Integrative Mouse and Human Studies Implicate ANGPT1 and ZBTB7C as Susceptibility Genes to Ischemic Injury

Abstract

Background and purpose: The extent of ischemic injury in response to cerebral ischemia is known to be affected by native vasculature. However, the nonvascular and dynamic vascular responses and their genetic basis are not well understood.

Methods: We performed a genome-wide association study in 235 mice from 33 inbred strains using the middle cerebral artery occlusion model. Population structure and genetic relatedness were accounted for using the efficient mixed-model association method. Human orthologs to the genes associated with the significant and suggestive single-nucleotide polymorphisms from the mouse strain survey were examined in patients with M1 occlusions admitted with signs and symptoms of acute ischemic stroke.

Results: We identified 4 genome-wide significant and suggestive single-nucleotide polymorphisms to be associated with infarct volume in mice (rs3694965, P=2.17×10(-7); rs31924033, P=5.61×10(-6); rs32249495, P=2.08×10(-7); and rs3677406, P=9.56×10(-6)). rs32249495, which corresponds to angiopoietin-1 (ANGPT1), was also significant in the recessive model in humans, whereas rs1944577, which corresponds to ZBTB7C, was nominally significant in both the additive and dominant genetic models in humans. ZBTB7C was shown to be upregulated in endothelial cells using both in vitro and in vivo models of ischemia.

Conclusions: Genetic variations of ANGPT1 and ZBTB7C are associated with increased infarct size in both mice and humans. ZBTB7C may modulate the ischemic response via neuronal apoptosis and dynamic collateralization and, in addition to ANGPT1, may serve as potential novel targets for treatments of cerebral ischemia.

Keywords: brain ischemia; cerebral infarction; genetics; models, animal; stroke.

Figure 1. Genome-wide association analysis using mouse stroke model

A) Infarct ratios of 33 strains of inbred mice measured 24 hours after occlusion of the left middle cerebral artery. Infarct ratio was calculated using the Swanson method. B) Genome-wide association analysis of infarct ratio in the middle cerebral artery occlusion model with genome-wide p-values using the efficient mixed model association (EMMA). Blue line is the suggestive significance line set at p = 10⁻⁵. C) Circle of Willis in A/J mouse and C57L/J mouse demonstrating large and small P1s, respectively. The vasculature is visualized with India ink perfusion. D) Correlation between infarct ratio and the completeness of the circle of Willis assessed with the ratio of the left P1 to left ICA diameters. Error bars are standard errors.

Figure 2. Infarct volume in human middle cerebral artery occlusion

A) Algorithm for obtaining human SNPs given mouse data. B) Infarct volume (cm³) from 33 Caucasian patients with M1 occlusions.

Figure 3. ZBTB7C expression

A) Representative immunohistochemical staining of normal and infarcted cortical regions for ZBTB7C (green), endothelial cell marker CD31 (red), astrocytic marker GFAP (red), and neuronal marker NeuN (red) in SWR/J control and MCAO mice. Scale bar is 100 µm. B) Strain-specific difference in expression of ZBTB7C after middle cerebral artery occlusion compared to control mice using quantitative PCR. C) Comparison of proportions of CD31+, GFAP+, and NeuN+ cells that also express ZBTB7C in SWR/J mice after MCAO in the ipsilateral hemisphere, contralateral hemisphere and in control mice. D) Quantitative PCR analysis showing relative expression of ZBTB7C in HUVECs (6 control, 9 OGD) and murine endothelial cell line (6 control, 10 OGD) under oxygen glucose deprivation compared to controls. There was a significant increase in expression in both HUVECs and murine endothelial cells. * indicates p<0.05. Error bars are standard errors.