Eiji Kohmura

Induction of aquaporin-4 water channel mRNA after focal cerebral ischemia in rat.

Aquaporin-4 (AQP4) is a member of a water-selective channel aquaporin-family and mainly expressed in the several structures of the brain and in the collecting duct of the kidney. Here we show its functional involvement in the water homeostasis of the ischemic brain. The expression of AQP4-mRNA is increased in the peri-infarcted cortex during the observation period ( approximately 7 days) after MCA-occlusion, maximally on day 3. The change corresponds to the generation and resolution of brain edema monitored by MRI. The signals for the mRNA are predominantly observed in glial cells in the molecular and outer granular layer of the peri-infarcted cortex. These results indicate that AQP4 plays a role in post-ischemic edema formation.

Basic fibroblast growth factor prevents thalamic degeneration after cortical infarction

In the focal infarction model of the rat middle cerebral artery (MCA), the thalamus of the occluded side becomes gradually atrophic, mainly because of retrograde degeneration. We determined whether basic fibroblast growth factor (bFGF) administered intracisternally could prevent this thalamic atrophy. We occluded the left MCA through a small cranial opening, and animals were then divided into two groups. One group received intra-cisternal injections of recombinant bFGF (1 μg dissolved in 0.1 ml of saline with 2% rat serum) starting 1 day after occlusion and repeated once a week to a total dose of 4 μg by four injections. The other group received vehicle solution by the same schedule. The animals were perfused and fixed at 28 days after occlusion, and histological examination was made at the level of the caudoputamen and thalamus. In the bFGF-treated rats, the area of the posterior ventral thalamus of the occluded side was 93% of that of the contralateral side, i.e., significantly larger than in the normal saline-treated rats (75%, p < 0.01). The infarction size was not statistically different in the two groups. Microscopic observation indicated that normal-saline-treated animals showed shrinkage and disappearance of thalamic neurons, whereas bFGF-treated groups showed preservation of thalamic neurons. Computerized analysis of the cell size substantiated this observation. To assess the effect of bFGF on astrocytes, bFGF or vehicle solution was injected into normal rats, and their histology was evaluated at 1, 2, and 4 weeks after injection. The bFGF-injected group showed a significant increase in glial fibrillary acidic protein-positive astrocytes in the brain tissue facing the ventriculocisternal system. The results suggest that bFGF prevents retrograde degeneration of thalamic neurons by working directly as a neurotrophic factor of these neurons or by activating astrocytes located in the thalamocortical pathway.

BRIEF EXPOSURE TO HYPOXIA INDUCES BFGF MRNA AND PROTEIN AND PROTECTS RAT CORTICAL NEURONS FROM PROLONGED HYPOXIC STRESS

We examined the hypoxic tolerance phenomenon in vitro. Brief exposure to hypoxia induced the production of basic fibroblast growth factor (bFGF) mRNA and protein in rat cortical neurons and protected them from hypoxic injury. Cortical neurons were cultured from 18th-day rat embryos in a serum-free medium and subjected to brief (4 h) and/or prolonged (24 h) hypoxia. Neuronal damage was assessed by quantifying lactate dehydrogenase (LDH) activity in the medium. After brief hypoxia, LDH release was identical to that of the controls, whereas prolonged hypoxia caused a significant increase in LDH release, indicating neuronal death. However, if brief hypoxia was applied 2 days prior to the prolonged hypoxia, no increase in LDH release was observed. The bFGF mRNA expression was assessed with Northern blot and protein immunoreactivity with Western blot analysis. The brief period of hypoxia caused a 2.5-fold increase in bFGF mRNA and considerable bFGF protein expression 1 day later, but prolonged hypoxia caused increase in the expression of bFGF mRNA at 2 days and no protein expression until 3 days after the start of the hypoxia. When cells were subjected to prolonged hypoxia 2 days after brief hypoxia, however, no increase in bFGF mRNA was observed, while bFGF protein was expressed continuously. We also observed that exogenously applied bFGF reduced neuronal injury produced by prolonged hypoxia. The results obtained with this model suggest that brief hypoxia induces bFGF protein and thus tolerance to subsequent lethal hypoxia. Basic FGF might play a role as a tolerance-associated factor in this process. Thus, an in vitro model is useful for assessing the response of cortical neurons to hypoxic stress and for researching new factors related to ischemic tolerance.

Hippocampal Neurons Become More Vulnerable to Glutamate after Subcritical Hypoxia: An in vitro Study

The neurotoxicity of glutamate and hypoxia was investigated in vitro on hippocampal neurons, which were obtained from 18-day-old rat fetuses and were maintained for 3 days in culture. Chemically defined medium without glutamate was used and the plating density was low enough that the effect of exogenously added glutamate could be directly evaluated. In the normal culture condition 1 mM glutamate was necessary to cause significant neuronal loss in the following 24 h. In marked contrast, when glutamate was added after subcritical hypoxic stress, a dose of glutamate as low as 10 μM could exhibit neurotoxicity. Administration of MK-801, a selective noncompetitive antagonist of the N-methyl-d-aspartate (NMDA) receptor, could in part reverse this increased susceptibility to low-dose glutamate after hypoxia, although MK-801 could not protect hippocampal neurons from high-dose glutamate. Therefore, both the NMDA receptor and other subclasses of the glutamate receptor may be involved in this neurotoxicity of glutamate. Different mechanisms of glutamate neurotoxicity with high and low doses are discussed. Our results showed that hippocampal neurons exposed to subcritical hypoxia become more vulnerable to glutamate than those without hypoxia. This increased susceptibility is of great interest to understanding the mechanism of slowly ongoing neuronal loss caused by ischemia or epilepsy.

Changes in growth inhibitory factor mRNA expression compared with those in c-jun mRNA expression following facial nerve transection.

We investigated growth inhibitory factor (GIF) mRNA expression within the rat facial nucleus with the aid of in situ hybridization. We found that GIF mRNA was expressed abundantly in the facial motoneurons of sham operated animals, and that this gene expression decreased after transection of the facial nerve. This decrease of GIF mRNA was first detected on the third day and was maintained for at least five weeks after transection of the nerve. Changes in c-jun, an immediate early gene, were also investigated with this model, and it was found that c-jun mRNA started to increase in the facial nucleus on the first day and that this increase was maintained for at least 5 weeks. These results suggest that the facial motoneurons, when their axons are transected, continuously respond to the injury and that GIF mRNA is actively suppressed to reduce the inhibition of neurite outgrowth in order to regenerate the axons.

Increased transcription of glutamate-aspartate transporter (GLAST/GluT-1) mRNA following kainic acid-induced limbic seizure.

Expression of mRNA for glutamate-aspartate transporter (GLAST/GluT-1/EAAT1) was studied in the brain of the rat which presented recurrent limbic seizure following systemic administration of kainic acid (KA) by in situ hybridization and Northern blot analysis. The expression of GLAST mRNA was markedly increased after 12 h and peaked after 48 h in animals which demonstrated limbic seizure. The induction of the mRNA were observed in the small non-neuronal cells in the hippocampus, especially around CA3 region and hilus. In contrast, there was no change in GLAST mRNA levels in KA injected seizure-free animals. These findings suggest that GLAST mRNA is induced by seizure and increased extracellular glutamate levels during seizure may be important for induction of GLAST mRNA.

Role of pyruvate in ischaemia-like conditions on cultured neurons.

In vitro, it is known that neurons in serum free mediums survive in the absence of glucose if pyruvate is present. We exposed cultured neurons on poly-L-lysine coated dishes to 6 different concentrations of pyruvate ranging from 0 to 1000 microM combined with glucose-supplement and normoxia (controls), glucose-deprivation and normoxia (hypoglycaemia), glucose-supplement and hypoxia (hypoxia), and glucose deprivation and hypoxia (ischaemia-like conditions) for 4 h. In all these conditions, lowering pyruvate below 250 microM led to a significant decrease of neuronal survival. Even in controls, there was no surviving neurons exposed to 0 microM of pyruvate. Glucose deprivation alone did not essentially influence the survival rate. Regardless of glucose deprivation, hypoxia led to an additional 23%-100% decrease of neuronal survival. When neurons were cultured above the astrocyte layers, neurons were much more tolerant with exposure to 0 microM of pyruvate for 4 h as compared to neurons on poly-L-lysine. The results indicate that pyruvate rather than glucose is essential for the energy supply to cultured neurons, probably because these neurons are immature and possess yet little capability to gain energy from glycolysis. Astrocytes may protect neurons from pyruvate deprivation by providing energy to neurons. In neuron rich culture, ischaemia is much better simulated by hypoxia-pyruvate deprivation rather than hypoxia-glucose deprivation.

BDNF atelocollagen mini-pellet accelerates facial nerve regeneration

We investigated the effect of BDNF mini-pellet on the GAP-43 mRNA expression and functional status of facial nerve in a rat model of facial nerve transection and immediate repair. The facial function started to recover at 17 days in the placebo group and 14 days in the BDNF group. BDNF group had shorter period of increased GAP-43 mRNA expression than the placebo group. Topically applied BDNF may accelerate the facial nerve regeneration.

Restoration of small bone defects at craniotomy using autologous bone dust and fibrin glue

BACKGROUND Bone gaps or burr holes often result in small but undesirable scalp or skin depressions after craniotomy. Whereas a number of reports have discussed cranioplasties to avoid large bone defects, little has been written about the problem of small bone defects which, despite their minor size, could result in bothersome cosmetic problems. METHODS This study was designed to assess a simple method to repair burr hole defects and bridge bone gaps with autologous bone dust and fibrin glue. Bone dust was collected when burr holes were created or craniectomy was performed. After replacement of the bone flap, the burr holes or bone gap were filled with a mixture of bone dust and fibrin glue. RESULTS AND CONCLUSIONS The mixture of bone dust and fibrin glue was easily shaped to fit bone defects, resulting in favorable cosmetic outcomes 1 to 5 years after operation.

Changes in brain organic osmolytes in experimental cerebral ischemia.

The cell volume is regulated not only by inorganic ions, but also by organic osmolytes, such as amino acids, methylamines, and polyhydric alcohols (polyols). Using proton nuclear magnetic resonance spectroscopy (1H-NMR), we measured the tissue concentrations of amino acids (alanine, aspartate, gamma-aminobutyric acid (GABA), glutamate, glutamine, N-acetyl-aspartate (NAA), taurine), methylamines (glycerophosphorylcholine (GPC), creatine+phosphocreatine (total creatine, tCr)), and polyols (myo-inositol) in the rat brain after middle cerebral artery occlusion (incomplete focal ischemia) or after decapitation (complete global ischemia). The total osmolytes expressed as a sum of total amino acids, total methylamines, and total polyols were significantly decreased at 24 h of focal ischemia (58.7% of control value, P=0.0025) whereas they were not changed following decapitation. The water content was increased from control value of 77.9%-84.1% after focal ischemia (P<0.0001) but not after decapitation. These results suggest that the brain organic osmolytes are involved in the process of edema formation following focal cerebral ischemia. Further elucidation of the cellular mechanisms regulating these organic osmolytes in cerebral ischemia may promote greater understanding of the pathophysiology involved in the evolution of brain edema.