Flemming Fryd Johansen

Selective dendrite damage in hippocampal CA1 stratum radiatum with unchanged axon ultrastructure and glutamate uptake after transient cerebral ischaemia in the rat

In this study of the rat hippocampal CA1 stratum radiatum 4 days after 20 min transient cerebral ischaemia, we demonstrated ultrastructural dendritic degeneration and loss, whereas most axons appeared undamaged. Autoradiographs 4 days after ischaemia showed unchanged Na+-dependent glutamate high affinity uptake. The role of glutamate and Ca2+ for this selective postsynaptic vulnerability is discussed.

Ischemia leads to apoptosis--and necrosis-like neuron death in the ischemic rat hippocampus.

Morphological evidence of apoptosis in transient forebrain ischemia is controversial. We therefore investigated the time sequence of apoptosis‐related antigens by immunohistochemistry and correlated it with emerging nuclear patterns of cell death in a model of transient forebrain ischemia in CA1 pyramidal cells of the rat hippocampus. The earliest ischemic changes were found on day 2 and 3, reflected by an upregulation of phospho‐c‐Jun in a proportion of morphologically intact CA1 neurons, which matched the number of neurons that succumbed to ischemia at later time points. At day 3 and later 3 ischemic cell death morphologies became apparent: pyknosis, apoptosis‐like cell death and necrosis‐like cell death, which were confirmed by electron microscopy. Activated caspase‐3 was present in the vast majority of cells with apoptosis‐like morphology as well as in a small subset of cells undergoing necrosis; its expression peaked on days 3 to 4. Silver staining for nucleoli, which are a substrate for caspase‐3, revealed a profound loss of nucleoli in cells with apoptosis‐like morphology, whereas cells with necrosis‐like morphology showed intact nucleoli. Overall, cells with apoptosis‐like morphology and/or caspase‐3 expression represented a minor fraction (<10%) of ischemic neurons, while the vast majority followed a necrosis‐like pathway. Our studies suggest that CA1 pyramidal cell death following transient forebrain ischemia may be initiated through c‐Jun N‐terminal kinase (JNK) pathway activation, which then either follows an apoptosis‐like cell death pathway or leads to secondary necrosis.

Ischemia induced changes in expression of the astrocyte glutamate transporter GLT1 in hippocampus of the rat.

Changes in cellular uptake of glutamate following transient cerebral ischemia is of possible importance to ischemia induced cell death. In the present study, we employed in situ hybridization and immunohistochemistry to investigate the influence of cerebral ischemia on expression of mRNA and protein of the astrocyte glutamate transporter GLT1, and of glial fibrillary acidic protein. Different subfields of CA1 and CA3 of the rat hippocampus were studied at various time-points after ischemia (days 1, 2, 4, and 21). In CA1, GLT1-mRNA was decreased at all time-points after ischemia except from day 2, whereas in CA3, decreases were seen only on day 1. Expression of GLT1-protein in CA1 was unchanged during the initial days after ischemia, but decreased markedly from day 2 to 4. In CA3, GLT1-protein increased progressively throughout the observation period after ischemia. Following the degeneration of CA1 pyramidal cells, a positive correlation between the number of CA1 pyramidal cells and expression of either GLT1-mRNA or -protein was evident selectively in CA1. Increases in expression of mRNA and protein of glial fibrillary acidic protein were present from day 2, most notable in CA1. The present data provide evidence that expression of GLT1 in CA1 of the hippocampus is not decreased persistently before the degeneration of CA1 pyramidal cells, but is downregulated in response to loss of these neurons. Since the reduction in GLT1 expression evolved concomitantly with the degeneration of CA1 pyramidal cells, it may contribute to the severity of CA1 pyramidal cell loss. A progressive postischemic increase in GLT1 expression in CA3 may be linked to the resistance of CA3 neurons to ischemic cell damage.

A role for mixed lineage kinases in granule cell apoptosis induced by cytoskeletal disruption

Microtubule disruption by colchicine induces apoptosis in selected neuronal populations. However, little is known about the upstream death signalling events mediating the neurotoxicity. We investigated first whether colchicine‐induced granule cell apoptosis activates the c‐Jun N‐terminal kinase (JNK) pathway. Cultured murine cerebellar granule cells were exposed to 1 µm colchicine for 24 h. Activation of the JNK pathway was detected by western blotting as well as immunocytochemistry using antibodies against phospho‐c‐Jun (p‐c‐Jun). Next, adult male rats were injected intracerebroventricularly with colchicine (10 µg), and JNK pathway activation in dentate granule cells (DGCs) was detected by antibodies against p‐c‐Jun. The second part of the study tested the involvement of mixed lineage kinases (MLK) as upstream activators of the JNK pathway in colchicine toxicity, using CEP‐1347, a potent MLK inhibitor. In vitro, significant inhibition of the JNK pathway, activated by colchicine, was achieved by 100–300 nm CEP‐1347, which blocked both activation of cell death proteases and apoptosis. Moreover, CEP‐1347 markedly delayed neurite fragmentation and cell degeneration. In vivo, CEP‐1347 (1 mg/kg) significantly prevented p‐c‐jun increase following injection of colchicine, and enhanced survival of DGCs. We conclude that colchicine‐induced neuronal apoptosis involves the JNK/MLK pathway, and that protection of granule cells can be achieved by MLK inhibition.

Taurine Neurons in Rat Hippocampal Formation are Relatively Inert to Cerebral Ischemia

It is well-known that transient cerebral ischemia causes selective loss of neurons in the hippocampus, striatum and cortical layers (14,24). It has been suggested that the difference in nerve cell vulnerability to ischemic damage is probably due to differences in the nature of their transmitters (14,22,27). For instances, GABAergic neurons in the striatum are more susceptible to transient ischemia than the cholinergic neurons (7) and dopaminergic neurons are more sensitive to ischemic damage than the GABAergic neurons in the gerbil brain (26). Since taurine has been demonstrated to have protective effects on the photoreceptor in the cat (12) and on the light-induced disruption of isolated frog rod outer segments (23), we decided to investigate the influence of transient cerebral ischemia on the taurine neurons in rat hippocampus. Taurine neurons were identified immunocytochemically with antibodies against cysteine sulfinic acid decarboxylase, the rate-limiting enzyme of taurine biosynthesis. Cysteine sulfinic acid decarboxylase has previously been used as a specific marker for identification of taurine neurons in the retina (10,17–19), cerebellum (3,4) and hippocampus (10,26,28).

Estimation of the hypothermic component in neuroprotection provided by cannabinoids following cerebral ischemia.

Cannabinoids have neuroprotective potentials, and the expression of endocannabinoids as well as cannabinoid receptors is induced after cerebral ischemia. They also induce hypothermia by lowering the hypothalamic set point. We have estimated the significance of such hypothermia in ischemic neuroprotection following systemic administration of WIN 55,212-2, a synthetic cannabinoid receptor agonist. Results showed that WIN 55,212-2 significantly reduced infarct volumes of rats subjected to focal cerebral ischemia (middle cerebral artery occlusion) and significantly decreased ischemic CA1 damage in rats subjected to global cerebral ischemia (two-vessel occlusion). A significant (approximately 50%) part of this neuroprotection was provided by WIN 55,212-2 induced hypothermia (33.7+/-1.1 degrees C/34.9+/-1.6 degrees C), because prevention of hypothermia by maintaining body core temperatures between 37.0 and 38.0 degrees C dissolved the neuroprotective effect into a hypothermic component and an unidentified component. Finally, the ability of WIN 55,212-2 to reduce levels of the proinflammatory cytokine IFNgamma in the infarcted hemisphere of rats subjected to focal cerebral ischemia required hypothermia. For the cannabinoid WIN 55,212-2, we have isolated and directly demonstrated that hypothermia is only part of, although significant, cannabinoid mediated neuroprotection in both global and focal cerebral ischemia. We conclude that cannabinoids are reliable candidates for drug-induced hypothermia and neuroprotection. These neuroprotective effects of cannabinoids could provide the basis for potential therapeutic uses of cannabinoids and/or endocannabinoids in stroke.

Inhibition in postischemic rat hippocampus: GABA receptors, GABA release, and inhibitory postsynaptic potentials

SummaryWe have investigated the GABAergic system in rat hippocampus at 1 hour and up to 21 days following 20 min of global cerebral ischemia. Distribution of 3H-GABA (in excess of unlabeled baclofen) and 3H-Ro-15-1788 (benzodiazepine antagonist) binding sites in hippocampus was studied utilizing quantitative autoradiography. The 3H-GABA binding was unchanged (p> 0.01) after ischemia, whereas the 3H-Ro-15-1788 binding decreased significantly (p< 0.01) in all hippocampal subfields 1–21 days after ischemia. Using microdialysis in CA1, we found that K+-stimulated GABA release at 1 hour and 1 day after ischemia was unchanged (p> 0.01) in comparison to preischemic controls. Electrophysiological recordings were made from CA1 of hippocampal slices prepared from rats sacrificed 1 hour, 1 day and 2 days after ischemia. Field potentials evoked by stimulation of the Schaffer collaterals showed no differences (p > 0.01) from those taken from controls. Postischemic intracellular recordings from the CA1 pyramidal cells showed that fast and slow inhibitory postsynaptic potentials were readily evoked on orthodromic stimulation. Together with our previous morphological results, demonstrating survival of hippocampal interneurons following ischemia, we conclude that hippocampal GABAergic interneurons preserve their inhibitory potential in the period preceding delayed CA1 pyramidal cell death. This conclusion taken together with the observation that postischemic 3H-Ro-15-1788 binding in hippocampus declined, suggest that benzodiazepines (by increasing the receptor affinity), GABA analogs, and GABA uptake inhibitors may be usefull in the treatment of ischemic CA1 pyramidal cell death in the rat.

Differences in Origin of Reactive Microglia in Bone Marrow Chimeric Mouse and Rat After Transient Global Ischemia.

Current understanding of microglial involvement in disease is influenced by the observation that recruited bone marrow (BM)-derived cells contribute to reactive microgliosis in BM-chimeric mice. In contrast, a similar phenomenon has not been reported for BM-chimeric rats. We investigated the recruitment and microglial transformation of BM-derived cells in radiation BM-chimeric mice and rats after transientglobal cerebral ischemia, which elicits a characteristic microglialreaction. Both species displayed microglial hyperplasia and rod cell transformation in the hippocampal CA1 region. In mice, a subpopulation of lesion-reactive microglia originated from transformed BM-derived cells. By contrast, no recruitment or microglial transformation of BM-derived cells was observed in BM-chimeric rats. These results suggest that reactive microglia in rats originate from resident microglia, whereas they have a mixed BM-derived and resident origin in mice, depending on the severity of ischemic tissue damage.

Ischemia as an excitotoxic lesion: protection against hippocampal nerve cell loss by denervation.

There are several indications for an involvement of neuroexcitatory mechanisms in ischemic neuron damage. Since we forwarded the hypothesis in 1982 that the transmitter glutamate is playing a key role, several lines of evidence have substantiated this: there is a pronounced transmitter release induced by ischemia and there is uptake of Ca++ via NMDA-operated calcium channels. Under certain circumstances postischemic neuron death can be impaired by administration of either NMDA-antagonists or calcium blockers. Further proof for the induction of harmful excitatory mechanisms by ischemia has been obtained by preischemic denervation of the vulnerable nerve cells. After transient cerebral ischemia in rats or gerbils, there are signs of irreversible damage (eosinophilia) of neurons in the dentate hilus (somatostatin-positive cells) after 2-3 hours and of hippocampal pyramidal neurons after 2-3 days (delayed neuron death). In the first case, removal of the (main) input to hilus cells by degranulation (colchicine selectively eliminates granule cells) protects these. In the case of pyramidal neurons removal of Schaffer collaterals/commisurals or input from the entorhinal cortex have a protective effect. Recently, we have measured glutamate and calcium in CA1 of denervated rats during 10 min of ischemia, and it turns out that there is almost no extracellular glutamate release or lowering of calcium in contrast to ischemic animals with intact innervation. Also in the postischemic period there are indications of a continuation of the damaging processes induced by ischemia. Besides the well known postischemic hypoperfusion, a prolonged release of glutamate has been reported, as well as burst firing in some models.(ABSTRACT TRUNCATED AT 250 WORDS)

GABAA receptor subunit interactions important for benzodiazepine and zinc modulation: a patch‐clamp and single cell RT‐PCR study

The expression of mRNAs for the GABAA receptor subunits α1, α6, β2, β3, γ2 and δ in single mouse cerebellar granule cells and cortical interneurons were analysed by RT‐PCR and correlated to their midazolam and zinc modulation of agonist‐induced receptor currents. The registration of molecular and electrophysiological data from each cell allowed us to estimate the significance of individual subunits and their two‐factor interaction for modulation. The presence of α6 decreased midazolam modulation, but statistical analysis also suggested interactions of α6 with β3 and γ2 with respect to midazolam modulation. Zinc modulation was decreased by the presence of γ2, and analysis points to an β3 effect as well as an interaction between γ2 and δ in zinc modulation. Thus, our model confirmed, in single native cells, the known effects of α6 in midazolam and γ2 in zinc modulation, and additionally pointed to significant subunit interactions that need to be further tested in recombinant receptors. The present study offers a method to identify subunit interactions in heteromeric receptor complexes.