Neuroprotection Mediated through GluN2C-Containing N-methyl-D-aspartate (NMDA) Receptors Following Ischemia

Abstract

Post-ischemic activation of NMDA receptors (NMDARs) has been linked to NMDAR subunit-specific signaling that mediates pro-survival or pro-death activity. Although extensive studies have been performed to characterize the role of GluN2A and GluN2B following ischemia, there is less understanding regarding the regulation of GluN2C. Here, we show that GluN2C expression is increased in acute hippocampal slices in response to ischemia. Strikingly, GluN2C knockout mice, following global cerebral ischemia, exhibit greater neuronal death in the CA1 area of the hippocampus and reduced spatial working memory compared to wild-type mice. Moreover, we find that GluN2C-expressing hippocampal neurons show marked resistance to NMDA-induced toxicity and reduced calcium influx. Using both in vivo and in vitro experimental models of ischemia, we demonstrate a neuroprotective role of GluN2C, suggesting a mechanism by which GluN2C is upregulated to promote neuronal survival following ischemia. These results may provide insights into development of NMDAR subunit-specific therapeutic strategies to protect neurons from excitotoxicity.

Figure 1. An increase in GluN2C expression, but not GluN2A or GluN2B, was observed following ischemia.

(a) Summary data showing quantitative real time PCR results of GluN2A, GluN2B and GluN2C mRNA levels extracted from acute hippocampal slices following 4 min OGD and 90 min reperfusion. Ct-values are normalized to β-actin internal control. (b,c) Representative Western blot showing GluN2C upregulation in membrane fraction of acute hippocampal slices (labeled Hp in figure) following 4 min OGD and 3 hr reperfusion (b) or 12 min GCI and 6 hr reperfusion (c) (full-length blots are presented in Supplementary Fig. S8). Cerebellum lysate (labeled Cb in figure) was used as positive control for GluN2C expression. Graphs depict mean ± SEM (n = 3, *p < 0.05, Student’s t-test).

Figure 2. GluN2C^−/− mice display less CA1 surviving neurons and hippocampal-dependent functional loss following global cerebral ischemia (GCI).

Both WT and KO (except sham groups) mice were induced with 15 min GCI. (a) Representative images of hippocampal sections comparing neuronal survival of WT sham, KO sham, WT GCI and KO GCI groups. Sections were stained with NeuN antibody (green). The left column depicts the hippocampus imaged using 10× objective and the right column images represent magnified CA1 areas (using 40× objective) highlighted by the white box. Cells that stained positively for NeuN staining were identified as surviving neurons. Scale bar, 50 μm. (b) Quantitiative summary of data showing the average number of NeuN+ cells (surviving neurons) per 320 μm length of medial CA1 region. (c) Average alternation percentage defined as “the number of alternating triads/(the total number of arm entries –2)” from Y-maze test at ischemia/reperfusion day 3 and 7 (n = 8 mice per GCI group, 3 mice per sham group, *p < 0.05, One-way ANOVA and Bonferroni test).

Figure 3. Overexpressed GluN2C mediates neuroprotection following NMDA-induced toxicity and surface expression levels remained unchanged in hippocampal neurons.

(a) Time-lapse images of DIV 13 primary hippocampal neurons co-transfected with DsRed (to visualize cellular morphology) and SEP-tagged GluN2 constructs. Cells were exposed to 200 μM NMDA and imaged for 25 min. Arrows depict structural hallmarks of excitotoxic effects (dendritic varicosities and cell body swelling) in neurons expressing DsRed alone, GluN2A, or GluN2B. GluN2C-expressing neuron images show resistance to NMDA-induced toxicity. Scale bar, 10 μm. (b) Quantitiative summary of data showing the average number of dendritic varicosities per 10 μm length of dendrite (Vector, n = 18 dendritic segments, 6 cells; GluN2A, n = 18 dendritic segments, 6 cells; GluN2B, n = 21 dendritic segments, 7 cells; GluN2C, n = 24 dendritic segments, 8 cells, *p < 0.05, ***p < 0.01, One-way ANOVA and Bonferroni test). (c) Time-lapse images of zoomed in dendritic regions of hippocampal neurons co-expressing DsRed and SEP-tagged GluN2A, GluN2B, or GluN2C (green) before (0 min) and following administration of 200 μM NMDA. Scale bar, 2 μm. (d) Time course distribution of receptor enrichment for spines and dendritic segments from hippocampal neurons treated with NMDA (SEP-GluN2A, n = 24 spines, 6 dendritic segments, 4 cells; SEP-GluN2B, n = 31 spines, 7 dendritic segmens, 4 cells; SEP-GluN2C, n = 20 spines, 11 dendritic segments, 7 cells, *p < 0.05, One-way ANOVA and Bonferroni test).

Figure 4. Reduced Ca²⁺ influx in GluN2C-expressing hippocampal neurons following NMDA treatment.

(a) Time-lapse fluorescent images of hippocampal neuronal cell bodies co-transfected with DsRed (channel not shown for visual purposes) and WT-GluN2A, GluN2B, GluN2C, or vector alone. Oregon Green is depicted in green and its signal intensity directly correlates with the intracellular Ca²⁺ concentration. Images were obtained before (0 min), during (every 5 sec) and after (5 min) NMDA treatment. Only 0 min and 5 min time points are shown. (b) Time course distribution of basal level Ca²⁺ signaling 1 m before NMDA stimulation comparing vector, WT-GluN2A, WT-GluN2B, and WT-GluN2C transfected hippocampal neurons. (c) Time course distribution of Ca²⁺ responses evoked by NMDA treatment. (d) Mean ΔF/F0 of peak intensity time points (140 s–240 s) examined in (c). (n = 3, *p < 0.05, Repeated Measures ANOVA and Bonferroni test).