Investigating Novel Chromosome 2 Genetic Variants Associated with Recurrent Stroke

As the fifth leading cause of death and leading cause of long-term disability , stroke affects 795 , 000 individuals annually in the United States alone. Recurrent stroke , which is far more deadly and disabling , accounts for 185 , 000 of the annual stroke cases. Stroke is a complex , multi-factorial disease that is influenced by environmental , lifestyle , and heritable factors. However , despite 40% of ischemic stroke risk being due to heritable factors , genetic contributors to stroke are not as well-documented as other risk factors. Utilizing the Vitamin Intervention for Stroke Prevention (VISP) clinical trial population , which was designed to determine if increased supplementation of B-vitamins reduced stroke vascular outcomes , we performed a genome-wide association study (GWAS) which identified a novel locus on chromosome 2 associated with recurrent stroke. Due to high linkage disequilibrium across this region , encompassing 7 known genes (SULT1C2 , SULT1C4 , GCC2 , LIMS1 , RANBP2 , CCDC138 , EDAR) , fine-mapping was performed to identify the most significant variants associated with recurrent stroke risk at this locus. Targeted DNA Next-Generation Sequencing (NGS) was performed , identifying 7 , 531 genetic variants of interest. After SNP prioritization , 28 high priority SNPs were genotyped and included in downstream analyses. In total , 164 SNPs (including 136 from the original GWAS) were analyzed for single-SNP and gene-based associations with stroke vascular endpoints and stroke-related phenotypes using Cox Proportional Hazards survival analyses , linear and logistic regressions , and SNP-Set Kernel Association Tests. Single-SNP analyses identified two novel associations for VISP recurrent stroke and composite endpoint , specifically within the low-dose trial arm. The most significant associations were observed between rs41280574 and days to VISP recurrent stroke (P=9.79x10-6) and days to composite endpoint (P=2.25x10-5). A second SNP , rs6542775 , was significantly associated with both VISP recurrent stoke (P=1.76x10-4) and days to VISP recurrent stroke (P=5.09x10-5). Gene-based analyses determined that the LIMS1 gene , containing rs6542775 , was significant for VISP recurrent stroke (P=5.79x10-4) , composite endpoint (P=1.21x10-3) , and disabling/fatal stroke , MI , or fatal CHD (P=5.95x10-3) in the low-dose treatment arm only. Secondary analyses of stroke-related biomarkers identified novel single-SNP and gene-based associations with phenotypes involved in blood clotting. Specifically , rs260632 , was associated with thrombomodulin in the total VISP population (P=2.67x10-5) and low-dose trial arm (P=2.87x10-4). RANBP2 was also found to be associated with thrombomodulin levels (P=3.26x10-3) within the entire VISP population. GCC2 , containing rs41280574 , was significant for total cholesterol levels (P=2.47x10-3) and SULT1C4 had associations with von Willebrand Factor (P=5.73x10-3) in the high-dose treatment arm. In total , this locus contains SNPs and genes that are strong future research candidates due to the novel significant associations with recurrent stroke and phenotypes related to blood clotting. Since the significant associations for stroke vascular outcomes were found solely in the low-dose treatment arm , it suggests a therapeutic response , highlighting potential treatment avenues. Furthermore , the associated variants with recurrent stroke both have implications in Wnt signaling , a regulator of fibronectin and E-cadherin , implicating this pathway with recurrent stroke risk. These results provide the foundation for future functional studies which could allow for improved risk assessment and targeted treatment for recurrent stroke