Gene-environment interaction reveals hidden heritability: plasma vitamin D concentration and its interaction with vitamin D receptor gene polymorphisms in Parkinson disease. L. Wang1,2, M. L. Evatt3, L. Maldonado1, W. R. Perry1, J. C. Ritchie4, G. W. Beecham1,2, E. R. Martin1,2, J. L. Haines5, M. A. Pericak-Vance1,2, J. M. Vance1,2, W. K. Scott1,2 1) Hussman Institute for Human Genomics, University of Miami Miller School of Medicine, Miami, FL; 2) Dr. John T. Macdonald Foundation Department of Human Genetics, University of Miami Miller School of Medicine, Miami, FL; 3) Department of Neurology, Emory University School of Medicine, Atlanta, GA; 4) Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA; 5) Center for Human Genetics Research, Vanderbilt University School of Medicine, Nashville, TN.

   Parkinson disease (PD) is a complex disease with both genetic and environmental risk factors. Low plasma vitamin D concentrations (vit D) have been associated with increased risk of PD in several studies. Single nucleotide polymorphisms (SNPs) in the vitamin D receptor (VDR) gene, which encodes the major mediator of vit Ds biological actions, are also associated with PD in multiple candidate gene studies. However, similar associations were not observed in other candidate gene and genome-wide association studies (GWAS). We hypothesize that these genetic effects are modified by vit D, and that joint analysis of genotype and vit D concentration may reconcile the disagreement across studies. Thus we jointly analyzed the association of vit D and VDR SNPs with PD in a GWAS sample of 484 PD cases and 409 controls (Edwards et al., 2010). LC-MS/MS was used to measure vit D concentration in stored plasma samples. Association between vit D and PD was evaluated by logistic regression. A 4 degree of freedom joint test of SNP genotype and interaction effects was conducted as suggested by Kraft et al. (2007). For each SNP, a full model containing the SNP (coded as genotypes), vit D deficiency (<20 ng/mL vs >=20 ng/mL), interaction terms, and three covariates (age, sex and sampling season) was compared to a restricted model containing only vit D deficiency and the covariates. Vit D deficiency was strongly associated with PD (Odds Ratio (OR)=2.7, 95% Confidence Interval (CI): (1.9, 3.7), P<0.0001). Joint tests were significant (likelihood ratio test P<0.05) for four SNPs: rs12721364, rs886441, rs2189480, and rs11574026. Analysis stratified by vit D deficiency found that the effect of VDR SNP genotypes on PD depended on vit D concentration. A significant dose-effect of the rare allele at rs886441 on PD is observed in the vit D non-deficient stratum (ORhet=1.6, CI=(1.1, 2.3); ORhom=3.7, CI=(1.6, 8.8)) but not in the vit D deficient stratum (ORhet=1.1, CI=(0.6, 2.0); ORhom=1.0, CI=(0.3, 4.2)). Similar patterns were observed at the other three SNPs. Our data not only strongly support vit D deficiency as a risk factor for PD but also, for the first time, demonstrate that the effect of VDR polymorphisms on PD depends on vit D concentration. This observation provides a potential explanation of the inconsistency across studies of VDR SNPs in PD where vit D concentration was not considered.

You may contact the first author (during and after the meeting) at