Delayed damage of striatal interneurons after cerebral ischemia in the gerbil

Selective neuronal vulnerability following mild focal brain ischemia in the mouse

The evolution of cellular damage over time and the selective vulnerability of different neuronal subtypes was characterized in the striatum following 30-minute middle cerebral artery occlusion and reperfusion in the mouse. Using autoradiography we found an increase in the density of [3H]PK11195 binding sites--likely reflecting microglial activation--in the lesion border at 3 days and in the whole striatum from 10 days to 6 weeks. This was accompanied by a distinct loss of [3H]flumazenil and [3H]CGP39653 binding sites from 10 days up to 6 weeks reflecting neuronal loss. Brain ischemia resulted in a substantial loss of medium spiny projection neurons as seen at three days by Nissl staining, TUNEL and immunocytochemistry using antibodies against microtubule-associated protein (MAP2), NeuN, mu-opioid receptors, substance P, L-enkephalin, neurokinin B, choline acetyltransferase, parvalbumin, calretinin and somatostatin. Both patch and matrix compartments were involved in ischemic damage. In contrast, the numbers of cholinergic, GABAergic, and somatostatin-containing interneurons in the ischemic striatum were not different from those in the contralateral hemisphere at 3 and 14 days. A low density of glutamate receptors, the ability to sequester calcium by calcium-binding proteins and other hitherto unidentified factors may explain this relative resistance of interneurons to acute ischemia.

Effect of cerebral ischemia on dopamine receptors and uptake sites in the gerbil hippocampus

Dopamine D1 and D2 receptors and uptake sites were studied in the gerbil hippocampus, parietal cortex and thalamus 1 h to 7 days after 10 min of cerebral ischemia using the occlusion of bilateral common carotid arteries. [3H]SCH23390 ([N-methyl-3H]R[+]-8-chloro-2,3,4,5-tetrahydro-3-methyl-5-phenyl-7-ol-be nzazepine) and [3H]mazindol were used as markers of dopamine D1 receptors and uptake sites, respectively. [3H]Nemonapride was used to label dopamine D2 receptors. No obvious alteration in [3H]SCH23390 and [3H]mazindol binding was found in the hippocampus up to 48 h after ischemia. These bindings showed a significant reduction in the hippocampus after 7 days of recirculation. In contrast, [3H]nemonapride binding was unaffected in the hippocampus during the recirculation periods. The parietal cortex and thalamus also exhibited no significant changes in [3H]SCH23390, [3H]nemonapride and [3H]mazindol binding after ischemia. MAP2 (microtubule-associated protein 2) immunoreactivity was unchanged in all regions up to 48 h after ischemia. Thereafter, a marked loss of MAP2-immunoreactive neurons was observed in the hippocampal CA1 and CA3 neurons 7 days after recirculation. These findings were consistent with histological observations with cresyl violet staining. Our results demonstrate that dopamine D1 receptors and dopamine uptake sites in the hippocampus are susceptible to cerebral ischemia, whereas dopamine D2 receptors in this region are particularly resistant. Furthermore, these findings suggest that dopamine transmission may not be major factor in producing ischemic hippocampal damage.

Cellular distribution of the NMDA receptor NR2A/2B subunits in the rat striatum

Using immunohistochemical double-labeling with a specific antibody recognizing both NR2A and NR2B subunits, we studied the cellular distribution of the NMDA receptor subunit NR2A/2B on all major known striatal neuron types. Among striatal interneurons, our results showed that none of somatostatin interneurons was labeled for NR2A/2B subunits, 56% of parvalbumin interneurons were double-labeled for NR2A/2B, and all identified cholinergic interneurons were labeled for NR2A/2B. Among striatal projections neurons, 95% of striatonigral neurons, 96% of enkephalin-containing neurons, and 98% of calbindin-containing striatal matrix neurons were double-labeled for NR2A/2B. Our studies demonstrate that there is a differential distribution of the NMDA receptor NR2A/2B subunits on striatal neuron types. The paucity of NR2A/2B subunits on NMDA receptors on striatal somatostatin interneurons may confer resistance to NMDA receptor-mediated excitotoxicity on these neurons.

Cellular expression of ionotropic glutamate receptor subunits on specific striatal neuron types and its implication for striatal vulnerability in glutamate receptor-mediated excitotoxicity

Glutamate receptors are composed of subtype-specific subunits. Variation in the precise subunit composition of a receptor may result in significant functional differences. Thus, a precise knowledge of subunit composition on striatal neurons is a prerequisite for understanding the selective vulnerability of striatal neurons to excitatory amino acids. In the present study, we used an immunohistochemical double-labelling approach to localize ionotropic glutamate receptor subunits (NMDAR1, GluR1, GluR2/3, GluR4 and GluR5/6/7) on specific striatal neuron populations. Our results showed that striatal cholinergic and somatostatin interneurons were not labelled for the alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionate, receptor subunits GluR1, GluR2/3 and GluR4. Most cholinergic and somatostatin interneurons (83.3% to 100%), however, were double-labelled for the N-methyl-D-aspartate receptor subunit NR1 and kainic acid receptor subunits GluR5/6/7. All parvalbumin interneurons were labelled for GluR1 and GluR4, and 96% GluR1 positive and 95% GluR4 positive neurons were also double-labelled as parvalbumin interneurons. About half of all parvalbumin interneurons co-localized with GluR2/3, and over 97% were labelled for NR1 and GluR5/6/7. Among striatal projection neurons, enkephalin-positive (mainly striatopallidal) neurons, striatonigral neurons (mainly substance P-positive) and calbindin-positive matrix neurons were not immunostained for GluR1 or GluR4. In contrast, 95% to 100% of each of these types of projection neurons were double-labelled for NR1, GluR2/3 and GluR5/6/7. Our results demonstrate that striatal neuron types differ in their expression of ionotropic glutamate receptor subunits and subtypes. The clear difference between striatal interneurons and projection neurons in ionotropic glutamate receptor subtypes/subunits supports the idea that differential glutamate receptor expression mechanism may account for the selective vulnerability of striatal projection neurons to excitotoxicity, and that glutamate receptor-mediated excitotoxicity may be involved in the striatal neurodegenerative diseases.

Post-ischemic changes of [3H]glycine binding in the gerbil brain after cerebral ischemia

Sequential changes of [3H]glycine binding in the gerbil were investigated in selectively vulnerable areas 1 h to 7 days after 10 min of cerebral ischemia. A significant reduction in [3H]glycine binding was found in the hippocampus and thalamus from as early as 1 h after ischemia. In contrast, the striatum and frontal cortex showed a significant decline in [3H]glycine binding from 5 h after recirculation. Thereafter, a severe reduction in [3H]glycine binding was observed in all regions 7 days after ischemia. MAP2 (microtubule-associated protein 2) immunoreactivity was unaffected in the hippocampus, frontal cortex and thalamus up to 48 h after ischemia. Thereafter, a severe loss of MAP2-immunoreactive neurons was found in these regions, especially in the hippocampal CA1 sector. However, the striatum showed a severe loss of MAP2 immunoreactivity from 24 h after ischemia. These results demonstrate that transient cerebral ischemia causes severe reduction in [3H]glycine binding throughout the brain, and this reduction precedes the neuronal damage in selectively vulnerable areas. These findings suggest that a neurotransmitter, glycine, may play a key role in the pathogenesis of post-ischemic neurodegeneration in selectively vulnerable areas.