Severity of middle cerebral artery occlusion determines retinal deficits in rats

Abstract

Middle cerebral artery occlusion (MCAO) using the intraluminal suture technique is a common model used to study cerebral ischemia in rodents. Due to the proximity of the ophthalmic artery to the middle cerebral artery, MCAO blocks both arteries, causing both cerebral ischemia and retinal ischemia. While previous studies have shown retinal dysfunction at 48h post-MCAO, we investigated whether these retinal function deficits persist until 9days and whether they correlate with central neurological deficits. Rats received 90min of transient MCAO followed by electroretinography at 2 and 9days to assess retinal function. Retinal damage was assessed with cresyl violet staining, immunohistochemistry for glial fibrillary acidic protein (GFAP) and glutamine synthetase, and TUNEL staining. Rats showed behavioral deficits as assessed with neuroscore that correlated with cerebral infarct size and retinal function at 2days. Two days after surgery, rats with moderate MCAO (neuroscore <5) exhibited delays in electroretinogram implicit time, while rats with severe MCAO (neuroscore ≥5) exhibited reductions in amplitude. Glutamine synthetase was upregulated in Müller cells 3days after MCAO in both severe and moderate animals; however, retinal ganglion cell death was only observed in MCAO retinas from severe animals. By 9days after MCAO, both glutamine synthetase labeling and electroretinograms had returned to normal levels in moderate animals. Early retinal function deficits correlated with behavioral deficits. However, retinal function decreases were transient, and selective retinal cell loss was observed only with severe ischemia, suggesting that the retina is less susceptible to MCAO than the brain. Temporary retinal deficits caused by MCAO are likely due to ischemia-induced increases in extracellular glutamate that impair signal conduction, but resolve by 9days after MCAO.

Keywords: Electroretinogram; Focal ischemia; Middle cerebral artery occlusion; Rat; Retina; Retinal ischemia.

Figure 1

Diagram of MCAO filament placement. The filament is fed through the external carotid artery (ECA) and into position along the internal carotid artery (ICA) to block the middle cerebral artery (MCA). Note the close proximity of the ophthalmic artery (OphA). CAs, Ciliary Arteries; CRA, Central Retinal Artery; ACA, Anterior Cerebral Artery; PCA, Posterior Cerebral Artery; BA, Basilar Artery; CCA, Common Carotid Artery. Modified from Steele et al., 2008.

Figure 2

Correlation of infarct size with behavioral responses. Representative sequential coronal slices of MCAO brains treated with TTC in an animal that received a neuroscore of 3 (A) or 6 (B). The infarct (white tissue) is larger in animals with higher neuroscores. C) Neuroscores showed a significant positive correlation with infarct size (R² = 0.805, p < 0.005).

Figure 3

Representative waveforms from moderate and severe MCAO eyes, contralateral eyes, and naïve controls at 2 days post-MCAO. For both ERG and oscillatory potential waveforms, severe animals showed decreases in amplitude in both MCAO and contralateral eyes, while moderate animals showed delays in latency in MCAO eyes. A) ERG waveforms in response to 137 cd s/m² flash stimuli. The two gray lines mark the control a- wave and b- wave, respectively. Arrows indicate delayed responses. B) Oscillatory potentials (OPs) from filtered ERGs in response to 137 cd s/m² flash stimuli. The gray line marks OP4 for the control trace. Arrows indicate delayed responses.

Figure 4

Mean ERG results at 2 days post-MCAO. A) A trend for reduction in dark-adapted a-wave amplitudes was observed in severe MCAO and contralateral eyes [Repeated measures ANOVA, F(4, 114) = 2.131, p < 0.01]. B) Significant delays were observed in a-wave implicit time in moderate MCAO eyes with a trend for delay in severe MCAO eyes [Repeated measures ANOVA, F(4, 111) = 5.650, p < 0.001]. C) A significant reduction in dark-adapted b-wave amplitudes was observed for severe MCAO and contralateral eyes while a trend for reduction in amplitude was observed in moderate MCAO eyes [Repeated measures ANOVA, F(4, 156) = 5.117, p < 0.002]. D) For b-wave implicit times, a trend for delay was observed in contralateral and MCAO eyes from the moderate group. E) A significant reduction in dark-adapted oscillatory potential amplitudes was observed in severe MCAO eyes with a trend for a decrease in severe contralateral and moderate MCAO eyes [Repeated measures ANOVA, F(4, 156) = 4.166, p < 0.007]. F) For oscillatory potential implicit times, we observed a significant delay in moderate MCAO eyes versus control and severe contralateral eyes [Repeated measures ANOVA, F(4,153) = 7.422, p < 0.001].

Figure 5

ERG b-wave responses at the brightest flash stimulus (137 cd s/m²) showed a significant correlation with neurological deficit scores (R² = −0.478, p < 0.05).

Figure 6

A–C) No significant differences were observed in number of rows of photoreceptor nuclei in the outer nuclear layer (A) or in cell counts in the inner nuclear layer (B). C) A significant reduction in retinal ganglion cells was observed in severe MCAO retinas versus moderate contralateral and controls [Repeated measures ANOVA, F(4, 180) = 2.940, p < 0.05]; ON, optic nerve. D) Representative micrographs of cresyl violet staining in a control retina, a moderate MCAO retina, and a severe MCAO.

Figure 7

Representative micrographs from sections immunostained for GS and GFAP. GFAP and GS were upregulated in the retina following MCAO. GFAP labeling was observed in Müller cells in severe and some moderate MCAO retinas at 3 days. In all other groups, GFAP staining was observed only in the nerve fiber layer (NFL), where GFAP-expressing astrocytes typically present. Increased GS labeling was observed in severe MCAO and severe contralateral eyes and in moderate MCAO eyes at 3 days. Moderate contralateral at 3 days and moderate MCAO and contralateral at 9 days showed levels similar to controls. GCL, Ganglion Cell Layer; IPL, Inner Plexiform Layer; INL, Inner Nuclear Layer; OPL, Outer Plexiform Layer; ONL, Outer Nuclear Layer; IS, Inner Segments; OS, Outer Segments.

Figure 8

Mean GS immunolabeling intensity as percent of naïve control for severe and moderate MCAO eyes versus contralateral eyes. A) A significant increase in GS intensity was observed between MCAO eyes and contralateral eyes within the severe group and moderate 3-day group but not the moderate 9-day group. Comparisons between severity and time points revealed that GS intensity was significantly increased in severe MCAO eyes compared with 9-day moderate MCAO eyes and in severe contralateral eyes compared with 3-day moderate contralateral eyes [Repeated measures ANOVA interaction effect, F(2, 13) = 8.006, p < 0.005]. B) A significant inverse correlation was observed between GS staining intensity and ERG b- wave at the bright flash stimulus (137 cd s/m²; R² = −0.627, p < 0.05).