyloid angiopathy should be considered.

ACCUMULATION OF PARKINSON'S-DISEASE-RELATED MOLECULES IN LEWY BODIES AND GLIAL CYTOPLASMIC INCLUSIONS

Tetsuro Murakami¹, Y. Imai², H. Inoue³, T. Kawarabayashi⁴, M. Nagai¹, T. Kurata¹, Y. Takehisa¹, Y. Harigaya⁵, M. Shoji⁴, R. Takahashi³, K. Abe¹

¹Department of Neurology, Okayama University Medical School, Okayama, Japan, ²RIKEN, Brain Science Institute, Saitama, Japan, ³Department of Neurology, Kyoto University Medical School, Kyoto, Japan, ⁴Department of Neurology, Hirosaki University of Mediciene, Aomori, Japan, ⁵Department of Neurology, Maebashi Red Cross Hospital, Gunma, Japan

Objective: To clarify the involvement of molecules that have been shown to be related to Parkinson's disease in the formation of Lewy bodies and glial cytoplasmic inclusions.

Background: The appearance of Lewy bodies (LBs), as well as neuronal loss in the substantia nigra, is a hallmark of Parkinson's disease (PD). Multiple system atrophy (MSA) is a progressive neurodegenerative disease characterized clinically by parkinsonism and ataxia, and pathologically by accumulation of alpha-synuclein in oligodendrocytes (glial cytoplasmic inclusions, GCIs). Recent progress in molecular biology has revealed several proteins that are genetically or experimentally related to PD. PD and MSA are now summarized as alpha-synucleinopathy, and a common cascade arising from the aggregation and accumulation of alpha-synuclein to the formation of LBs or GCIs leading to cell death is suggested.

Methods: We immunohistopathologically examined the localization of Parkinson's-disease-related molecules (alpha-synuclein, Pael-R, Parkin, Glup, and ubiquitin) in LBs and GCIs in PD, MSA, and normal controls (n=6, respectively).

Results: All these molecules examined were shown to be accumulated in LBs of PD. In GCIs of MSA, however, only alpha-synuclein and ubiquitin were shown to be present, and Pael-R, Parkin, and Glup were not accumulated.

Conclusions: Alpha-synuclein, Pael-R, Parkin, Glup, and ubiquitin are actually accumulated in LBs, and might be involved in the inclusion formation. However, in GCIs of MSA, only alpha-synuclein and ubiquitin was present, suggesting that the presence of a complete distinct pathway involved in the formation of LBs and GCIs in PD and MSA.

DISTRIBUTION OF AMYLOID DEPOSITION IN ALZHEIMER'S DISEASE AND NORMAL AGING DEMONSTRATED BY [11C]PIB PET

Hitoshi Shinotoh, Kiyoshi Fukushi, Shigeki Hirano, Hitoshi Shimada, Noriko Tanaka, Tsuneyoshi Ota, Koichi Sato, Tetsuya Suhara, Toshiaki Irie

Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan

Background and aims: Beta-amyloid plaques are the hallmark of Alzheimer's disease (AD) and the prevalence of AD increases in general population as aging. A previous [11C]PIB PET study showed that four of the 41 non-demented subjects already had amyloid deposition in the brain similar to those in AD patients1). We investigate age-associated amyloid deposition in middle-aged to elderly healthy subjects in an attempt to explore the initial distribution of amyloid deposition in presymptomatic phase of AD by [11C]PIB PET.

Methods: The subjects consisted of 15 healthy subjects ranging in age from 48 to 91 years with a mean age of 66 years and 12 patients with AD ranging in age from 58 to 85 years with a mean age of 69 years. The mean MMSE scores of the AD group was16.5 ± 5.3. [11C]PIB -PET acquisition was a dynamic sequence of 19 PET scans for 90 min. Spatial normalization of PET images were performed using a specific [11C]PIB template, which was created from 24 [11C]PIB scans. The Logan graphical analysis using arterial input function was employed to calculate distribution volume (DV). The target tissue to the cerebellum DV ratio (DVR) was calculated as an index of amyloid deposition in each voxel. Voxel-based statistical analysis was performed using SPM2.

Results: Two-tailed t-test showed significant amyloid deposition in the posterior cingulate, precunate, lateral parietal and temporal cortices, and frontal cortex in the AD patients compared with healthy controls (p<0.001, extent >100 voxels). Simple regression analysis showed a significant correlation between age and amyloid deposition in the posterior cingulated, precunate, upper part of parietal cortex, lower lateral temporal cortex, gyrus rectus, and occipital cortex (p <0.01, extent >100 voxels).

Discussion and Conclusion: The age-associated amyloid deposition may represent the amyloid deposition in the early phase of presymptomatic AD. The distribution of the age associated amyloid deposition mostly overlapped with those in AD. However, the age-associated amyloid deposition was seen in the occipital cortex, where no significant amyloid deposition was seen in the AD patients. The results suggest that the occipital cortex may be one of brain regions with amyloid deposition in the early phase of AD but the amyloid may not deposit further in the occipital cortex in the subsequent phases of AD.

CHANGES IN CBF DISTRIBUTION DURING ADOLESCENCE:A STUDY IN NEUROLOGICALLY NORMAL PATIENTS BY MEANS OF BRAIN SPECT AND SPM99

Taku Fujimoto, Kentaro Fujii, Hironori Otomune, Takuya Uehara, Yasuhiro Osaki, Ashik Bin Ansar, Yasuyuki Kimura, Katsufumi Kajimoto, Hiroki Katoh, Makiko Tanaka, Eku Shimosegawa, Jun Hatazawa

Department of Nuclear Medicine and Tracer Kinetics, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

Objectives: The changes in regional cerebral blood flow (CBF) distribution during adolescence have been little studied because of ethical reasons. We analyzed regional CBF distribution in neurologically normal patients age ranged from 8 to 41 years.

Patients and Methods: Twenty-two patients were enrolled in the present study. They were suspected to have neurological defects, and it was finally concluded that they were neurologically normal. All underwent brain perfusion SPECT with 99mTc-ECD under resting condition and no medications. The patients were subdivided into the following age groups: the 8-11 years-old group (early adolescence, 4 men and 3 women), the 16-18 years-old group (late adolescence, 4 women), and the 20-40 years-old group (Adult, 5 men and 6 women). By means of the SPM99, the images were anatomically normalized and smoothed. The between-group differences in CBF distribution were estimated.

Results: Between early and late adolescence, significant relative CBF difference (late-early) was found in right hippocampus (x=24, y=6, z=20, p<0.01, uncorrected) and anterior cingulate gyrus (x=10, y=26, z=-12, p<0.05, uncorrected). Between late adolescence and adult group, significant CBF difference (adult-late) was found in bilateral precuneus and posterior cingulate gyrus (x=22, y=76, z=40, p<0.01, uncorrected).

Discussion: The CBF distribution in the right hippocampus and anterior cingulate gyrus was increased during early adolescence, and the CBF distribution in the bilateral precuneus and posterior cingulate gyrus was increased during late adolescence. The hippocampus is responsible for storage and retrieval of general declarative memory [1]. The precuneous plays main part in retrieval of episodic memory [2]. The anterior cingulate gyrus mediates processing selection of extraordinary stimulation such as visual or olfactory stimulation [3]. The posterior cingulate cortex is involved in spatial orientation and memory, and connections between this area and parahippocampal cortices contribute to these processes [4]. Our findings indicate that the functional developments of memory system in human brain occur during adolescence.

DYSFUNCTION OF MITOCHONDRIAL OXIDATIVE PHOSPHORYLATION IN ALS RAT MODEL

Makiko Nagai¹, Tetsuro Murakami¹, Yasuyuki Ohta¹, Yasushi Takehisa¹, Nobutoshi Morimoto¹, Kazunori Miyazaki¹, Tomoko Kurata¹, Tatsushi Kamiya¹, Koji Abe¹, Serge Przedborski²

¹Department of Neurology, Okayama University, Okayama, Japan, ²Center for Motor Neuron Biology and Disease, New York, NY, USA

Background and aims. Amyotrophic lateral sclerosis (ALS) is caused by the selective degeneration of motoneurons in spinal cord, brain stem and cerebral cortex. The degeneration of motoneurons leads to skeletal muscle atrophy, paralysis and death. About 10 % of ALS cases are familial and mutation of Cu, Zn superoxide dismutase (SOD1) has been demonstrated to be a culprit among 20 % of the familial cases. Although contraversial, mitochondrial abnormalities have been reported in ALS-patients and in mouse model such as mutated human SOD1 transgenic mice and mnd mouse. To address this issue we assess mitochondrial function in transgenic rats overexpressed a human mutated SOD1 associated with ALS.

Material and methods. Spinal cords and cerebellum from G93A and non-transgenic rats were homogenized in isolation buffer. The homogenate was centrifuged at 1,000 g for 10 min, and the resulting supernatant was layered onto 7.5% Ficoll medium on top of the 10% Ficoll medium and centrifuged at 79,000 g for 30 min. The mitochondiral pellet was resuspended in isolation buffer. 10 µg isolated mitochondria were separated on 15% SDS-PAGE and transferred to a nitorocellulose membrane. The membrane was immunostained rabbit anti-human SOD1 antibody or mouse anti-cytochrome c oxidase antibody. Secondary antibodies conjugated with HRP were used. Mitochondria were suspended in respiration buffer at 28 oC, and oxygen-consumption rates were measured in a closed-chamber cuvette with a mini-stirring bar using Clark-type electrode. To assess mitochondrial respiration mediated through complex I, mitochondria were preincubated with 10 mM glutamate and 5 mM for 2.5 min before 500 µM ADP was added to induce state 3 respiration. For complex II mediated respiration, mitochondria were preincubated for 2.5 min with 10 mM succinate before 500 µM ADP was added. For complex IV mediated respiration, mitochondria were preincubated for 2.5 min before 10 mM ascorbate and 0.2 mM N,N,N',N'-tetramethyl-benzendiamine were added to induce oxygen consumption.

Results. Significant level of human mutated SOD1 was detected in purified mitochondria of G93A rat, but not endogeneous rat SOD1. SOD1 aggregates were detected in G93A spinal cord mitochondria. Significant reduction of O2 consumption using complex I substrates were observed in the spinal cord mitochondria from G93A rats. On the other hand, enzyme activities, which are normalized against citrate synthase activities showed no significant changes in complex I. Polarographical study of mitochondrial respiration via complex II or complex IV showed no significant difference between nontransgenic and G93A rat mitochondria. difference between non-transgenic and G93A rat mitochondria.

Conclusion. In transgenic G93A rat model of ALS, we showed a specific reduction of mitochondria respiration mediated through complex I, but not complex II and IV. This functional abnormality was only detected in intact mitochondria of G93A spinal cord, but not in freeze-thawed lysed mitochondria. This study supports the role of mitochondial defects in the pathogenesis of ALS.

EXTENDED POLYGLUTAMINE REPEATS TRIGGER A FEEDBACK LOOP INVOLVING THE MITOCHONDRIAL COMPLEX III, THE PROTEASOME, AND HUNTINGTIN AGGREGATES

Hirokazu Fukui¹, Carlos Moraes^1,2,3

¹Neuroscience Graduate Program, University of Miami Miller School of Medicine, Miami, FL, USA, ²Department of Neurology, University of Miami Miller School of Medicine, Miami, FL, USA, ³Department of Cell Biology and Anatomy, University of Miami Miller School of Medicine, Miami, FL, USA

Mitochondrial abnormalities represent a major cytopathology in Huntington's disease, a fatal neurodegenerative disease caused by CAG repeat expansions in the gene encoding huntingtin. In the present study, we investigated whether defects in the mitochondrial respiratory function are consequences of the expression of mutant huntingtin or whether they promote the formation of huntingtin aggregates. To take advantage of existing mitochondrial DNA mutants, we developed human osteosarcoma 143B cells expressing mutant huntingtin in an inducible manner and found that cells expressing mutant huntingtin but not wild type huntingtin exhibited a reduced activity of Complex II + III and an increased activity of Complex IV. Conversely, pharmacological treatments that inhibited Complex III activity significantly promoted the formation of huntingtin aggregates. This effect of Complex III inhibition on the huntingtin aggregates appeared to be mediated by the inhibition of proteasome activity but not by ATP depletion or production of reactive oxygen species. Accordingly, complex III mutant cells also showed decreased proteosome activity. These results suggest the presence of a feedback system connecting the mitochondrial respiratory Complex III and the production of huntingtin aggregates. Our results suggest that therapeutic interventions targeting complex III and/or proteasome could ameliorate the progress of Huntington's disease.

DIFFERENCES OF NEURAL INTERACTIONS BETWEEN EARLY AND LATE ONSET ALZHEIMER'S DISEASE: A NETWORK ANALYSIS

Sang Soo Cho¹, Heejung Kim¹, Hey Jin Kang¹, Eun Joo Kim², Dong Soo Lee¹, Yu-kyeong Kim¹, Duk L. Na², Sang Eun Kim¹

¹Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, South Korea, ²Department of Neurology, Sungkyunkwan University School of Medicine, Seoul, South Korea

Background and aims: Early-onset (EO) and late-onset (LO) Alzheimer's diseases (ADs) are known to have different cognitive decline profiles and pathophysiology. The present study was designed to determine and compare the neural interactions underlying general cognitive function for the two subtypes of Alzheimer's disease using FDG PET and path analysis.

Methods: FDG PET imaging was performed in 73 patients with EO AD (AD onset before 65 yr) (age 59.1± 5.7 y, 49 females/24 males, MMSE score 17.5±6.9), 47 patients with late onset AD (age 74.1±4.7 y, 33 females/14 males, MMSE 18.9±6.5) and 15 young (54.0±2.8 y, 6/9) and 20 aged (69.0±4.1 y, 12/8) healthy controls. For path analysis, five brain regions (middle and inferior frontal, inferior temporal gyri, inferior parietal lobule, and posterior cingulate) were selected in each hemisphere based on the results of the correlation analysis between MMSE score and regional brain glucose metabolism using SPM99. Hippocampus was selected additionally because this area is important for the pathology of AD. Voxel values for each coordinate were extracted from the data set of each subject. For the model construction, anatomical connection was defined based on the previous researches. First, we tested the goodness-of-fit to assess the suitability of the constructed model. Second, using this model, we evaluated the differences in the path model between two patients groups. All the model construction and path analysis processing was done by AMOS 5.0 (SmallWaters Co.)

Results: Goodness-of-fit of final constructed model was selected [X2 (4) =3.12, P=0.54 in left and X2 (4) = 1.88, P=0.76 in right]. In both hemispheres, the intra-hemispheric functional networks proved to be significantly different between the two AD groups [left: X2 diff (13) = 28.38, P < 0.01; right: X2 diff (13) = 27.65, P < 0.05]. Comparing with functional network of age-matched normal controls, several inter-regional connections were weakened in both AD group. In EO AD, especially the neural connections from frontal to lateral and medial temporal area were changed while in LO AD, those of frontal to posterior cingulate and hippocampus to posterior cingulate were changed.

Conclusion: Our data demonstrate the different patterns of dysorganization in functional network between EO and LO AD patients. They provide additional insight into the pathophysiology of AD subtypes and neural functioning for cognitive processes.

EFFICACY OF MINOCYLINE FOR DELAYED TISSUE PLASMINOGEN ACTIVATOR TREATMENT AFTER EMBOLIC FOCAL ISCHEMIA IN RATS

Yoshihiro Murata¹, Anna Rosell¹, Xiaoying Wang¹, Robert H. Scannevin², Kenneth Rhodes², Yoichi Katayama³, Eng H. Lo¹

¹Neuroprotection Research Laboratory, Departments of Neurology and Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA, ²Johnson and Johnson Pharmaceutical Research and Development, Spring House, PA, USA, ³Department of Neurological Surgery, Nihon University School of Medicine, Tokyo, Japan

INTRODUCTION: Tissue plasminogen activator (tPA) therapy for acute stroke patients must be administrated within 3 hours after symptom onset. The narrow therapeutic time window (TTW) may be related to the elevated risks of intracranial hemorrhage after thrombolysis. In the present study, we investigated whether minocycline can prevent delayed tPA- associated intracranial hemorrhage and extend the TTW of tPA treatment. METHODS: Male spontaneously hypertensive rats were subjected to embolic middle cerebral artery (MCA) occlusion. Regional cerebral blood flow was monitored by laser-Doppler flowmetry. Five experimental groups were used. Gtoup1; Saline at 1hour after ischemia (untreated group), Group2; tPA at 1hour after ischemia, Group3; tPA at 6hr after ischemia (delayed tPA), Group4; minocycline at 4hours after ischemia (minocycline alone), Group5; minocycline and tPA were administrated respectively at 4hours and at 6hours (combination treatment with minocycline and delayed tPA). All drugs were administrated intravenously. At 24hours after ischemia, we evaluated mortality rates, infarct volumes with TTC (2,3,5-triphenyltetrazolium chloride) staining, and hemorrhage volumes with spectrophotometric assay. Blood samples for MMP zymography were taken from Group 3 and Group 5 to investigate the effect of minocycline for delayed tPA treatment. RESULTS: Early 1 hr thrombolysis with tPA restored blood flow and significantly reduced infarct volumes. However, late 6 hr tPA did not work, and instead increased mortality and induced hemorrhagic conversion. Treatment with minocycline re-established late tPA's protective effects by decreasing mortality, reducing infarction volumes in surviving rats, and ameliorating brain hemorrhage severity. Quantitative data are presented in the Table below:

In Group 3, plasma MMP-9 was upregulated after ischemia and tPA treatment further amplified these ischemic responses in MMP-9. Combination treatment with minocycline plus tPA significantly reduced the increment of MMP-9 post-ischemia and post-tPA, in concert with the improved outcomes described above. CONCLUSIONS: Our initial results indicate that (1) minocycline can reduce infarct volume even though tPA is administrated at 6 hours after ischemia, (2) minocycline can prevent intracranial hemorrhage associated with tPA treatment, (3) minocycline can reduce the mortality rate associated with delayed tPA treatment, and (4) these improvements in outcome after delayed tPA plus minocycline therapy may be related to the ability of minocycline to ameliorate ischemic and tPA-induced MMP-9 responses. Therefore, minocycline may be a promising candidate to extend the TTW of tPA treatment in part by preventing tPA-associated intracranial hemorrhage. Supported in part by NIH grants R01-NS37074, R01-NS48422, R01-NS56548, and P50-NS10828.

ELEVATION OF INTERLEUKIN-6 LEVELS IN THE ASPIRATION FROM THE STENOTIC LESION AFTER PERCUTANEUS CAROTID ARTERY STENTING

Yuko Abe¹, Kazuo Kitagawa¹, Manabu Sakaguchi¹, Taku Hoshi¹, Shigetaka Furukado¹, Hidetaka Hogaku¹, Masatsugu Hori², Toshiyuki Fujinaka¹, Toshiki Yoshimine¹

¹Stroke Center, Osaka University Hosipital, Suita, Osaka, Japan, ²Department of Cardiovascular Medicine, Osaka University, Suita, Osaka, Japan

Background: Inflammation plays a fundamental role in the initiation, development, progression, and eventual rupture of atherosclerotic plaques. A variety of circulating markers of inflammation such as high sensitivity C-reactive protein (hsCRP), interleukin-6 (IL-6) and interleukin-18 (IL-18) have been predictors of future vascular events. However, it remains unclear as to which markers of inflammation originate locally from the atherosclerotic plaque. Carotid artery stenting (CAS) with a distal protection device causes the mechanical plaque rupture and releases the components of atheromatous plaque. Therefore, we investigated inflammatory marker levels by aspiration at the site of the stenotic lesion after CAS together with those in the descending aorta.

Methods: The study involved 23 patients with 27 severe carotid artery stenosis vessels that underwent CAS. In all patients, a balloon based embolization protection device and aspiration catheter were used during CAS. We took the preprocedual blood sample from the descending aorta and the postprocedual blood sample at the site of the plaque under distal balloon occlusion. Serum concentrations of IL-6, IL-18, and hsCRP were measured from these samples.

Results: The preprocedual concentrations of IL-6, IL-18, and hsCRP amounted to 2.14 ng/L (interquartile ranges (IQR), 1.09 to 4.51 ng/L), 239 ng/L (IQR, 192 to 319 ng/L), and 0.09 mg/dL (IQR, 0.05 to 0.60 mg/dL). The postprocedual concentrations of IL-6, IL-18, and hsCRP became 3.34 ng/L (IQR, 1.99 to 5.56 ng/L), 250 ng/L (IQR, 181 to 330 ng/L), and 0.09 mg/dL (IQR, 0.03 to 0.46 mg/dL). The local level of IL-6 significantly increased at the site of plaque after CAS (P<0.001). In contrast, local hsCRP slightly decreased (P=0.002) while IL-18 remained unchanged (P=0.355).

Conclusion: The fact that local level of IL-6 increased after CAS relative to the systemic circulation supports our hypothesis that local IL-6 production was present in carotid atherosclerotic plaque. These results also suggested that serum IL-18 and hsCRP levels may be a systemic origin.

IMPLANTED HUMAN NEURAL STEM CELL PROMOTE NEUROGENESIS OF ENDOGENOUS STEM CELLS AFTER FOCAL CEREBRAL ISCHEMIA

Sun Ryu¹, Seung-Hoon Lee¹, Miyeoun Song¹, Seung up Kim², Byung-Woo Yoon¹

¹Department of Neurology, Seoul National University and Neuroscience Research Institute, Clinical Research Institute SNU Hospital, Seoul, South Korea, ²Brain Disease Research Center, Ajou University, Seoul, South Korea

Background and Purpose Transplantation of human neural stem cell (hNSC) have been previously shown to improve functional recovery of ischemic rats. To extend these findings, we examined the endogenous stem cell proliferation in the presence of intraventricular implanted hNSC after focal cerebral ischemia.

Methods Transient focal ischemia was induced by middle cerebral artery occlusion (MCAo) for 60 minutes with a nylon filament in rat. Twenty four hours after MCAo, either 6 × 105 hNSC or phosphate-buffered saline was injected into the lateral ventricle on the ischemic side. All animals received daily injection of bromodeoxyuridine (BrdU; 100mg/kg, i.p.) for 7days after treatment for identification of newly synthesized DNA. On the 14th day after MCAo, we evaluated infarct volume and endogenous neurogenesis histologically. Functional recovery was serially evaluated by the modified Neurological Severity Score after MCAo.

Result At 14 days after transplantation, there was a significant recovery in behavioral performance (p < .05). Cerebral infarcts in hNSC-implanted rats were reduced in size (p < .05). hNSCs were found to survive and preferentially localize to ipsilateral ischemic hemisphere. Cell proliferation and endogenous neurogenesis were assessed with BrdU labeling and immunostaining for cell type specific markers.

Significantly more BrdUrd-labeled endogenous cells were located in the the subventricular zone and hippocampus (p < .05), and more newly generated cells in both region were immunoreactive for Dcx, NeuN and GFAP in hNSC- implanted rats than in vehicle-injected rats.

Conclusion Implantation of human NSCs promoted the proliferation of endogenous neural stem cells, and that the proliferated cells differentiated into more mature neural cells after stroke in rats.

EFFECTS OF LONG-TERM ADMINISTRATION OF HMG-COA REDUCATESE INHIBITOR, ATROVASTATIN, ON STROKE EVENTS AND L-CBF IN SHRSP

Noriko Tanaka, Yasuo Katayama, Toshiya Katsumata, Tatsuo Otori, Yutaka Nishiyama

Division of Neurology, Department of Internal Medicine, Nippon Medical School, Tokyo, Japan

[Background and aims] To determine the effect of long-term administration of a HMG-CoA reductase inhibitor, atorvastatin, on stroke events in stroke-prone spontaneously hypertensive rats (SHRSPs), and its possible neuroprotective effect.

[Methods] Atorvastatin (2 mg/kg, 20 mg/kg) or vehicle was orally administered to 8-week-old SHRSPs for 11 weeks. The survival ratio and stroke incidence were calculated, and plasma lipids and plasma levels of ADMA (asymmetric dimethylarginine, a marker of endothelial dysfunction) were measured after sacrifice. The effect of atorvastatin on local cerebral blood flow (l-CBF) was also determined in 13-week-old SHRSPs after treatment with 20 mg/kg atorvastatin daily for 5 weeks.

[Results] Survival ratios at 19 weeks of age were 15, 30 and 50% in the vehicle, low-dose (2 mg/kg) and high-dose groups (20 mg/kg), respectively (Fig.1). Survival ratio was significantly higher in the high-dose group compared to the vehicle group. Stroke incidence was significantly lower in the high-dose group compared to the vehicle group. ADMA levels were 0.81±0.18 (mean± S.D.), 0.62±0.09 and 0.61±0.06 µmol/L in the vehicle, low-dose and high-dose groups, respectively. Atorvastatin significantly reduced the ADMA levels without affecting plasma lipid levels. L-CBF tended to be higher in the treated group compared to control, but without significance. The l-CBF decreases in SHRSPs as hypertension develops. Statin administration may attenuate the reduction of CBF, arterial endothelial injury and disruption of the blood-brain barrier in SHRSPs. Thus, atorvastatin was determined to provide a protective effect against hypertension-based stroke. The data suggest that increased NO production may take part in the efficacy of the statin for stroke protection. In addition to the lipid-lowering effects of statins, pleiotropic effects may play important roles in stroke prevention.

[Conclusion] Statins may exert protective effects, not only against atherosclerosis-based stroke, but also hypertension-based stroke.

POSTISCHEMIC ADMINISTRATION OF THE CANDESARTAN REDUCES INFARCT SIZE AFTER TRANSIENT CEREBRAL ISCHEMIA IN HYPERTENSIVE RATS

Emi Omura-Matsuoka, Kazuo Kitagawa, Yoshiki Yagita, Tsutomu Sasaki, Yasukazu Terasaki, Naoki Ohyama, Kenichi Todo, Masatsugu Hori

Department of Cardiovascular Medicine, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

Background and aims: In the present study, we investigated the ability of the orally administered angiotensin II type 1 receptor blocker, candesartan, to ameliorate stroke outcome in spontaneously hypertensive rats after focal cerebral ischemia.

Methods: Adult male spontaneously hypertensive rats weighing 250 to 350 g were used in this study. Focal cerebral ischemia was induced by middle cerebral artery occlusion permanently or transiently, lasting for 60 minutes following 48 hours reperfusion. In rats with permanent MCA occlusion, vehicle (n = 8) or candesartan (n = 8 per group) was applied orally by gavage at three different doses (0.1, 1 and 10 mg/kg), three hours after ischemia. In rats with transient MCA occlusion, vehicle (n = 12) or candesartan (n = 12 per group) were administered twice (3 and 24 hours after MCA occlusion). We measured systolic arterial blood pressure in conscious rats with permanent MCA occlusion by the tail-cuff method before ischemia and 3, 6 and 24 hours after ischemia. In rats with transient MCA occlusion, time point of 48 hours after ischemia was added. Evaluation of neurological deficits was performed before ischemia and 3, 6, 24 and 48 hours after ischemia. Twenty-four or forty-eight hours after MCA occlusion, the brain was dissected into coronal 2-mm sections and the brain slices were incubated in saline containing 2% 2,3,5-triphenyltetrazolium chloride (TTC). The ischemic volume was determined by integrating the appropriate area and the section thickness.

Results: 1. Blood Pressure; In both permanent and transient MCAO groups, treatment with 1 and 10 mg/kg candesartan significantly reduced systolic blood pressure as early as 3 hours after administration compared with vehicle, and the reduction was maintained till 21 hours after treatment. In transient MCAO group, the reduction in systolic blood pressure was more pronounced after 24 hours after second administration in a dose-dependent manner. There was no significant difference in blood pressure between candesartan 0.1 mg/kg-treated and vehicle-treated rats. 2. Ischemic Lesions; A. Permanent ischemia; Lesion areas detected with TTC staining were occupied almost of all left hemisphere supplied by the MCA (vehicle, 315.4 ± 56.1 mm3). There was no difference in lesion size among the treatment groups (0.1 mg/kg, 309.3 ± 54.9 mm3; 1 mg/kg, 346.2 ± 48.1 mm3; 10 mg/kg, 322.1 ± 31.5 mm3). B. Transient ischemia; Cerebral cortex and caudoputamen were also affected in transient MCA occlusion, but the extent of ischemic lesion were smaller than that of permanent MCA occlusion group. The total lesion volume was significantly smaller in rats treated with candesartan at dose of 0.1 mg/kg than those treated with vehicle (109.6 ± 86.6 mm3 versus 234.9 ± 59.1 mm3; P < 0.05). There was no difference between candesartan 1 or 10 mg/kg treated-rats and vehicle-treated rats. 3. Neurological Findings; In both permanent and transient MCAO groups, there was no difference among the treatment groups.

Conclusions: Our results demonstrate that the angiotensin II type 1 receptor blocker, candesartan orally applied after transient focal ischemia reduced infarct volume, at doses with no effect on blood pressure.

TISSUE-TYPE PLASMINOGEN ACTIVATOR ENHANCED THE PERMEABILITY OF BLOOD BRAIN BARRIER

Yasuhiro Suzuki¹, Nobuo Nagai², Desire Collen², Kazuo Umemura¹

¹Department of Pharmacology, Hamamatsu University School of Medicine, Hamamatsu, Japan, ²Center for Molecular and Vascular Biology, University of Leuven, Leuven, Belgium

[Background] Although thrombolytic therapy of ischemic stroke with tissue-type plasminogen activator (tPA) improves clinical outcome, it increases a risk of intracerebral hemorrhage, which has been thought to be via plasmin activation. Recently, however, it has been reported that tPA affects the permeability of blood brain barrier (BBB) without plasmin activation, and then that may be result in intracerebral hemorrhage and brain damage.

[Objectives] We investigate whether tPA may affect the permeability of BBB in a middle cerebral artery (MCA) occlusion model in mice.

[Methods] The permanent MCA occlusion was performed by double ligations and cutting the MCA between them. Permeability of BBB was measured by leakage of Evans Blue (EB) to the brain tissue. EB was intravenously injected 2 hours before sacrifice. An intravenous bolus of 10 mg/kg of tPA was given at various time (either 1, 4 or 70 hours) from MCA occlusion. Instead of tPA, same dose of inactivated tPA obtained by incubation with D-Phe-Pro-Arg-choloromethylketone was also administered. To investigate possible mechanism(s) of tPA effects, one or two mg/kg of receptor associated protein (RAP), an antagonist of low-density lipoprotein (LDL) receptor family, was intravenously administered 5 min before injection of tPA.

[Results] tPA given 4 hours after MCA occlusion significantly increased EB extravasation 2 hours after tPA administration (1.0 [<0.01?6.8] µg in median and range; n=8, P<0.05 vs. 0.44 [<0.01?0.84] in controls, n=8), whereas tPA given at 4 hours did not increased EB extravasation at 5 hours after tPA treatment (0.27 [<0.01?0.77] µg; n=6, NS vs. 0.06 [<0.01?0.58] µg in controls; n=6). Furthermore, tPA given 1 hour after MCA occlusion did not alter EB extravasation 2 hours after tPA administration (0.08 [<0.01?6.8] µg; n = 6, NS vs. 0.01 [<0.01?0.66] in controls, n=6), whereas tPA given 70 hours after MCA occlusion significantly increased EB extravasation 2 hours after tPA administration (4.1 [1.2?8.0] µg; n=6, P<0.05 vs. 0.96 [0.03?0.84] in controls, n=8). Administration of inactivated tPA 4 hours after MCA occlusion did not increase the EB extravasation 2 hours after tPA treatment (0.17 [<0.01?0.48]; n=6). Co-treatment with 1 or 2 mg/kg of RAP 4 hours after MCA occlusion significantly suppressed to the EB extravasation 2 hours after tPA treatment in a dose-dependent manner; a median value of 0.88 µg (0.08–2..9) with 1 mg/kg or 0.10 µg (<0.01?0.91) with 2 mg/kg of RAP (P<0.05 in each vs. without RAP treatment, n=7).

[Conclusions] tPA given 4 hours after MCA occlusion transiently enhanced the permeability of BBB 2 hours after the treatment, which reduced to the basal level at 5 hours. The enhancement by tPA treatment was specific beyond 4 hours after MCA occlusion. The increase of BBB permeability by tPA required its protease activity and was through LDL receptor family. Therefore, the temporal increase of BBB permeability after delayed tPA treatment may be related to the intracranial hemorrhage and brain damage.

DETECTION OF MISERY PERFUSION BY TRANSCRANIAL COLOR SONOGRAPHY IN CAROTID OCCLUSIVE DISEASE: COMPARISON WITH POSITRON EMISSION TOMOGRAPHY

Yukio Sugiyama¹, Takuya Hayashi², Kazuyuki Nagatsuka¹, Hiroshi Moriwaki¹, Kotaro Miyashita¹, Susumu Miyamoto³, Kazuo Minematsu¹, Hiroaki Naritomi¹

¹Department of Cerebrovascular Medicine, National Cardiovascular Center, Osaka, Japan, ²Department of Investigative Radiology, National Cardiovascular Center Research Institute, Osaka, Japan, ³Department of Cerebrovascular Surgery, National Cardiovascular Center, Osaka, Japan

TRENDS IN THE INCIDENCE OF STROKE IN YAMAGATA, JAPAN

Rei Kondo, Yasuhiko Matsumori, Daisuke Tsuchiya, Naoki Kato, Yasuaki Kokubo, Shinya Sato, Takamasa Kayama

Department of Neurosurgery, Yamagata University Faculty of Medicine, Yamagata, Japan

The purpose of this study was to clarify the incidence and the features of acute cerebrovascular diseases in Yamagata Prefecture where the population is about 1.22 million, by comparing with those in other prefectures reported in previous studies.

Consecutive 13,641 cases of acute stroke were prospectively registered to Yamagata Society in Treatment for Cerebral Stroke (YSTCS) between January 1, 1998 and December 31, 2002. Cerebral infarction, intracerebral hemorrhage, subarachnoid hemorrhage and transient ischemic attack were observed in 58.4%, 25.8%, 11.4% and 4.4% of the patients, respectively. The frequencies of intracerebral hemorrhage and subarachnoid hemorrhage (37.2%) in Yamagata Prefecture were significantly higher than those reported in other studies (p < 0.01). In regards to the time from the onset to medial contact, the rate of the time within 3 hours and the time from 3 to 6 hours were 21% and 15%, respectively. There was the significant longer time in Yamagata Prefecture than in others (p < 0.01).

This study was one of the largest stroke registration studies in Japan enrolling 13,641 patients. In all patients, computed tomography (CT) and/or magnetic resonance image (MRI) was performed on admission. In addition, diagnosis was especially made by experienced neurologists or neurosurgeons in stroke based on CT and/or MRI findings. Our results could show current status of stroke patients in Yamagata Prefecture.

RISK FACTORS OF CEREBRAL INFARCTION: COMPARISON BETWEEN LACUNAR THROMBOSIS AND ATHEROSCLEROTIC THROMBOSIS

Kiwami Kidana, Mieko Otsuka, Akira Ueki

Department of Neurology, Jichi Medical University Omiya Medical Center, Saitama-City, Saitama, Japan

Purpose: The major risk factor for lacunar thrombosis is hypertension, and those for atherosclerotic thrombosis are diabetes, and hyperlipidemia. Incidence of lacunar thrombosis has been decreased in recent Japan mainly due to treatment of hypertension. On the contrary, rapidly growing incidence of diabetes and/or hyperlipidemia in Japan may contribute the increased incidence of atherosclerotic thrombosis. To test this explanation, we compared risk factors between patients with lacunar thrombosis, atherosclerotic thrombosis, and cardiogenic embolism.

Methods: Fifty nine patients who admitted to our neurological ward from July, 2005 to August, 2006 were recruited. Clinical diagnosis of three groups was made by history of present illness (the mode of onset, clinical course), neurological examination, and cranial CT, MRI, and MRA. Frequencies of hypertension, diabetes, hyperlipidemia, obesity, smoking, dehydration., alcohol consumption were compared between three subgroups.

Results: We picked out 59 patients in all- lacunar thrombosis was 22 patients (37.3%), atherosclerotic thrombosis was 14 patients (23.7%), and cardiogenic embolism was 23 patients(38.9%). The average age of embolic infarction was 72.2±9.4 years old, lacunar was 62.2±8.9 years old, and atherosclerotic thrombosis was 62.6±11years old. The male ratio of lacunar thrombosis was 90.9%, atherosclerotic thrombosis was 64.3%, and cardiogenic embolism was 60.9%. The ratio of diabetes was 50% in lacunar thrombosis, 14.3% in atherosclerotic thrombosis, and 34.8% in cardiogenic embolism. The ratio of hypertension was 86.4% in lacunar thrombosis, 57.1% in atherosclerotic thrombosis, and 52.2% in cardiogenic embolism. The ratio of hyperlipidemia was 45.5% in lacunar thrombosis, 28.6% in atherosclerotic thrombosis, and 21.7% in cardiogenic embolism. The average of BMI was 23.6±3.0 in lacunar thrombosis, 20.7±2.9 in atherosclerotic thrombosis, and 23.0±5.5 in cardiogenic embolism.

Conclusion: Unexpectedly, frequencies of diabetes, hyperlipidemia, and obesity was lower in atherosclerotic thrombosis than lacunar thrombosis. Higher consumption of alcohol and probably concomitant malnutrition might contribute the genesis of atherosclerotic thrombosis.

REGULATION OF FOXO3A TRANSCRIPTION FACTOR BY PHOSPHOINOSITIDE 3-DEPENDENT SIGNALING IN ISCHEMIC NEURONAL DEATH

Gyeong-Moon Kim, Kyung-Sil Jin, Kwang-Yeol Park

Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea

Backgound and aims: FOXO3a, a forkhead family of transcription factor, has been known to regulate apoptosis, cell cycle, cell metabolism, or the activation of genes involved in DNA repair and oxidative stress. Akt deactivates FOXO3a by direct phosphorylation and cytoplasmic retention. Survival factor withdrawal leads to FOXO3a dephosphorylation, nuclear translocation, target gene activation and apoptosis eventually. We sought to evaluate the relationship between FOXO3a activation and its target gene expression in neuronal cell death after oxygen-glucose deprivation (OGD). Methods: Cultured neurons from rat embryonic cortex were subjected to OGD for 120 minutes. At periods of 30 minutes, 2, 6, and 24 hours after reoxygenation, proteins were purified and subcellular fractionization was performed. The levels of Akt, phospho-Akt, Bim and Fas ligand (FasL) expression were measured by Western blot and immunohistochemistry. DNA binding activity of FOXO3a was determined by electromobility gel shift assay (EMSA). We also evaluated the amount of cell death after specific Akt/PI3K inhibitor LY294,002 and evaluate the changes of protein binding of FOXO3a and the expression of Bim and FasL. Results: p-Akt activity was increased in the early phase of reoxygenation period in both cytoplasmic and nuclear extracts. FOXO3a target genes, Bim and FasL, showed increased expression after OGD. EMSA revealed significant elevation of DNA binding activity for FOXO3a after OGD and it peaked at 3–6 hours of reoxygenation. In addition, subcellular localization of FOXO3a by double immunofluorescence staining with neuronal marker (NeuN) revealed nuclear translocation of FOXO3a in morphologically apoptotic neurons. Furthermore, LY294,002 treatment promoted ischemic neuronal death through increased DNA binding of FOXO3a and enhanced expression of FasL. Conclusions: These results suggest that the activation of FOXO3a transcription factor may be mediated, at least in part, by the PI3K-dependent signaling pathway in ischemic neuronal death. Transcriptional regulation of FOXO3a maybe targeted as therapeutic approach to prevent neuronal apoptosis after acute ischemic insult.

OXIDATIVE STRESS INCREASES THE PHOSPHORYLATION OF SER10 HISTONE H3 THROUGH IKAPPAB KINASE ALPHA IN ENDOTHELIAL CELLS AFTER ISCHEMIA AND REPERFUSION

Yun Seon Song, Purnima Narasimhan, Carolina M. Maier, Pak H. Chan

Departments of Neurosurgery, Neurology and Neurological Sciences, and Program in Neurosciences, Stanford University School of Medicine, Stanford, CA, USA

Introduction: Histone modification is one of the key regulatory steps essential for transcriptional activation. Phosphorylation of histone H3 activates the transient expression of immediate-early genes. Stimulation of both the ERK pathway and the stress-activated p38 pathway also results in an increase in phosphorylation of histone (Ser10) during activation of immediate-early gene expression (Strahl and Allis 2000). It has been reported that IKK alpha has histone H3 kinase activity after cytokine exposure. The cytokine-induced phosphorylation of Ser10 by IKK alpha in histone H3 seems to be especially important for the subsequent acetylation of Lys14 by cAMP-response element binding protein for transcriptional activation (Yamamoto et al. 2003). However, the cellular role of IKK alpha associated with oxidative stress in cerebral ischemia is unknown.

Materials and methods: To clarify the role of IKK alpha in oxidative stress, we investigated the IKK alpha-related proteins after transient focal cerebral ischemia (tFCI) and oxygen-glucose deprivation (OGD) of endothelial cells. Samples were taken 0, 1, and 2 hours after tFCI and OGD. Cytosolic and nuclear fractions were prepared from ischemic brain and endothelial cells using ProteoExtract(Calbiochem) to study the nuclear translocation of protein. Changes in protein were investigated with the use of Western blot analysis and immunohistochemical analysis.

Results: We observed nuclear accumulation of IKK alpha 1 hour after 30 minutes of transient focal ischemia using Western blotting. Immunohistochemistry also showed nuclear accumulation of IKK alpha, and showed increased expression and phosphorylation of IKK alpha in endothelial cells after tFCI. To gain further insight into nuclear function of IKK alpha, we examined nuclear changes in histone H3 and its modified H3 after in vitro OGD of endothelial cells. Immediately after OGD of endothelial cells, the phosphorylated Ser10 histone H3 was increased 1.8-fold compared with the control, in accordance with nuclear accumulation of IKK alpha in endothelial cells after OGD. Acetylated (Lys14) and phospho (Ser10) histone H3 also increased 4.8-fold compared with the control. An upstream kinase of IKK alpha, NF-kappaB-inducing kinase (NIK), was also increased up to 3.5-fold in endothelial cytosol after OGD. We found increased expression of pro-inflammatory COX-2 and MMP-9 after OGD of endothelial cells.

Conclusion: Our results show that the histone H3 kinase activity of IKK alpha is strongly induced via NIK after ischemic oxidative stress. Nuclear IKK alpha may play a role in phosphorylation of Ser10, which might be important for the subsequent acetylation of Lys14 in histone H3. It is critical in facilitating the rapid oxidative stress-induced expression of harmful downstream genes, such as COX-2 and MMP-9, which cause neuronal cell death. Blocking IKK alpha activation by oxidative stress is a new potential therapeutic target for clinical application in stroke.

HYPOXIA-INDUCED UPREGULATION OF ENDOTHELIAL SMALL G PROTEIN RHOA AND RHO-KINASE/ROCK2 INHIBITS ENOS EXPRESSION

Hong Guo Jin¹, Hiroshi Yamashita¹, Yoshito Nagano¹, Hiromasa Fukuba¹, Masanori Hiji¹, Toshiho Ohtsuki¹, Tetsuya Takahashi¹, Tatsuo Kohriyama¹, Kozo Kaibuchi², Masayasu Matsumoto¹

¹Department of Clinical Neuroscience and Therapeutics, Hiroshima University Graduate School of Biomedical Sciences, Hiroshima, Japan, ²Department of Cell Pharmacology, Nagoya University Graduate School of Medicine, Showa, Nagoya, Japan

The small G protein RhoA and its downstream effector Rho-kinase/ROCK2 play an important role in regulation of various vasculature cellular functions. Nitric oxide (NO) produced by endothelial NO synthase (eNOS) is an important mediator of vascular homeostasis and cerebral blood flow. Using the human endothelial cell line HUVEC, the present study investigated the role of RhoA and Rho-kinase in endothelial eNOS protein expression under hypoxic conditions as an in vitro model of ischemia. RhoA protein levels in HUVEC were low under normoxic conditions, but were significantly increased after 5 h of hypoxia. Endothelial Rho-kinase expression was not detected until after 3 h of hypoxia; such expression remained significantly increased after 5 h. On the other hand, endothelial eNOS expression was similar after 3 h of hypoxia, but was significantly decreased after 5 h. The hypoxia-induced decrease in eNOS expression was significantly enhanced by expression of the constitutively active form of RhoA and significantly inhibited by suppression of RhoA expression by small interfering RNA. The hypoxia-induced decrease in eNOS expression was significantly inhibited when endogenous Rho-kinase activation was inhibited by Rho-binding domain expression. Furthermore, the hypoxia-induced decrease in eNOS expression was significantly enhanced by expression of the constitutively active form of Rho-kinase. Since expression and activation of RhoA and Rho-kinase inhibit eNOS expression in endothelial cells, attempts to down-regulate RhoA and Rho-kinase by multiple drugs, such as statins or Rho-kinase inhibitors, might provide endothelial and cardiovascular benefits through upregulation of eNOS.

ALPHA-AMINO-3-HYDROXY-5-METHYL-4-ISOXAZOLE PROPIONIC ACID (AMPA) RAPIDLY INDUCES DNA FRAGMENTATION OF THE HIPPOCAMPAL NEURONS IN THE INITIAL STAGE OF NECROSIS

Kyoko Ibaraki, Seiji Yamamoto, Yong Wang, Atsuo Miyakawa, Takashi Sakurai, Susumu Terakawa

Photon Medical Reseach Center, Hamamatsu University School of Medicine, Hamamatsu, Japan

INTRODUCTION: Under the video enhanced contrast-differential interference contrast (VEC-DIC) microscope, the intact cultured neurons show large nuclei containing an amorphous nucleoplasm, and an application of glutamate induces granular structure inside the nucleus within 20 minutes (1). This nuclear change corresponds to DNA fragmentation, indicating that the extracellular stimuli would affect the nuclear DNA in the very early stage of excitotoxicity (2). However, whether alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) could produce the comparable change remains unclear. In the present study, we sought to determine the sole effects of AMPA application on the hippocampal neurons.

METHODS: Dissociated hippocampal cultures were prepared from one-day old Wister rats, and used for the experiment three weeks later. Morphological changes in hippocampal neurons induced by AMPA were observed under the VEC-DIC microscope. To detect the type of cell death, that is, necrosis or apoptosis, hippocampal neurons were stained following AMPA application with propidium iodide (PI) and fluorescein-4-isothiocyanate (FITC) conjugated annexin V (annexin V-FITC), and examined under the fluorescence microscope. DNA fragmentation was assessed using comet assay, a single cell gel electrophoresis assay. To assess the severity of DNA fragmentation, we classified DNA damage from grade 1 (no damage) to 5 (severe damage) by visual scoring.

RESULTS: Under the VEC-DIC microscope, continuous exposure to higher concentration of AMPA (100 µM, 1 mM) dose-dependently showed the nuclear granulation and the cellular swelling within 5 minutes. Although 10 µM-AMPA also induced the cellular swelling, it did not produce the marked nuclear granulation, suggesting that the pathway inducing cellular swelling may be distinct from that inducing nuclear granulation. Non-competitive AMPA receptor antagonist, GYKI-52466 (50 µM), blocked these morphological changes after continuous exposure to 100 µM-AMPA for 1 h, while N-methyl-D-aspartate (NMDA) receptor blocker, MK-801 (20-200 µM), did not do so, indicating that AMPA mediated toxicity independently of NMDA receptor activation. The staining with PI and annexin V-FITC revealed that AMPA (100 µM, 1 mM) induced necrosis but not apoptosis in dose dependent manner. The treatment with AMPA (100 µM, 1 mM) for 1 h significantly increased the percentage of grade 4, severer DNA fragmentation in comet assay.

COMMENTS: In hippocampal neurons, AMPA receptor stimulation alone, independently of NMDA receptor, induces the cellular swelling and the nuclear granulation in the very early stage of necrosis. The granular change in the nucleus observed under the VEC-DIC microscope corresponds to DNA fragmentation. Our results suggest that: 1) even in the process of necrosis, the extracellular stimuli to the AMPA receptor rapidly produce the nuclear signal to induce DNA fragmentation; and 2) the pathway inducing DNA fragmentation may be less sensitive to AMPA receptor activation as compared with that inducing cellular swelling.

Supported by the grants from JSPS (#18591584), and from Japanese COE program.

SELECTIVE VULNERABILITY OF MOTOR NEURON AND ENDOPLASMIC RETICULUM STRESS RESPONSE AGAINST TRANSIENT ISCHEMIA IN SPINAL CORD

Takashi Yamauchi¹, Masahiro Sakurai², Koji Abe³, Yoshiki Sawa⁴

¹Department of Cardiovascular Surgery, Sakurabashi Watanabe Hospital, Osaka, Osaka, Japan, ²Department of Cardiovascular Surgery, National Hospital Organization Sendai Medical Center, Sendai, Miyagi, Japan, ³Department of Neurology, Okayama University Graduate School of Medicine, Okayama, Okayama, Japan, ⁴Department of Cardiovascular Surgery / Osaka Graduate School of Medicine, Suita, Osaka, Japan

Background and aims: Delayed paraplegia after operation of thoracic aorta is considered to be related to vulnerability of motor neuron against ischemia. However, the mechanisms of such vulnerability are not fully understood. Recently, endoplasmic reticulum (ER) stress has been reported to participate in neuronal cell death. In the present study, we investigated the expression of ER stress related molecules and discussed the relationship between the vulnerability and ER stress after transient ischemia in spinal cord.

Methods: Rabbit spinal cord ischemia model was applied with use of a balloon catheter. In this model, spinal motor neuron shows selectively delayed neuronal death whereas other spinal neuron such as inter neurons survive. Histological changes with hematoxylin-eosin staining and immunohistochemical analysis and western blotting for ER stress related molecules, which are glucose regulated protein 78 (GRP78), phosphorylated eukaryotic initiation factor 2 alpha (p-eIF2α) and inositol requiring ER transmembrane RNAse α isoform (IRE1α), were examined.

Results: The immunohistochemical study showed that GRP78, which is supposed to be molecular chaperone, was induced strongly both in inter neuron and motor neuron at early stage of reperfusion, but the prolonged expression was observed in inter neuron. P-eIF2α, which inhibits protein synthesis and modulate ER stress, was induced only in inter neuron at 6 hr of reperfusion. IRE1α, which is supposed to transduce ER stress related death, was more strongly and prolonged expressed in motor neuron.

Conclusions: These results indicate that the vulnerability of motor neuron of spinal cord might be partially attributed to ER stress response for transient ischemia.

SIAH-1 FACILITATES UBIQUITINATION AND DEGRADATION OF FACTOR INHIBITING HIF-1ALPHA (FIH)

Hiromasa Fukuba, Hiroshi Yamashita, Yoshito Nagano, Jin Hong Guo, Masanori Hiji, Toshiho Ohtsuki, Tetsuya Takahashi, Tatsuo Kohriyama, Masayasu Matsumoto

Department of Clinical Neuroscience and Therapeutics, Hiroshima University Graduate School of Biomedical Sciences, Hiroshima, Japan

Hypoxia-inducible factor-1alpha (HIF-1alpha) has a central role in neuroprotective responses to hypoxia in the brain. Hydroxylation of HIF-1alpha by prolyl-hydroxylase PHD and aspargynyl-hydroxylase FIH (factor inhibiting HIF-1alpha) causes proteasomal degradation and transcriptional inhibition of HIF-1alpha. Siah ubiquitin ligases regulate the abundance of PHD via targeting for proteasomal degradation. The present study identified Siah-1 as a binding partner for another hydroxylase FIH. Siah-1 and FIH coimmunoprecipitated each other in mammalian cells. Siah-1 was found both to interact with the JmjC domain of FIH through its substrate-binding domain and to specifically ubiquitinate FIH via its RING finger domain. Siah-1 facilitated FIH degradation via the ubiquitin-proteasome pathway under hypoxic conditions. Such findings suggest that Siah ubiquitin ligases might play a role as up-stream regulators of both hydroxylases for HIF-1alpha, i.e., PHD and FIH, by targeting them for proteasomal degradation, leading to increased HIF-1alpha abundance, and transcriptional activity in hypoxia.

CYCLING PROPAGATION OF SPREADING DEPOLARIZATION AROUND ISCHEMIC FOCI IN THE GYRENCEPHALIC BRAIN OF CATS

Hajime Nakamura¹, Anthony J. Strong², Stefan Vollmar¹, Michael Sue¹, Andrew K. Dunn³, Rudolf Graf¹

¹Max-Planck Institute for Neurological Research, Cologne, Germany, ²Department of Clinical Neurosciences (Neurosurgery), King's College London, London, UK, ³Department of Biomedical Engineering, University of Texas at Austin, Austin, TX, USA

WILD-TYPE BONE MARROW TRANSPLANTATION AGGRAVATES BRAIN INFARCT AFTER MIDDLE CEREBRAL ARTERY OCCLUSION/REPERFUSION IN MMP-9 KNOCK-OUT MICE

Qingmin Guo¹, Mohammed Hossain², Damir Janigro², Marc R. Mayberg³, Shobu Namura¹

¹Department of Anatomy and Neurobiology, Neuroscience Institute, Morehouse School of Medicine, Atlanta, GA, USA, ²Department of Neurological Surgery, Cerebrovascular Research Center, Cleveland Clinic Foundation, Cleveland, OH, USA, ³Seattle Neuroscience Institute, Seattle, WA, USA

[Background and aims] Matrix metalloproteinases (MMPs) are implicated in post-ischemic brain injury. Although MMP-9 activity is demonstrated in brain tissue after focal ischemia, its critical cellular source remains unclear. In this study, we examined MMP-9 activity and brain tissue damage in the following four chimeric mouse groups with combinations of wild-type (WT) and MMP-9 knockout (KO) mice as donor and recipient: WT/WT, WT marrow transplanted into WT recipients; WT/KO, WT marrow transplanted into KO recipients; KO/WT, KO marrow transplanted into WT recipients; KO/KO, KO marrow transplanted into KO recipients.

[Methods] The chimeric animals were generated by irradiation (9 Gy) and bone marrow transplantation (3-5 million cells). Six weeks after bone marrow transplantation, these mice were subjected to 90 min of transient middle cerebral artery occlusion (filamentous MCAO) and reperfusion under general anesthesia with isoflurane. Brain was collected at 24 hr after reperfusion and cut into coronal sections by using a cryostat. Brain infarct was assessed by Nissl staining and expressed as % of contralateral hemispheric volume. Data were expressed as mean +/− SD. Statistical comparisons were performed using Kruskal-Wallis non-parametric test followed by Mann-Whitney test. P < 0.05 was considered as significant. In separate animals, each brain hemisphere was collected at 24 hr after reperfusion, homogenized, and thereafter analyzed for gelatinolytic enzymatic activity by gel zymography.

[Results] The ischemic hemispheres of WT/KO demonstrated comparable MMP-9 activity levels to those in WT/WT. In contrast, KO/WT and KO/KO were devoid of MMP-9 activity. MMP-2 activity levels were similar in all groups. KO/KO demonstrated significantly smaller infarct (41.1+/−12.5, n=5) compared with those in WT/WT (58.9+/−5.9, n = 5) and WT/KO (56.1+/−10.6, n=5). KO/WT demonstrated comparable infarct size (50.4+/−7.8, n=5) to that in WT/KO.

[Conclusions] These findings suggest that bone marrow-derived cells contribute as the main cellular source of MMP-9 to brain infarct formation.

Support Contributed By: National Institute of Neurological Disorders and Stroke Grant NS048532, NS034194, NS046513.

IMPACT OF SELECTIN LIGANDS ACTIVITIES ON LEUKOCYTE RECRUITMENT AND ISCHEMIC DAMAGE AFTER TRANSIENT FOCAL CEREBRAL ISCHEMIA

Hiroharu Kataoka^1,2, Seong Woong Kim², Nikolaus Plesnila²

¹Department of Neurosurgery, Kyoto University, Kyoto, Japan, ²Institute for Surgical Research, Ludwig-Maximilians University, Munich, Germany

Background and aims: Leukocytes contribute to secondary brain damage after transient focal cerebral ischemia, and the mechanism of leukocyte accumulation in the ischemic brain is not fully understood. Because selectins mediate rolling, the first step of leukocyte endothelium interaction (LEI), we investigated the effect of the deficiency of selectin ligand activity on LEI and ischemic damage after transient focal cerebral ischemia in the present study.

Methods: We used alpha (1,3) fucosyltransferase IV and VII double knockout (FucT-IV-/−/FucT-VII-/−) mice, in which terminal fucosylation of selectin ligands is impaired and thus all selectin-dependent rolling disappear. FucT-IV-/−/FucT-VII-/− and the wild type mice were subjected to 60 minutes of transient middle cerebral artery occlusion (MCAO) by an intraluminal filament. Leukocyte-endothelium interactions were assessed by intravital microscopy (IVM) 5 hours after reperfusion. Infarction volume, neuroscore and leukocyte infiltration into the brain parenchyma were examined after 24 hours of MCAO.

Results: IVM disclosed almost no rolling leukocytes in pial venules of FucT-IV-/−/FucT-VII-/−mice (1.3±2.0/100mm × min, P<0.01 vs. control). However, the decrease of rolling lead to neither reduced leukocyte adhesion to endothelium nor reduced leukocyte infiltration in the brain parenchyma. And infarction volume and motor function after 24 hours of MCAO were not different between FucT-IV-/−/FucT-VII-/− mice and wild type mice.

Conclusion: The current results suggest that selectin blockade is not enough to prevent leukocyte recruitment and to ameliorate ischemic damage after transient focal cerebral ischemia.

SINGLE-CELL RESOLUTION MAPPING OF POTASSIUM METABOLISM IN THE HIPPOCAMPAL FORMATION AFTER GLOBAL ISCHEMIA IN RATS

Monika Riek-Burchardt¹, Petra Henrich-Noack², Claudia Pforte², Henning Scheich², Klaus G. Reymann^1,2, Juergen Goldschmidt²

¹FAN-Research Institute for Applied Neurosciences GmbH, Magdeburg, Germany, ²Leibniz Institute for Neurobiology, Magdeburg, Germany

Transient global ischemia results in delayed death of vulnerable neurons in the hippocampal formation. Different methods exist for detecting and mapping ischemia-induced biochemical and morphological alterations, but is has remained impossible so far to map, with single-cell resolution, changes in metabolism and neuronal activity in the post-ischemic hippocampal formation.

We here used a novel technique, thallium autometallography (TlAMG), for mapping patterns of neuronal activity and metabolism in the hippocampal formation of rats at different time points after ischemia (6h, 1d, 2d, 3d, 7d, 14d and 8weeks after transient two-vessel occlusion with hypotension). TlAMG is a histochemical method for non-radiaoactive high-resolution mapping of the uptake and distribution of the monovalent thallium ion (Tl+) in the brain. Tl+ is a well-established K+-probe. Since in neurons K+-uptake strongly depends on Na,K-ATPase activity, which in turn is increased in active versus less active neurons, TlAMG provides information about both, potassium metabolism and neuronal activity.

With TlAMG we find complex patterns of neuronal and glial Tl+ uptake in the post-ischemic hippocampal formation. At time points later than 3d after ischemia we find pyramidal neurons in CA1 which are no longer able to accumulate Tl+ indicating the breakdown of Na,K-ATPase activity and K+-gradients. In CA1 neurons with morphological characteristics of interneurons, in contrast, Tl+-uptake can be high even at very late time points after ischemia. At time points earlier than three days, and before morphological alterations are detectable, there is an increase in Tl+-uptake in CA1 pyramidal neurons. Interestingly in both cases ? the increased uptake before cell death and the early phase of decreased uptake reflecting breakdown of K+-gradients ? pathological alterations in Tl+-uptake appear to affect spatially clustered groups of neighboring pyramidal cells in CA1 to different degrees.

CYSTATIN C EXPRESSION IN ISCHEMIC WHITE MATTER LESIONS

Atsushi Nagai¹, Nobuyuki Umegae², Masaharu Terashima³, Seung U. Kim⁴, Junichi Masuda¹, Shuhei Yamaguchi²

¹Deparment of Laboratory Medicine, Shimane University School of Medicine, Izumo, Shimane, Japan, ²Deparment of Internal Medicine III, Shimane University School of Medicine, Izumo, Shimane, Japan, ³Deparment of Biochemistry and Molecular Medicine, Shimane University School of Medicine, Izumo, Shimane, Japan, ⁴Deparment of Neurology, University of British Columbia, Vancouver, BC, Canada

Background and aims: Cystatin C, an inhibitor of cysteine proteases including cathepsins B, H, L and S, is secreted from various kinds of cells in the body and involved in the regulation of local inflammation, tumor invasion and metastasis. It also exists in cerebrospinal fluid (CSF) 5.5 times higher concentration than in the serum and might function as a major cysteine-protease inhibitor in CSF. There is growing evidence that cystatin C level in CSF is one of the indices to estimate the condition of cystein protease system in various neurological diseases. To study whether cystatin C is involved in the progression of ischemic white matter lesions (WMLs), cystatin C levels in CSF were analyzed in patients with leukoencephalopathy. Furthermore, expression of cystatin C was investigated pathologically in the white matter tissues with severe WMLs as well as in the established neuronal culture system.

Materials and Methods: Cystatin C levels in CSF were measured in 23 cases with severe WMLs of Fazekas' grade 3 (14 with hypertension; W/HT(+) and 9 without hypertension; W/HT(−)) and compared with those obtained in 38 cases with slight WMLs of grade 0-1. In the pathological study, cystatin C immunoreactivity was investigated in the white matter of 3 patients with severe WMLs compared to 3 patients with slight WMLs and 4 patients without WMLs. Expression and secretion of cystatin C were analyzed in the neuronal culture system by using the stimuli of some proteases and cytokines.

Results: Cystatin C levels in CSF in patients with W/HT(+) and W/HT(−) were lower than the patients with grade 0–1 (p < 0.005 and p < 0.0001, respectively). The immunohistochemical study showed that expression of cystatin C was detected in astrocytes, and the number of astrocytes in the severe WMLs was decreased compared to the controls (p < 0.005) and slight WML group (p < 0.005). In neuronal cultures, the stimuli with thrombin, matrix metalloproteinases and interleukin 1beta increased the mRNA expression of cystatin C in human astrocytes and hybrid neuron cultures, but ELISA revealed that only thrombin significantly increased production and secretion of cystatin C in astrocytes. The stimulus with more than 10 U/ml of thrombin induced cell death of astrocytes.

Conclusion: These results suggest that low cystatin C levels in CSF of patients with ischemic WMLs might depend on the decreased number of astrocytes that secrete cystatin C in response to the stimuli of proteases and inflammatory cytokines. Cystatin C secretion from astrocytes may influence the level in CSF and modify the progressing process of WMLs.

MILD EMBOLIC ISCHEMIA-INDUCED NEUROVASCULAR INJURY IS ASSOCIATED WITH ACCUMULATION OF BETA-AMYLOID IN THE RAT CEREBRAL MICROVESSELS

Feng Han¹, Kohji Fukunaga^1,2

¹Department of Pharmacology, Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai, Japan, ²Tohoku University 21st Century COE Program, Comprehensive Research and Education Center for Planning of Drug Development and Clinical Evaluation, Sendai, Japan

Dysfunction of neurovascular system contributes to neurodegeneration and cognitive decline in Alzheimer's disease (AD) patients. In the present study, immunohistochemical studies revealed that immunoreactivity against β-amyloid peptide, which is consider to be causal factor of AD, significantly increased in the microvessels 12 weeks after mild embolic ischemia injury in aged male rats. Likewise, Western blotting analyses using extracts from microvessels demonstrated that β-amyloid peptide level elevated one week after mild ischemic injury in aged rats, and more pronounced increase was found 8 and 12 weeks after ischemia. The β-amyloid peptide deposition in microvessels following embolic ischemia was apparently in part due to increase in peroxynitrite formation after ischemic injury. Indeed, protein tyrosine nitration in microvessels observed 6-24 hours after ischemia, preceded production of β-amyloid peptide in the microvessels. In 8-12 weeks after ischemic injury, β-amyloid peptide deposition also coincided with GSK3β activation, tau hyperphosphorylation and Fas-ligand overexpression. At the same period, CaMKII activity which required for memory formation was also largely attenuated. Taken together, mild ischemia-induced peroxynitrite formation in the microvessels likely triggers the early events of AD such as β-amyloid accumulation, thereby leading to neurodegeneration.

Key words: β-amyloid peptide; Microvascular; Peroxynitrite; Cerebral ischemia; Alzheimer's disease

IN VIVO 19F CSI DETECTION OF FDG AND FGD-6P IN RAT BRAIN

Daniel Coman¹, Basavaraju Sanganhalli¹, David Cheng¹, Timothy McCarthy², Douglas Rothman¹, Fahmeed Hyder¹

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA, ²Pfizer Inc., Groton, CT, USA

IN VIVO MAPPING OF ACUTE TISSUE HYPOXIA AFTER MIDDLE CEREBRAL OCCLUSION USING 18F-FLUOROMISONIDAZOLE AND POSITRON EMISSION TOMOGRAPHY IN THE RAT

Masashi Takasawa^1,2, John S. Beech³, Tim D. Fryer⁴, P. Simon Jones¹, Young T. Hong⁴, Rob Smith⁴, David K. Menon³, Franklin I. Aigbirhio⁴, Jean-Claude Baron¹

¹Stroke Research Group, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK, ²Centre for Brain Repair, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK, ³Division of Anaesthesia, University of Cambridge, Addenbrooke's Hospital, Cambridge, UK, ⁴Wolfson Brain Imaging Centre, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK

Background: 18F-fluoromisonidazole (FMISO) is gaining momentum as a PET marker of tissue hypoxia after stroke in man. We previously showed in rat microPET studies that FMISO is specific for the hypoxic viable tissue (1). However, no previous study has attempted to map the topography of hypoxia after stroke. Invasive PO2 probe and electron paramagnetic resonance studies have reported a gradient of tissue PO2 following acute occlusion, showing that the hypoxia is severe in the ischemic core but less so in peripheral areas (2), (3). However, the actual 3D topography of hypoxia was not provided by these point measurements. In the present study, we aim to map the topography of hypoxia using FMISO microPET in the acute stage of middle cerebral artery occlusion (MCAo).

Methods: As described in detail elsewhere (1), we used Buchan's distal MCA clip model in male spontaneously hypertensive rats (SHRs) (n=3), and also studied normal SHRs (n=5). Around 80MBq 18F-FMISO was iv injected mean 64.7 minutes after the clip was placed, and PET scanning (microPET P4 scanner, Concorde Microsystems, USA) was continued for three hours. The brain was harvested two days later for histology. Corrections for randoms, dead time, background, normalization, attenuation, sensitivity and decay were applied to the data during reconstruction; final image resolution was ~2.3mm FWHM. We created 3D standardized uptake value (SUV) maps using the average of the last ten frames (130-180mins) in each rat, which was coregistered onto normal SHR T2 MRI template using SPM2. The mean SUV distribution map for each rat group was produced by averaging the SUV maps. A MCA cortex ROI was created based on the anatomy of the MRI template and MCA infarct literature.

Results: In all three stroke rats, significant trapping was found over the affected MCA cortex, with continuous increases of PET activity over time until end of scan. The SUV of MCA cortex ROI in stroke rats was 1.15±0.33 (mean ± SD) in affected side as compared to 0.62±0.35 in the unaffected side (p<0.05), and 0.61±0.10 in normal rats (p<0.05).

In the mean SUV map, FMISO trapping was highest in the anterior MCA cortex (proximal relative to clipping site), and gradually decreased towards posterior areas of the MCA territory (SUV: 1.36 and 0.93 in peak and posterior areas respectively). This distribution was identical in each rat. Preliminary data from one rat suggest the acute-stage distribution of FMISO trapping matched the final infarct topography.

Conclusion: The topography of FMISO trapping observed here, with a clear antero-posterior gradient, is consistent with the classical core-penumbra concept and with available point measurements of tPO2 (2),(3). The assumption that the degree of tracer trapping reliably represents tPO2 will need to be addressed. Serial FMISO studies would be of interest to understand the temporal profile of hypoxia, and to relate acute hypoxia to final tissue outcome.

LACTATE IS OXIDIZED BY THE HUMAN BRAIN FOLLOWING TRAUMATIC BRAIN INJURY

Thomas Glenn¹, Gareth Wallis², Daniel Hirt¹, Paul Vespa¹, Josh Dusick¹, George Brooks², David Hovda¹, Neil Martin¹

¹Department of Neurosurgery, UCLA, Los Angeles, CA, USA, ²Department of Integrative Biology, UC Berkeley, Berkeley, CA, USA

VISUALIZATION OF THE PENUMBRA: COMBINATION OF IMAGING TECHNIQUES PROVIDES INSIGHTS INTO THE PATHOPHYSIOLOGY OF INFARCT DEVELEPMENT IN FOCAL ISCHEMIA

Axel Heimann, Kolamal Nageswari, Beat Alessandri, Oliver Kempski

Institute for Neurosurgical Pathophysiology, Johannes Gutenberg-University, Mainz, Germany

THE IMPACT OF ANESTHETICS AND NORMOBARIC HYPEROXIA ON SUSCEPTIBILITY TO CORTICAL SPREADING DEPRESSION

Chiho Kudo, Michael A. Moskowitz, Cenk Ayata

Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, USA

Cortical spreading depression (CSD) is a transient neuronal and glial depolarization and disruption of membrane ionic gradients that propagates slowly across the cerebral cortex. CSD susceptibility can be experimentally assessed by recording the frequency of CSDs evoked upon continuous topical application of KCl (1M). We systematically studied the influence of different anesthetic types and the impact of normobaric hyperoxia on CSD susceptibility in this model. Male Sprague-Dawley rats were intubated and mechanically ventilated, and femoral arteries were cannulated for arterial pressure and blood gas measurements. CSD occurrence and propagation speed were recorded using two intracortical glass micropipettes (350 micron depth). One of 3 different anesthetics was tested in each rat: isoflurane (1%), alpha-chloralose (60-150mg/kg/h with nitrous oxide, 100-180mg/kg/h without nitrous oxide), urethane (0.4–0..65g/kg/h with nitrous oxide, 0.6–0..76g/kg/h without nitrous oxide). Each anesthetic was used either alone (30%O2, 70%N2) or in combination with nitrous oxide (30%O2, 70%N2O) or normobaric hyperoxia (100%O2). We found that highest CSD frequency was observed under urethane anesthesia (p<0.05 vs. all other groups except alpha-chloralose+100%O2; Table). Isoflurane significantly reduced the frequency of CSDs compared to urethane (p<0.05), especially when combined with nitrous oxide (p<0.05 vs. all other groups). Nitrous oxide also suppressed CSDs under urethane (p<0.05) but not alpha-chloralose anesthesia. Normobaric hyperoxia tended to increase CSD frequency under alpha-chloralose (p=0.1) but not isoflurane anesthesia. We were unable to test the impact of nitrous oxide alone on CSD susceptibility, because, nitrous oxide does not provide surgical depth of anesthesia. The propagation speed, amplitude and duration (at half-maximal amplitude) of CSDs did not differ among groups. In summary, CSD susceptibility is modulated by the choice of anesthetic. The mechanism of CSD suppression by isoflurane may involve blockade of gap junctions, whereas the inhibitory effect of nitrous oxide likely involves blockade of NMDA receptors or voltage-gated calcium channels. Our data underscore the importance of proper anesthetic choice for experimental studies of CSD susceptibility (Guedes and Barreto 1992; Verhaegen, Todd et al. 1992; Saito, Graf et al. 1995; Piper and Lambert 1996), and suggest that nitrous oxide may have efficacy as an abortive agent in the clinic for acute migraine, as previously reported (Triner, Bartfield et al. 1999). Furthermore, the lack of CSD suppression by normobaric hyperoxia argues against a pathophysiological role for CSD in cluster headache where normobaric hyperoxia is an effective therapy.

Acknowledgements: This study was supported by NIH Migraine Program Project (1P01NS35611, MAM) and AHS/Merck US Human Health Migraine and Women's Health Research Award (2006, CK).

MILD HYPOTHERMIA (33°C) AMELIORATES BLOOD BRAIN BARRIER DYSFUNCTION AND IMPROVES NEUROLOGICAL OUTCOME FOLLOWING SUBARACHNOID HEMORRHAGE IN RATS

Elisabeth Torok^1,3, Andreas Trinkl², Stefan Zausinger³, Nikolaus Plesnila^1,3, Karsten Scholler³

¹Institute for Surgical Research, University of Munich, Medical Center Groshadern, Munich, Germany, ²Department of Neurology, University of Munich, Medical Center Groshadern, Munich, Germany, ³Department of Neurosurgery, University of Munich, Medical Center Groshadern, Munich, Germany

Objective: Acute brain injury is the primary cause of mortality after SAH. Therapeutic options for the inhibition of the early pathophysiological mechanisms after SAH are quite limited. Hypothermia has a neuroprotective effect in severe head injury and cerebral stroke. Therefore the aim of the current study was to investigate acute and chronic effects of mild hypothermia after SAH.

Methods: Male Sprague-Dawley rats were subjected to SAH induced by an endovascular filament. Animals were randomly assigned to one of the following three groups: 1) normothermia, 2) two hours of hypothermia starting 60 minutes after SAH, and 3) two hours of hypothermia starting 180 minutes after SAH. Cortical laser Doppler flux (rCBF) and ICP were continuously measured up to 6 hours after SAH. Brain water content was measured 24 hours after SAH by the wet-dry method. Neurological deficits were assessed from day 1?7. On Day 7 we quantified the permeability of the BBB by Bovine serum albumin Western Blot, the alteration of the microvascular basal lamina by Collagen-type-IV-Western Blot, and the activity of matrix metalloproteinases (MMP) 9 & 2 by gelatine zymography.

Results: 2 hours of mild hypothermia, induced 60 minutes after SAH improved the ipsilateral rCBF (4h, p<0,05), the neurological outcome (24h, p<0,05), and the integrity of the BBB (7d, p<0,05) and the microvascular basal lamina (7d, p<0,05). Mild hypothermia also reduced the activity of MMP-2. Induction of hypothermia 180 minutes after SAH had no neuroprotective effect. Mild hypothermia had no effect on post-ictal ICP and brain water content.

Conclusion: Two hours of mild hypothermia, induced 60 minutes after SAH have a positive effect on the neurological function and the integrity of the BBB, although it does not affect brain edema formation. Activation of MMP-2 may play a role for the damage of the microvascular basal lamina. Over all hypothermia seems to be an effective strategy for the treatment of early brain injury after SAH.

MICROARRAY ANALYSIS OF BBB-DAMAGED VESSELS EXTRACTED BY MICRODISSECTION

Masaki Ueno¹, Yasuyuki Iwanaga², Masaaki Ueki², Haruhiko Sakamoto¹

¹Department of Pathology and Host Defense, Fac. of Medicine, Kagawa University, Kagawa, Japan, ²Department of Anesthesilogy and Emergency Medicine, Faculty of Medicine, Kagawa University, Kagawa, Japan

Background and aims: We previously reported (1) that the blood-brain barrier (BBB) function was deteriorated in vessels located along hippocampal fissures in 3-month-old stroke-prone spontaneously hypertensive rats (SHRSP). In this study, vascular samples were microdissected from the hippocampi of SHRSP and WKY as a control and the difference in gene expression between the BBB-damaged vessels in SHRSP and vessels without BBB damage in WKY was examined by a microarray.

Methods: Three-month-old SHRSP and WKY rats were used in this study. Microarray analysis was performed in vascular samples extracted by microdissection. In addition, the differences in the expression of some genes and proteins between brain tissues in the two strains of rats were examined using real-time PCR, Western blotting and immunohistochemistry.

Results: The microarray assay revealed that the ratio of osteopontin gene expression in the vascular tissue of the hippocampi of SHRSP to that of WKY was the highest among 8,435 genes examined. Real-time PCR analysis revealed that the gene expression of osteopontin was significantly upregulated in the hippocampal samples of SHRSP compared with that in the hippocampal samples of WKY rats or with that in the cortical samples of SHRSP. Immunohistochemical and Western blot analyses showed that the osteopontin protein expression was seen in perivascular ED1-positive macrophages/microglial cells located around hippocampal fissures and significantly upregulated in the hippocampi of SHRSP compared with that of WKY.

Conclusions: These findings indicate that the expression of osteopontin is upregulated in BBB-damaged vessels in hypertensive SHRSP, suggesting a novel role of osteopontin in BBB function.

LOSS OF CO2-REACTIVITY IN CEREBRAL RESISTANCE VESSELS AFTER SUBARACHNOID HEMORRHAGE IN MICE

Benjamin P. Friedrich², Sergej Feiler², Sebastian Sonanini², Nikolaus Plesnila^1,2

¹Laboratory of Experimental Neurosurgery, Department of Neurosurgery, University of Munich Medical Center, Grosshadern, Munich, Germany, ²Institute for Surgical Research, University of Munich Medical Center, Grosshadern, Munich, Germany

Introduction: Severe impairment of cerebral blood flow (CBF) is one of the major acute sequels of subarachnoid hemorrhage (SAH). Underlying mechanisms are, however, poorly understood. Because CO2 plays a major role in the regulation of CBF we investigated the CO2-reactivity of cerebral resistance vessels in the brain following experimental SAH by intravital microscopy (IVM).

Methods: Male C57/Bl6 mice were anesthetized with medetomidine, fentanyl, and midazolam (i.p.), intubated, and mechanically ventilated. Endtidal pCO2 and arterial blood pressure were measured by microcapnometry and a pressure transducer connected to a femoral line, respectively. SAH was induced by an intraluminal filament. Three or 24 hours later the cerebral microcirculation was visualized by intravital fluorescence microscopy through an opened cranial window using FITC-dextran. In four groups (no surgery, sham surgery, SAH 3h, SAH 24h; n=6 each) vessel diameters of cerebral arterioles were assessed 1 h before and 15 min after ventilation with 10% CO2. In two other groups (n=6 each) animals where successively normo-/hyper-/normo- and finally hypoventilated (15 min each) in order to assess vessel reactivity to hypo- and hypercapnia.

Results: Under baseline conditions arterial diameters were constant in all groups (baseline diameters were normalized to 100%). Ventilation with 10% CO2 (paCO2 = 274 +/− 11 mmHg) induced rapid and significant vasodilation in control and sham animals (164 +/− 16%; p<0.001). After SAH CO2 did not induce any vascular reaction (99 +/− 10% and 107 +/− 4% after 3 and 24 h, respectively). In the hyper- and hypoventilated sham operated animals arterial diameters shrank after hyperventilation (86 +/− 8%; p<0.05; paCO2 = 22 +/− 2 mmHg), returned to normal when animals were normoventilated (102 +/− 12%), and dilated after hypoventilation (134 +/− 18%; p<0.05; paCO2 = 49 +/− 2 mmHg). After SAH vessels did not show any reaction to different paCO2 levels (101 +/− 8% after hyperventilation; 105 +/− 8% after hypoventilation; p>0.05 each). All data are given as means +/− SD.

Discussion: Already within the first few hours following SAH - when no vasospasm of the large arteries is observed - cerebral resistance vessels, i.e. cerebral arterioles, do not react to CO2. Our data therefore suggest, that quite rapidly after SAH cerebral vasoreactivity is dramatically impaired thereby possibly contributing to the early decrease of CBF and high mortality after SAH. The mechanisms of the abrogated CO2-reactivity after SAH, however, are yet unclear and require further investigation.

UNILATERALLY SYMPTOMATIC MOYAMOYA DISEASES IN CHILDREN: LONG-TERM FOLLOW-UP OF 20 PATIENTS WITH PET AND SPECT

Shinji Nagata¹, Koichiro Matsukado¹, Takato Morioka¹, Futoshi Mihara², Tomio Sasaki¹

¹Department of Neurosurgery, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan, ²Departmetn of Radiology, Fukuoka Higashi Medical Center, Fukuoka, Japan

Background and aims: In unilaterally symptomatic moyamoya disease in children, it remains controversial whether bypass surgery should be performed on the asymptomatic side. Here, we verify the validity of our strategy of only performing bypass surgery on the symptomatic side.

Methods: Among 91 pediatric moyamoya patients who underwent bypass surgery in our department between 1980 and 2004, 20 patients with unilateral ischemic symptoms who were followed up for more than 60 months were analyzed in the present study. Initially, we only performed bypass surgery on the symptomatic side for all 20 patients. Among these 20 patients, 5 developed frequent transient ischemic attacks (TIAs) in the initially asymptomatic side and underwent a second bypass surgery on that side (Group A), 8 developed sporadic TIAs and were followed up without surgery (Group B), and 7 did not experience any ischemic symptoms on the asymptomatic side (Group C).

Results: In total, 18 patients progressed well without cerebral infarctions after their last surgery, although some of them showed deterioration of angiographic stenosis and a transient decrease in the regional cerebral blood flow and/or cerebral perfusion reserve. One patient in Group A had an intraventricular hemorrhage 5 years after the second operation and one in Group B had a minor stroke on the initially asymptomatic side.

Conclusion: In unilaterally symptomatic moyamoya disease, we can delay bypass surgery for the asymptomatic side until the development of ischemic symptoms.

LACK OF GLUTAMATE EXCITOTOXICITY IN PRIMARY RAT AND PIGLET CEREBROVASCULAR ENDOTHELIAL CELL CULTURES

Ferenc Domoki^1,2, Bela Kis¹, Tamas Gaspar¹, Ferenc Bari², David W. Busija¹

¹Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston-Salem, NC, USA, ²Department of Physiology, University of Szeged, Szeged, Hungary

Background and aims: Cerebrovascular endothelial cells (CECs) have recently been implicated as targets of excitotoxic injury, contributing to stress-related blood brain barrier (BBB) disruption. Thus, we sought to determine whether primary cultures of CECs are indeed sensitive to a pharmacological dose of L-glutamate (L-glut) or N-methyl-D-aspartate (NMDA).

Methods: Primary CEC cultures were obtained from 2 week old Wistar rat pups and 1 day old piglets. Cells were seeded onto collagen type IV/fibronectin coated dishes, 12 mm circular cover slips, 96-well plates, or 24mm diameter, 3µM pore size TranswellR inserts to study cell morphology, intracellular Ca2+ levels (Fluo4-fluorescence, confocal microscopy), CEC viability (MTS-formasan assay), or transendothelial electrical resistance (TER), respectively. CECs produced confluent cultures and were used within 4–5 days after isolation.

Results: L-glut (Fig A) and NMDA (1-1mM) did not elevate Ca2+ levels in rat CECs in contrast to the profound effect of the Ca2+ ionophore calcimycin. Neither L-glut nor NMDA (1-1mM, 8h) caused visible changes in the neat cobblestone appearance of CECs. They also unaffected rat (Fig B) or piglet CEC viability and did not augment cell death induced with 8h medium withdrawal (Eagle's balanced salt solution (EBSS) supplemented with 1g/l glucose) or 12h oxygen-glucose deprivation (OGD). TER was also unaffected by both L-glut (Fig C) and NMDA application.

Conclusions: Our results indicate that primary rat and piglet CEC cultures are resistant to glutamatergic stimuli and maintain prominent in vitro BBB features even at doses exceeding the physiological range. Our findings are consistent with previous studies in piglets showing intact pial arteriolar responses in vivo after prolonged exposure to L-glut/NMDA, and no effect of NMDA on pial arterioles in vitro. The present findings challenge previous studies claiming CECs are vulnerable to excitotoxicity.

Support: HL-50587, HL-65380, HL-77731, and OTKA T046531.

BENEFIT OF CONTINUOUS CARDIAC OUTPUT MONITORING IN NEUROINTENSIVE CARE OF TRAUMATIC BRAIN INJURED PATIENTS

Tadashi Kaneko¹, Kotaro Kaneda¹, Yasutaka Oda¹, Hiroyasu Koizumi², Takeshi Inoue¹, Ryosuke Tsuruta¹, Shunji Kasaoka¹, Michiyasu Suzuki², Tsuyoshi Maekawa¹

¹Advenced Medical Emergency and Critical Care Center, Yamaguchi University Hospital, Ube, Yamaguchi, Japan, ²Department of Neurosurgery, Yamaguchi University Hospital, Ube, Yamaguchi, Japan

Introduction: It is reported that maintenance of cerebral blood flow (CBF) is essentioal in neurointensive care. The guidelines of traumatic brain injury (TBI) remarked only mean arterial pressure (MAP), intracranial pressure (ICP) and cerebral perfusion pressure (CPP). If CBF should be maintained, more advanced hemodynamic monitoring is required. In this study, we investigated whether hemodynamic parameters could be risk factors in TBI patients.

Methods: Hemodynamic parameters were retrospectively investigated in 13 severe TBI patients who were started pulmonary artery catheter monitoring within 24 hours after admission. Two cases caused cerebral herniation (CH group), in spite of initial ICP were low. Hemodynamic parameters were collected up to 72 hours after admission, and were compared between CH and non-CH groups.

Results: Table shows the average of 72 hours sequential changes of MAP, ICP, CPP, SvO2, CCI and Hb in CH and non-CH group. MAP, ICP, CPP and CCI had the significantly difference between the two groups. TBI treatment guidelines show that CPP, calculated by MAP and ICP, is emphasized on hemodynamic treatment of severe TBI patients. In our study, average CPP was kept more than 60 mmHg defined by TBI guidelines in both, CH group showed significantly lower CCI. These data suggests that both CPP and CCI should be maintained to keep enough CBF in severe TBI patients. Recently, it was reported that hypertonic saline reduced intracranial pressure and improved cardiac index. There is a possibility that improving CPP and CCI without hypertonic saline could be maintain intracranial pressure.

Conclusion: In severe TBI patients, CCI monitoring has essential role in neuro-intensive care. From our retrospective data, improving both CPP and CCI could keep sufficient CBF and prevent cerebral herniation. Eligible CCI is probably more than 3.0 l/min/m2 from our data.

ENHANCEMENT OF CEREBRAL BLOOD FLOW BY SYSTEMIC HYPERTONIC SALINE THERAPY CORRELATES WITH OUTCOME IN PATIENTS WITH POOR-GRADE SPONTANEOUS SUBARACHNOID HEMORRHAGE

Ming-Yuan Tseng, Pippa G. Al-Rawi, Marek Czosnyka, Peter J. Hutchinson, Hugh Richards, John D. Pickard, Peter J. Kirkpatrick

Department of Neurosurgery, Addenbrooke's Hospital, University of Cambridge, Cambridge, UK

Background: Systemic administration of 23.5% hypertonic saline enhances cerebral blood flow (CBF) in patients with poor-grade spontaneous subarachnoid hemorrhage (SAH). Whether the increment in CBF correlated with changes in autoregulation indices (Mx, PRx) or outcome at discharge is unknown.

Methods: Thirty-five poor-grade aSAH patients received intravenous 23.5% hypertonic saline, 2 mL/Kg, and bedside monitoring, including transcranial Doppler (TCD) and intracranial pressure (ICP) monitoring. Seventeen of them underwent xenon-enhanced computerized tomography (XeCT) for quantifying CBF. Outcome was measured by modified Rankin Scale (MRS) at discharge. Data were analyzed using repeated measurement ANOVA and Dunnett's correction. Comparison was made between patients with favorable/unfavorable outcome using multivariate logistic regression.

Results: A maximum rise in blood pressure by 10.27% (p<0.05) and cerebral perfusion pressure (CPP) by 21.24% (p<0.01) at 30 minutes, followed by a maximum fall in ICP by 93.06% (p<0.01) at 60 minutes was observed. Changes in ICP and CPP persisted for a minimum of 180 minutes. TCD showed that the baseline autoregulation (Mx) was impaired on the ipsilateral side of ruptured aneurysm and increments in flow velocities were higher and longer on the contralateral side (48.75 vs. 31.96%, p=0.045; 180 vs. 90 minutes, p<0.05). The autoregulation was briefly impaired on the contralateral side during the infusion. XeCT showed a dose-dependent effect of CBF increments on favorable outcome (odds ratio for MRS 1 to 3, 1.27/ml/100gm tissue*min, p=0.045, Figure)

Conclusions: Bolus systemic hypertonic saline therapy may be used for rescuing cerebral ischemia in patients with poor-grade subarachnoid hemorrhage.

EXPERIENCE WITH ALPHA-CHLORALOSE IN RAT MIDDLE CEREBRAL ARTERY OCCLUSION

Janos Luckl, Joel Greenberg

Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA

Introduction: Alpha-chloralose is widely used as an anesthetic in studies of the cerebro-vasculature because of its minimal depression of autonomic function. However, there have been no controlled studies of focal ischemia in rat under alpha-chloralose anesthesia because of its potential convulsant properties and its long-lasting action. In the current study we demonstrate that alpha-chloralose (AC) has a fairly wide margin of safety and we compare its effect with isoflurane (ISO) on blood flow, infarct size, EEG and peri-infarct depolarization (PID) during transient middle cerebral artery (MCA) occlusion in the rat.

Methods: Artificially ventilated rats were prepared using 1% ISO anesthesia for occlusion of the right MCA. Anesthesia was either switched to AC (60 mg/kg bolus, 30 mg/kg/hr; n=7) or was maintained on ISO (n=8). The MCA was occluded for 90 minutes using the filament technique and physiological parameters were maintained and monitored during ischemia and the reperfusion (30 min.) phase. EEG was monitored from a screw electrode and changes in cerebral blood flow (rCBF) from a laser Doppler probe (5 mm lateral, 2 mm posterior to bregma) placed over the ischemic cortex. The animals of the AC group required 3 hrs post-surgical observation including weaning from the respirator, extubation and monitoring core temperature. Neurological examinations were performed at 24 and 72 h after ischemia at which point all animals were sacrificed and the brains removed for determination of infarct size.

Results: The experimental protocol and treatment with the two different anesthetics did not cause significant alterations in most of the physiological variables. Significantly higher pre - and post-ischemic blood pressure were detected in the AC animals. While the initial drop in flow and the intraischemic rCBF were almost identical in the two groups (AC:33¬+/−9.7%, ISO:36+/−7.6%), the animals under AC anesthesia showed a significantly better reperfusion (AC:154+/−35%, ISO:113+/−38%; p<0.05). Analysis of infarct areas at six stereotaxic levels and subsequent calculation of infarct volumes revealed that the cortical and total ischemic injury were larger under AC (177+/−35.1 and 238+/−40 mm3) than under ISO (137+/−16.2 and 194+/−16.9 mm3) (p<0.05). The reduction in the EEG power was greater in the AC group early into ischemia but by 60 min this difference disappeared. The incidence and total duration of non-convulsive (electric) seizures were similar in both groups, and occurrence of clinically manifested tonic/clinic seizures on the contralateral limbs was also similar. The number and duration of peri-infarct depolarizations were calculated from the “hyperemic” flow changes. The averaged duration (132+/−14 sec.) and frequency (8/90 min.) of these “flow peaks” under isoflurane anesthesia is consistent with electrophysiological data in the literature. The total duration and frequency of PID was significantly higher in the AC animals supporting the hypothesis that isoflurane inhibits not only the formation of spreading depression, but also PID.

Conclusion: We conclude that alpha-chloralose is a safe anesthetic in stroke studies with excellent postoperative recovery and survival. Its features of not depressing cardiovascular and cerebrovascular reflexes and presumably not exerting neuroprotective effects make it an excellent anesthetic for stroke studies in the rat.

PREISCHEMIC NOT POSTISCHEMIC INSULIN AT A SINGLE CLINICALLY RELEVANT DOSE PROTECTS AGAINST ISCHEMIC SPINAL CORD INJURY IN RABBITS

Mishiya Matsumoto, Daisuke Nagamizo, Shunsuke Tsuruta, Hiroaki Matayoshi, Atsuo Yamashita, Takefumi Sakabe

Department of Anesthesiology, Yamaguchi University Graduate School of Medicine, Ube, Yamaguchi, Japan

Background and aims: A recent clinical trial of intensive insulin therapy in critically ill patients has aroused interest in brain protection by insulin in acute ischemic stroke. However, it has not been well determined whether insulin protects against ischemic spinal cord injury. We examined the effects of single dose of insulin that prevents mild hyperglycemia either in the preischemic or postischemic period on the outcome after transient spinal cord ischemia in rabbits.

Methods: We assigned rabbits to 4 groups (n = 8 in each); untreated control (C) group, preischemic insulin (Pre-I) group, preischemic insulin with glucose (GI) group, and postischemic insulin (Post-I) group. Insulin (0.5 IU/kg) was administered 30 min before ischemia in the Pre-I and GI groups, and just after reperfusion in the Post-I group. The blood glucose concentrations in the GI group were maintained comparable to those in the C group. Spinal cord ischemia was produced by occluding the abdominal aorta for 13 min under general anesthesia with isoflurane. Rabbits were neurologically assessed daily for 7days after reperfusion using 5-point score system: 4 = normal motor function, 3 = ability to draw legs under body and hop but not normally, 2 = some lower-extremity function with good antigravity strength but inability to draw legs under body and/or hop, 1 = poor lower-extremity function but weak antigravity movement only, 0 = paraplegic with no lower-extremity function. After final neurological assessment, the lumbar spinal cord was histopathologically evaluated by hematoxylin and eosin staining.

Results: The mean blood glucose concentration before ischemia in the Pre-I group (118 ± 17 mg/dL, mean ± SD) was significantly lower than in the C (159 ± 21 mg/dL), GI (180 ± 30 mg/dL), and Post-I (158 ± 16 mg/dL) groups and those of 30 min after reperfusion in the Pre-I (92 ± 17 mg/dL) and Post-I (100 ± 12 mg/dL) groups were significantly lower than in the C (148 ± 31 mg/dL) and GI (140 ± 22 mg/dL) groups. The mean motor function score in the Pre-I group (3.8 ± 0.7) was significantly better than in the C (1.1 ± 0.4), GI (1.6 ± 1.5), and Post-I (1.6 ± 1.6) groups. The number of normal neurons in the anterior lumber spinal cord in the Pre-I group (60 ± 14) was significantly greater than in the C (18 ± 18), GI (29 ± 26), and Post-I (24 ± 31) groups.

Conclusions: These results suggest that a relatively small dose of preischemic insulin protects against ischemic spinal cord injury. Although insulin has been reported to have a direct neuroprotective effect, the protective effects demonstrated in the present study are mainly due to its glucose-lowering effect.

NITROUS OXIDE ADMINISTRATION INCREASED NEURONAL DAMAGE DURING CEREBRAL ISCHEMIA WITHOUT AFFECTING EXTRACELLULAR GLUTAMATE CONCENTRATION IN GERBILS

Hideki Taninishi, Yoshimasa Takeda, Motomu Kobayashi, Minako Arai, Toshihiro Sasaki, Kiyoshi Morita

Department of Anesthesiology and Resuscitology, Okayama University Medical School, Okayama City, Japan

[Background] It has been conflicting that nitrous oxide affected deteriorative or neuroprotective during cerebral ischemia, no conclusion has been evaluated. Therefore, in the present study, the effect of nitrous oxide on neuronal cell damage during cerebral ischemia was quantitatively evaluated by the ischemic duration causing 50% neuronal damage, and the relationships between ischemic depolarization and histological outcome were determined. Furthermore, the effect of nitrous oxide on dynamic change in extracellular glutamate concentration was also determined. [Materials and methods] Under 1% halothane anesthesia in 30% oxygen, fifty gerbils (weighing 71.0±5.6 g) were randomly assigned to a group receiving 70% nitrous oxide (N2O group, n=25) and a group receiving 70% nitrogen (N2 group, n=25). In the first experiment, forebrain ischemia was performed by occlusion of bilateral common carotid arteries for 3, 5 or 7 minutes (n=6 in each group). During ischemia, direct current potential was recorded from the bilateral hippocampal CA1 regions. Histological evaluation of bilateral hippocampal CA1 regions was performed 5 days after the ischemia. Relationships of neuronal damage with ischemic duration and duration of ischemic depolarization were determined by logistic regression curves. In the second experiment, forebrain ischemia was performed for 5 minutes by the same procedure as that used in the first experiment (n=7 in each group). Right hippocampal extracellular glutamate was measured using microdialysis and high-performance liquid chromatography every minute from 5 minutes before to 30 minutes after the ischemia. During the experimental period, nitrous oxide or nitrogen administration was continued, cerebral and rectal temperatures were maintained at 37.0±0.5 °C. [Results] Onset time of ischemic depolarization was shorter in the N2O group (1.65±0.46 minutes) than in the N2 group (1.92±0.38 minutes, p<0.01). In the case of 5-minute ischemia, the percentage of damaged neurons in the N2O group (81.5±22.4 %) was higher than that in the N2 group (38.9±13.3 %, p<0.01). Logistic regression curves in the N2O group shifted to the direction that neuronal damage was increased, ischemic durations causing 50% neuronal damage in the N2O group and N2 group were 4.43 minutes and 5.29 minutes, respectively (p<0.05). The 95% confidence intervals did not overlap from 3.80 to 5.85 minutes of ischemic duration and from 5.35 to 7.97 minutes of ischemic depolarization. There was no difference in both groups on dynamic change in extracellular glutamate concentration. Peak extracellular glutamate concentrations in the N2O group and N2 group were 61.0±39.6 mcM and 70.0±31.3 mcM, respectively (p=0.64). [Conclusion] Nitrous oxide accelerated neuronal damage during cerebral ischemia. In the N2O group, onset time was shorter and logistic regression curves were shifted to the direction that neuronal damage was increased compared with those in the N2 group. However, dynamic changes in extracellular glutamate concentration had no difference in the both groups. These results suggested that the acceleration of neuronal damage by nitrous oxide was predominantly affected by a factor other than the amount of extracellular glutamate release during ischemic depolarization.

EFFECTS OF FOCAL ISCHEMIA ON ENDOTHELIUM-DEPENDENT RELAXATION AND GENE EXPRESSION IN CEREBROARTERIAL ENDOTHELIAL CELLS: FOCUS ON THE ENDOTHELIN SYSTEM

Lothar Schilling, Benedikt Brosch, Gerold Koch, Nina Weinzierl, Clemens Heiser, Ralf Erber

Division of Neurosurgical Research, Medical Faculty Mannheim, University of Heidelberg, Heidelberg, Germany

Background and aims: In the rat middle cerebral artery (MCA) endothelin (ET)-1 may induce contraction by activating ET(A) receptors located on smooth muscle cells and relaxation activating by ET(B) receptors on the endothelium. This relaxation is mediated by release of nitric oxide. Transient focal brain ischemia in rats has been described to result in a de novo gene expression of the ET(B) receptor in the smooth muscle cells mediating contraction [1]. The aim of the present study was to determine whether there is also an increase of ET(B) receptor-mediated relaxation and an upregulation of the ET(B) receptor-mRNA in the endothelial cells.

Methods: Adult male Sprague-Dawley rats underwent a 2 h transient MCA occlusion (MCAO) and were sacrificed 22 or 46 h later. Naive animals were used for control. Cerebrovascular reactivity to sarafotoxin 6c (S6c, a selective ET(B) receptor agonist) was studied in ring segments of the MCA. Gene expression was studied selectively in endothelial cells which were disrupted in situ using a microperfusion protocol. The total mRNA was sampled and underwent reverse transcription followed by quantitative polymerase chain reaction (qPCR) analysis of the mRNA for Tie-2 and smooth muscle specific actin, for ET(B), ET(A), ET-1, endothelin converting enzyme (ECE), and endothelial nitric oxide synthetase (eNOS) using elongation factor (EF)-1 as house-keeping gene. Analysis was performed using the ΔΔCT methodology.

Results: After ischemia/reperfusion endothelium-dependent relaxation was significantly enhanced. The pD2 values (mean±SD) obtained for S6c are: 9.3±0.4 (24h) and 9.7±0.4 (48h) vs. 8.9±0.3 for controls; the maximum relaxation (% of precontraction, induced by U46619, 3E-7 M) obtained after MCAO are 76.0±20.3% (24h) and 79.7±18.0% (48h) vs. 43.6±20.7% (control). However, S6c did not induce contraction when applied on resting tone, neither in control animals nor after ischemia/reperfusion injury.

Analysis of gene expression in the endothelial cells showed that compared with the intact vessel wall the Tie-2 mRNA was approximately 3.5 fold increased and actin mRNA content was only 6.5%. The ΔΔCT values for the remaining genes are listed in the table (*p<0.05 vs. control).

Conclusion: The present results indicate a significant upreguation of the gene expression of the ET-system in cerebral endothelial cells following ischemia/reperfusion injury. These alterations, underlining the importance of the vascular ET-system in the pathophysiology of ischemia/reperfusion injury are more pronounced after 24 than after 48 h. Furthermore, eNOS expression is transiently upregulated at 24h after MCAO. However, ET(B) receptor-mediated relaxation is significantly enhanced even at 48h after ischemia. Therefore, endothelium-dependent relaxation may involve more than the NO pathway in regulation of vessel tone after ischemia/reperfusion injury.

PERMANENT MCAO IN SHEEP - A NOVEL MODEL OF FOCAL CEREBRAL ISCHEMIA

Johannes Boltze^1,2, Annette Foerschler³, Henryk Barthel⁴, Andrea Barbara Lang¹, Christiane Maria Boltze¹, Osama Sabri⁴, Frank Emmrich^1,2, Uwe Gille^5,6

¹Fraunhofer Institute of Cell Therapy and Immunology, Leipzig, Germany, ²Institute of Clinical Immunology and Transfusion Medicine, University of Leipzig, Leipzig, Germany, ³Department of Neuroradiology, University Clinic of Leipzig, Leipzig, Germany, ⁴Department of Nuclear Medicine, University of Leipzig, Leipzig, Germany, ⁵Institute for Veterinary Anatomy, University of Leipzig, Leipzig, Germany, ⁶VITA34 INc., Leipzig, Germany

POSTISCHEMIC NORADRENERGIC STIMULATION WITH NOREPINEPHRINE REUPTAKE INHIBITORS REDUCES BRAIN INJURY AND FACILITATES RECOVERY FOLLOWING PHOTOTHROMBOTIC CORTICAL ISCHEMIA IN MICE

Sohee Park, Jooyeon Lee, Jinho Jang, Wonsuk Lee

Department of Pharmacology, Pusan National University School of Medicine, Busan, South Korea

This study aimed to investigate whether norepinephrine (NE) reuptake inhibitors reduces brain injury and facilitates the recovery following photothrombotic cortical ischemia in mice. Male ICR mice were anesthetized and systemically administered Rose Bengal. Permanent focal ischemia was induced in the medial frontal cortex and somatosensory cortex by irradiating the skull with a cold light laser. Animals were subacutely treated with NE reuptake inhibitor (desipramine or maprotiline) for 10 days starting on the second day after ischemia induction. The light-irradiation after Rose Bengal injection caused well-demarcated tissue damage in the cerebral cortex, which was located in the sensorimotor cortex and extended to the corpus callosum with no evidence of striatal damage. The infarct size including infarct area and infarct volume was significantly reduced at 14 days after the first dose of desipramine as well as maprotiline, and this infarct-reducing ability was further preserved until 21 days after the first dose. The Evans blue extravasation index of desipramine- and maprotiline-treated group was significantly lower than that of vehicle-treated group. The lower limit (LL) of mean arterial blood pressure for cerebral blood flow autoregulation was shifted toward a normal value by subacute treatments with desipramine and maprotiline. This LL-restoring ability of desipramine and maprotiline was preserved until 21 days after the first dose. Taken together, it is suggested that postischemic noradrenergic stimulation with NE reuptake inhibitors could facilitate the recovery following photothrombotic cortical ischemia, and thereby indicating advantages in the therapeutic strategy for cerebral ischemia. This work was supported by the Korea Science and Engineering Foundation (KOSEF) grant funded by the Korea government (MOST) (No. R01-2004-000-10570-0).

A COMPARISON OF GENE AND PROTEIN REGULATION CHANGES BETWEEN HUMAN AND A RAT MCAO MODEL OF STROKE USING CDNA MICROARRAYS

Nick Mitsios¹, Mohamad Saka¹, Roberta Pennucci¹, Jerzy Krupinski², Coral Sanfeliu³, John Gaffney¹, Pat Kumar¹, Shant Kumar⁴, Matthew Sullivan¹, Mark Slevin¹

¹Department of Biological Sciences, Manchester Metropolitan University, Manchester, Lancs., UK, ²Department of Neurology, University Hospital De Bellvitge, Barcelona, Spain, ³Department of Pharmacology and Pathology, Institute of Biomedical Investigations, Barcelona, Spain, ⁴Department of Pathology, Manchester University and Christie Hospital, Manchester, Lancs., UK

Altered gene expression is an important feature of ischemic cerebral injury and affects proteins of many functional classes. An analysis of gene expression profiles after stroke in humans and a comparison with those in animal models could identify novel gene expression changes and lead to a better understanding of stroke pathophysiology.

We have used microarrays to investigate the dynamic changes in gene expression at various timepoints after MCAO in human and rat brain.

Our results demonstrated a significant difference in the total numbers of genes affected and the time-course of gene expression between the two models. The total number of deregulated genes in the rat model was 335 versus 126 in the human, while out of a total number of 393 overlapping genes, 184 were changed only in the rat and 36 only in the human with a total of 41 genes deregulated in both models. Interestingly, the mean fold changes were much higher in the human stroke study. Some genes have not been previously studied in association with ischemia and their expression patterns were further analyzed by RT-PCR, Western blotting and immunohistochemistry at various time points for the validation of the cDNA array results. These genes included cyclin-dependent kinase 5, p21-activated kinase 1, matrix metalloproteinase 11 and leukaemia inhibitory factor.

These findings confirmed previous studies reporting that parallel screening of gene expression can detect both previously known and novel transcriptional features of cerebral ischemia, leading to a better understanding of this complex process. Moreover, our work highlights the importance of research using human stroke tissue in the search for novel therapeutic agents, which may be useful in the treatment of stroke.

LINKAGE STUDIES FOR FAMILIAL ARTERIOVENOUS MALFORMATION IN THE TAKAYAMA COMMUNITY

Katsunobu Takenaka¹, Sumiko Inoue², Wanyang Liu², Kayoko Inoue², Youhei Mineharu², Hiroyasu Yamakawa³, Masamitsu Abe⁴, Jafar J. Jafar⁵, Roman Herzig⁶, Hidenori Kitajima¹, Katsuhiko Hayashi¹, Akio Koizumi²

¹Department of Neurosurgery, Takayama Red Cross Hospital, Takayama, Gifu, Japan, ²The Department of Health and Environmental Sciences, Kyoto University Graduate School of Medicine, Kyoto, Kyoto, Japan, ³Department of Emergency Medicine, Chunou Kousei Hospital, Seki, Gifu, Japan, ⁴Department of Neurosurgery, Faculty of Medicine, Saga University, Saga, Saga, Japan, ⁵Department of Neurosurgery, New York University Medical Center, New York, NY, USA, ⁶Stroke Center, Department of Neurology, Palacky University Medical School and University Hospital, Olomouc, Czech Republic

Background and aims: Genetic factors for brain arteriovenous malformation (AVM) are unexplored due to the low incidence of familial cases, albeit local and familial clustering 1. To circumvent such difficulties, we conducted a combination of a pedigree-based linkage analysis and a population-based association analysis in the Takayama community2. Subjects: The study was approved by the Ethics Committee of Kyoto University Institutional Review Board. A total of 12 patients with AVM, from 6 families, participated in the linkage analysis. We recruited 26 unrelated cases (17 men and 9 women) and 30 controls (14 men and 16 women) for the association analysis. Two sets of identical twins from Takayama, with discordant phenotypes, also participated. Methods: Linkage Analysis: A genome-wide linkage analysis was performed in 12 patients from 6 unrelated families using the GENEHUNTER program3. Association Analysis: A genome-wide association analysis of 26 cases and 30 controls was performed using a GeneChip 10K mapping array. Statistical Criteria for the Selection of Candidate Genes: In the linkage analysis, we considered p<0.05 to be an evidence of suggestive linkage. In the single SNP association analysis, p<0.0001 was considered to be statistically significant. Results: With a dominant model, the linkage analysis revealed 7 candidate regions (Table 1). With a recessive model, a significant p value was only obtained at 6q25 (p=0.005, data not shown). The highest LOD score of 1.88 was obtained at D6S1581 with a dominant model (Table 1). Association analyses revealed 4 SNPs and 2 haplotypes with a significant association (Table 2). Conclusions: We minimized the population stratification by selecting a study population within the Takayama community, where clusters of sporadic AVM and familial AVM have been reported4. However, the present study failed to identify genetic factors for AVM, although the low statistical power may have resulted in such evidence being missed.

CHINESE HERBAL MEDICINE SERVE AS NEUROPROTANT FOR NEURODEGENERATIVE DISEASES

Hsin-Hsueh Lee¹, Yi-Hsuan Lee², Ling-Ling Yang¹

¹Department of Pharmacognosy, College of Pharmacy, School of Pharmacy, Taipei Medical University., Taipei, Taiwan, ²Department of Physiology, Taipei Medical University, Taipei, Taiwan

Recently the herbal nutrition supplements are very popular applied for chronic disease prevention. In this study, we want to assess the role of the Chinese herbal medicines on neuroproective effects in rat cortical neurons. Controlling or modifying risk factors hypertension, high blood cholesterol, smoking, overweight/obesity, diabetic, and physical inactivity can prevent stroke. The polyphenols isolated from Scutellaria baicalensis Georgi, including baicalin and baicalein were investigated for their neuroprotective effects against glutamate/NMDA(Glu/NMDA) stimulation and glucose deprivation in Sprague-Dawley (SD) rat primary cultured rat brain neurons. Cell death was accessed by lactate dehydrogenase (LDH) release assay for necrosis, and mitochondrial activity was accessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) reduction activity assay. The Ca2+ influx was assayed as described using 45Ca and investigated by confocal microscopy. S. baicalensis exerted neuroprotective effects. Baicalein is the most effective polyphenolic compound tested in preventing neurotoxicity induced by both glutamate and GD, whereas baicalin was only effective in preventing glutamate toxicity. Chinese herbal medicines exerted neuroprotective polyphenolic compound with novel actions on both activity-dependent of [Ca2+]i activity. Further studies are needed not only to elucidate its neuroprotective mechanism, but also to reveal its role in intracellular calcium signaling for cortical functions.

Key words: S. baicalensis, Baicalein, glutamate, glucose-deprivation, neuroprotection

PREVENTION OF SPINAL MOTOR NEURON DEATH BY IGF-1 ASSOCIATING WITH THE SIGNAL TRANSDUCTION SYSTEMS IN SODG93A TRANSGENIC MICE

Hisashi Narai¹, Yasuhiro Manabe¹, Tetsuro Murakami², Makiko Nagai², Tatsushi Kamiya², Koji Abe²

¹National Hospital Organization Okayama Medical Center, Okayama, Japan, ²Department of Neurology, Graduate School of Medicine and Dentistry, Okayama University, Okayama, Japan

COUPLING OF GLYCOLYSIS TO SYNAPTIC TRANSMISSION: ROLE OF SYNAPTIC VESICLE-ASSOCIATED PYRUVATE KINASE IN NEUROTRANSMITTER UPTAKE INTO SYNAPTIC VESICLES

Atsuhiko Ishida^1,5, Yasuko Noda¹, David G. Bole¹, Hisashi Umemori¹, Tetsufumi Ueda^1,2,3,4

¹Molecular and Behavioral Neuroscience Institute, Medical School, University of Michigan, Ann Arbor, MI, USA, ²Ann Arbor, MIDepartments of Biological Chemistry, Medical School, University of Michigan, Ann Arbor, MI, USA, ³Department of Pharmacology, Medical School, University of Michigan, Ann Arbor, MI, USA, ⁴Department of Psychiatry, Medical School, University of Michigan, Ann Arbor, MI, USA, ⁵Department of Biochemistry, Asahikawa Medical Col., Asahikawa, Japan

Glucose metabolism is of vital importance in synaptic transmission. We have focused on the potential role of glycolysis on glutamate accumulation into synaptic vesicles. Previously, we showed that the ATP-generating glycolytic enzymes glyceraldehyde phosphate dehydrogenase and 3-phosphoglycerate kinase, which were found to be bound to synaptic vesicles, provide energy required for glutamate transport into synaptic vesicles (J. Biol. Chem. 278, 5929-5940, 2003). Here we report that ATP generated by the vesicle-associated pyruvate kinase, another ATP-generating glycolytic enzyme, is harnessed to transport glutamate into synaptic vesicles. When synaptic vesicles isolated by sucrose-density gradient centrifugation were incubated with [3H]Glu in the presence of phosphoenol pyruvate (PEP) and ADP, a marked incorporation of [3H]Glu was observed. The uptake was dependent on a low concenrtaration of Cl- in the medium, and was inhibited by FCCP, Bafilomycin, Rose Bengal, and trans-ACPD, but not by cis-ACPD. These results show that it occurs through vesicular glutamate transporters (VGLUT), in the presence of an electrochemical proton gradient generated by V-type H+-ATPase (V-ATPase). The pyruvate kinase activators K+ and Tl+ stimulated both vesicle-associated pyruvate kinase activity and PEP/ADP-dependent glutamate uptake. The highly selective pyruvate kinase inhibitor (Z)-3-chlorophosphoenolpyruvate diminished both vesicle-associated pyruvate kinase activity and PEP/ADP-dependent glutamate uptake. Two-dimensional Western blotting analysis revealed that a unique subunit species of pyruvate kinase, not detected in the cytosol, is associated with synaptic vesicles. These data demonstrate that glutamate uptake is supported by activation of vesicle-associated pyruvate kinase. ATP-trapping experiments using hexokinase have indicted that ATP endogenously produced by the vesicle-associated pyruvate kinase is used more efficiently by V-ATPase than is exogenously added ATP. This suggests that pyruvate kinase is functionally coupled to V-ATPase. In support of these biochemical studies indicating the occurrence of pyruvate kinase in synaptic vesicles, we also provide immunocytochemical evidence that a population of synaptic vesicles is associated with pyruvate kinase. Furthermore, we have found that not only Glu but also other neurotransmitters such as GABA, dopamine, and serotonin were taken up into crude synaptic vesicles in an ADP/PEP-dependent manner. These observations support the notion that glycolysis is coupled to synaptic transmission through its linkage to neurotransmitter accumulation in synaptic vesicles. Supported by NIH Grant 042200 and NIMH Grant 071384 (T.U.)

GLP-1 REDUCES BLOOD-BRAIN GLUCOSE TRANSPORT

Susanne Lerche¹, Birgitte Brock¹, Joergen Rungby¹, Hans E. Boetker¹, Niels Moeller¹, Anders Rodell², Jens J. Holst³, Ole Schmitz¹, Albert Gjedde²

¹Department of Pharmacology, Aarhus University, Aarhus, Denmark, ²PET Center, Aarhus University Hospital, Aarhus, Denmark, ³Department of Medical Physiology, Copenhagen University, Copenhagen, Denmark

Glucagon-like-peptide-1 (GLP-1) has many effects regarding glucose homeostasis. It stimulates insulin secretion in a glucose dependent manor, inhibits glucagon secretion and gastric emptying and has trophic effects on pancreatic β-cells, making it a potent antidiabetic hormone. GLP-1 receptors are broadly represented in the brain. Recent research in rodents has suggested a potentially protective effect of GLP-1 in the brain. The mechanism behind this effect is unknown. The possible role of GLP-1 in the human brain glucose metabolism is not well characterized. To gain more insight into changes of glucose delivery and/or metabolism, which may explain neuroprotective effects, we explored glucose transport and consumption in 10 healthy men in a randomized, double-blinded placebo-controlled cross-over experimental design. The acute effect (independent of insulin) of GLP-1 on glucose uptake in the brain was visualized by Positron Emission Tomography (PET) during a pituitary-pancreatic normoglycaemic (plasma glucose ? 4.5mM) clamp with FDG (18-fluoro-deoxy-glucose) as a tracer of glucose. Insulin infusion rate was 0.12mU/kg/min, growth hormone and glucagon were replaced to baseline levels. Total GLP-1 levels were 258 ± (SEM) 23 vs. 16 ± 0.09 pmol/l (GLP-1 vs. placebo). In total cerebral grey matter, GLP-1 reduced initial glucose clearance, K1(glc) by 29%, 21±2 ml/(100cm3)/min on placebo vs. 15±1 ml/(100cm3)/min on GLP-1 (P= 0.04). In the cerebral cortex, reduction was 29%, 21±2 vs. 15±1 (P=0.04), in white matter 33%, 12±1 vs. 8±0.3 (P=0.03), in thalamus 29% 24±3 vs. 17±1 (P=0.07), in striatum 29%, 21±2 vs. 15±1 (P=0.05), in cerebellar cortex 28%, 25±3 vs. 18±1 (P=0.07) and in brainstem 31%, 16±2 vs. 11 ± 0.6 (P=0.02). The efflux coefficient, k2(glc) showed similar reduction. The regions showed a uniform trend towards reduced cerebral glucose metabolic rate and increased lumped constant. As a result, the intracerebral glucose concentration remained the same in all regions, with and without GLP-1. For the first time, we demonstrate that a hormone involved in postprandial glucose regulation also limits glucose delivery to brain tissue and hence provides evidence that GLP-1 has an important role in brain glucose regulation. As GLP-1 reduces glucose transport across the intact blood brain barrier at normoglycaemia, GLP-1 may also protect the brain by limiting intracerebral glucose fluctuations when plasma glucose is increased.

CEREBRAL BLOOD FLOW IN CIRRHOTIC PATIENTS WITH ACUTE HEPATIC ENCEPHALOPATHY

Peter Iversen¹, Per Borghammer², Michael Sorensen¹, Susanne Keiding¹

¹PET Centre and Department of Medicine V, Aarhus University Hospital, Aarhus, Denmark, ²PET Centre, Aarhus University Hospital, Aarhus, Denmark

The study compared cerebral blood flow (global and regional) in cirrhotic patients (n=17) with an episode of acute hepatic encephalopathy (HE), New Haven criteria grade I (n=6), grade II (n=5), grade III (n=4), or grade IV (n=2) with cirrhotic patients without HE (n=20), and age and sex matched healthy controls (n=15). The aim was to test the hypothesis that the neurological symptoms defined by the New Haven criteria are related to changes in cerebral blood flow.

Global and regional cerebral blood flow rates (CBF) were measured with dynamic 15O-water PET. Parametric maps of CBF were created using measured arterial inputs. The maps were registered to an MRI-template in a common stereotactic space and smoothed with a Gaussian filter to a final resolution of 14 mm (FWHM). Voxel-based statistical analysis was used to test differences of CBF in selected regions of interest in the grey matter tissue.

When compared to controls, both groups of patients with cirrhosis had significantly reduced global perfusion rates. There was no significant difference between the group of patients with HE and without HE (Table 1). After correcting for effects of global decreases by normalization of individual scans to the global grey matter perfusion rate, the regional CBF rates between the patient groups were found significantly reduced in the frontal lobe, occipital lobe, and the posteriomedial part of the parietal lobe, with exception of sensorimotor cortex (M-I and S-I) and supplementary motor area (SMA) (data not shown).

Due to hyperventilation, the patients with HE were in a state of respiratory alkalosis with reduced arterial partial pressure of CO2 (PCO2) (Table 1); in patients with cirrhosis with no HE, PCO2 was normal. However, there was no significant correlation between CBF and the degree of HE, suggesting that hyperventilation had minimal influence on CBF in the HE patient group.

In conclusion, the findings do not support the hypothesis that low cerebral blood flow is exclusive to the complication of cirrhosis.

REGION SPECIFIC EFFECTS OF HYPOXIC PRECONDITIONING ON CEREBRAL BLOOD FLOW IN THE ADULT RAT BRAIN

Susanne Wegener^1,2, Khaled Restom¹, Eric C. Wong³

¹Department of Radiology, University of California San Diego, La Jolla, CA, USA, ²Department of Neurology, Berlin Neuroimaging Center, Berlin, Germany, ³Department of Radiology and Psychiatry, University of California San Diego, La Jolla, CA, USA

Background and aims: Neuroprotection induced by hypoxic preconditioning (HP) results in a more efficient cellular and vascular response to injury. As part of this response, the attenuation of a cerebral blood flow (CBF) decrease has been identified in hypoxic/ischemic injury models in the immature rat brain. Not much is known, however, about the mechanisms of HP in the mature brain. Our goal was to characterize HP induced changes in 1) baseline CBF, 2) vasoreactivity to a 5% CO2 challenge and 3) the CBF response to a global hypoxic stress.

Methods: 14 adult spontaneously breathing Wistar rats were exposed to either room air (sham) or 8% O2 (HP) for 3h. Two days later, rats were anesthetized with 1.5% Isoflurane and CBF maps were acquired using quantitative pulsed arterial spin labeling methods along with anatomical images on a 3T MRI scanner during inhalation of room air, 12% O2 or 5% CO2 in air. Arterial blood gases and blood pressures were obtained under the same anesthetic regimen. Region of interest as well as voxel wise statistics (corrected for multiple comparisons) were applied for CBF analysis.

Results: CBF showed previously described regional variations when rats were breathing room air. Interestingly, discrete shifts in the distribution of CBF in air were observed in HP as compared to sham animals. 12% O2 lead to mild increases in blood flow to subcortical structures, e.g. the hippocampus, thalamus, caudate putamen, and less to the parietal cortex in sham animals (Fig.1). In HP animals, the CBF response to hypoxia was attenuated. Clusters of voxels exhibiting significant differences in hypoxia CBF maps in HP animals were detected in the ventral caudate putamen and insular/parietal cortex. The response to 5% CO2 was a robust CBF increase in most brain regions and was not different between groups.

Discussion: The increase in CBF to vulnerable brain regions such as the hippocampus with hypoxia most likely reflects a compensatory mechanism to redistribute blood flow for maintenance of energy metabolism. The region dependent attenuation of this response in HP suggests that the hypoxic stimulus imposed less stress on certain areas of the brain, possibly due to cellular or vascular adaptation processes. Furthermore, preconditioning induced discrete local changes in baseline CBF might indicate a role of these brain regions in cerebral tolerance formation.

FLOW / METABOLISM COUPLING IN NEURODEGENERATIVE DISORDERS: A PRELIMINARY REPORT

Eric A. Schmidt¹, Pierre Payoux^2,3, Julien Darreon³, Thierry Voisin⁴, Jean Paul Esquerre², Pierre Celsis^2,3

¹Service De Neurochirugie, Hopital Purpan, Toulouse, France, ²Centre TEP, Hopital Purpan, Toulouse, France, ³Unite INSERM 825, Hopital Purpan, Toulouse, France, ⁴Federation De Geriatrie, Casselardit Toulouse, Toulouse, France

CEREBRAL AUTOREGULATION REMAINS INTACT DURING AVM-RESECTION

Carsten Stueer¹, Toshiki Ikeda², Michael Stoffel¹, Carlo Schaller², Bernhard Meyer¹

¹Neurosurgical Department, Munich Tech University, Munich, Germany, ²Neurosurgical Department, Bonn Medical Center, Bonn, Germany

Background and aims: To test the hypothesis that function of cerebral autoregulation in AVM patients undergoing resective surgery differs to the one observed in nonvascular patients.

Methods: Dynamic autoregulation (AR) via correlation coefficient index method (CCI) was assessed before and after resection in n = 12 patients with small/medium-sized AVM and transsylvian dissection in n = 15 patients with nonvascular lesions (= controls). MAP, CBF (thermal dilution method), and routine Intraoperative physiological parameters were continuously measured. All data were pooled according to groups and times, and compared via ANOVA (P<0.05) and post-hoc Wilcoxon correction for multiple comparisons. All data (given below as means ± SD) were collected in the context of a prospective study.

Results: CCI did not show any significant within-group or between-group differences regarding the controls (control pre: 0.06 ± 0.4, post: −0.04 ± 0.5) and AVM-group (pre: 0.04 ± 0.4, post: 0.04 ± 0.5)), respectively.

Conclusions: Dynamic autoregulation results indicate that the status of autoregulation is not altered in patients with uncomplicated small or medium AVM compared to non-AVM conditions. During resective surgery, autoregulation remains intact in AVM patients.

ADENOSINE (ADO) IS NOT INVOLVED IN THE INCREASE IN CBF INDUCED BY HYPERCARBIA

Grzegorz J. Miekisiak, Alexander Beylin, David K. Kung, H. Richard Winn

Department of Neurosurgery, Mount Sinai School of Medicine, New York, NY, USA

Background and Aims: ADO has been suggested as contributing to cerebrovascular dilatation during hypercarbia. Consequently, we evaluated the CBF response to 60 s. of hypercarbia (7%) in WT and A2a Knock-Out (KO) mice in the absence and presence of the non-specific Ado receptor antagonist, theophylline (THEO, as aminophylline= 50mg/kg, ip) as well as the specific A2a antagonist, ZM 241385 (1 mg/kg, iv). Ado induced vasodilatation of cerebral vessels occurs primarily through activation of the A2a receptor.

Methods: Mice (22-30 gm) were intubated, anesthetized, and ventilated. We continuously monitored end-tidal CO2, arterial blood pressure (BP), & CBF (laser Doppler). WT and KO animals were blindly challenged with 7% CO2 (40%O2 balanced with nitrogen) for 60s and CBF measured by laser Doppler.

Results: As indicated in Fig 1, inhalation of 7% CO2 for 60 s. resulted in a 29.5 ± 2.4 % increase in the WT animals and a 34.55 ± 3.3 % in the KO animals. Pre-treatment of each group with nonspecific Ado receptor antagonist, THEO, did not statistically affect the CBF response. Similarly, pre-treatment with ZM 241385, the specific A2a receptor antagonist, did not alter the increase in CBF during hypercarbic challenge (data not shown).

Conclusions: These data are not consistent with the hypothesis that adenosine is involved in hypercarbic cerebral vasodilatation.

[Supported by NS 021076]

HYPERTENSION ? SEQUENCE OF STRUCTURAL CHANGES IN THE ARTERIOLES OF THE BRAIN MICROCIRCULATION

Roland Auer¹, Garnette Sutherland²

¹Department of Pathology, University of Calgary, Calgary, AB, Canada, ²Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada

BACKGROUND: Hypertension is common. The brain microvasculature undergoes sequential structural changes which have not been clearly delineated.

METHODS: 20 sequential autopsies of hypertensives were examined, with heart weights ranging from 450 to 650 grams.

RESULTS: Arterioles initially undergo smooth muscle hyperplasia. Tortuosity ensues, producing glomeruloid vascular formations. Eventually smooth muscle cells die by apoptosis, either slowly or rapidly, with macrophage accumulation. The resultant acellular wall consists totally of collagen which is laid down in reaction to smooth muscle cell loss. Two changes can then occur ? ectasia or stenosis. Ectasia occurs according to P=2T/R and becomes self-sustaining. If collagen deposition is insufficient, eventual rupture occurs as thinning of a microaneurysm leads to a wall rent and a brain hemorrhage. If collagen deposition is exuberant, stenosis and lacunar infarction can occur. This sequence of arteriolar changes in the white matter is associated with imaging alterations in the centrum semiovale. Lacunes occur in spinal cord and hippocampus, not just deep telencephalon. Bleeds can be treated with recombinant factor VIIa because the size of the leaking vessel is so small, unlike berry aneurysms.

ABNORMAL DIAZOXIDE-INDUCED VASODILATION OF MIDDLE CEREBRAL ARTERIES FOLLOWING TRANSIENT FOCAL CEREBRAL ISCHEMIA IN RATS

Prasad Katakam¹, Gabor Lanzser^1,2, Keita Mayanagi¹, James A. Snipes¹, David W. Busija¹

¹Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winsto Salem, NC, USA, ²Institute of Clinical Experimental Research and Human Physiology, Semmelweis University, Budapest, Hungary

Diazoxide, a mitochondrial KATP channel activator, elicits arterial vasodilation by increasing the generation of mitochondria-derived reactive oxygen species that stimulate calcium (Ca2+) sparks and calcium activated K+ (KCa) channels. We evaluated the diazoxide-induced vasodilation in rat middle cerebral arteries (MCA) following transient focal cerebral ischemia. Male Wistar rats were subjected to cerebral ischemia by MCA occlusion (MCAO, 90 minutes) followed by reperfusion (48 hours). Vascular reactivity of isolated MCAs was assessed by videomicroscopy. Diazoxide (100 micromol/L) induced vasodilation was intact in ipsilateral MCAs while it was diminished in contralateral MCAs (% relaxation: 35±6 in contralateral MCAs versus 62±7 in ipsilateral MCAs and 69±6 in MCAs from Wistar rats not exposed to MCAO, p<0.05). Similarly, vasodilation to bradykinin (10 micromol/L) was reduced in contralateral MCAs (27±8%, p<0.05) compared to ipsilateral (68±6%). Nitroprusside (100 micromol/L) induced vasodilation was preserved in all MCAs (95±3 in ipsilateral versus 96±3 in contralateral MCAs, p<0.05). In contrast, pressure induced constriction was decreased in ipsilateral MCAs compared to contralateral (% constriction at 80 mmHg: 4±2 in ipsilateral versus 22±6 in contralateral MCAs, p<0.05). Similarly, endothelin-1 (10 nanomol/L) induced constriction was also diminished in ipsilateral MCAs (23±9%, p<0.05) compared to contalateral MCAs (56±6%). Thus, cerebral ischemia diminishes Diazoxide induced as well as endothelium-dependent vasodilation in contralateral MCAs despite intact myogenic tone and smooth muscle dependent vasodilation. Diminished generation of mitochondria-derived ROS and/or Ca2+ sparks or impaired coupling of Ca2+ sparks to KCa channels may mediate diminished vasodilation to diazoxide.

MULTI-MODAL STUDY OF STIMULUS HETEROGENEITY IN RAT SOMATOSENSORY CORTEX

Peter Herman^1,2, Basavaraju Sanganahalli¹, Fahmeed Hyder^1,3

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA, ²Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Budapest, Hungary, ³Department of Biomedical Engineering, Yale University, New Haven, CT, USA

EFFECT OF CELL-FREE HEMOGLOBIN TRANSFUSION AND 20-HETE SYNTHESIS INHIBITION ON PIAL ARTERIOLAR DIAMETER DURING MIDDLE CEREBRAL ARTERY OCCLUSION

Suyi Cao¹, Herman Kwansa¹, Richard Roman², Raymond Koehler¹

¹Department of Anesthesiology, Johns Hopkins University, Baltimore, MD, USA, ²Department of Physiology, Medical College of Wisconsin, Milwaukee, WI, USA

Background and aims: Hemodilution can increase cerebral blood flow during middle cerebral artery occlusion (MCAO), but neuroprotection is limited by the decrease in oxygen carrying capacity. Exchange transfusion of cell-free hemoglobin polymers (Hb) during MCAO can decrease infarct volume. Without MCAO, exchange transfusion of a 5% albumin solution produces pial arteriolar dilation, whereas exchange transfusion with a Hb solution produces pial arteriolar constriction that is blocked by the 20-HETE synthesis inhibitor HET0016 (Qin et al. J Appl Physiol 2006; 100:336-42). Because 20-HETE synthesis is oxygen dependent, the constriction after Hb transfusion may represent an autoregulatory response to limit over-oxygenation at decreased blood viscosity. We tested whether HET0016 would augment pial arteriolar dilation selectively after Hb exchange transfusion compared to albumin transfusion during MCAO.

Methods: In isoflurane-anesthetized and mechanically ventilated rats, cranial windows were constructed over the MCA-ACA border region for measurement of pial arteriolar diameter. MCAO was produced for 2 h by the intraluminal filament technique. At 15 min of MCAO, the window was superfused with 0.1% ethanol vehicle or 1 uM HET0016 for the remainder of the protocol. At 35-50 min of MCAO, an exchange transfusion with a 5% albumin solution or 5% albumin + 6% Hb solution was performed. The Hb polymer had a nominal molecular weight of ~20 MDa and a P50 of 4 mmHg.

Results: MCAO initially produced 36±10% (±SD) dilation which subsided to 19±6% dilation at 30 min of MCAO in the MCA border region. Subsequent exchange transfusion with the albumin solution resulted in no further change in diameter (22±16% of preischemia at 2h MCAO; n=8). In a group with initial dilation of 36±16%, HET0016 superfusion did not attenuate the loss of dilation at 30 min (22±14%) or the level of dilation after albumin transfusion (27±22% at 2 h MCAO; n=8). In a third group in which initial dilation after MCAO (25±9%) subsided to 12±12% at 30 min with superfusion of vehicle, subsequent exchange transfusion of Hb resulted in a loss of sustained dilation (4±19% of preischemia at 2h MCAO; n=8). In a fourth group with initial dilation of 35±12%, superfusion of HET0016 failed to block this loss of dilation (6±17%; n=9) after Hb transfusion.

Conclusion: Initial vasodilation after MCAO is not fully sustained by either albumin or Hb transfusion. However, loss of vasodilation is greater after Hb transfusion than after albumin transfusion and may limit efficacy of this oxygen carrier during MCAO. Blunted vasodilation after albumin or Hb transfusion is not dependent on 20-HETE synthesis, possibly because low oxygenation during MCAO limits 20-HETE synthesis. (Supported by NIH NS38684)

TIME VARYING MULTIVARIATE SYSTEM IDENTIFICATION OF CEREBRAL AUTOREGULATION

Alexander Brian Rowley^1,2, T. Peng¹, S.J. Payne¹

¹Department of Engineering Science, University of Oxford, Oxford, UK, ²University Computing Laboratory, University of Oxford, Oxford, UK

QUANTIFYING NONLINEAR COUPLING IN CEREBRAL HAEMODYNAMICS USING INFORMATION TRANSFER

Alexander Brian Rowley^1,2, S.J. Payne¹

¹Department of Engineering Science, University of Oxford, Oxford, UK, ²University Computing Laboratory, University of Oxford, Oxford, UK

HETEROGENOUS RESPONSE IN CBF DURING AUTOREGULATION: A NONINVASIVE LASER SPECKLE STUDY IN THE RAT BRAIN CORTEX

Peter Herman, Laszlo Kocsis, Istvan Portoro, Andras Eke

Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Budapest, Hungary

THE INFLUENCE OF HYPOGLYCEMIA ON REGIONAL CEREBRAL BLOOD FLOW IN CASES OF HYPOGLYCEMIC COMA

Yuzo Terakawa¹, Naohiro Tsuyuguchi¹, Katsuyuki Nunomura², Naoaki Murayama², Masahito Fujishige², Akinori Yamamura², Toshio Nakagawa², Kazuo Hashi²

¹Department of Neurosurgery, Osaka City University Graduate School of Medicine, Osaka, Japan, ²Department of Neurosurgery, Shinsapporo Neurosurgical Hospital, Hokkaido, Japan

Background: Recently, diffusion-weighted imaging (DWI) has been reported to show hyperintensity lesions in the acute stage of hypoglycemia. However, changes of regional cerebral blood flow (CBF) in hypoglycemic coma in human patients have received scant attention.

Methods: We report two cases of hypoglycemic coma in which both DWI and single photon emission computed tomography (SPECT) were used.

Results: Case 1. A 74-year-old woman with a history of diabetes mellitus controlled using oral medication was admitted to our hospital after being found unconscious at home. Her family had noticed the patient had been drowsy the day before admission. On admission, the patient had slight hypertension (blood pressure 158/80mmHg), regular heart rate, and irregular respiration but no hypoxia. Neurologically, the patient was in a deep coma (Japan Coma Scale 200, Glasgow Coma Scale 5) with nonreactive pinpoint pupils. Oculocephalic reflex was intact and Babinski reflex was positive bilaterally. The patient showed tetraplegia with decerebral posturing. The DWI obtained immediately after admission revealed bilaterally symmetrical hyperintensity lesions in the middle cerebral peduncles, temporoparietal cortex, corona radiate, and the splenium of the corpus callosum. The blood glucose level on admission was 20 mg/dl. Although the blood glucose was rapidly corrected to within the normal range by intravenous administration of glucose, the level of consciousness remained unchanged. SPECT obtained on the following day revealed diffuse hyperperfusion in both hemispheres, especially in the bilateral temporoparietal regions. Case 2. An 89-year-old woman with medically controlled diabetes mellitus and hypertension was admitted to our hospital for consciousness disturbance. On admission, the patient had slight hypertension (blood pressure 150/70mmHg) but otherwise normal vital signs. Neurological examination on admission showed the patient was in a coma (Japan Coma Scale 100, Glasgow Coma Scale 6) and tetraplegic. Babinski reflex was positive bilaterally. The DWI obtained on admission revealed bilaterally symmetrical hyperintensity lesions in the temporoparietal cortex (greater on the right side), internal capsules, deep white matter at the level of centrum semiovale, and the splenium of the corpus callosum. The laboratory data on admission showed the blood glucose level was 45 mg/dl. Despite the glucose infusion, the level of consciousness did not recover. SPECT obtained two days after admission revealed relative hyperperfusion in the temporoparietal area.

Conclusions: These radiological findings indicate that severe hypoglycemia may lead to an increase in CBF in the acute stage of hypoglycemic coma.

HEMODYNAMIC RESPONSE TO FOREPAW STIMULATION IN ANESTHETIZED MICE SOMATOSENSORY CORTEX

Kazuto Masamoto, Michelle Tasker, Ping Wang, Seong-Gi Kim

Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA

Aim: Genetically altered mice are a potentially powerful tool to elucidate the molecular basis of neuro-vascular coupling mechanism. However, neural activity-induced hemodynamic responses in anesthetized murine cortex remains uncharacterized, which hampers the use of a mice model for functional imaging studies (e.g., fMRI and optical imaging). In the present study, hemodynamic responses to forepaw stimulation in the mice somatosensory cortex were characterized under pentobarbital, ketamine-xylazine, and isoflurane anesthesia, all of which are generally used for repeated animal experiments.

Materials and methods: A total of thirteen male C57B6J mice (23-35 g) were divided into three groups. First group (N = 4) was induced with pentobarbital anesthesia (an initial dose of 90 mg/kg i.p. and supplemental dose of 30-45 mg/kg/h i.p.). Second group (N = 6) was induced with a cocktail of ketamine and xylazine (an initial dose of 60 mg/kg and 10 mg/kg i.m., respectively, and supplementally injected with only ketamine 20 mg/kg i.m. every 20-40 min.). The animals in the third group (N = 3) were ventilated with a mixture of isoflurane (4% for induction and 1.3–1..5% for experiments) and air with supplemental oxygen (a total of 30–35%). Animals in the isoflurane group were intubated and mechanically ventilated, whereas the other two groups breathed spontaneously. After the skin covering the skull was removed, oil was immediately placed to preserve the integrity of the skull. First, intrinsic optical imaging (620-nm wavelength) was performed for the localization of the forepaw area in the primary somatosensory cortex. Then, hemodynamic response was measured with laser-Doppler flowmetry (LDF) on the activation focus induced by electrical forepaw stimulation (rectangular pulses with 0.5-ms pulse width and 0.8-mA current). Stimulation duration was fixed at 10 sec for all experiments and stimulation rate (frequency) was varied (2, 4, 6, 8 and 12 Hz). Preliminary experiments determined a pulse width and current that would not elicit a pain response from the animal.

Results and discussion: All animal groups showed clear localization of forepaw area in 620-nm optical imaging. The clearest contrast of activation area was observed with ketamine-xylazine group. Under pentobarbital anesthesia, the hemodynamic response increased with an increase in stimulation frequency. The peak of 10 ± 12% (Mean ± SD, N = 4) signal change in LDF was observed at stimulation with 12 Hz. Under ketamine-xylazine anesthesia, the highest signal change (13 ± 4% at peak) was observed at stimulation with 6 Hz. These results indicate that the optimum frequency differs depending on the anesthesia used, which was consistent with our previous results obtained in rats (Masamoto et al., 2006). In isoflurane-anesthetized group, however, no significant hemodynamic response was observed for any stimulation frequency. This result indicates that the isoflurane may not be a good agent for functional imaging studies with mice, which was not the case in rats anesthetized with isoflurane (Masamoto et al., 2006). In conclusion, mice anesthetized with pentobarbital or ketamine-xylazine can be used for repeated functional imaging studies, which make possible to probe the mechanism of neurovascular coupling with a specific molecular basis.

FUNCTIONAL UNCOUPLING OF PERIVASCULAR NITRIC OXIDE PRODUCTION FOLLOWING NEONATAL HYPOXIA/ISCHEMIA

Thomas Kent¹, Regino Perez-Polo², Roderic Fabian¹

¹Department of Neurology, Baylor College of Medicine / Michael E Debakey VAMC, Houston, Texas, USA, ²Department of Human Biochemistry and Genetics, University of Texas Medical Branch, Galveston, Texas, USA

Background and Aims: Compromise of cerebral vascular function due to a loss of vascular reactivity may be a cause of cerebral injury after hypoxia-ischemia. A phenomenon that has gained increasing attention as a possible determinant of blood flow and a cause of vascular injury under pathological conditions is the uncoupling of endothelial cell nitric oxide synthase (eNOS, NOS3), which reduces NO production and results in the production of superoxide anion (O2-) by the enzyme. There have been few direct assessments of this phenomenon in the CNS, although eNOS coupling appears to be developmentally regulated in the neonate in other vascular beds. Using a histofluorescence method, we characterized acute changes in cerebral blood flow, vascular O2- anion, and nitric oxide (NO) in neonatal rat cerebral hypoxic/ischemic injury and their response to pharmacological manipulation and the eNOS substrate, tetrahydrobiopterin (BH4).

Methods: Neonatal P7 rats were subjected to cerebral hypoxia ischemia by unilateral carotid ligation followed by hypoxia with hyperoxic resuscitation. Laser Doppler flowmetry was used to monitor CBF, and cerebral vascular wall NO and O2- were determined using i.p.-injected fluorescent dyes (DAF-2AC and DHE respectively) and stimulation with acetylcholine, assessed with fluorescence microscopy. The effect of the NADPH inhibitor apocyanin and the NOS inhibitor, L-NAME, injected i.p. at the onset of resuscitation on fluorescence patterns, and the functional effect of tetrahydrobiopterin supplementation were determined.

Results: We found a progressive decline in perivascular NO production and an increase in perivascular O2- in the hours following hyperoxic resuscitation (Figure: ratio between ischemic/contralateral hemisphere for DAF-2AC and DHE fluorescence; p=0.01 at 24 hours). Reperfusion CBF declined by up to 30% in the first 3 hours relative to baseline. The ratio between perivascular O2- and NO fluorescence was restored by both apocyanin and L-NAME. Administration of tetrahydrobiopterin increased vascular wall NO, reduced vascular wall superoxide anion levels (Figure), and normalized CBF.

Conclusions: These results demonstrate vascular dysfunction follows hyperoxic resuscitation after neonatal hypoxic/ischemic injury manifested by reduced CBF, a reduction in perivascular production of NO and an increase in O2-. Inhibitor studies implicate both NADPH oxidase and NOS and indicate these enzymes functionally disrupt vascular function. That tetrahydrobiopterin normalized these abnormalities suggests a possible role of substrate oxidation leading to uncoupling of eNOS due to tetrahydobiopterin deficiency with resultant implications for propagation of injury.

THE SPHINGOSINE-1-PHOSPHATE ANALOGUE FTY720 IS NEUROPROTECTIVE IN A RODENT MODEL OF BRAIN ISCHEMIA

Hyung-Hwan Kim², Hwa Kyoung Shin¹, Nicolas Blondeau³, James Liao², Christian Waeber¹

¹Department of Radiology, Massachusetts General Hospital, Charlestown, MA, USA, ²Department of Vascular Medicine Research, Brigham & Women's Hospital, Cambridge, MA, USA, ³I.P.M.C., UMR 6097, C.N.R.S, U.N.S.A., Valbonne, France

Sphingosine-1-phosphate (S1P) is a lipid mediator exerting pleitropic effects on a wide variety of cell types, including glial cells, neurons and endothelial cells. In particular, S1P is known to be a survival factor for various cell types and is a key player in protective mechanisms against hypoxia- or ischemia-mediated insults. FTY720 has shown significant protective effects after controlled bilateral kidney ischemia, and attenuates ischemia-reperfusion injury in both normal and cirrhotic livers by activation of cell survival Akt signaling. The purpose of this study was to evaluate the effect of FTY720 administration on in vivo and in vitro models of brain ischemia. Halothane-anesthetized C57Bl/6 mice (20-25 g) underwent 90-min focal cerebral ischemia induced with an 8.0 nylon monofilament coated with a silicone resin/hardener mixture, introduced into the left internal carotid artery up to the anterior cerebral artery. Regional cerebral blood flow was measured using laser-Doppler-flowmetry. Mice treated with 3 mg/kg FTY720 orally once a day for 3 days starting after reperfusion showed a significantly reduced infarct 3 days after reperfusion, as visualized by TTC staining (108 ± 14 vs. 68 ± 8 mm3 for vehicle-treated mice; p=0.04), and significantly attenuated neurological deficit (5-point scale) 1, 2 and 3 days after reperfusion (Q1, median, Q3: 1, 1, 1.5 vs. 2, 2, 3 on day 3; p=0.02). Intravenous 0.5 mg/kg FTY720 did not affect systemic blood pressure and CBF (assessed with laser speckle flowmetry) for up to 60 minutes. We then tested whether the effects of FTY720 on infarct size and neurological deficit were due to increased cerebral blood flow during ischemia using noninvasive laser speckle flowmetry. FTY720 (3 mg/kg) did not alter the CBF deficit when administered orally to isoflurane-anesthetized mice 1h before distal middle cerebral artery occlusion (dMCAO): the area of severely ischemic cortex (? 20% residual CBF) was 3.7±2.0mm2 in FTY-treated mice, compared to 4.2±0.9 mm2 in the saline-treated group. Because S1P is a potent anti-apoptotic factor, we then examined the effects of FTY720 in vitro, using a oxygen-glucose deprivation (OGD) cell death model. Mouse primary cortical neurons were exposed to 2-hr OGD/24-hr reoxygenation, decreasing neuron viability by more than 40% with respect to basal cell death (as assessed using Hoechst staining). 3 microM S1P and 100 nM FTY (added at the time of re-oxygenation) restored viability to control levels. Similarly, in rat brain primary endothelial cells subjected to 4-hr OGD/24hr recovery, 3 microM S1P and 30 nM FTY720 increased cell viability to levels observed in control cells. The presence of an Akt inhibitor (20 microM 1L-6-Hydroxymethyl-chiro-inositol 2-(R)-2-O-methyl-3-O-octadecylcarbonate) or NOS inhibitor (100 microM L-NAME) reversed the protective effects of 3 microM S1P on rat brain endothelial cells following 4hr OGD/24hr re-oxygenation. Taken together, these results suggest that the S1P receptor/Akt/NOS pathway may play an important role in neuroprotection following brain ischemia.

NEUROPROTECTIVE EFFECT OF THE ROOTS POLYGONUM CUSPIDATUM ON TRANSIENT FOCAL CEREBRAL ISCHEMIA IN RATS

Youngmin Bu¹, Jinmo Kim¹, Mi-Yeon Kim², Yun Tai Kim², Jinhee Jung², Young Mi Park³, Zhen Hua Jin³, Kiok Sung³, Hoyoung Choi², Hocheol Kim^2,3

¹Department of Herbal Pharmacology, College of Pharmacy, Woosuk University, Chonbuk, South Korea, ²Korea Institute of Science and Technology of Eastern Medicine (KISTEM), NeuMed Co. Ltd., Seoul, South Korea, ³Department of Herbal Pharmacology, College of Oriental Medicine, KyungHee University, Seoul, South Korea

Objectives: The roots of Polygonum cuspidatum has been widely used in traditional Korean medicine for circulatory diseases. It contains significant amount of phenols, such as resveratrol and is known to have various biological activities. The purpose of the present study is to development of neuroprotective antioxidant agents. For the purpose, we investigated the neuroprotective effect of antioxidant herb, the root of Polygonum cuspidatum on transient focal cerebral ischemia in rat.

Methods: The roots of Polygonum cuspidatum were extracted by 85% MeOH (PC). PC was fractionated by hexane (PCH), ethyl acetate (PCE), butanol (PCB), and water (PCW). Radical scavenging effects were investigated using DPPH assay and MDA assay in brain homogenates. Neuroprotective effect was invesigated using transient focal cerebral ischemia rat model (2 h of ischemia, 22 h of reperfusion). Sensory motor function was investigated by rotarod test, balance beam test at 24 h after ischemia. The brain slices were stained by 2 % 2, 3, 5-triphenyltetrazolium chloride and the infarct volume was measured by graphic analyzer at 24 h after ischemia.

Results: PC showed dose-dependent radical scavenging effects on DPPH and MDA assay. PCE was most effective fraction among fractions (Figure). Oral administration of PC 300 mg/kg reduced brain infarct by 30.8% and improved sensory motor function. PCE was aloso most effective fraction on focal cerebral ischemia rat model. PCE improved the rotarod test by 33%, the balance beam test by 32% and reduced brain infarct volume by 35.8% compared with vehicle treated group. Quantative study of phenol contents in fractions showed that PCE contained phenols more than other fractions.

Conclusion: Polygonum cuspidatum could be a candidate herb for the development of neuroprotective agents. According to the active guided fractionation and HPLC analysis, polyphenols like resveratrol could be the major compounds of neuroprotection.

Grant support: This work was supported by grant from the Brain Korea 21 Project (Ministry of Education, Korea), and by a grant (PF002102-00) from the Plant Diversity Research Center of the 21st Frontier Research Program funded by the Korean Ministry of Science and Technology.

ALTERED EXPRESSIONS OF IL-1BETA AND IL-1RA BY ESTROGEN IN ORGANOTYPIC HIPPOCAMPAL SLICES EXPOSED TO OXYGEN-GLUCOSE DEPRIVATION

Ji-Seung Choi, Soo-Jeong Kim, Kyoung-Eun Lee, Eun-Mi Park

Department of Pharmacology, Ewha Medical Research Institute, College of Medicine, Ewha Womans University, Seoul, South Korea

Background and aims : Anti-inflammatory mechanisms have been implicated in beneficial effects of estrogen. However, it is not clearly defined how estrogen regulates mediators related to inflammation. The balance of IL-1β and IL-1 receptor antagonist (IL-1ra) plays a role in the development of inflammation. This study was performed to evaluate whether estrogen affects the expression of IL-1β and IL-1ra in ischemic hippocampus.

Methods : Hippocamapal slices obtained from 10 days old rats were exposed to oxygen-glucose deprivation (OGD) for 30 min and then were reperfused for 72 h. 17-β estradiol (E2, 1 nM) was treated for 7 days before OGD. CA1 neuronal death was quantified by propidium iodide (PI) staining after reperfusion. Expressions of IL-1β and IL-1ra in slices were studied by real-time PCR and immunofluorescence.

Results : PI intensity of CA1 in slices treated with E2 was significantly reduced at 24 h and 72 h after reperfusion (p<0.05 vs. no treatment, fig. 1). E2 also reduced mRNA expression of IL-1β at 2h and 24h after reperfusion (p<0.05, fig.2). Fewer IL-1β positive cells were detected in E2 pretreated slices compared to no treatment. However, mRNA expression of IL-1ra was markedly enhanced by E2 up to 72h after reperfusion (p<0.05 vs. no treatment, fig.3).

Conclusions : These data demonstrate not only suppression of IL-1β but also enhancement of IL-1ra by estrogen pretreatment in ischemic hippocampus, suggesting that modulation of the balance between IL-1β and IL-1ra may be associated with anti-inflammatory mechanisms of estrogen, in part.

“This work was supported by the Korea Research Foundation Grant funded by the Korean Government(MOEHRD)” (KRF-2005-204-E00075).

ISOFLURANE PRECONDITIONING NEUROPROTECTION IN EXPERIMENTAL FOCAL STROKE IS GENDER-SPECIFIC IN YOUNG AND MIDDLE-AGED MICE

Hideto Kitano¹, Lan Wang², Vanessa R. Anderson², Jennifer M. Young², Ann E. Meinke², Patricia D. Hurn², Stephanie J. Murphy²

¹Department of General Internal Medicine, Hyogo College of Medicine, Hyogo, Japan, ²Department of Anesthesiology and Peri-Operative Medicine, Oregon Health and Science University, Portland, OR, USA

Background and Aims: We have previously shown that neuroprotection due to isoflurane preconditioning (IsoPC) 24 hours before experimental focal cerebral ischemia is specific to young males. However, it is unknown if males and females respond differently to IsoPC in experimental stroke at different ages. We evaluated IsoPC in male and female ischemic brain in young (8 to 14 weeks old) and middle-aged (35 to 40 weeks old) mice and determined if gender-and age-specific responses are linked to differences in Akt activation.

Methods: Young and middle-aged male and female C57BL/6 mice were preconditioned for 4 hours with air (sham preconditioning, Sham PC) or 1.0% IsoPC. Mice subsequently underwent either 2 hours of reversible middle cerebral artery occlusion or were used for Akt determinations 24 hours after preconditioning. For infarction volume analysis, brains were collected at 22 hours reperfusion. Cortical and striatal infarct volumes (% contralateral structure) were determined by digital image analysis of 2 mm thick coronal brain slices stained with 2,3,5-triphenyltetrazolium chloride. For Akt analysis, cortices and striata were subdissected from the left hemisphere of preconditioned brains and pooled together from 3 mice (n=4 pooled samples per group) for western blot analysis. Akt activation was expressed as the ratio of normalized phosphorylated Akt (p-Akt) to total Akt (t-Akt).

Results: In contrast to corresponding Sham PC mice, cortical and striatal infarct volumes following IsoPC were decreased in young and middle-aged males; no significant differences in middle-aged females; and increased in young females. As compared to corresponding Sham PC mice, cortical and striatal p-Akt/t-Akt was increased by IsoPC only in young males, with no changes in p-Akt/t-Akt by IsoPC in middle-aged males and in females irrespective of age (Table).

Conclusions: Regardless of age, IsoPC neuroprotection in experimental stroke is male-specific, with the neuroprotective benefits of IsoPC being lost in females. Gender and age can alter cortical and striatal Akt activation in response to IsoPC. However, further study is needed to determine other possible divergent mechanisms underlying the sex-and age-specific responses to IsoPC in ischemic stroke.

NEUROPROTECTIVE AND ANTIOXIDANT EFFECTS OF TFM-007, A TAIWAN FOLK MEDICINE

Hsin-Hsueh Lee, Ling-Ling Yang

Department of Pharmacognosy, College of Pharmacy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan

Oxidative stress is implicated in neuronal apoptosis that occurs in physiological settings and in neurodegenerative disorders. Taiwan is located in subtropical area and abundant of natural resources and ethnobotanies. In this study, two hundred of the common anti-inflammatory ethnobotanies of ethanol extracts were detected. From anti-oxidantive activities survey, we found the ethanol extract of TFM-007 is one of the most potential bio-resource. The dried flower of TFM-007 has been traditionally used in oriental medicine for the anti-tussive and anti-inflammatory. In the present study, we investigated the effects of free radical scavenging and anti-oxidative activities of TFM-007. Its antioxidant activities were evaluated by cell-free bioassays. TFM-007 with IC50=65 µg/ml) was shown to inhibit lipid-peroxidation initiated by Fe2+/ascorbic acid in rat brain, liver, heart and kidney homogenates. In addition, TFM-007 (IC50=10 µg/ml) on FRAP and scavenge DPPH radicals ability. At the same time, the neuroprotective activity will be displayed on different insults in Wistar cortical neurons. Furthermore, the results showed TFM-007 at micro morality concentration attenuated the arachidonic acid and glutamate-induced neurotoxicity. This is the first demonstration of neuroprotective and anti-oxidative effects of TFM-007. Although complex mechanisms may be involved in the neuroprotective actions, TFM-007 may be useful for the management of neurodegenerative disorders associated with excitotoxicity and oxidative stress and the effects with anti-apoptosis and survival related proteins will further investigated.

Key words: Ethnobotany, TFM-007, arachidonic acid, glutamate, oxidative stress, neuroprotection

GELDANAMYCIN PROTECTS RAT BRAIN THROUGH OVEREXPRESSION OF HSP70 AND REDUCING BRAIN EDEMA AFTER CEREBRAL FOCAL ISCHEMIA

Hyung-Min Kwon¹, Yong-Ju Kim², Seung-Hoon Lee², Byung-Woo Yoon²

¹Department of Neurology, Seoul National University Boramae Hospital, Seoul, South Korea, ²Neuroscience Research Institute, SNUMRC and Clinical Research Institute, Seoul National University Hospital, Seoul, South Korea

Background and aims: Geldanamycin (GA), a benzoquinone ansamycin, binds HSP90, releases heat shock factor 1 and induces heat shock proteins (HSPs). HSP70, a major molecular chaperone, protects the brain against ischemic injury through inhibition of apoptotic pathways in vivo and reduced matrix metalloproteinase 9 (MMP-9) after ischemia in vitro. We hypothesized that GA would protect brain from focal ischemia via induction of HSP70 and MMP suppression in vivo.

Methods: GA or vehicle was injected into the lateral cerebral ventricles of adult male Sprague-Dawley rats using the stereotatic frame 24 hour before ischemia. Rats were subjected to 2-hour middle cerebral artery occlusions using the suture technique followed by 22-hour reperfusions. One day after ischemia, we evaluated infarction volume, brain swelling, behavioral scores, and immunohistochemistry.

Results: Western blots showed that GA at 2 µg/kg induced HSP70 by 24 hour following administration. GA decreased infarct volumes (219.8 ± 42.8 mm3 vs 63.5 ± 23.5 mm3)and brain edema (20.4 ± 6.3% vs 10.9 ± 3.8%), and improved behavioral outcomes (P < 0.05). Immunohistochemistry showed that GA induced HSP70 and decreased MMP-9.

Conclusions: GA protects brain from focal ischemia and reduces edema. This may be due, at least in part, to GA overexpression of HSP70.

ASIATIC ACID IS NEUROPROTECTIVE IN A MOUSE MODEL OF STROKE

Rajanikant Krishnamurthy, Daniel Zemke, Arshad Majid

Department of Neurology and Ophthalmology, Michigan State University, East Lansing, MI, USA

Background and aims - Asiatic acid is a plant-derived compound with anti-inflammatory and anti-oxidant properties that is currently used as a wound healing agent. Recent studies have shown that asiatic acid protects cultured neurons against both glutamate and beta-amyloid induced toxicity by reducing oxidative stress. Oxidative stress is believed to play an important role in neuronal damage associated with stroke, and many anti-oxidant drugs have been shown to be neuroprotective in models of stroke. Furthermore, some derivatives of asiatic acid have been found to have anti-apoptotic effects, and oral administration of asiatic acid enhances nerve regeneration in animals. Both of these effects may be beneficial for the treatment of stroke as well. The purpose of this study was therefore to determine if asiatic acid is protective in permanent focal cerebral ischemia in mice, a model of stroke.

Methods - Mice were administered 4 doses of 75 mg/kg asiatic acid or vehicle orally at 1 hour before and 3, 10, and 20 hours after the induction of ischemia by permanent occlusion of the middle cerebral artery.

Results - Mice treated with asiatic acid showed an approximately 50% decrease in infarct size at 24 hours after induction of ischemia compared with vehicle-treated controls.

Conclusions - These results suggest that asiatic acid may be an effective treatment for the reduction of damage after stroke.

PYRUVATE PROTECTION IN EXPERIMENTAL STROKE INVOLVES AN ANTI-INFLAMMATORY MECHANISM VIA INHIBITION OF NF-KB AND MMP9

Qing Wang^1,2, Xiannan Tang^1,2, Raymond A. Swanson¹, Midori A. Yenari¹

¹Department of Neurology, University of California, San Francisco and The San Francisco Veterans Affairs Medical Center, San Francisco, CA, USA, ²Department of Anesthesia, Stanford University School of Medicine, Stanford, CA, USA

Stroke is the third leading cause of death in the United States after cancer and heart disease. Existing therapy for ischemic stroke, intravenous rt-PA, is limited. Increasing attention for protective therapy is focused on preserving the ischemic penumbra and inhibiting post-ischemic inflammation. Pyruvate is a key intermediate in glucose metabolism, but is also a free radical scavenger. Prior work has shown that ethyl pyruvate is neuroprotective in a rat model of middle cerebral artery occlusion (MCAO). The purpose of this study is to assess the neuroprotective and anti-inflammatory effects of pyruvate following transient MCAO in rats. Male Sprague-Dawley rats (n=8/group) were subjected to 2 hours of MCAO, and pyruvate (500mg/kg, IP) was administered 10 minutes before reperfusion. 24 hours later, infarct size, neurological deficits, physiological parameters, NF-kB translocation, microglial activation and neutrophil infiltration were evaluated. Pyruvate did not affect physiological parameters but significantly reduced infarct volume by 71.25 % (46.20 ± 10.1 mm2 vs 160.49 ± 21.6 mm2 P<0.01). This was accompanied by improvement in behavioral tests, including forelimb-placement, elevated body swing test, cylinder test, and Bederson score (P<0.05). Pyruvate treatment also reduced the number of neutrophils stained by MPO by 45% (46±2.68/HPF vs 25± 1.69/HPF, P<0.001, n=7/group) compared with the vehicle-treated group. Microglial activation as measured by ED1 staining was reduced by 40% (30± 2.5/HPF vs 18±1.2/HPF, P<0.001). Isolectin B4 positive cells (to identify all microglia) were also reduced by 52% (52±2.57/HPF vs 25±1.7/HPF, P<0.001). Immunohistochemistry showed that pyruvate reduced numbers of brain cells with nuclear NF-kB staining from 31% (MCAO) to 7% (MCAO+pyruvate) (P<0.001). As a positive control, a separate set of rats (n=3/group) was given LPS (5mg/kg, IV) to cause brain inflammation without cell death. LPS increased nuclear staining of NF-kB and this was almost completely abolished by pyruvate (19% for LPS vs 0.5% for LPS+pyruvate, P<0.001). A NF-kB DNA binding assay was used to test activity. NF-kB's DNA binding capacity increased by 4.57 and 4.80 folds for LPS and MCAO treatment respectively compared with control, uninjured rats (7.68% for control vs 39.13% and 44.56% for LPS and MCAO, respectively, P<0.001). Pyruvate treatment significantly reduced NF-kB activation in LPS and MCAO groups (18.7 ± 1.5% for MCAO + pyruvate and 17.53 ± 1.4% for LPS+ pyruvate vs 44.56 ± 3.9% for MCAO and 39.13 ± 2.4% for LPS, P<0.001, n=7/group). Western blots of MMP9, a NFkB regulated gene, showed increased MMP9 protein by 4 and 3.70 fold for LPS and MCAO treatment, respectively compared with shams (9.36±0.73 % for shams vs.37.39±11.97% and 34.86±8.21% for LPS (P<0.01) and MCAO (P=0.01), respectively, n=3/grp). Pyruvate treatment profoundly reduced MMP9 increases due to LPS administration and MCAO (11.54 ± 1.27% for MCAO + pyruvate and 10.78 ± 1.32% for LPS+ pyruvate, P<0.01 vs. no treatment). Taken together, these results strongly suggest that neuroprotective effects of pyruvate on MCAO may be closely correlated to its anti-inflammatory effect, possibly at the transcriptional level.

Key Words: middle cerebral artery occlusion, pyruvate, cerebral ischemia, stroke, inflammation, neuroprotection.

This work was funded by NINDS, AHA.

Specific grants:

R01 NS40516

P50 NS14543

AHA EIA 0540066N.

CAPSAICIN PRE-TREATMENT IN EXPERIMENTAL CEREBRAL ISCHAEMIA

Renee Turner, Robert Vink

Discipline of Pathology, University of Adelaide, Adelaide, SA, Australia

Background and Aims: Few studies have examined the role of neuropeptides following cerebral insults such as stroke and traumatic brain injury. However, our laboratory has previously shown that the neuropeptide substance P (SP) is increased following experimental stroke, indicative of neurogenic inflammation, and is associated with significant functional deficits, cerebral oedema and breakdown of the blood brain barrier. Blocking the action of SP with a SP receptor antagonist significantly improves outcome, in particular, functional outcome. Capsaicin, an agent isolated from chilli peppers, causes the release of neuropeptides, including SP, neurokinin A (NKA) and calcitonin gene-related peptide (CGRP), amongst others, to the point of depletion. Given the beneficial effects of SP receptor blockade, the removal of neuropeptides may also provide significant protection from the ischaemic event. In the present study, we investigated whether capsaicin pre-treatment was more protective than a SP antagonist following stroke.

Methods: Animals were randomly assigned to treatment and control groups. Capsaicin (125mg/kg) or equal volume of vehicle was administered subcutaneously over a 3 d period. From our trauma studies, 14 d was shown to be the time-point where neuropeptides were depleted, as such at14 d following pre-treatment experimental stroke was induced under Isoflurane anaesthesia using the reversible thread model of middle cerebral artery occlusion. Reperfusion was achieved at 2 h post-stroke onset. Those animals not exhibiting anti-clockwise circling were excluded. The motor function of the animals was assessed using a rotarod device for 7 d following stroke.

Results: Animals pre-treated with capsaicin demonstrated a significantly improved motor function, as assessed by the rotarod, compared to vehicle pre-treated animals. However, neither group achieved normal functional levels during the post-stroke assessment period. We compared the performance of capsaicin-treated animals to SP antagonist-treated animals. Interestingly, the performance of SP antagonist-treated animals was superior to capsaicin-treated animals.

Conclusions: It was expected that ablation of neuropeptides would provide a greater deal of protection form the ischaemic event than a SP antagonist. However, capsaicin pre-treatment does not provide any more protection from the ischaemic event. Given the fact that the capsaicin group performed below that of the SP antagonist group suggests that some neuropeptides may in fact have beneficial roles following stroke. CGRP, a known potent vasodilator, may be one such neuropeptide.

EFFECTS OF LACTATE LOADING ON BRAIN GLUCOSE METABOLISM: INVESTIGATION WITH [18F]FDG-PET IN CONSCIOUS RATS

Mikako Ogawa¹, Y. Ouchi², T. Fuchigami¹, T. Torizuka², M. Futatsubashi^2,3, Y.B. Jin¹, Y. Magata¹

¹Photon Medical Research Center, Hamamatsu University School of Medicine, Hamamatsu, Japan, ²Positron Medical Center, Hamamatsu Medical Center, Hamamatsu, Japan, ³PET Medical Application Group, Hamamatsu Photonics K.K., Hamamatsu, Japan

Backgrounds and aims: Glucose has been thought as a principal metabolic substrate of the brain. However, Magestritti et al. proposed an astrocyte-neuron lactate shuttle hypothesis, which claims the metabolic role of lactate in the brain, especially in neurons(1). This model hypothesizes that lactate produced in astrocytes is then transferred to the neuron, which then uses the lactate as an energy substrate aerobically. Recently, Kemppainen et al. reported that high intensity exercise decreased brain glucose uptake(2). They speculated that lactate is produced during the exercise and utilized in the brain. To elucidate this phenomenon in terms of the role of lactate in brain energy metabolism, we investigated the effect of lactate administered systemically on the brain glucose metabolism in conscious rats by [18F]FDG-PET.

Methods: Each conscious rat (n=8) was firmly placed in a cone-shaped thin polyethylene holder on the scanner bed. Rats were fasted 12hr before the PET scan. Dynamic PET scans were performed after a bolus intravenous injection of 7.4 MBq [18F]FDG for 45 min using small animal PET scanner (FX, Gamma Medica-Ideas Inc.). Arterial blood sampling was performed throughout the scanning period, and plasma was separated by centrifugation and counted by a γ-counter. Blood glucose concentration was measured at 45 min after [18F]FDG injection. For the lactate group (n=4), 100 mM lactate was infused intravenously (0.65 mL/hr) throughout the measurement from 10min prior to [18F]FDG injection. The regions of interest (ROIs) were placed on the coronal sections of serial PET images generated every 1 min. The rate constants (K1, k2, k3, k4) were estimated by OLS to calculate CMRglc. We used 0.625 for LC (3).

Results: Lactate infusion reduced the value of Kcomp (= K1*k3/(k2+k3)) (control group: 0.043 min-1, lactate group: 0.034 min-1), and the CMRglc level was found to be higher in the lactate group (52.8 µmol/100g/min) than in the control group (42.7 µmol/100g/min). The blood glucose concentration was elevated after lactate treatment.

Conclusion: The decrease of Kcomp may indicate a possibility of greater utilization of lactate than glucose in the experimental setting of intravenous lactate infusion. Thus, it was suggested that lactate might be a favorable fuel for brain energy metabolism than glucose in such circumstances. The unexpected finding of CMRglc elevation in the lactate group might be due to an increase of serum glucose through glyconeogenesis in the liver activated by the lactate loading. Using this technique, the dynamics of lactate in the brain ischemia can be further clarified using living rats with MCA occlusion.

A ROLE OF DORSOLATERAL PREFRONTAL CORTEX IN DRUG-CUE INDUCED NEURAL RESPONSE

Takuya Hayashi^1,2, Ji Hyun Ko¹, Antonio Strafella¹, Bruce Pike¹, Alain Dagher¹

¹McConnell Brain Imaging Center, Montreal Neurological Institute, Montreal, QC, Canada, ²Investigative Radiology, National Cardiovascular Center Research Institute, Osaka, Japan

Background and aims: The drug cue-reactivity paradigm has been among the most prominent tools for investigating craving and relapse. Previous neuroimaging studies have associated cue-reactivity with activations of frontal areas, such as anterior cingulate (ACC) and dorsolateral prefrontal (DLPFC) cortices, in a manner dependent on the context of drug use. We had previously shown that an area of DLPFC responded to cigarette cues in a manner that was modulated by drug availability, and suggested that this region may modulate the response of other brain areas to cues. The current study used repeated transcranial magnetic stimulation (rTMS) to reduce excitability of this DLPFC area and investigated the context-dependent neural response to smoking cues using fMRI.

Methods: Ten healthy smokers (8 men; mean age: 23 years; average smoking: 19 cigarettes/day) were recruited. Each subject was scanned for structural MRI to localize the target of rTMS, followed by four fMRI scans on different days, during which they viewed video clips of smoking-cue (individuals smoking cigarettes) or control (without smoking). Each of four fMRI scans was performed under different conditions of expectancy (i.e. non-expectant: no smoking allowed for 4 hours after the scan, verified by CO monitoring; expectant: cigarette available immediately after the scan) and rTMS (true vs. sham) targeting the left DLPFC region identified in our previous study. The target for the rTMS was identified by linearly transforming the MNI template brain into each user's native space. Using frameless stereotaxy, we placed the TMS coil over each subject's target, and gave 1800 pulses of 1Hz rTMS with intensity of 59% (of stimulator maximum) which corresponded to the average resting motor threshold. Immediately after receiving true or sham rTMS, subjects underwent a 15-min fMRI scan while they watched alternate 2-minute video clips of the smoking-cue and control. They answered a craving questionnaire after each clip. Image data was analyzed using FSL.

Results: The self-reporting cue-induced craving tended to be lower in the true rTMS condition compared to sham rTMS. In the sham rTMS conditions, we found larger cue-induced BOLD response in the left DLPFC and ACC during the expectant compared to the non-expectant condition, as well as a correlation between craving and BOLD signal in the left DLPFC, as in our previous study2. In contrast, in the true rTMS conditions, neither cue-induced BOLD response, nor correlation between BOLD and craving were affected by expectancy in either frontal area. Analysis of effective connectivity revealed that the stimulated part of left DLPFC became less responsive to input from the contralateral DLPFC and medial temporal regions, but not to ACC (thresholded at uncorrected p = 0.001).

Conclusions: These findings indicate that in cue-reactivity the DLPFC facilitates ACC activity based on inputs from areas other than ACC, suggesting a key role of DLPFC in associating perceived drug availability with craving in a positive feed-forward manner.

DELINEATION OF VOI'S AT HMPAO SPECT IMAGES FOR AUTOMATIC CLINICAL EVALUATION OF NEW IMAGES

Claus Svarer¹, Gerda Thomsen¹, David Erritzoe¹, Steen G. Hasselbalch², Lisbeth Marner¹, Vibe Frokjaer¹, Lars H. Pinborg¹, Karine Madsen¹, Gitte M. Knudsen¹

¹Neurobiology Research Unit, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark, ²Memory Disorders Research Unit, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark

Background and aims: In daily clinical routine, evaluation of 99mTc-HMPAO SPECT images is generally performed by simple visual inspection. However, this evaluation is critically dependent on experienced reviewers. In this study an automatic method for delineation of Volumes of Interest (VOI's) at HMPAO SPECT images is presented. The method does not require a MR image but works directly at the HMPAO images. It can be used for automatic quantification of flow values for different brain regions in clinical SPECT scanning and may be a helpful additional tool for HMPAO SPECT evaluation.

Methods: SPECT HMPAO images together with high resolution (?1.2 mm) T1-weighted MRI were acquired for 10 normal healthy subjects. Using the approach described in (Svarer, 2005) a VOI set consisting of 11 VOI's (left and right) was generated automatically for each of the HMPAO images. These HMPAO images together with the VOI's were used as templates. For a normal database of 36 HMPAO images (age 42-84 year) the spatial transformation between HMPAO template spaces and the individual images was estimated by the method described by Svarer et al. (2005) though using HMPAO templates. A VOI set was therefore available for each of the 36 images in the database. Semi-quantitative estimates of rCBF relative to cerebellum were obtained by dividing each VOI value with the corresponding cerebellar mean VOI value. For ten subjects (five healthy subjects and five patients) the same approach was used for extraction of mean VOI values from their respective HMPAO images. The database mean VOI values was detrended for age and similar transformations were performed for the VOI values for the ten subjects. Student's t-test was used to test if the mean VOI values were significant different (p<0.05) from the normal database, and similar tests were performed between left and right side for five of the VOI's and between a posterior and anterior cortical VOI.

Results: The outcome of the automatic approach was in perfect accordance with the overall assessment of three trained clinicians. One of five presumably healthy subjects unexpectedly proved to have an abnormal rCBF pattern; this image was also judged as abnormal on the basis of the automatic approach. In addition, the automatic approach identified a slight left-right side difference in one of the other healthy subjects, that was left unnoticed by the observers. Further, the images for the five patients were all identified as abnormal by the automatic approach where one of these was rated as normal by visual inspection.

Conclusion: Using this automatic approach it was possible objectively to get semi-quantitative estimates of rCBF for a number of brain regions, thus quantitatively assessing abnormalities in relation to a database of age-matched healthy subjects. In the limited number of patients presented, the automatic approach performed equally well to experienced observers and, in addition, gave rCBF values relative to healthy controls.

ANALYSIS OF BRAIN SPECT IMAGE OF MOYAMOYA DISEASE BY STEREOTACTIC EXTRACTION ESTIMATION (SEE). -COMPARISON OF ISCHEMIC AND HEMORRHAGIC PATIENTS-

Masaru Yamada¹, Izumi Yuzawa¹, Sachio Suzuki¹, Kazuhisa Iwamoto¹, Kiyotaka Fujii¹, Yuji Asano²

¹Department of Neurosurgery, Kitasato University School of Medicine, Sagamihara, Kanagawa, Japan, ²Department of Radiology, Kitasato University School of Medicine, Sagamihara, Kanagawa, Japan

Background and aims: Cerebral blood flow (CBF) characteristics and differences of Moyamoya disease in ischemic and hemorrhagic patients are not fully understood. To elucidate CBF characteristics between different onset symptoms of Moyamoya disease, we analyzed SPECT data by using stereotactic extraction estimation (SEE).

Subjects and Methods: Preoperative 32 moyamoya disease patients were subjected to quantitative 123I-IMP-SPECT (ARG method) with one-time arterial sampling at rest and after acetazolamide-challenge. Data were analyzed by 3D-SSP method using SEE software. All the SPECT data were given anatomical coordinates. “Extent” of CBF dectrease is defined as number of pixels whose value is below 2SD of the normal database in each identical anatomical area. “Severity” of CBF decrease is defined as mean value of such pixels.

Results: Extent of CBF decrease was statistically larger in patients with ischemic presentation who needed bypass surgery compared to those of conservatively treated ischemic patients and hemorrhagic patients. Severity of CBF decrease was statistically larger in patients with ischemic presentation who needed bypass surgery compared to those of hemorrhagic patients. Limit of this method was difficulty in precise exclusion of SPECT data of damaged brain.

Conclusion: SEE analysis using whole brain SPECT data clearly showed CBF decrease in extent and severity in hemorrhagic patients with moyamoya disease is significantly smaller than that of ischemic patients. This may suggest pathophysiological mechanism for hemorrhage is not directly related to CBF decrease.

DIAGNOSTIC VALUE OF KINETIC ANALYSIS USING DYNAMIC FDG PET IN PATIENTS WITH MALIGNANT BRAIN TUMOR

Yoshihiro Nishiyama¹, Yuka Yamamoto¹, Nobuyuki Kawai², Motoomi Ohkawa¹

¹Department of Radology, Faculty of Medicine, Kagawa University, Kita-Gun, Kagawa, Japan, ²Department of Neurological Surgery, Faculty of Medicine, Kagawa University, Kita-Gun, Kagawa, Japan

Background and aims: The most frequent malignant primary brain tumors are high grade astrocytoma (anaplastic astrocytoma and glioblastoma) and central nervous system (CNS) lymphoma. MRI detects small intracranial lesions, but has difficulties in making a differential diagnosis. FDG PET is widely applied to the study of brain tumors. However, several studies have demonstrated diagnostic limitations of FDG PET for imaging brain tumors because of the high basal glucose metabolic rate of normal brain tissue. We evaluated the quantitative imaging of glucose metabolism of malignant primary brain tumors with PET using kinetic analysis.

Methods: Twenty-eight lesions in 25 patients with malignant primary brain tumor (4 anaplastic astrocytoma, 10 glioblastoma and 14 CNS lymphoma were examined by FDG dynamic PET. Dynamic emission data were acquired for 60 min immediately following injection of FDG. Applying a three-compartment four-parameter model, metabolic rate constants [K1 (from plasma to brain), k2 (from brain to plasma), k3 (phosphorylation)] and cerebral metabolic rate of glucose (CMRGlc) were quantitatively measured in tumor lesions and the contralateral gray matter.

Results: Tumor K1 was lower than in the contralateral gray matter in all 4 anaplastic astrocytomas, in 8 glioblastomas (80%) and in 7 CNS lymphomas (50%). Tumor k2 was lower than in the contralateral gray matter in 2 anaplastic astrocytomas (50%), in 7 glioblastomas (70%) and in 11 CNS lymphomas (79%). Tumor k3 was higher than in the contralateral gray matter in 1 anaplastic astrocytoma (25%), in 6 glioblastomas (60%) and in 12 CNS lymphomas (86%). Tumor CMRGlc was higher than in the contralateral gray matter in no anaplastic astrocytoma, in 4 glioblastomas (40%) and in 12 CNS lymphomas (86%). The CMRGlc value (mean±S.D.) of CNS lymphoma (69.6±32.6) was significantly higher (p<0.03) than that of anaplastic astrocytoma (35.7±7.7) and glioblastoma (43.3±23.5).

Conclusions: The kinetic analysis, especially CMRGlc, from dynamic FDG PET might be useful for differential diagnosis of high grade astrocytoma and CNS lymphoma.

FUNCTIONAL NEUROIMAGING OF AUTONOMIC NERVOUS RESPONSES DURING LAVENDER-AROMA UNING [18F]FDG PET

Xudong Duan¹, Manabu Tashiro¹, Tomoyuki Yambe², Sayuri Seto³, Masatoshi Itoh¹

¹Division of Nuclear Medicine, Cyclotron and Radioisotope Center, Tohoku University, Sendai, Japan, ²Department of Medical Engineering and Cardiology, Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan, ³Teikoku Seiyaku CO., Ltd., Kagawa, Japan

THE THERAPEUTIC INTERVENTION FOR PERSONS WITH PROFOUND MULTIPLE DISABILITIES - CASE STUDY USING NEAR-INFRARED SPECTROSCOPY -

Daisuke Hirano¹, Takamichi Taniguchi², Kotaro Takeda³, Hiroyuki Iwasaki¹, Hideo Shimoizumi¹, Motoko Sugihara²

¹International University of Health and Welfare Rehabilitation Center, Otawara, Tochigi, Japan, ²Department of Occupational Therapy, The School of Health Science, International University of Health and Welfare, Otawara, Tochigi, Japan, ³CREST, Japan Science and Technology Agency, Kawaguchi, Saitama, Japan

DISTAL MIDDLE CEREBRAL ARTERY OCCLUSION MODEL FOR REAL TIME OPTICAL MONITORING OF CEREBRAL BLOOD FLOW AND METABOLISM

Hwa Kyoung Shin¹, Sean I. Savitz¹, Christian Waeber¹, Michael A. Moskowitz¹, Cenk Ayata^1,2

¹Stroke and Neurovascular Regulation Laboratory, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ²Stroke Service and Neuroscience Intensive Care Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA

Background and aims: Optical imaging of cerebral cortex has recently gained momentum with the advent of novel technologies such as laser speckle flowmetry (LSF), multispectral reflectance imaging (MSRI), intrinsic optical signal imaging, and multiphoton confocal microscopy. Here we describe the technique and characterize the hemodynamic, electrophysiological, histological and behavioral outcome features of a distal middle cerebral artery occlusion (dMCAO) model where ischemia and reperfusion can be reproducibly achieved during real-time optical imaging.

Methods: Isoflurane-anesthetized, intubated and ventilated mice (C57BL/6) with arterial blood gas and pressure monitoring are placed in a stereotaxic frame, and dorsal skull surface is exposed. After stripping of the temporalis muscle, distal MCA is exposed via a small burr hole in the temporal bone just above the zygomatic arch, and occluded using a microvascular clip. LSF and MSRI are simultaneously and continuously performed before and during dMCAO. Reperfusion is achieved 60 min later by careful removal of the clip, and confirmed by imaging. Infarct volume and tissue swelling are assessed at 24 and 48 hr (TTC), and 1 month (H&E). Neurological deficits are longitudinally assessed using 5-point, corner and cylinder tests (n=5).

Results: Infarcts were limited to the dorsolateral cortex sparing the striatum. Infarct volumes were significantly larger at 48h compared to 24 hr after reperfusion (21+/−2 vs. 10+/−2 mm3, n=5 and 8, respectively, p<0.05; +/−SEM). By overlying the dorsal view of the cortical infarct at 48h (TTC) and the dorsal CBF map 60 min after dMCAO (LSF), we determined that residual CBF less than 28+/−2% of preischemic baseline resulted in cerebral infarction (Fig). Ischemic tissue swelling (ipsilateral-contralateral hemisphere) was detectable at 24h (6.2+/−1.0 mm3) and tended to increase at 48h (8.1+/−2.4 mm3). There were 2.8+/−0.4 peri-infarct depolarizations/h, as determined by spreading hypoperfusion waves on LSF and confirmed by electrophysiological recordings using glass micropipettes. Neurologically, cylinder test showed reduced use of contralateral paw during a full rear; this motor deficit fully recovered between days 7 and 14. Corner test was more sensitive at later time points, where an ipsilateral preference was detected at 28d. The traditional 5-point grading scale did not demonstrate deficits at any time point after occlusion.

Conclusions: Real time optical and electrophysiological monitoring of focal cerebral ischemia is feasible in a dMCAO mouse model and may be useful for examining the impact of specific genes on cerebrovascular function.

This work was supported by the American Heart Association (0335519N, Ayata; 0475008N, Savitz), NIH (P50 NS10828, Moskowitz).

DOPAMINERGIC REGULATION OF CONSCIOUS EXPERIENCE

Hans Lou¹, Pedro Rosa¹, Joshua Skewes¹, Troels Kjaer², Albert Gjedde¹

¹CFIN, Aarhus University, Aarhus, Denmark, ²Department Clinical Neurophysiol, Copenhagen University, Copenhagen, Denmark

In attention deficit hyperactivity disorder (ADHD), both conscious experience and dopaminergic activity are low (1), and in yoga nidra meditation high (2). We propose that the link is causal, and that dopaminergic activity regulates conscious experience. We show with [11C]raclopride PET in eight healthy, young, right-handed female volunteers that visuo-verbal conscious experience is linked to dopamine release in the striatum: Presenting each of 24 words 12 times for 43 ms did not result in conscious experience from viewing words, as ascertained by interview after the scanning. Conscious experience arose when subjects viewed each of 24 words six times for 86 ms during 60 min scanning session. The order was counterbalanced, and word presentation was terminated by masking. The mean pB of the five striatal regions tested was 3.47 for the “unaware” condition and 2.38 for the “aware” condition, reflecting a rise in dopamine D2 receptor occupancy from 12.5 % in the unaware condition to 16% in the aware condition (P=0.03, two-tailed t-test).

The effect on visual awareness of D1/D2 receptor stimulation with oral pergolide was examined in a double blind study of 14 healthy, young, right-handed female volunteers. We recorded how well 3–4 letter words were subjectively experienced as seen on a four-point scale from 0 (not seen at all) to 3 (completely and clearly seen), with 1 and 2 in between. Recorded immediately after each test, the ratings for 16, 30, and 50 ms presentation times, terminated by masking (placebo in brackets), were 1.62 (1.44), 2.11 (1.90), and 2.58 (2.38). Pergolide differed from placebo with P= 0.05 (repeated measures ANOVA). Each data point included 40 words. The forced choice measure of recognition memory was unchanged on pergolide. The test upheld the hypothesis of dopaminergic regulation of conscious experience. The hypothesis explains how disturbed conscious experience with delusions and hallucinations is prevalent in conditions with dysregulation of synaptic dopamine concentration, such as may be the case in schizophrenia and drug abuse.

PATHOPHYSIOLOGY OF TRANSIENT DIFFUSION-WEIGHTED MRI LESIONS IN POST-STROKE SEIZURES: CHANGES IN LOCAL CEREBRAL BLOOD FLOW AND GLUCOSE METABOLISM

Hiroshi Moriwaki¹, Naoaki Yamada², Takuya Hayashi², Yukio Sugiyama¹, Kazuhito Fukushima², Kotaro Miyashita¹, Hiroaki Naritomi¹

¹Department of Cerebrovascular Medicine, National Cardiovascular Center, Suita, Osaka, Japan, ²Department of Radiology, National Cardiovascular Center, Suita, Osaka, Japan

BACKGROUND AND AIMS: Following seizures, focal abnormality may be observed transiently on diffusion-weighted MRI (DWI). However, detailed mechanisms of such transient high signal intensity lesion (THL) manifestation have yet remained unclear. In order to clarify the clinical significance and pathogenetic mechanism of THL, we performed SPECT/PET studies in association with MRI in patients with post-stroke seizures.

METHODS: During the last 8 years, 75 patients with post-stroke seizures underwent MRI studies in association with Tc-99m HMPAO SPECT and/or F-18 FDG PET in the ictal and/or peri-ictal phases. In 21 of them (12 males, mean age 69.8 yrs), THL was observed on DWI. Thirteen of 21 patients had status epilepticus. MRI was performed repetitively after adequate control of seizures. Tc-99m HMPAO SPECT was performed at the same period as initial MRI in all the patients, and F-18 FDG PET was undertaken in 9 of them. On both HMPAO and FDG images, the count ratio of abnormal high-uptake area to the contralateral homologous area (L/N ratio) was calculated.

RESULTS: A total of 43 THLs were observed in 21 patients. In 10 patients, single THL was located in areas surrounding the old cerebral infarction or hemorrhage. In the other 11 patients, multiple THLs were observed; 29 of 33 lesions were located remote from the old stroke area, such as the thalamus (n=6), hippocampus (n=8), cingulate gyrus (n=5) and cerebral cortices (n=10). Multiple THLs were more frequently associated with status epilepticus than single THL (10/11 vs. 3/10, P<0.01) and eventually resulted in infarct-formation at the hippocampus in 4 patients. HMPAO demonstrated flow increase in all patients. The L/N ratio of FDG was significantly higher than that of HMPAO (1.47+/−0.21 vs. 1.17+/−0.14, P<0.001).

CONCLUSION: The present results suggest that multiple THLs following seizures usually manifest in the limbic system such as the thalamus or hippocampus relating with the severity of seizures. The manifestation of transient DWI lesions may be attributable to excessive metabolic increase accompanied by flow increase.

SPECTRAL CONTENT OF THE ONGOING BOLD FMRI ACTIVITY IS MODULATED BY RESTING STATE-TYPE

Mark McAvoy, Giovanni d'Avossa, Marcus E. Raichle, Linda Larson-Prior

Washington University School of Medicine, Saint Louis, MO, USA

ELECTROPHYSIOLOGICAL EVIDENCE FOR LONG_TERM DEPRESSION IN THE CONTRALATERAL, INTACT BRAIN FOLLOWING TRANSIENT FOCAL CEREBRAL ISCHEMIA IN RATS

E-Jian Lee¹, Sheng-Yang Huang¹, Yu-Hsiang Kuan¹, Tsung-Ying Chen², Hung-Yi Chen³, Tian-Shung Wu⁴

¹Department of Surgery, National Cheng Kung University Medical Center, Tainan, Taiwan-R.O.C., ²Department of Anesthesiology, Buddhist Tzu-Chi University and Buddhist Tzu Chi General Hospital, Hualien, Taiwan-R.O.C., ³Institute of Pharmacy, China Medical University, Taichung, Taiwan-R.O.C., ⁴Department of Chemistry, National Cheng Kung University, Tainan, Taiwan & National Research Institute of Chinese Medicine, Taipei, Taiwan-R.O.C.

Introduction: Focal brain injury may induce synapto-functional, metabolic and electrophysiological dysfunction in the contralateral areas [1]. We hypothesize that contralateral electrophysiological diaschisis may represent a form of long-term depression after injury.

Methods: Male Sprque-Dawley rats, weighing 240-280 gm, were subjected to intraluminal suture occlusion for 60 minutes or sham-occlusion. Physiological parameters, including core temperature, local cortical perfusion and arterial blood gases, were measured, whereas the somatosensory potentials (SSEPs) evoked from individual forepaw and hindpaw regions were recorded prior to the ischemic insult and at 1, 3, 7, 21 or 28 days after reperfusion. Postmortem brain ischemic damage was determined by quantitative image analysis of Nissl-stained sections.

Results: The sham-operated group did not cause a significant change in the physiological parameters or the SSEP waveforms recorded either in the ipsilateral or the contralateral brain hemispheres at various intervals of recording. There was no difference in brain infarction volumes among each time interval of the ischemic groups (p > 0.05). Relative to the baseline data, at 24-72 hours of reperfusion the SSEPs recorded from ischemic fore- and hindpaw cortical fields were depressed and the amplitudes decreased to 36.4–41..3% and 41.9–44..5% of baseline, respectively (p < 0.001). On contrast, the SSEPs recorded from non-ischemic fore- and hindpaw cortical fields decreased. The amplitudes decreased to 76.0–83..0% and 74.5–81..8% of baseline, respectively (p < 0.001). At 7-28 days of reperfusion improved SSEPs recorded from ischemic fore- and hindpaw cortical fields were noted. The amplitudes recovered to 49.9–54..7% and 45.5–58..5% of baseline, respectively (p < 0.001). On contrast, the SSEPs recorded from non-ischemic fore- and hindpaw cortical fields were also noted to improve. The amplitudes recovered to 88.6–96..2% and 85.5–96..3% of baseline, respectively (p < 0.001 up to 21 days of reperfusion). Relative to the pooled sham-operated controls, the amplitudes of SSEPs recorded from both ischemic fore-and hindpaw cortical field in the ischemic animals significantly but gradually improved by 18.3% (p < 0.001) and 16.6% (p < 0.001) of baselines, respectively, during a course of 28-day recovery. On contrast, the amplitudes of SSEPs recorded from both non-ischemic fore- and hindpaw cortical field in the ischemic animals also significantly improved by 17.9% (p < 0.001) and 10.4% (p < 0.001) of baselines, respectively, up to 21 days of reperfusion.

Conclusions: Cerebral ischemia-reperfusion induces time-dependent electrophysiological depression not only in the ischemic, but also in the contralateral, intact brain, and both the events, however, improve over time. This electrophysiological evidence suggests long-term depression in the contralateral, intact brain and functional re-organization following cerebral ischemia-reperfusion.

NON-MAGNETIC WHISKER STIMULATION FOR FMRI AND NEUROPHYSIOLOGICAL STUDY

Basavaraju G. Sanganahalli, Peter Herman, Fahmeed Hyder

Department of Diagnostic Radiology, Yale University, New Haven, CT, USA

The use of animal models for simultaneous fMRI and neurophysiology has become important for neuroscience studies. In the rat, many studies have utilized different stimulation devices to characterize the whisker deflection to barrel response for fMRI and elctrophysiological studies. It is extremely important to generate stimuli with identical sensory effects both inside (fMRI) and outside (neurophysiology) the magnet. Air-puffs were generated from pulses of compressed air, which could be delivered in a computer-controlled way by inbuilt solenoid unit. Under these conditions, the air puff stimuli deflected all the whiskers by 2 mm (rostral-caudal) which were glued with a tiny adhesive tape. Whiskers stimuli were presented at 4-40Hz for 30 s duration. All fMRI data were obtained on a modified 11.7T Bruker horizontal-bore spectrometer. The images were acquired with gradient echo EPI sequence (TR/TE= 1000/12 ms). Whisker stimulation induced positive BOLD response in the contralateral barrel cortex (Fig) with a maximum response at 12 Hz (top; 4, 8, 12 Hz and bottom; 20, 30 and 40 Hz). For approximately the same location in the brain, the same stimulus (each frequency) produced similar activation patterns across different subjects. The fMRI results were in agreement with neurophysiology data from pO2 and LDF probes reflecting tissue oxygenation and perfusion changes, which together give rise to the BOLD signal change. While this device provides easy control of frequency of stimulation, we are currently investigating amplitude variations. In summary, unilateral whiskers stimulation with air puffs resulted in significant increases in neurophysiologic signals in the contralateral barrel field in agreement with prior findings.

WHICH IS CORRELATED WITH CLINICAL FEATURES, SEVIERITY OR EXTENT OF LESION IN ALZHEIMER'S DISEASE IN SPECT?

Akiko Ishiwata¹, Sunao Mizumura², Shin Kitamura¹, Yasuo Katayama¹

¹Department of Neurology, Nippon Medical School, Tokyo, Japan, ²Department of Radiology, Nippon Medical School, Tokyo, Japan

BACKGROUND AND AIMS: We have reported perfusion decline in the posterior cingulate gyrus, the precuneus and the hippocampus in the very early stage of Alzheimer's disease (AD) as a predictive marker for progression. This study is to evaluate whether clinical severity is correlated with extent and severity of these lesions in SPECT in AD. METHODS: Forty-one patients (13 males, age 71 +/− 7.8 yrs, mean +/− SD) with AD according to NINCDS-ADRDA underwent 123I-IMP SPECT and neuropsychological examinations including Mini Mental State Examination (MMSE), Alzheimer's Disease Assessment Scale Japanese version (ADAS-J cog), Wechsler Adult Intelligence Scale-Revised (WAIS-R) at initial visit. Image sets were transformed to the bicommissural stereotactic coordinate system, and regional anatomical activities were examined by three-dimensional stereotactic surface projections (3D-SSP). An abnormal Z-score was defined as >1.64 (corresponding to p<0.05). The extent of lesions was defined as a total number of the coordinates with abnormal Z-score in the cerebral hemisphere, and severity as an average Z-value in the same region. Spearman correlation coefficients were used to evaluate whether clinical severity was correlated with extent and severity of these lesions on 3D-SSP. RESULTS: All patients underwent MMSE (mean score 16.3, range 3-25) and WAIS-R (mean score 79.3, range 43-110), and 21 patients underwent ADAS-J cog (mean score 21.0, range 8.3–40..7). The extent was significantly correlated with ADAS-J cog score (r = 0.62, p = 0.02), however, the severity showed no correlation (r = 0.31, p = 0.24). CONCLUSIONS: Diffuse rather than focal perfusion abnormality reflects progression of clinical symptom in AD.

SELECTIVE NEURONAL DEATH AND PARAMAGNETIC EFFECT AFTER ISCHEMIC STRESS

Masayuki Fujioka¹, Toshiaki Taoka², Yoshiyuki Matsuo³, Kenichi Mishima⁴, Kumiko Ogoshi⁵, Yoichi Kondo⁶, Masakazu Tsuda⁴, Michihiro Fujiwara⁴, Takao Asano⁷, Toshisuke Sakaki⁸, Akihiro Miyasaki⁸, Akio Asai¹, Keiji Kawamoto¹, Darren Park⁹, Bo Siesjo¹⁰

¹Department of Neurosurgery, Kansai Medical University, Hirakata, Osaka, Japan, ²Department of Radiology, Nara Medical Univeristy, Kashihara, Nara, Japan, ³Developmental Research Laboratories, Shionogi & Co., Ltd., Toyonaka, Osaka, Japan, ⁴Department of Neuropharmacology, Faculty of Pharmaceutical Sciences, Fukuoka University, Fukuoka, Japan, ⁵Department of Public Health, Nara Medical Univeristy, Kashihara, Nara, Japan, ⁶Department of Neuroscience, Okayama University Medical School, Shikata, Okayama, Japan, ⁷Department of Neurosurgery, Saitama Medical School and Center, Kawagoe, Saitama, Japan, ⁸Department of Neurosurgery, Nara Medical University, Kashihara, Nara, Japan, ⁹Neuroscience Center, The Queen'S Medical Center, University of Hawaii, Honolulu, HI, USA, ¹⁰Neuroscience Center, Lund University, Lund, Sweden

The delayed neuronal death after brain ischemia has been well investigated in the pyramidal neuronal death in the hippocampal CA1 region and in the neurodegeneration of the thalamus and substantia nigra after cerebral cortical ischemia (known as remote effect). Recently, we have reported the novel type of delayed ischemic neuronal death that is indicated by magnetic resonance imaging (MRI) (Ann Neurol 2003;54:732-747, Stroke 2000;31:797-798, Stroke 1999;30:1975-1977, Stroke 1999;30:1038-1042, Stroke 1999;30:1043-1046, Stroke 1997;28:584-587, Stroke 1994;25:2091-2095).

Transient global brain ischemia, hypoglycemic coma, and brief focal ischemia lead to the primary neuronal damage in the areas that are subjected to the transient energy failure in humans and rats (hippocampus, striatum, cerebral cortex, substantia nigra and thalamus). Subsequently, the slowly-progressive secondary neurodegeneration and chronic glial activation (astrocyte and microglia) develop in these areas. The chronic brain inflammation, long-lasting oxidative stress and extracellular accumulation of amyloid precursor protein (APP) are involved in this slowly-progressive neurodegeneration. Furthermore, manganese (Mn), a paramagnetic metal, increases in the ischemic areas. The Mn is neurotoxic and can provoke the secondary neurodegeneration. MRI detects the on-going process of this delayed ischemic neurodegeneration associated with paramagnetic effect. We report the experimental and clinical data regarding this delayed ischemic neurodegeneration that results in the cognitive decline of late-onset in humans and rats.

We suggest that 1) the hyperintensity on T1-weighted MRI after ischemic stress involve tissue Mn accumulation accompanied by Mn-superoxide dismutase and glutamine synthetase inductions in reactive astrocytes, 2) the MRI changes correspond to slow neurodegeneration with a chronic inflammatory response and signs of oxidative stress, and 3) the subjects with these MRI changes are at risk for showing a late impairment of brain function even though the transient ischemia is followed by total neurological recovery.

A PET [18-F]ALTANSERIN STUDY OF 5-HT2A RECEPTOR BINDING IN THE HUMAN BRAIN AND RESPONSES TO PAINFUL HEAT STIMULATION

Ron Kupers^1,2, Vibe Frokjaer³, Arne Naert¹, David Erritzoe³, Henrik Kehlet¹, Gitte Knudsen³

¹Department for Surgical Pathophysiology, Rigshospitalet, Copenhagen, Denmark, ²PET Unit, Rigshospitalet, Copenhagen, Denamrk, ³Neurobiology Research Unit, University Hospital Rigshospitalet, Copenhagen, Denmark

Background: There is strong evidence that serotonin plays an important role in pain regulation. One source of serotonin pathways in the CNS arises from the raphe nuclei from which the major descending pain modulatory pathways originate. A second source arises from the medullary, pontine and mesencephalic reticular formation and innervates an extensive (sub)cortical network, including the thalamus, basal ganglia, anterior cingulate and prefrontal cortex, amygdala and hippocampus. This places the serotonin pathways in a very strategical position to regulate pain processing. No PET study has yet investigated possible correlations between serotonin binding and pain responsiveness. In this study we assessed the relation between baseline 5HT2A binding and response to noxious heat stimulation in a group of young healthy volunteers.

Methods: Twenty-two healthy subjects (32.2+8.9 y), 12 males and 8 females underwent a 40 min PET scan using a steady-state regime with the 5-HT2A antagonist, [F-18]altanserin. PET images were co-registered to 3 Tesla MRI's that were acquired for each individual subject, and ROIs were applied automatically by applying a predefined MRI-based template onto the individual MRIs and PET images. Cerebellum was used as a reference region valid for non-specific binding alone. The binding potential of specific tracer binding (BP1) was used as the outcome measure and calculated for 19 regions of interest (ROI). Response to painful stimulation was tested in a separate quantitative sensory testing session, carried out several months after the PET scanning. Responses to both phasic and tonic heat stimulation were assessed. Heat stimuli were administered to the upper leg with a peltier element-based thermal stimulator (probe surface 3×3cm). In each participant we measured the pain threshold, pain tolerance, response to a series of phasic heat stimuli of 5 s and response to a 7-minute tonic painful heat stimulus. Pain intensity and pain unpleasantness were rated following (phasic) or during (tonic) stimulation.

Results: None of the regions showed a significant correlation between [F-18]altanserin binding and pain threshold. In contrast, [F-18]altanserin BP in sensory-motor (r=0.59;P=0.01), parietal (r=0.48;P=0.05) and posterior cingulate (r=0.52;P=0.03) cortices correlated significantly with pain ratings on the 47oC phasic stimulus. In addition, a negative correlation (r=-0.69;P=0.002) was found between [F-18]altanserin BP in the caudate nucleus and pain tolerance. Significant correlations were also found between tonic pain ratings and [F-18]altanserin binding in orbitofrontal (r=0.52;P=0.03) and sensory-motor (r=0.56;P=0.02) cortices or response to tonic heat pain.

Conclusion: Our data show suggest that 5-HT2A receptors in both cortical and subcortical structures may co-determine an individual's response to pain. The correlation between [F-18]altanserin binding in the caudate nucleus and pain tolerance provides a neurochemical basis for behavioral findings of a role of this area in pain tolerance. The correlation between [F-18]altanserin binding and tonic pain ratings suggests a possible role of this area in pain coping mechanisms.

IN VIVO CHARACTERIZATION OF NEURONAL METABOLISM AFTER STROKE IN RATS AS STUDIED BY 1H/13C MRS

Jet P. van der Zijden¹, Robin A. De Graaf², Rick M. Dijkhuizen¹

¹Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands, ²Department of Diagnostic Radiology, Magnetic Resonance Research Center, Yale University School of Medicine, New Haven, CT, USA

DOPAMIINERGIC FUNCTIONS IN IDIOPATHIC NORMAL PRESSURE HYDROCEPHALUS

Yasuomi Ouchi¹, Teiji Nakayama², Toshihiko Kanno¹, Etsuji Yoshikawa³, Masami Futatusbashi³, Yasuo Tanizaki¹, Genichi Sugiura⁴, Tatsuo Torizuka¹

¹Positron Medical Center, Hamamatsu Medical Center, Hamamatsu, Japan, ²Department of Neurosurgery, Hamamatsu Medical Center, Hamamatsu, Japan, ³Hamamatsu Photonics KK, Hamamatsu, Japan, ⁴Department of Psychiatry, Hamamatsu Medical University, Hamamatsu, Japan

Idiopathic normal pressure hydrocephalus (iNPH) is a brain disorder with unique entity, characterized by the clinical triad; gait impairment, dementia and urinary incontinence. Gait disturbance is sometimes reminiscent of parkinsonism, and some patients are indeed responsive to levodopa to some degree. These findings indicate the presence of dopaminergic dysfunction in iNPH, and if so, non-surgical therapeutic approach may be replaced for shunting. Here, we investigated alterations in dopaminergic function by examining the levels of binding of pre- and postsynaptic dopamine radiotracers using PET before and after drainage of cerebrospinal fluid (CSF).

MULTI-MODAL IMAGING OF ACUTE AND CHRONIC UP-REGULATION OF HYPOXIA-INDUCIBLE FACTOR-1 IN TUMORS

Tadashi Miyagawa, Hiroki Namba, Ronald Blasberg

Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA

Objective: Detection and assessment of tumor hypoxia, blood flow dependent hypoxia (acute) and oxygen diffusion dependent hypoxia (chronic), are important to understand tumor biology for diagnosis and treatment. This study addresses whether in vivo multi-modal imaging systems can be used for assessment of acute and chronic up-regulation of hypoxia-inducible factor-1 (HIF-1) in tumors.

Methods: Multi-modality reporter gene system for imaging using fluorescence, bioluminescence (BLI) and nuclear imaging technique controlled by hypoxia responsive elements (HRE/TGL) was developed and transduced to C6 and RG2 glioma cells. After selection by FACS, the tumor cells were implanted into mice subcutaneously. Constitutively TGL expressing tumor cells were also implanted in the same animals as a positive control. Acute tumor hypoxia was induced by i.p. injection of Hydralazine (HYZ). To investigate the feasibility of BLI to image up-regulation of HIF-1, time-scheduled sequential BLI was performed with or without HYZ treatment. Thereafter, autoradiographic imaging study was applied in the same animals with [18F]-FDG, [3H]-Xanthine and [14C]-FIAU in order to investigate quantitative and special relationship among glucose metabolism, tumor blood flow and HIF-1 expression.

Results: Regional assessment of BLI expressed by relative photon (ratio to 15min value) in HRE/TGL C6 tumor in vivo showed no differences between large and small tumors and plateau phase 10 min after i.p. injection of luciferase, lasting 10 min and decreasing thereafter. HYZ treatment increased the BLI relative photon in the HRE/TGL tumors whereas an increase of BLI relative photon in positive control tumor was marginal. Quantitative evaluation revealed that HYZ reduced tumor blood flow measured by Xanthine from 0.36 to 0.10 (ml/min/g), suggesting that HYZ can induce acute tumor hypoxia by decreasing blood flow. Autoradiographic images showed there was a good spatial relationship between FDG and FIAU uptake in the tumors, indicating that glucose metabolism was enhanced in hypoxic lesion and correlated with up-regulation of HIF-1 in the tumors. Immunostaining with pimonidazole and oxygen regulating protein 150 also revealed tumor hypoxia in the lesion of both increased FDG and FIAU uptake.

Conclusion: Cells transduced with the multimodality reporter system can be used to visualize hypoxic conditions with fluorescence, BLI and radiotracer. Under acute and/or chronic hypoxic conditions, HIF-1 expression in tumors measured by FIAU uptake was observed in blood flow dependent manner. FDG uptake was spatially correlated with HIF-1 expression.

PARIETAL ACTIVATION MODULATES TIME DELAY IN INTER-TEMPORAL CHOICES

Jon S. Wegener^1,2, Kristoffer Madsen^1,3, Mark S. Christensen^1,4, Julian Jamison⁵

¹Danish Research Centre for MR, Copenhagen University Hospital, Hvidovre, Denmark, ²Center for Integrated Molecular Brain Imaging, Copenhagen University Hospital, Hvidovre, Denmark, ³Informatics and Mathematical Modelling, Technical University of Denmark, Denmark, ⁴Institute of Exercise and Sport Sciences, Copenhagen University, Copenhagen, Denmark, ⁵Brain and Creativity Institute, University of Southern California, Los Angeles, CA, USA

Humans make many choices that involve evaluating outcomes occurring at different points in time. These evaluations of inter-temporal choices have large individual differences and have many implications for outcomes in health and elsewhere. Although inter-temporal choices have been studied behaviorally in several species, the neural underpinnings have only recently been addressed in two fMRI studies (McClure et al. 2004, 2006). Our aim was to investigate the pre-decision (evaluation) phase of inter-temporal choices to elucidate how the time delay component, independent of choice and reward, is processed in the brain. We hypothesized a graded activation of bilateral parietal areas in response to the processing of longer time delays. We used a simple event related fMRI paradigm with binary choices. Option A was a fixed 700 DKK (1USD=approx. 6 DKK) reward immediately after the scanning session, and option B was either a control condition with an immediate reward of 700 + random noise (+/− 10 DKK), or an equal distribution across three different time delays: one week, one month, or one year. For each time point independently, the amounts for the future options varied adaptively according to the previous choices made by the subject. Subjects were instructed that the computer would randomly pick one of their approx. 200 choices. This amount would be transferred to their bank account at the associated point in time. 20 healthy subjects (15F/5M) were scanned in a Siemens 3T Trio (3×3×3 mm voxels, 40 slices, TR=2.35), measuring pulse and respiration. The paradigm consisted of 2 s trials followed by a 1–6 s. ISI. The scanning session consisted of a structural scan followed by an approx. 20 min trial run. Individual data were analyzed using separate regressors specifying (1) the onsets of the presentation of the control condition, one week, one month, and one year, and (2) first and second order parametric modulations of time and reward, and (3) a set of nuisance regressors based on pulse, respiration and residual movement. Results from group level random effects analysis of the first order modulations of time (n=19, p<=0.05 (FDR)), confirmed our hypothesis, showing graded activations in parietal areas bilaterally. Additional activations were observed in R temporal areas, inferior frontal areas bilaterally, and PCC. The activations in these areas got progressively stronger the shorter the time delay was between option A and B. Finally, we performed exploratory correlation analyses between the discount factor at one year (i.e. the inverse of the option B reward level at which the subject was indifferent to option A) and the slope of the graded activations in the regions identified above. This was significant for multiple areas, including the R parietal (r=-0.74), L parietal (r=-0.44), and the R temporal (r=-0.43). In conclusion, our results indicate that the parietal areas participate both in the evaluation of time delays in inter-temporal choices and that the steepness of activation correlates with the steepness of the individual time-discount function.

OPTICAL PHARMACOKINETICS OF INTRAARTERIAL MITOXANTRONE

Shailendra Joshi¹, Roberto Reif², Mei Wang¹, Irving J. Bigio²

¹Department of Anesthesiology, Columbia University, New York, NY, USA, ²Department of Biomedical Engineering, Boston University, Boston, MA, USA

Introduction: Intraarterial drug delivery is associated with rapid changes in tissue drug concentrations. The drug transits the cerebral circulation in 1–5 seconds, depending on the animal species. Measurements of the changes in tissue drug concentrations are beyond the time resolution of conventional laboratory techniques such as microdialysis or tissue biopsy. Optical Pharmacokinetics (OP) is a specialized form of diffuse reflectance spectroscopy, which enables the noninvasive real-time measurement of drug concentrations in-situ. We hypothesized that intraarterial delivery of mitoxatrone (MTO) will be enhanced if the drug was injected during cerebral hypoperferfusion (CPH).

Method: The OP method determines drug concentrations by measurement of the wavelength-dependent optical absorption coefficient of the tissue, and can be used for drugs that have an optical absorption band within the wavelength range from near UV to near-IR (300-1900 nm). An optical fiber probe, comprising separate illumination and collection fibers, is placed in gentle contact with the tissue surface, and the tissue beneath the surface is sampled by the diffusely back-scattered light.

Experimental protocol: Preliminary experiments in four rabbits each revealed that there was a need to disrupt the BBB for effective delivery. Ten rabbits were anesthetized by intravenous propofol infusions. Surgical preparation included the cannulation of the ear lobe vein, femoral and internal carotid arteries, skull shaving for the placement of OP and Laser Doppler probes and skull screws to attached the EEG leads. Hemodynamic data was continuously recorded. After the baseline measurements, the BBB of the animals was disrupted by intracarotid injection of mannitol (8 ml, 25% over 40 seconds). Thereafter animals were randomized to receive 3mg of MTO injection with normal blood flow (normal perfusion group) or when the blood flow was significantly decreased by contralateral ICA occlusion and severe systemic hypotension (CPH group). with bolus intravenous injection of esmolol (30 mg) and adenosine(30 mg).

Results: Figure 1 illustrates the changes in hemodynamics and tissue concentrations of MTO in the two groups of animals. Bolus injection of MTO during cerebral hypoperfusion increased the area under the concentration time curve but did not result in a sustained increase in tissue concentrations.

Discussion and Conclusions: The experiments demonstrate the feasibility of measuring tissue drug concentrations using diffuse reflectance spectrometry. OP measurements revealed that tissue concentration of MTO did not significantly increase with induced cerebral hypoperfusion. This is in contrast to dramatic increase in tissue concentrations of lipid soluble drugs like propofol, pentothal, and carmustine that we have observed in the past. Reduction of arterial pressure could adversely effect tissue deposition of water soluble compound by decreasing or even revering the fluid efflux in the capillary bed.

EVALUATION OF SIGMA1 RECEPTORS IN ALZHEIMER'S DISEASE USING [11C]SA4503 PET

Masahiro Mishina^1,2,3, Kiichi Ishiwata², Masashi Ohyama², Shin Kitamura^3,4, Yuichi Kimura², Kei-ichi Oda², Shiro Kobayashi¹, Yasuo Katayama³, Kenji Ishii²

¹Neurological Institute, Nippon Medical School Chiba-Hokusoh Hospital, Imba-Gun, Chiba, Japan, ²Positron Medical Center, Tokyo Metropolitan Institute of Gerontology, Itabashi-Ku, Tokyo, Japan, ³Department of Neurology, Nephrology and Rheumatology, Nippon Medical School, Bunkyo-Ku, Tokyo, Japan, ⁴Department of Internal Medicine, Nippon Medical School Musashi Kosugi Hospital, Kawasaki-Shi, Kanagawa, Japan

OBJECTIVE: To investigate the mapping of sigma1 receptor in Alzheimer's disease (AD) using [11C]SA4503 PET.

METHODS: We studied five AD patients and seven volunteers. A dynamic series of decay-corrected PET data acquisition was performed for 90 min starting at the time of the injection of 500 MBq of [11C]SA4503. A two-tissue two-compartment model was used to estimate K1, k2, k3 and k4. The ratio of k3 to k4 was computed as the binding potential (BP). Unpaired t-tests were used to compare the K1 and BP in patients with AD and normal subjects.

RESULTS: To compare with normals, BP of AD was significantly lower in the frontal, temporal and occipital lobe, cerebellum and thalamus (Figure B), while K1 was significantly lower in parietal lobe (Figure A).

DISCUSSION: Distribution of cortical sigma1 receptors was reduced in patients with AD.

CONTEXTUAL MODULATION OF THE HIGH-RESOLUTION 3T GE BOLD FMRI RESPONSE

Cheryl Olman

Departments of Psychology and Radiology, University of Minnesota, Minneapolis, MN, USA

Background and Aims. Inhibitory neural processes play a key role in human behavior: it is hard to find a perception or behavior that is not modulated by context, and the mechanisms of contextual modulation are frequently, if not dominantly, inhibitory. In human vision, contextual modulation is apparent in the suppression of visual features that are not distinguished from a scene background, those which lack contrast along low-level dimensions such as brightness, orientation or color. We have begun a series of experiments measuring the hemodynamic response to low-level contextual modulation in early visual areas.

Methods. In this particular experiment, four target stimuli (Gabor patches) were presented at 3 degrees eccentricity (one in each quadrant of the visual field). Subjects were required to perform a contrast discrimination task on the target stimuli: the contrast of one of the four Gabors was augmented during one of two stimulus presentation intervals (a standard two-interval forced-choice task). In preparatory studies before the fMRI experiment, this psychophysical measurement allowed us to infer the early visual contrast response function to the target under three conditions: targets alone, targets with parallel flanking Gabor patches that suppressed the early visual response (measured as increased discrimination thresholds), and targets with orthogonal flanking Gabor patches (which did not suppress the early visual response). Responses to the targets were suppressed by 30% when targets presented at 35% contrast were surrounded by 50% contrast parallel flankers. In the corresponding fMRI experiment, the contrast discrimination task motivated a constant level of effort and alertness during all three experimental conditions and allowed experimental verification of the presence of surround suppression during the scanning. For the fMRI experiment, the three experimental conditions were presented in an event-related design with an average inter-trial-interval of 4.5 s. The posterior occipital cortex was imaged using a gradient echo (GE) EPI pulse sequence at 3 Tesla, with 1.5 mm isotropic resolution and a volume repetition time of 1.5 s. A general linear model was used to estimate the event-related response in each voxel to each stimulus type.

Results. The amplitude of the BOLD response to the targets with parallel flankers was not decreased, even though both the psychophysical experiments and the electrophysiological literature indicate neural activity in this region of primary visual cortex is suppressed.

Conclusions. Even considering the possible confounds of hemodynamic blurring, this finding questions the utility of standard GE BOLD techniques for studying contextual modulation on a fine scale. Although there was no observed correlation between BOLD peak amplitude and inferred neural activity in primary visual cortex, the different stimulus configurations elicited interesting differences in the shape of the hemodynamic response: the addition of suppressive flankers modulated the relative contributions of positive and negative portions of the response. Further experiments are investigating what the shape of the hemodynamic response function can tell us about contextual interactions in high resolution BOLD fMRI.

CLINICAL USEFULNESS OF THREE-DIMENTIONAL PREOPERATIVE PLANNING IN NEURONAVIGATION SYSTEM WITH MULTIMODAL IMAGE FUION TECHNIQUE

Noritaka Masahira, Shoko Yamada, Naoki Ikawa, Shinichi Toyonaga, Hiromichi Nakabayashi, Kae Chang Park, Keiji Shimizu

Department of Neurosurgery, Kochi University Medical School, Nankoku, Kochi, Japan

Background: Surgical approach to a brain tumor is usually decided preoperatively based on conventional computed tomography (CT) scan, magnetic resonance imaging (MRI), and angiography, but individualized preoperative assessment of the relationship between critical structures is difficult because of poor visualization of these examinations.

Methods: MRI was SIGNA Horizon 1.5T (GE, Milwaukee, WI) and scanning conditions were as follows; T1-weighted-imaging (T1WI), scanning sequence of Fast SPGR, 1.5mm slice, slice gap of 0mm, TE/TR = 2.8/12 or 4.2/13, Matrix of 256×256, and 240mm field of view. CT scan was Aquilion TSX-101A (TOSHIBA medical systems, Tokyo, Japan) and scanning conditions were as follows; helical multi-slice scanning, 1.0mm slice, helical pitch of 15.0. Navigation system was Vecter Vision2 with plannging software of iPlan Cranial version 2.5 (Brain LAB, Heimstetten, Germany).

Case presentation: Intracranial metastasis from renal cell carcinoma was identified at the pituitary stalk compressing chiasma upward in a 72-year-old woman who complained of visual disturbance. The patient underwent tumor resection in the left pterional approach considering the tumor location of the left side dominancy, however, the surgery finished without touching the tumor due to too small working space. Then, we used preoperative three-dimensional (3D) planning of neuronavigation system with multimodal image fusion soft. The 3D image showed right side deviation of anterior communicating artery with elevated right A1 portion causing enough working space for tumor removal surrounded by right internal carotid artery, A1 segment, and optic nerve. The 3D image also proved the necessity to remove the right orbital roof and zygomatic arch for recognization of the highest attachment of the tumor to the brain tissue, and we performed right orbitozygomatic approach for this case. The operation finished as we planned preoperatively and total resection of the tumor was completed. The visual acuity of the patient recovered very well when she was discharged our hospital one month after the surgery.

Conclusions: Neuronavigation system is important part of both planning and performing intracranial surgery.

THROMBIN AND INTRACEREBRAL HEMORRAHGE-INDUCED NEUROGENESIS IN ADULT RATS

Shuxu Yang, Shuijiang Song, Guohua Xi, Takehiro Nakamura, Richard F. Keep, Julian T. Hoff, Ya Hua

Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA

Background and Aims: Neurogenesis is modulated by both physiological stimuli and pathological conditions, where it may contribute to brain recovery. However, neurogenesis in intracerebral hemorrhage (ICH) has not been investigated. Thrombin formation causes acute brain injury after ICH, but thrombin also can stimulate cell proliferation. The present study examined if neurogenesis takes place in ICH and the role of thrombin in ICH-related neurogenesis.

Methods: Male adult Sprague-Dawley rats were used in this study. This study was divided into four parts. In the first part, rats received either an intracaudate injection of 100-µl autologous whole blood (ICH) or a needle insertion (sham). The rats were killed 1, 3, 7, 14, 30 and 60 days later for Western blot analysis and immunohistochemistry of doublecortin (DCX). In the second part, rats had an ICH or a sham operation, and then received intraperitoneal injections of 5-bromo-2′-deoxyuridine (BrdU, 50 mg/kg) at day 7 and day 9 after ICH or sham operation. Brains were perfused at 14 days after ICH to identify BrdU-positive cells. In the third part, rats had an intracaudate injection of 50-µl thrombin (1 U) and brains were sampled for Western blots 3, 7, 14 and 30 days later. In the last part, rat received an injection of 100-µl blood with or without a thrombin inhibitor, hirudin (5 U). The brains were sampled at 14 days for DCX quantitation.

Results: The timecourse of DCX was examined by Western blot analysis. DCX levels in the ipsilateral basal ganglia started to increase as early as seven days after ICH (170 ± 49 % of sham), peaked at 14 days (805 ± 241 % of sham, p < 0.01) and then gradually decreased at one month (437 ± 161 %, p < 0.05) and two months (360 ± 104 %, p < 0.05). At two weeks after ICH, DCX protein levels in the ipsilateral basal ganglia were strongly increased (2722 ± 814 vs. 383 ± 50 pixels in sham, p < 0.01). Immunohistochemistry also demonstrated that DCX immunoreactivity was markedly increased in the ipsilateral subventricular zone and basal ganglia at two weeks after ICH. To confirm if the neuroblasts were newly generated after ICH, we performed immunofluorescence doublestaining for DCX and BrdU. We found that DCX positive cells were BrdU positive. To test the role of thrombin in neurogenesis, one unit thrombin, which does not cause marked brain injury, was injected into the caudate. Thrombin increased DCX levels in the ipsilateral basal ganglia at day 7 (p<0.05). We also found hirudin blocked ICH-induced upregulation of DCX in the ipsilateral basal ganglia (151±94 vs. 2689±837 pixels in the vehicle-treated group, p<0.05).

Conclusions: Our results demonstrated that neurogenesis occurs in the brain following ICH and that thrombin may play a role in ICH-induced neurogenesis.

NOTCH SIGNALING IS ACTIVATED IN ADULT HIPPOCAMPAL NEURONAL REGENERATION AFTER THE TRANSIENT GLOBAL ISCHEMIA

Soichi Oya¹, Gakushi Yoshikawa¹, Nobuhito Saito¹, Takaaki Kirino^2,3, Nobutaka Kawahara^1,3

¹Department of Neurosurgery, The University of Tokyo Graduate School of Medicine, Tokyo, Japan, ²Research Institute International Medical Center of Japan, Tokyo, Japan, ³SORST, Japan Science and Technology Corporation (JST), Tokyo, Japan

Introduction. Notch signaling has been shown to regulate neural stem cells in the developing brain. In general, activation of Notch signaling has two main effects. One is proliferation and maintenance of neural stem cells. The other is inhibition of neural differentiation of neural stem cells (NSC) in favor of their taking glial fate. While these facts have been confirmed first in in vitro settings and developing brains, the function of Notch signaling in the adult neurogenesis is recently becoming a big issue to be explained. Along with this idea, we were encouraged to investigate whether Notch signaling is involved in adult neurogenesis in the hippocampal CA1 after transient global ischemia. We also studied whether we could boost the therapeutic effect of growth factor treatment by manipulation of Notch signaling.

Materials and methods. Adult male Wister rats were subjected to 6-minute global ischemia using 4-vessel occlusion and hypotention. Epidermal growth factor (EGF) and fibroblast growth factor-2 (FGF2) were injected into the lateral ventricles from Day2 to Day5 to promote proliferation of NSCs responding to the ischemic insult. BrdU was injected intraperitoneally to label proliferating cells during this period. Subsequent intraventricular administration of the gamma secretase inhibitor was performed from Day5 to Day11 aiming at Notch signaling inhibition and augmented neuronal differentiation of NSCs. Rats were sacrificed at Day5 and Day28 to evaluate the change of Notch1 positive cells around the posterior periventricle and the NeuN positive cells in the CA1 region, respectively.

Results. The number of NeuN positive cells in the postischemic CA1 region decreased to 6% of that of control rats. The EGF/FGF2 treated rats, however, showed significant 22% increase in the number of NeuN positive cells in the CA1 region. At Day5, triple-fold increase of Notch1 positive cells were observed between the posterior periventricle and CA1 region in treated rats compared to non-treated rats. These cells co-expressed BrdU, Nestin, and Notch1. In rats subjected to gamma secretase therapy subsequent to EGF/FGF2, the number of NeuN positive cells showed small but significant increase of 28% compared to rats treated only by EGF/FGF2.

Conclusion. Increase of Notch1 positive cells along with growth factor treatment in the acute phase of ischemia indicates Notch signaling is involved in neuronal regeneration in the rat hippocampal CA1 after transient global ischemia. We also confirmed administration gamma secretase inhibitor in the subacute stage leads to improvement of cell regeneration. We further need to clarify how this treatment affects neuronal regeneration in adult neurogenesis in the CA1 region to assess whether Notch signaling could be a possible target for the cell regeneration therapy.

CELL THERAPY OF STROKE IN SMALL AND LARGE ANIMALS

Johannes Boltze^1,2, Annette Foerschler³, Uwe Richard Schmidt^1,2, D.C. Wagner^1,2, Daniel Christoph Schwarz⁴, Christiane Maria Boltze¹, Larissa Bulawina², Andrea Barbara Lang¹, Osama Sabri⁵, Henryk Barthel⁵, Frank Emmrich^1,2, Uwe Gille^6,7

¹Fraunhofer Institute of Cell Therapy and Immunology, Leipzig, Germany, ²Institute of Clinical Immunology and Transfusion Medicine, University of Leipzig, Leipzig, Germany, ³Department of Neuroradiology, University Clinic of Leipzig, Leipzig, Germany, ⁴Department of Neurology, University Clinic of Leipzig, Leipzig, Germany, ⁵Department of Nuclear Medicine, University Clinic of Leipzig, Leipzig, Germany, ⁶Institute of Veterinary Anatomy, University of Leipzig, Leipzig, Germany, ⁷VITA 34 INc., Leipzig, Germany

THE LOCALIZATION OF PHOSPHO-CREB IN NEUROBLASTS IN ISCHEMIA-INDUCED STRIATAL NEUROGENESIS

Tsutomu Sasaki, Kazuo Kitagawa, Emi Omura-Matsuoka, Kenichi Todo, Yasukazu Terasaki, Shiro Sugiura, Yoshiki Yagita, Masatsugu Hori

Division of Stroke Research, Department of Cardiovascular Medicine, Osaka, Japan

Background: Stroke leads to markedly enhanced neuronal progenitor proliferation in the subventricular zone and induce migration toward injured striatum. In hippocampal neurogenesis, the cAMP-CREB pathway regulates multiple aspects of adult hippocampal neurogenesis. Also, in the normal SVZ-Olfactory bulb, the implication of cAMP-CREB cascade has been elucidated. Neurotrophins play crucial roles in adult neurogenesis following ischemia. Brain-derived growth factor and insulin-like growth factor promote adult neurogenesis, and both growth factors possess the CRE sequence in their promoters. Therefore, we examined the profiles of phospho-CREB (pCREB) expression in stroke-induced striatal neurogenesis.

Subjects: 1) Transients focal ischemia: Middle cerebral artery was occluded by the intraluminal filament for 30 min. 2) BrdU labeling protocols: BrdU was used to label proliferating cells. BrdU was daily injected starting 2 days after ischemia for 2 weeks. Mice were killed at 1, 7, 14, and 42 days after BrdU injection, transcardially perfused with 4% PFA and immunohistochemistry was performed. Expression of phospho-CREB was evaluated. 3) Rolipram treatment: Rolipram (3 mg/kg, or vehicle, i.p.) was administered 3 days after BrdU labeling for 3 weeks. CREB-CRE cascade activation by rolipram was evaluated by CRE-lacZ transgenic mice. At 42 days after BrdU labeling, the striatal and hippocampal neurogenesis were compared.

Results: Msi-1-positive progenitor cells in the SVZ did not stain for pCREB. Most DCX- or PSA-NCAM-positive neuroblasts in the SVZ showed colocalization of pCREB staining. Also, the marked enhancement of phospho-CREB immunoreactivity was observed in the migrating neuroblasts toward the injured striatum. The treatment of rolipram activated the CREB-CRE signaling, and significantly enhanced the survival of nascent BrdU-positive neurons in the hippocampus, but rolipram only did not significantly enhance the striatal neurogenesis.

Conclusion: The enhanced phospho-CREB expression in neuroblasts plays a crucial role in both hippocampal and injured striatal neurogenesis. These results indicate that activation of cAMP-CREB signaling in combination with neurotrophins may have implications for therapeutic intervention to enhance neurogenesis after ischemia.

INTRAVENOUS TRANSPLANTATION OF MESENCHYMAL STEM CELLS DERIVED FROM PERIPHERAL BLOOD PROTECTS AGAINST CEREBRAL ISCHEMIA

Osamu Honmou, Ryo Ukai, Kiyohiro Houkin

Department of Neurosurgery, Sapporo Medical University, Sapporo, Hokkaido, Japan

Transplantation of mesenchymal stem cells prepared from adult bone marrow (BMSCs) has been reported to ameliorate functional deficits in several CNS diseases in experimental animal models. Despite the fact that the stem/precursor cells in peripheral blood are widely used for reconstruction in the hematopoietic system, it is not fully understood whether peripheral blood-derived precursor/stem cells (PMSCs) can differentiate into a neural lineage. To test the hypothesis that treatment with PMSCs may have a therapeutic benefit in stroke, we compared the efficacy of systemic delivery of BMSCs and PMSCs. A permanent middle cerebral artery occlusion in rat was induced by intraluminal vascular occlusion with a microfilament. Rat BMSCs and PMSCs were prepared in culture and intravenously injected into the rats 6 hours after MCAO. Lesion size was assessed at 6 hours, 1, 3 and 7 days using MR imaging and histology. The hemodynamic change of cerebral blood perfusion on stroke was assessed the same times using perfusion weighted image (PWI). Functional outcome was assessed using the treadmill stress test. Both BMSCs and PMSCs treated groups had reduced lesion volume, improved regional cerebral blood flow, and functional improvement compared to the control group. The therapeutic benefits of both MSC-treated groups were similar. These data suggest that PMSCs derived from peripheral blood could be an important cell source of cell therapy for stroke.

EXPRESSION OF NETRIN-1 AND ITS RECEPTORS DCC AND NEOGENIN IN RAT BRAIN AFTER ISCHEMIA

Atsushi Tsuchiya, Takeshi Hayashi, Kentaro Deguchi, Yoshihide Sehara, Toru Yamashita, HanZhe Zhang, Violeta Lukic, Makiko Nagai, Tatsushi Kamiya, Koji Abe

Department of Neurology, Graduate School of Medicine, Dentistry and Pharmacy, Okayama University, Okayama, Japan

Background: Once the central nervous system (CNS) was severely damaged, neurological deficit would permanently continue because tissue regeneration is quite limited. However, even if new neurons were not provided, a functional recovery should take place if axons grew after injury. It is therefore very important to investigate the mechanism of axonal growth in the ischemic brain in order to consider a novel mean of therapy for stroke. Chemotropic factors are important for axon pathfinding, and a growing number of chemotropic factors have been identified in the past several years. Netrins are chemotropic factors for axon with chemoattractant or chemorepellant guidance activities, and deleted in colorectal cancer (DCC) and neogenin are receptors for netrins.

Method: We examined expression of netrin-1, DCC, and neogenin in rat brain after 90 min of transient middle cerebral artery occlusion (tMCAO). The animals were sacrificed at 1, 7, 14, and 28 days after the ischemia, and immunohistochemistry for these molecules were carried out. In order to identify the cell phenotype which expressed them, we performed double staining for these molecules and cell type markers, which are NeuN, GFAP, and NAGO.

Results: Netrin-1 was expressed in the peri-ischemic area with a peak at 14 days (fig 1). Because axonal regeneration takes place at 14 days after injury, such expression could be involved in promotion of axonal growth after tMCAO. Although the origin of netrin-1 in normal adult brain had remained poorly understood, we showed it was neuron but not astroglial cell after the tMCAO. DCC was expressed both in neurons and astrocytic feet which peaked at 14days (fig 2). Although DCC expressed in neurons plays a role in promoting axonal growth, the role of DCC expressed in astrocytes remains uncertain. It could also be implicated in axonal growth under ischemia. The expression of netrin-1 and its receptor DCC were overlapped both in the spatial and temporal patterns, which indicate that netrin-1 bound to DCC and then could work for axonal growth. Neogenin was expressed in endothelial cells at the MCA territory, which was different from the cell phenotype with netrin-1 expression (fig 3). This suggests that neogenin's angiogenic activity may not have relevance with netrin-1, but rather some other molecules might be implicated.

Conclusion: Netrin-1-DCC system was activated in the ischemic brain, and should be involved in axon growth after injury.

SUBVENTRICULAR ZONE-DERIVED NEUROBLASTS MIGRATE AND DIFFERENTIATE INTO MATURE NEURONS IN THE POST-STROKE STRIATUM

Toru Yamashita¹, Mikiko Ninomiya², Takeshi Hayashi¹, M. Nagai¹, Tatsushi Kamiya¹, Nobuo Araki², Hideyuki Okano³, Kazunobu Sawamoto⁴, Koji Abe¹

¹Department of Neurology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan, ²Department of Neurology, Saitama Medical School, Saitama, Japan, ³Department of Physiology, Keio University School of Medicine, Tokyo, Japan, ⁴Bridgestone Laboratory of Developmental AndRegenerative Neurobiology, Keio University School of Medicine, Tokyo, Japan

In the normal brain, GFAP-expressing cells in the subventricular zone (SVZ) include a neurogenic cell population that gives rise to olfactory bulb neurons only. Herein, we report evidence that these cells can produce new neurons outside the olfactory bulbs after a stroke. SVZ GFAP-expressing cells labeled by a cell-type-specific viral infection method were found to generate neuroblasts that migrated toward the injured striatum. Long-term tracing experiment using a Cre-loxP system revealed the SVZ-derived neuroblasts differentiated into mature neurons in the striatum, in which they formed synapses with neighboring striatal cells.

TRANSFECTINON OF NOTCH-ICD INTO BONE MARROW STROMAL CELLS LEADS TO FUNCTIONAL RECOVERY IN A RAT MODEL OF STROKE

Makoto Hayase^1,2, Yasushi Takagi¹, Mari Dezawa², Kazuhiko Nozaki¹, Nobuo Hashimoto¹

¹Department of Neurosurgery, Kyoto University Graduate School of Medicine, Kyoto, Kyoto, Japan, ²Department of Anatomy and Neurobiology, Kyoto University Graduate School of Medicine, Kyoto, Kyoto, Japan

OBJECTIVE: We previously demonstrated that Neurons can be specifically induced from bone marrow stromal cells (MSCs) with extremely high efficiency using gene transfection of the Notch intracellular domain (Notch-ICD) and trophic factor administration. We investigated the efficacy of Notch-ICD transfected MSCs transplantation into rats subjected with focal cerebral ischemia. METHODS: Three days before transplantation, permanent focal cerebral ischemia was induced on Wistar rats. Notch-ICD transfected MSCs were locally transplanted into ischemic lateral striatum and cortex. The behavioral analyses were performed on days 4, 7, 10, and 14 after transplantation. RESULTS: transplantation of Notch-ICD transfested MSCs improved the score of modified limb placing test. Histological analysis revealed that transplanted cells expressed Tuj-1 and NeuN as neuronal marker. CONCLUSION: These results suggest that the use of MSCs transfected with Notch-ICD may be a promising therapeutic strategy for cerebral infarction.

CRITICAL ROLE OF PCREB SIGNALING FOR NEUROGENESIS AFTER CHRONIC CEREBRAL HYPOPERFUSION

Ryota Tanaka, Nobukazu Miyamoto, Yuji Ueno, Ning Zhang, Takao Urabe

Department of Neurology, Juntendo University School of Medicine, Tokyo, Japan

VASCULAR ENDOTHELIAL GROWTH FACTOR PROMOTES BRAIN TISSUE REGENERATION WITH A NOVEL BIOMATERIAL POLYDIMETHYLSILOXANE-TETRAETHOXYSILANE

HanZhe Zhang¹, Takeshi Hayashi¹, Kanji Tsuru², Kentaro Deguchi¹, Mitsuyuki Nagahara², Satoshi Hayakawa², Makiko Nagai¹, Tatsushi Kamiya¹, Akiyoshi Osaka², Koji Abe¹

¹Department of Neurology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan, ²Biomaterial Laboratory, Graduate School of Natural Science and Technology, Okayama University, Okayama, Japan

Background: Injury of the central nervous system (CNS) of mammals may cause irreparable damage due to the limited regeneration capacity of the tissue. In order for the CNS tissue to regenerate, to supply a scaffold for the newly produced cells to adhere is necessary. Application of biodegradable polymers such as collagen, gelatin, fibrin and alginates has been investigated for organ damage other than the CNS. In this context, we recently produced a novel biomaterial from polydimethylsiloxane (PDMS) and tetraethoxysilane (TEOS). We added the most potent angiogenic factor vascular endothelial growth factor (VEGF) and observed the difference in brain regenerative process in the PDMS-TEOS scaffold.

Methods: Male Wistar rats of 12 weeks old were anesthetized and placed in a stereotaxic apparatus. A piece of bone over the right frontal cortex was excised, and 2 × 5 mm defect in the brain was made by cutting and aspiration. The defect was positioned from 1.0 mm anterior to 4.0 mm posterior of the bregma and 2.0 to 4.0 mm lateral to the midline. The depth of the lesion was 2.0 mm from the brain surface. The block of PDMS-TEOS scaffold, with or without VEGF, was placed into the lesion immediately after making the defect. The animals were sacrificed at 30 days after PDMS-TEOS scaffold implantation. For histological analysis of the newly formed tissue, the sections were stained with HE. In order to identify cell types in the newly formed tissue, we performed immunohistochemical analysis for glial fibrillary acidic protein (GFAP), an astrocyte marker, and N-acetylglucosamine oligomer (NAGO), an endothelial cells marker. In order to determine whether the cells in PDMS-TEOS scaffold were under mitosis, we carried out double fluorescent study for cell phenotype markers and Ki67.

Result: The PDMS-TEOS scaffold remained at the implanted site for 30 days and kept the integrity of brain shape (Fig. A and B). HE staining showed that a small number of cells infiltrated into the PDMS-TEOS scaffold even without VEGF at 30 days after the implantation, but that addition of VEGF significantly increased the number of infiltrated cells NAGO staining showed only a few positive cells in the PDMS-TEOS scaffold without VEGF, but significant increase of endothelial cells with VEGF was confirmed. Immunohistochemical study for GFAP demonstrated that only a few astrocytes were found in the PDMS-TEOS scaffold without VEGF, which was again significantly increased with VEGF. Double staining with proliferation maker Ki67 demonstrated that VEGF significantly increased newly formed astrocytes and endotheial cells, indicating that addition of VEGF accelerated tissue restoration and angiogenesis.

Conclusions: The present study showed that a new porous PDMS-TEOS is a good candidate biomaterial for brain tissue restoration. Endothelial as well as glial cells successfully infiltrated into this biomaterial. Addition of VEGF further promoted new tissue formation.

ENCAPSULATED BDNF-PRODUCING CELL GRAFT HAS A NEUROPROTECTIVE EFFECT AND ENHANCES NEUROGENESIS IN THE DAMAGED HIPPOCAMPUS

Tetsuro Shingo, S. Kuramoto, W. Yuan, A. Kondo, T. Kadota, N. Tajiri, T. Morimoto, H. Yoshida, Y. Miyoshi, I. Date

Department of Neurological Surgery, Okayama University Graduate School, Okayama, Japan

[Background] It has been well known for that brain-derived neurotrophic factor (BDNF) influences the birth of granule cells in the dentate gyrus, which is one of the areas in the brain that demonstrates neurogenesis throughout life, and also contributes to the development of temporal lobe epilepsy (TLE). Encapsulation of cells has the potential to provide a protective barrier against host immune cell interactions after grafting and to supply the neurotrophic factors or neurotransmitters in the stable fashion throughout the brain. We investigated the effect of the continuous and low dose BDNF supply to the hippocampus of the epilepsy model rat using the encapsulated BDNF-producing cells that were implanted in the striatum.

[Methods] We introduced BDNF complementary DNA into baby hamster kidney (BHK) cells and established a cell line that produces BDNF continuously. BHK-BDNF cells and BHK cells that had been transfected with an expression vector that did not contain BDNF cDNA (BHK-control) were encapsulated into the hollow fibers. Each capsule was implanted into the striatum of Fisher 344 rat. Seven days after capsule implantation, bromodeoxyuridine (BrdU), which labeled the proliferated cells in the subventricular zone or dentate gyrus of hippocampus, was administered seven times in a day. Next day we injected kainate 1 microgram solved in PBS, or PBS only into the rat lateral ventricle stereotactically. The evolutions of epilepsy following kainate treatment were monitored by video-EEG. Two weeks after lesion, rats were processed immunohistochemically.

[Results] Prior to implantation, we checked, using ELISA, to see whether BDNF was continuously released at a rate of more than 5 ng/capsule/day from encapsulated BHK-BDNF cells. Although the level of BDNF from the BHK-BDNF capsules following explantation was lower than that prior to implantation, the retrieved BHK-BDNF capsules persisted in producing approximately 3 ng of BDNF per day up to 3 weeks after transplantation. BDNF-producing capsule transplantation enhanced the proliferation of BrdU/PSA-NCAM positive cells in the subventricular zone and subgranular zone of the normal rats. In epilepsy rats, low dose of BDNF protected hippocampal neurons (CA1 and 3) against kainate administrations, and epileptic spikes were improved on the EEG. BrdU positive cells also increased in BHK-control group, but most of the divided cells did not differentiate into neurons or neuronal progenitor cells. Some of BrdU positive cells in the granule cell layer migrated toward the CA3 regions.

[Conclusion] These results suggest encapsulated BDNF-producing cells promote neurogenesis and assist the repair of hippocampal neurons from damage.

A NOVEL STIMULANT OF NEUROGENESIS FOLLOWING MOUSE BRAIN ISCHEMIA VIA PI3K/AKT AND ERK ACTIVATION IN THE SUBGRANULAR ZONE OF HIPPOCAMPUS

Kohji Fukunaga, Norifumi Shioda

Department of Pharmacology, Graduate School of Pharmaceutical Sciences and The 21st Century COE Program, CRESCENDO, Tohoku University, Sendai, Japan

Neurogenesis is well documented in the subgranular zone (SGZ) of hippocampus. Especially, in the hippocampal, crucial role of neurogenesis in learning and memory formation has been addressed. The neurogenesis in the hippocampus is regulated by environmental stress, seizure and stroke. However the mechanism underlying neurogenesis in the adult brain remains to be unclear. We here assessed whether a novel stimulant of phosphatidylinositol 3-kinase (PI3K)/Akt and extracellular signal-regulated kinase (ERK) pathways promotes brain ischemia-induced neurogenesis in a mouse transient middle cerebral artery occlusion (MCAO) model. Intraperitoneal administration of a vanadyl compound, which is a potent inhibitor for protein tyrosine phosphatases, stimulated neurogenesis in SGZ following brain ischemia. The drug led to activation of PI3K/Akt and ERK pathways in SGZ and those activations were inhibited by treatment with specific inhibitor, wortmannin or U-0126, respectively. Likewise, wortmannin or U-0126 treatment totally abolished both ischemia- and the vanadyl compound-induce neurogenesis. In behavioral pharmacological studies, the vanadyl compound-induced neurogenesis was closely associated with improvement of memory deficits following brain ischemia. Taken together, peripheral administration of a novel vanadyl compound stimulates neurogenesis following brain ischemia through PI3K/Akt and ERK activation in the SGZ, thereby improving ischemia-induced memory deficits.

DISTRIBUTION OF INDUCIBLE NITRIC OXIDE SYNTHASE AND CELL PROLIFERATION IN RAT BRAIN AFTER TRANSIENT MIDDLE CEREBRAL ARTERY OCCLUSION

Yoshihide Sehara, Takeshi Hayashi, Kentaro Deguchi, Atsushi Tsuchiya, Toru Yamashita, Hanzhe Zhang, Violeta Lukic, Tatsushi Kamiya, Koji Abe

Department of Neurology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan

Background: Nitric oxide (NO) is a physiological messenger synthesized by NO synthase (NOS), including neuronal nNOS, endothelial eNOS, mitochondrial mtNOS, and inducible iNOS. All central nervous system (CNS) cells synthesize NO by oxidation of L-arginine, in a reaction catalyzed by NOS. NO can be neuroprotective or neurotoxic in cerebral ischemia, depending on the NOS isoform involved. In addition to direct neurotoxic effect in ischemic brain, iNOS adversely affect neurological outcome by blocking neurogenesis. In the present study, therefore, we studied the chronological and spatial change of the distribution of iNOS and cell proliferation in subventricular zone (SVZ) after transient focal cerebral ischemia. Methods: Adult male Wistar rats (12 weeks old) were used in this study. The right middle cerebral artery (MCA) was transiently occluded through the common carotid artery for 90 min (tMCAO). In order to estimate the number of proliferating cells, 50 mg/kg of bromodeoxyuridine (BrdU) were intraperitoneally injected at 12 and 24 hours before decapitation. The rats were sacrificed at 1, 3, 7 and 21 days after tMCAO (n = 6, each). Sham-operated rats were sacrificed at 7 days after the sham operations (n = 6). The brains were quickly removed, frozen in powdered dry ice and coronal sections of 10 µm thickness were prepared. These sections were obtained from caudoputamen level, and were identical among the investigated animals. Results: After 90 min of tMCAO, iNOS positive cells decreased in the ischemic core at 1 to 21 d, and increased in the ipsilateral periischmic area at 1 and 3 d. BrdU positive cells appeared in the ischemic core at 3 to 21 d, in the periischemic area at 3 and 7 d, and increased in the ipsilateral SVZ at 7 d. ED-1 positive cells appeared in the ischemic core at 3 to 21 d, and some of them were double positive for BrdU or iNOS. It was thus indicated that the major source of iNOS in the ischemic core was macrophages. In the periischemic region, on the other hand, iNOS expressing cells were positive either for NeuN or GFAP, indicating that neurons and astrocytes expressed iNOS. Conclusions: The chronological change of iNOS expression was summarized in the Figure. iNOS expression preceded appearance of BrdU positive cells. It is possible to consider that iNOS inhibits cell proliferation in the ischemic brain.

UPREGULATION OF SYNTAXIN1 IN THE ISCHEMIC CORTEX FOLLOWING PERMANENT FOCAL ISCHEMIA IN RATS

Fang Cao, Ryuji Hata, Pengxiang Zhu, Masahiro Sakanaka

Department of Functional Histology, Ehime University Graduate School of Medicine, Toon, Ehime, Japan

THERAPEUTIC EFFECT OF PIGF GENE-MODIFIED HUMAN MESENCHYMAL STEM CELLS ON STROKE MODEL

He Liu¹, Osamu Honmou¹, Kuniaki Harada¹, kiminori Nakamura², Kiyohiro Houkin¹, Hirofumi

Hamada², J.D. Kocsis³

¹Department of Neurosurgery, Sapporo Medical University School of Medicine, Sapporo, Japan, ²Department of Molecular Medicine, Sapporo Medical University School of Medicine, Sapporo, Japan, ³Neuroscience Research Center, VA Medical Center, West Haven, CT, USA

Background: Intravenous delivery of mesenchymal stem cells (MSCs) into rat cerebral ischemia models has demonstrated reduced infarction size and improved functional outcome. Placental growth factor (PlGF) is angiogenic to impaired non-neural tissue. We transfected hMSCs with the PlGF gene using a fiber-mutant F/RGD adenovirus vector and investigated whether PlGF gene-modified hMSCs (PlGF-hMSCs) could contribute to the functional repair in a rat permanent middle cerebral artery occlusion (MCAO) model.

Method: hMSCs and PlGF-hMSCs were intravenously infused into the rats 3 hours after MCAO. MRI studies, histological examination and behavioral studies were performed to evaluate the therapeutic benefits with treatment.

Results: ELISA analysis of the infarcted hemisphere revealed an increase in PlGF in both hMSC groups, but a greater increase in the PlGF-hMSC group. After 7 days of MCAO, both hMSCs and PlGF-hMSCs reduced lesion volume and apoptosis, induced angiogenesis, and elicited functional improvement compared to the control sham group, but the effect was greater in the PlGF-hMSC group.

Conclusion: Thus we suggest that the hypothesis that PlGF contributes to neuroprotection and angiogenesis in cerebral ischemia and cellular delivery of PlGF to the brain can be achieved by intravenous delivery of hMSCs.

CELL THERAPY FOR CEREBRAL ISCHEMIA IN SUBACUTE STAGE: TRANSPLANTATION OF EMBRYONIC STEM CELLS-DERIVED NEURAL PROGENITORS SORTED BY FLUORESCENCE-ACTIVATED CELL SORTING

Motoaki Fujimoto, Hideki Hayashi, Yasushi Takagi, Jun Takahashi, Kazuhiko Nozaki, Nobuo Hashimoto

Department of Neurosurgery, Kyoto University Graduate School of Medicine, Kyoto, Japan

Cerebral ischemia causes neuronal death and disruption of neural circuit in the central nervous system. Various neurological disorders caused by cerebral infarction can be potentially fatal and keep impairing the quality of life. Recent treatment strategy for ischemia focuses just on early reperfusion and prevention of neuronal death in acute stage of infarction. Functional recovery in chronic stage mainly depends on physical treatment and rehabilitation. We aim to establish the cell therapy for cerebral ischemia, using embryonic stem (ES) cells. ES cells have self-renewaling and pluripotentive capacity. We reported the transplanted monkey and mouse-ES cells could survive and differentiate into various types of neurons and glial cells, and make the neuronal network in basal ganglia. In this report, we transplanted the neural progenitors derived from ES cells into the ischemic mouse putamen at 1week after transient middle cerebral artery occlusion. We inspired the differentiation of the neural progenitors from mouse ES cells using the serum-free suspension culture (SFEB culture) method. In this method, telencephalic neural progenitors could be effectively induced from the floating aggregates of ES cells. Furthermore, to prevent tumor formation, we purified the progenitors expressing the neural progenitor marker Sox1 by fluorescence-activated cell sorting (FACS). At 1 month after transplantation we analyzed the survival and differentiation of the transplanted cells. In addition we estimated the network formation and the functional recovery from ischemic damage.

ENDOGENOUS THROMBOSPONDINS 1 AND 2 ARE NECESSARY FOR SYNAPTIC PLASTICITY AND SPONTANEOUS FUNCTIONAL RECOVERY AFTER STROKE

Jason Liauw¹, Stan Hoang¹, Cagla Eroglu², Michael Choi¹, Raphael Guzman¹, Tonya Bliss¹, Ben

Barres², Gary K. Steinberg¹

¹Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA, ²Department of Neurology and Neurological Sciences / Stanford University School of Medicine, Stanford, CA, USA

Background/Aims: Thrombospondins 1 and 2 (TSP-1/2) belong to a family of secreted multimeric extracellular glycoproteins and have angiostatic as well as synaptogenic properties. While the robust expression of TSP-1/2 that occurs after stroke has been postulated to drive the resolution of post-ischemic angiogenesis, little is known about the role TSP-1/2 may play in synaptogenesis and subsequent functional recovery after stroke. In this study, we investigate whether TSP-1/2 is necessary for the spontaneous synaptic and motor recovery that occurs after stroke.

Methods: An ischemia model was generated in 8 to 12 weeks old wild-type (n=10) and TSP-1/2 knockout (KO) mice (n=8) by unilateral occlusion of the distal middle cerebral artery. Spontaneous recovery of motor functions was assessed through rotarod, tongue protrusion, sticky-tape, and limb asymmetry behavioral paradigms for four weeks after stroke. At 4 weeks post-stroke, synaptic quantification was performed by staining sections with mouse anti-synaptophysin (pre-synaptic) and rabbit anti-PSD-95 (post-synaptic). Synaptic puncta, defined as co-localization of pre-and post-synaptic markers, were identified using ImageJ. Synaptic densities were compared in the penumbra region between wild-type and TSP-1/2 knock out mice (n=5 per group). Biotinylated dextran-amine (BDA) was injected into the contralesional homotopic cortex to quantify axonal sprouting from the contralesional homotopic cortex to the ipsilesional penumbra and striatum.

Results: TSP-1/2 KO mice exhibited significant deficits in their ability to recover motor function following ischemic brain injury as compared to their WT counterparts in the tongue protrusion test at week 4 (p<0.05). Moreover, synaptic staining revealed reduced synaptogenesis in the penumbra of TSP-1/2 KO mice as compared to that seen in homologous areas of WT mice (p<0.05). BDA staining also showed reduced axonal sprouting in the penumbra region from the contralesional cortex in TSP-1/2 KO mice compared to that in wild-type.

Conclusion: These results demonstrate that functional recovery after stroke requires TSP-1/2 expression. TSP-1/2 expression may also play a necessary role in neuroplasticity after stroke given that a deficiency of TSP-1/2 results in a reduction of synaptogenesis and axonal sprouting. An understanding of the mechanisms underlying post-stroke neuroplasticity may improve on therapies aimed at augmenting recovery after stroke.

IMPROVED BRAIN OXYGEN UTILIZATION DURING GRADED ASPHYXIA IN NEWBORN PIGLETS WITH INTRAUTERINE GROWTH RESTRICTION

Reinhard Bauer^1,3, Bernd Walter², Ulrich Brandl³

¹Institute of Molecular Cell Biology, University Hospital, Friedrich Schiller University, Jena, Germany, ²Department of Neurosurgery, Zentralklinik Bad Berka, Bad Berka, Germany, ³Department of Neuropediatrics, University Hospital, Friedrich Schiller University, Jena, Germany

Background and aims: There is insufficient knowledge referring to the effects of intrauterine growth restriction (IUGR) on metabolic regulation of cerebral blood flow (CBF) in newborns. Recently, we have shown that IUGR resulted in an improved ability to withstand critical periods of gradual oxygen deficit as shown by improved cerebrovascular autoregulation at hemorrhagic hypotension in an animal model of naturally occurring IUGR in term born piglet*. Furthermore, IUGR is associated with an increased incidence of perinatal asphyxia, which results in cardiocirculatory redistribution. However, it is unknown whether IUGR can affect brain O2 availability under asphyxic conditions.

Methods: Studies were conducted to examine the effects of IUGR on CBF regulation and oxygen consumption (CMRO2) at different stages of asphyxia (hypercapnic hypoxemia) and pure hypoxia (normocapnic hypoxemia). We used 1-day old anesthetized piglets (n=85). Animals were divided into normal-weight piglets (NW, n=42; BW:1522 ± 125g) and IUGR piglets (n = 43, BW:826 ± 51g) according to their birth weight. Correct allocation was confirmed afterwards by estimation of the brain to liver ratio. CBF was measured by colored microspheres. Brain AVDO2 was determined using arterial and sagittal sinus blood samples (CMRO2 = CBF × AVDO2). Different stages of hypoxemia were induced for 1-h by appropriate FiO2 lowering (moderate hypoxia: paO2=31-34 mmHg, severe hypoxia: paO2=20-22 mmHg). Fourteen NW and sixteen IUGR piglets received additionally 6% CO2 to the breathing gas, so that a paCO2 of 74-80 mmHg resulted (asphyxia groups). Eight NW and nine aIUGR animals served as untreated control. In addition, three NW and three IUGR piglets were used for brain COX-2 and eNOS estimation by Western blotting. Furthermore, affinity of hemoglobin for oxygen under hypoxic and asphyxic conditions were measured in blood samples from fifteen NW and fourteen IUGR newborn piglets by estimation of individual oxygen dissociation curves and calculation of halfsaturation oxygen pressure (P50) values from linear interpolations using the Hill plot.

Results: Systemic cardiovascular and blood gas parameters, as well as hypoxia- related effects on alteration in CBF regulation and CMRO2 were similar in NW and IUGR piglets: At pure moderate and severe O2 deficit a marked CBF increase, maintained O2 availability and aligned O2 utilization resulted in maintained O2 consumption. In contrast, during asphyxia cerebral oxygen extraction was markedly increased in IUGR animals (P < 0.05). This resulted in a significantly diminished CMRO2-related increase of CBF at gradually reduced arterial oxygen content (P < 0.05). A marked difference in brain COX-2 and eNOS content could not be verified between NW and IUGR piglets. However, blood derived from IUGR piglets showed an increased O2 affinity by 17% at asphyxic conditions (P < 0.05).

Conclusions: An enhanced effectivity in oxygen availability appeared in newborn IUGR piglets under graded asphyxia by improved cerebral oxygen utilization. This is connected with altered O2 affinity of hemoglobin leading to increased oxygen transport capacity under asphyxic conditions. We suggest that IUGR newborns are more capable of protecting the brain against O2 loss during asphyxia than NW neonates.

POSTNATAL MATURATION ENHANCES CALCIUM INFLUX, AND REDUCES MYOFILAMENT CALCIUM SENSITIZATION IN MYOGENIC RESPONSES TO CEREBRAL ARTERY STRETCH

Renan Sandoval, Elisha Injetti, James Williams, William Pearce

Department of Physiology, Center for Perinatal Biology, Loma Linda University School of Medicine, Loma Linda, CA, USA

CEREBRAL BLOOD FLOW, CEREBRAL OXYGEN METABOLISM, AND ATROPHY IN HEALTHY AGING

Manouchehr S. Vafaee¹, Per Borghammer¹, Joel Aanerud¹, Paul Cumming², Albert Gjedde^1,2

¹PET Center, Aarhus University Hospital, Aarhus, Denmark, ²CFIN, Aarhus University, Aarhus, Denmark

Introduction: It is debatable where and to what extent cerebral blood flow (CBF) and cerebral oxygen metabolism (CMRO2) decline with healthy aging. It is possible that neocortical atrophy could affect the measurements of these physiological factors in addition to the direct effects of aging on blood flow and metabolism. To tackle these issues and to shed some light on the cerebrometabolic changes following normal aging, we completed a series of CBF/CMRO2 PET studies on a healthy group.

Material and methods: 12 healthy young males and females, aged (26±4 yrs), and 16 elderly healthy subjects, aged (58y±5 yrs), underwent one CBF and two identical CMRO2 quantitative 3-min PET scans in a state-of-rest. The cortical grey matter was extracted from the subjects MRI scans by means of an automated segmentation algorithm. We also performed a voxel-based subtraction of young and old and a voxel-wise regression of CBF and CMRO2 versus age. The magnitudes of CMRO2 and CBF were determined with the two-compartment weighted-integration methods. Mean subtracted image volumes (elderly minus young) were converted to z-statistic volumes. Significant focal differences of CBF and CMRO2 and grey matter intensity were identified on the basis of the 3D Gaussian random field theory and cluster analysis.

Results: The average whole brain CBF in young subjects (53±5ml/min.hg) did not differ from the elderly population (52±6ml/min.hg). However, the average global CMRO2 in aged subjects (154±24µmol/min.hg) was lower than in the young subjects (166±13µmol/min.hg) but did not reach significance (p=0.1). Oxygen consumption had decreased more progressively than CBF in some regions and predominantly so in the neocortex, although clusters of CBF and CMRO2 decline overlapped extensively. CBF and CMRO2 decreased in most of the frontal lobe (except orbitofrontal cortex, M1, PMA and SMA), and L/R lateral parietal cortex and L posterior cingulate. Oxygen decline was more pronounced in the pre-central gyrus than CBF. A somewhat similar and significant pattern of decrease in grey matter intensity was seen, primarily in the frontal and lateral parietal cortices.

Discussion: While it is accepted that many changes occur in the brain as a result of normal aging, it is not fully agreed to what degree and in which regions of the brain those changes take place. Our data indicate that global CMRO2 decreases (although not significantly) without concomitant decrease in global CBF. If so, the finding confirms that oxygen consumption is a more accurate index of neuronal work than blood flow. We observe overlapping areas of decreased CBF and CMRO2 and decline in grey matter intensity. It is thus likely that neocortical atrophy contributes to the decreased signal in these PET scans. However, the data also revealed non-overlapping areas. The decreases of CBF and CMRO2 in healthy aging are most likely due to combinations of atrophy and direct changes, such as functional deafferentation, changes in neurotransmitter systems, and cell death following healthy aging. The metabolic depression as a result of healthy aging suggests a dysfunction of neuronal circuits that underlie some age-related changes in cognitive function.

EVIDENCE OF ENDOTHELIAL DYSFUNCTION IN PRIMARY LATERAL SCLEROSIS (PLS) WITH COGNITIVE IMPAIRMENT AS MEASURED BY CT PERFUSION

Matthew Murphy^1,2,3, Gloria Grace⁴, Michael Strong⁴, Carmela Tartaglia⁴, J.B. Orange⁵, Xiaogang Chen², Ann Rowe⁴, Karen Findlater⁴, Ting-Yim Lee^1,2,3

¹Department of Medical Biophysics, University of Western Ontario, London, ON, Canada, ²Imaging Research Laboratories, Robarts Research Institute, London, ON, Canada, ³Imaging Division, Lawson Health Research Institute, London, ON, Canada, ⁴Department of Clinical Neurological Sciences, London Health Sciences Centre, London, ON, Canada, ⁵School of Communication Sciences and Disorders, University of Western Ontario, London, ON, Canada

BACKGROUND: Primary lateral sclerosis (PLS) is a rare neurodegenerative disorder of the upper motor system that may also present with clinically significant cognitive impairment. We propose that mean transit time (MTT), or the ratio of cerebral blood volume (CBV) to cerebral blood flow (CBF), a surrogate indicator of cerebral perfusion pressure and hence vascular tone, can be used to investigate endothelial dysfunction in PLS patients with cognitive impairment.

METHODS: Nine male and nine female PLS patients aged 44 to 72 (mean, 59.11 ± 8.67) years were studied with CT Perfusion, a contrast enhanced dynamic CT scanning method which measures cerebral blood flow (CBF), blood volume (CBV) and MTT, on the same day of neuropsychological testing. Eight of these patients had concomitant cerebrovascular disease and/or diabetes and were excluded from the analysis. The remaining 10 patients, (5 men, 5 women) with an age range of 44 to 69 years (mean, 58.4 ± 9.6 years), were then stratified into two groups based on neuropsychological testing: (1) cognitively-intact PLS patients (PLSn; those having zero or one abnormal scores (n=7)); and, (2) cognitively-impaired PLS patients (PLSci; those having greater than 2 abnormal test scores (n=3)). These data were then compared with 4 male and 2 female normal controls aged 34 to 63 (mean, 52.9± 9.9) years. CT images were segmented into the four lobes of the brain with each lobe further subdivided into grey and white matter (GM and WM, respectively). CBF, CBV and MTT for GM and WM in each lobe were compared across groups using ANOVA with post-hoc Tukey tests and significance declared at P < 0.05.

RESULTS: PLSci showed increased CBV and MTT with reduced CBF relative to PLSn. MTT was significantly increased in the PLSci group in all regions and the increase affected both grey and white matter, with the exception of sparing of the temporal lobe white matter. No differences in MTT existed upon inclusion of the full cohort of patients including those with concomitant cerebrovascular disease and/or diabetes.

CONCLUSIONS: As demonstrated by the current study, cognitive impairment in the background of neurodegeneration found in PLS may affect neurovascular coupling leading to disturbances in local cerebral hemodynamics and the increase in MTT of PLSci patients. MTT, being the ratio of CBV to CBF, is sensitive to changes in vascular tone. The observed increase in MTT, therefore, may signify abnormal vascular tone and hence endothelial dysfunction in PLSci patients.

SEROTONIN TRANSPORTER BINDING IN HEALTHY TWINS AT HIGH RISK OF DEVELOPING MOOD DISORDER IN COMPARISON WITH TWINS AT LOW RISK

Vibe G. Frokjaer¹, Maj Vinberg², David Erritzoe¹, William Baare³, Jacob Madsen⁴, Lars V. Kessing², Gitte M. Knudsen¹

¹Neurology Research Unit, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark, ²Psychiatric Department, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark, ³Danish Magnetic Resonance Center, Hvidovre University Hospital, University of Copenhagen, Copenhagen, Denmark, ⁴PET and Cyclotron Unit, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark

STATISTICAL PARAMETRIC MAPPING ANALYSIS OF RELATIONSHIP BETWEEN REGIONAL BLOOD FLOW AND SYMPTOM CLUSTERS OF DEPRESSIVE MOOD IN CHRONIC KIDNEY DISEASE

Sang Heon Song¹, Seong-Jang Kim², Ji Hoon Kim³

¹Department of Internal Medicine, Pusan National University Hospital, Busan, South Korea, ²Department of Nuclear Medicine, Pusan National University Hospital, Busan, South Korea, ³Department of Psychiatry, Pusan National University Hospital, Busan, South Korea

Background: To investigate the relationship between regional cerebral blood flow (rCBF) and symptom clusters of depressive mood in pre-dialytic chronic kidney disease (CKD). Methods: Twenty seven patients with stage 4–5 CKD were performed statistical parametric mapping (SPM) analysis of brain SPECT. Correlation analyses between separate symptom clusters of depressive mood and rCBF were done. Results: The first factor, depressive mood, was negatively correlated with rCBF in right insula, posterior cingulated, and left superior temporal gyrus, and positively correlated with rCBF in left fusiform gyrus. The second factor, insomnia, was negatively correlated with rCBF in right middle frontal gyrus, both cingulate gyrus, right insula, right putamen, and right inferior parietal lobule, and positively correlated with rCBF in left fusiform gyrus and both cerebellar tonsil. The third factor, anxiety and psychomotor aspect, was negatively correlated with rCBF in left inferior frontal gyrus, right superior frontal gyrus, right middle temporal gyrus, right superior temporal gyrus, and left superior frontal gyrus, and positively correlated with rCBF in right ligual gyrus and right parahippocampal gyrus. Conclusion: In this study, the separate symptom clusters were correlated with specific rCBF pattern similar to areas that involved in major depressive disorder patients without CKD. However, discordant areas with rCBF of major depressive disorder patients were revealed, too. Further investigation of larger scale should be needed.

VOXEL BASED STATISTICAL ANALYSIS OF REGIONAL CEREBRAL GLUCOSE METABOLISM IN CHILDREN WITH ATTENTION-DEFICIT HYPERACTIVITY DISORDER

Jihoon Kim¹, Seong-Jang Kim², Jung Sub Lee³

¹Department of Child-Adolescent Psychiatry, Pusan National University Hispital, Busan, South Korea, ²Department of Nuclear Medicine, Pusan National University Hispital, Busan, South Korea, ³Department of Orthopedic Surgery, Pusan National University Hispital, Busan, South Korea

Voxel based statistical analysis was conducted to identify cerebral metabolic differences between children with Attention-deficit Hyperactivity disorder (ADHD) and normal controls. F-18 FDG-PET brain scans were compared between 14 drug- naive children with ADHD and 15 drug-naive normal controls using statistical parametric mapping (SPM). Decreased glucose metabolism were found in left parahippocampal gyrus, left hippocampus, left anterior cingulate gyrus, right anterior and posterior lobes of cerebellum, left superior temporal gyrus, left insula, left medial and middle frontal gyri, right medial frontal gyrus, left basal ganglia (putamen, amygdala and caudate nucleus) in children with ADHD. Also, ADHD patients showed hypermetabolism in both supramarginal gyri. Current study suggests that children with ADHD have more dysfunction relatively large area of brain compared to normal controls.

DOES ALZHEIMER'S DISEASE REALLY EXIST?

Frederick Peng

Department of Neurosurgery & Neurological Institute, Taipei VGH, Taipei, Taiwan-R.O.C.

Background and Aims: This presentation aims to rectify the historical mistake committed by Kraepelin; it has come down as Alzheimer's disease for almost a century. The eponym was not based on scientific merits but, rather, on power play. There were pieces of evidence: (1) the patient had at least four organic brain diseases; (2) Alzheimer did not treat the patient, nor did he perform the autopsy and prepare the autopsied materials for oral presentation and publication, but (3) Oskar Fischer (1907) had a far superior contribution which has been either overlooked or misunderstood.

Methods: I have undertaken a critical literature review from a historical perspective, including Alzheimer (1907) and Fischer (1907)along with Bonfiglio (1908), Perusini (1910;1911), Kraepelin (1910) and many other selected publications

Results: Alzheimer's disease (AD) has been known since 1910. It is claimed to consist of two hallmarks: senile plaques and neurofibrillary tangles. Upon reviewing the literature, I have found that this claim is utterly false; instead, the patient had at least four brain diseases: arteriosclerosis, evenly atrophic brain, senile plaques, and neurofibrillary tangles; I also suspect that she had DM and decubitus angina.

The false claim has been taught at every medical school throughout the world. However, the historical facts indicate that Alzheimer NEVER made any original histopathologic description of senile plaques in the patient, although he did briefly describe “the transformation of fibrils.” In fact, he did not even treat the patient; instead, he assigned her to his junior colleague, Gaetano Perusini, who treated her for just seven months (November 1901-June 1902), although most subsequent investigators believe that Alzheimer treated her from 1901 till her death in 1906. The reason is that Alzheimer quit his job in Frankfurt in June 1902 to join Kraepelin in Heidelberg; the two then moved to Munich in 1903. In other words, the materials for his oral presentation in November, 1906 at a conference of German psychiatrists in Tubingen and for his 1907 paper were entirely collected and prepared by Perusini. On the other hand, Oskar Fischer made the most extensive and detailed description in 1907 of 16 cases with eight histopathologic illustrations of both diseases.

My thorough examination of the two publications in 1907 by Fischer, whose article was a lengthy one of 12 pages, and Alzheimer, whose case report was barely two-page long, has enabled me to compare them point-by-point in this presentation. I declare that it is unfair to Fischer and the general public to allow the false claim to stand unchallenged. Even Perusini changed his mind in 1911 and followed Vedrani (1910;1911), by attributing senile plaques and presbyophrenia to Fischer as Fischer's Plaques and Fischer's Presbyophrenia.

Conclusions: The eponym, AD, is a misnomer, created by Kraepelin for his self interest, which has misled the entire medical field for almost a century. In fairness to Fischer and to do justice to the scientific community, I call for a new term to replace AD which never existed, does not exist, and will never exist.

SIGNIFICANT ASSOCIATION BETWEEN LEUKOARAIOSIS AND METABOLIC SYNDROME IN HEALTHY PEOPLE

Kaechang Park, Shoko Yamada, Hiromichi Nakabayashi, Shinichi Toyonaga, Noritaka Masahira,

Eiiichi Nakai, Naoki Ikawa

Department of Neurosurgery, Kochi Unversity Medical School, Kochi, Japan

Introduction: LA has been considered as an intermediate substitute of ischemic brain damages. MetS attracts attention as a risk factor for cardiovascular and cerebrovascular diseases. However, no study has described the relationship of MetS to LA in healthy individuals derived from various age groups.

Methods: We studied 3,070 persons at ages between 28 and 78 years (mean, 52.7 years) with no past history of stroke who visited a healthcare facility in Kochi prefecture, Japan, from September 2005 through December 2006 and underwent a questionnaire survey, blood analysis, and magnetic resonance imaging of the brain as part of their routine health checkup at the facility. MetS was defined using the criteria of the National Cholesterol Education Program Adult Treatment Panel III. LA was assessed using the visual rating scale of the Atherosclerosis Risk in Communities study. Multiple logistic regression analysis was performed to examine associations between LA and MetS and its components while considering influences of potential confounding variables.

Results: 752 (24.2%) subjects had LA on MRI. MetS was significantly associated with the presence of LA (multivariate adjusted odds ratio [OR], 3.40; 95% confidence interval [CI], 2.34–4..95). When five components were entered simultaneously into a model, significant associations were observed for elevated blood pressure (BP) (adjusted OR, 2.13; 95% CI, 1.54 −2.95), impaired fasting glucose (IFG) (adjusted OR, 1.16; 95% CI, 1.14–2..41), and hypertriglyceridemia (hyper-TG) (adjusted OR, 1.55; 95% CI, 1.07–2..26). Adjusted ORs of LA for individuals with MetS were relatively constant across grades of LA; the adjusted OR obtained from a multinomial logistic regression model were 3.41 (95%CI, 2.30–5..06) for minimal LA, 3.07 (95%CI, 1.75–5..38) for mild, moderate, and severe LA, and 3.33 (95%CI, 2.30–4..84) for all grades of LA.

Conclusions: MetS was significantly associated with all grades of LA, including the minimal grade of LA. Elevated BP, IFG, and hyper-TG of MetS components were independent risk factors of LA. MetS can play an important role in identifying healthy individuals who have an increased risk of LA.

HEMODYNAMIC CHANGES IN FOCAL BRAIN ISCHEMIA OF RATS ASSESSED BY LASER SPECKLE FLOWMETRY: ROLE OF ISCHEMIC DEPOLARIZATIONS

Hajime Nakamura^1,3, Anthony J. Strong², Guenter Mies¹, Heike Endepols¹, Stefan Vollmar¹, Michael Sue¹, Toshiki Yoshimine³, Andrew K. Dunn⁴, Rudolf Graf¹

¹Max-Planck Institute for Neurological Research, Cologne, Germany, ²Department of Clinical Neurosciences (Neurosurgery), King's College London, London, UK, ³Department of Neurosurgery, Osaka University, Osaka, Japan, ⁴Department of Biomedical Engineering, University of Texas at Austin, Austin, TX, USA

A UNIFIED NEUROBIOLOGICAL FRAMEWORK FOR FUNCTIONAL NEUROIMAGING USING A MODEL INTEGRATING COMPARTMENTALIZED ENERGY METABOLISM AND HEMODYNAMICS

Agnes Aubert¹, Luc Pellerin², Pierre Magistretti³, Robert Costalat⁴

¹URA CEA CNRS 2210, Service Hospitalier Frederic Joliot, Orsay, France, ²Departement De Physiologie, Universite De Lausanne, Lausanne, Switzerland, ³Brain and Mind Institute, Ecole Polytechnique Federale De Lausanne and Centre De Neurosciences Psychiatriques, Department of Psychiatry, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland, ⁴Universite Pierre Et Marie Curie-Paris6, Institut National De La Sante Et De La Recherche Medicale U678, Paris, France

Background: Brain functional imaging methods, including functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), have dramatically evolved in recent years. However, their biochemical and physiological bases remain unclear, especially the cellular and molecular mechanisms taking place upon activation in neurons and astrocytes. On the basis of a previously developed modeling approach, we developed a coherent neurobiological framework that not only explains several in vitro and in vivo experimental results but also provides a basis to interpret imaging signals, especially fMRI and Magnetic Resonance Spectroscopy (MRS) data.

Methods: System of ordinary differential equations has been numerically solved using MATLAB software version 7.

Results: First, using a model of compartmentalized energy metabolism between astrocytes and neurons, we show that complex NADH kinetics observed in vitro by Kasischke and colleagues (2004, Science 305, 99-103) can be accounted for by distinct metabolic responses in neurons and astrocytes. Second, extension of this model to an in vivo situation, including hemodynamic aspects, allowed us to predict the cerebral metabolic rate of oxygen consumption, and, combined with the changes in the cerebral blood flow, we could reproduce the time evolution of intraparenchymal oxygen concentration upon activation as measured experimentally in animal, for instance by Ances and colleagues (2001, Neuroscience Letters 306, 106-110). Third, applying the same kind of model to fMRI in humans, we could determine that the initial dip of the blood oxygen level dependent (BOLD) signal in fMRI, is critically dependent on the oxidative response of highly oxidative cells, for instance highly oxidative neurons, while the late features of the BOLD signal correspond to a combination of responses from two cell types: highly oxidative cells ('red’ neurons) and poorly oxidative cells (astrocytes and ‘white’ neurons). Moreover, the model allows to suggest a set of rules for (i) the presence of more or less oxidative cells within a voxel, (ii) the influence of the kind of stimulation, upon the fMRI signal.

Conclusion: Our results, using a versatile modeling approach, suggest new rules for the interpretation of fMRI signals, which could be tested by comparing biochemical and physiological data obtained in vitro and in vivo in animals to fMRI and MRS data in humans.

ACUTE TREATMENT WITH ROSUVASTATIN PROTECTS OBESE MICE AGAINST TRANSIENT CEREBRAL ISCHEMIA

Keita Mayanagi, Prasad V. Katakam, Tamas Gaspar, Ferenc Domoki, David W. Busija

Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston-Salem, NC, USA

ATTENUATION OF ISCHEMIC BRAIN INJURY AFTER FOCAL CEREBRAL ISCHEMIA IN RAT BY DOUBLE-STRANDED ADENO-ASSOCIATED VIRUS 2-DELIVERED VASCULAR ENDOTHELIAL GROWTH FACTOR

Kai-Yun Chen, Cheng-Fu Chang, Yung-Hsiao Chiang

Department of Neurological Surgery, Tri-Service General Hospital, Taipei, Taiwan-R.O.C.

Background and aims: Vascular endothelial growth factor (VEGF) has been shown to display neuroprotective effects in hypoxic-ischemic injury. We investigated the neuroprotective capacity of VEGF in a rat model of stroke.

Methods: To increase the in vivo transduction efficiency in brain, we used a double-stranded adeno-associated virus 2 vector to deliver VEGF. Seven days after gene transfer, the rats underwent 60 minutes of middle cerebral artery occlusion (MCAO) followed by 24 hours of reperfusion. Infarct volume was determined by TTC staining, neuronal injury was examined by TUNEL, cleaved caspase3 and fluoro-Jade B immunostaining. To further examined the signaling pathway of VEGF's neuroprotective activity, the expression and phosphorylation of Akt, ERK-1/−2 and p38 were analyzed by western blot.

Results: A single administration of dsAAV2 vector carrying VEGF (dsAAV2-VEGF) 7 days before the onset of ischemia, effectively decreased the infarct volume, TUNEL-positive, cleaved caspase3-positive and fluoro-Jade B-positive neurons. Next, we explored whether the dsAAV2-VEGF protected primary cultures of cortical neurons from hypoxia. After infection of dsAAV2-VEGF in primary cortical neurons 3 days before 24 h of hypoxia, the number of surviving cells was significantly increased and the activation of caspase3 was reduced. Both ischemic brains and cortical neurons infected with dsAAV2-VEGF showed increased phosphorylated Atk, while the expressions of phosphorylated ERK-1/−2 and p38 were without differences.

Conclusion: Exogenous expression of VEGF by dsAAV2 vector could attenuate ischemic brain injury through activation of Akt. In vitro experiments demonstrated that dsAAV2-VEGF could protect cortical neurons from hypoxic injury. Our results suggest that dsAAV2-VEGF is a highly potent neuroprotective tool for stroke therapy.

PRECLINICAL SAFETY PROFILE OF THE RECOMBINANT DIRECT-ACTING FIBRINOLYTIC, ALFIMEPRASE

Yuko Kurosawa¹, Joseph Clark¹, Aigang Lu¹, Kenneth Luskey³, Gail Pyne-Geithman¹, Thomas Tomsick²

¹Department of Neurology, University of Cincinnati, Cincinnati, OH, USA, ²Department of Neuroradiology, University of Cincinnati, Cincinnati, OH, USA, ³Nuvelo Inc., San Carlos, CA, USA

Purpose. To assess safety of alfimeprase in a rat acute ischemic stroke reperfusion model. Alfimeprase, a rapid direct-acting fibrinolytic, was supplied by Nuvelo. Safety was compared to tissue plasminogen activator (tPA) with hemorrhagic transformation (HT) as primary endpoint and behavioral changes and mortality as secondary endpoints. Hemorrhage is evaluated to provide hemorrhage rate and volume in the brain.

Methods. Five SD rat groups were studied in blinded fashion: alfimeprase at doses of 0.03 (N = 8), 0.1 (N = 11), and 0.3 mg/kg (N = 8); tPA at 1 mg/kg (N = 9); and carrier infused controls (N = 9). The carrier control was the carrier for the alfimeprase. The middle cerebral artery was occluded using a 3–0 suture for 5 hours followed by removal of the suture for reperfusion. Drug or carrier solutions were infused immediately following reperfusion through PE 10 tubing over a 10 minute period. The animals were anesthetized, decapitated and the brains rapidly harvested and frozen approximately 24 hours later. Serial brain sections were obtained (typically 6 sections per brain) and inspected for the number and volume of hemorrhages.

Results. The HT number, severity and the mortality rate for the 0.03 mg/kg dose of alfimeprase were significantly lower than for tPA. In a pooled analysis of all doses, alfimeprase was not significantly different than control for all parameters studied.

Conclusion. Alfimeprase at 0.03 mg/kg dose caused less HT and lower mortality than tPA. When compared to tPA, all other doses of alfimeprase were not significantly worse for all parameters tested. Alfimeprase is a new fibrinolytic agent, with a favorable safety profile based on phase II clinical data, that is appropriate for further evaluation in human stroke revascularization studies.

This work supported by Nuvelo.