Investigating Genetic Variants of the Folate One Carbon Metabolism (FOCM) Pathway for Associations with Recurrent Stroke and Measures of Homocysteine

Stroke is currently the fifth leading cause of death and the leading cause of long-term disability in the United States. Stroke accounts for over 140 , 000 deaths annually and according to the Centers for Disease Control and Prevention , approximately 795 , 000 individuals will suffer a stroke each year. More than 185 , 000 of these strokes will be a second or recurrent stroke. Stroke is a polygenic disease with multiple genetic and environmental risk factors. Scientific literature has been able to identify many of the environmental risk factors for stroke , including obesity , history of smoking , and hypertension. There are also heritable risk factors , as indicated by race and family history of stroke , that contribute to the disease , however many of the genetic determinants have not been fully elucidated. Due to advancements in next-generation sequencing methods , geneticists have devised techniques to identify genetic risk factors for complex diseases such as stroke. The Genome Wide Association Study (GWAS) is a method commonly utilized by geneticists to scan single nucleotide polymorphisms (SNPs) across the entire genome to detect associations between genotypes and phenotypes. An ideal population to investigate possible disease causing variants for stroke and recurrent stroke is the Vitamin Intervention for Stroke Prevention (VISP) clinical trial. VISP was a double blinded , randomized , multi-center clinical trial that enrolled 3 , 680 ischemic stroke patients to determine if vitamins B6 , B12 , and folic acid supplementation could prevent recurrent stroke or other cardiovascular related deaths. Multiple GWAS have been performed using the VISP population , identifying SNPs associated with numerous traits including recurrent stroke and metabolites of the folate one carbon metabolism (FOCM) pathway. The FOCM pathway is a critical regulatory network that contains many genes that have been implicated in cardiovascular disease risk. The FOCM pathway has been studied extensively for its role in DNA methylation , purine synthesis , and other essential metabolic reactions. Using the VISP population and the GWAS approach , we have identified numerous SNPs located in the FOCM pathway genes that have been associated with stroke related phenotypes. The purpose of this study is to use fine-mapping approaches to identify the likely causal SNPs driving the associations detected from our GWAS finding in five genes from the FOCM pathway genes (MTR , MTHFR , GNMT , CBS , and TCN2). Illumina based sequencing was conducted on a subset of 96 VISP individuals , identifying a total of 2 , 632 SNPs across the five candidate genes. High priority SNPs were identified based on novelty and their location within genes (coding regions and non-synonymous amino acid changes were of the highest priority) , for subsequent genotyping in the full VISP population , consisting of 2 , 100 individuals who provided consent for inclusion in genetic studies. Of the 53 high priority SNPs identified , only 38 had available TaqMan SNP genotyping probes. SNP genotyping was conducted using TaqMan SNP genotyping assays and allelic discrimination. Linear and logistic regression analyses were performed for 27 of the 38 SNPs using unadjusted , minimally adjusted (age , sex , and PC 1-10) and fully adjusted (age , sex PC 1-10 , and treatment) models for association with 10 phenotypes from the VISP study (baseline homocysteine , post-methionine load homocysteine , delta-post methionine load homocysteine , baseline vitamin B6 , baseline vitamin B12baseline folate , previous number of strokes , VISP recurrent stroke , VISP vascular end points , and recurrent stroke ever). We identified associations for three SNPs with measures of vitamin B12 for the unadjusted model , but these findings did not remain significant after adjusting for additional covariates. We also identified several SNPs that approached genome wide significance for multiple phenotypes , including measures of post-methionine load homocysteine , delta-post methionine load homocysteine , and VISP recurrent stroke. In conclusion , none of the SNPs identified in this study displayed evidence of association more significant than the SNPs previously identified in our GWAS. These results suggest that the novel and coding SNPs selected for this study may not influence recurrent stroke and stroke related phenotypes. Furthermore , additional fine-mapping may be required to definitively identify the likely causal genetic variant(s) across these five loci