Importance, Challenges, and Opportunities of Gene-Environment Interactions (GxE) Research: A Study of Parkinson's Disease
MetadataShow full item record
University of Washington Abstract Importance, Challenges, and Opportunities of Gene-Environment Interactions (GxE) Research: A Study of Parkinson's Disease Nirupama Nini Shridhar Chair of the Supervisory Committee Dr. Karen Edwards Department of Epidemiology Objective. This interdisciplinary dissertation comprised a normative analysis: social justice in genomics research, and a statistical and functional analyses: evaluating gene-environment interactions (GxE) in Parkinson's disease (PD) research. The normative analysis examined the potential for unfair health distribution created by the genomics research agenda, and evaluated research methodologies that may be more socially just. The statistical and functional analyses evaluated gene-environment interactions GxE in PD. GxE are thought to be causal in complex disease risk. Understanding the mechanisms by which GxE alter risk in complex diseases represents a leverage point by which to modify disease biology. The intent here, was to find new susceptibility genes that are mutable (by environmental exposures), and in the causal pathway to PD risk. Smoking has been consistently shown to be associated with reduced risk of developing PD. One of the biggest effects of smoking is DNA methylation. This research used a novel gene selection method by following the environment into the body, by identifying genes that are known to be methylated by smoking. This set up an a priori hypothesis that somewhere along the network of genes known to be methylated by smoking is an interaction with gene networks involved in PD risk. The research also hypothesized that regulatory single nucleotide polymorphisms (rSNPs) on genes that are differentially methylated by smoking act as genetic determinants to epigenetic modifications. Methods. Normative Methods: A normative bioethical analysis was conducted in looking at the question of “what would be a socially just genomics research agenda?''. The analysis used Sen's 'capabilities approach', and work by other capabilities theorists, to define the framework for social justice. This framework was then applied to genomics research to examine social justice in genomics research. Statistical and Functional Methods: GxE association analysis were performed using case-control data from the NeuroGenetics Research Consortium (NGRC). The NGRC data contain 2000 PD cases and 1986 unrelated controls (n= 3986). The dataset, included genotype information for the SNPs on the 39 genes selected (n= 2281), and the environmental factor (smoking use). The genes were tiered based on evidence: Tier 1 (3 genes and 747 SNPs), and Tier 2 (36 genes and 1534 SNPs). The threshold for statistical significance (after adjusting for multiple comparisons) were: Tiers 1 and 2 (p=1.13 x 10-4 and p=5.67 x 10-5). Based on the results of the association analysis, the SNP results were then further characterized for functional consequence by synthesizing multiple lines of evidence to identify PD susceptibility genes from the perspective of the environmental exposure. Results. Normative Analysis: Gene-centric approaches do not meet the demands of social justice. GxE approaches have the potential to meet the demands of social justice. It is a socially just method because it could to lead to translational outcomes with public health utility. Association and Functional Analyses: No interactions were statistically significant (in either tier). They were however, statistically suggestive. Layering the functional evidence over the statistical evidence, a single gene emerged in Tier 2 that was statistically suggestive and with significant functional evidence. RARA, a gene in the retinoic acid signaling pathway coding for the protein RARα. The two SNPs showing evidence for functional consequences, (rs2120200 and rs36030243) are SNPs that alter gene expression and bind transcription factors. Results from the association analysis also had two SNPs from the same gene among the top 10 SNPs in the sensitivity analysis (rs12103711 and rs2715553). Rs12103711 is an intronic SNP of RARA (OR = 1.80, 95% CI: 1.23- 2.65, p = 0.002802) and rs2715553 is a cSNP of RARA (OR = 0.72, 95% CI: 0.57 - 0.91, p = 0.00563). Linkage disequilibrium(LD) was evaluated between the four SNPs in RARA (two from the functional analysis, and two from the statistical analysis) to determine if these SNPs were all pinpointing a single signal with evidence for regulation, and found that the four SNPs are in strong LD with each other, although the rs2715553, a cSNP is physically some distance away (35 kb) from the other three SNPs. We found rs2120200 and rs36030243 are likely rSNPs on the RARA gene that interact with smoking and are in LD with a coding SNP on the same gene. Conclusion. To our knowledge, no prior analysis has used a capabilities approach to examine social justice in genomics research. We developed a framework by which to analyze social justice in genomics research, and using a case study of Parkinson's disease (PD), we demonstrated that gene-environment interaction (GxE) methods meet the demands of social justice in genomics research. This exploratory analysis demonstrated that selecting genes by following the effect of the environmental exposure is a valid method of identifying susceptibility genes for complex diseases. We have used a novel approach to identifying genes and characterizing SNP results that warrant further study in understanding the interaction between smoking and PD. For genomic research to be effective and socially just, we need to gain a better understanding of how the environment interacts with genes to modify complex disease risk.