Mendelian randomization identifies FLCN expression as a mediator of diabetic retinopathy

The goal of the study was to identify genes whose aberrant expression can contribute to diabetic retinopathy. We determined differential response in gene expression to high glucose in lymphoblastoid cell lines derived from matched type 1 diabetic individuals with and without retinopathy. Those genes exhibiting the largest difference in glucose response between diabetic subjects with and without retinopathy were assessed for association to diabetic retinopathy utilizing genotype data from a meta-genome-wide association study. All genetic variants associated with gene expression (expression QTLs; eQTLs) of the glucose response genes were tested for association with diabetic retinopathy. We detected an enrichment of the glucose response gene eQTLs among small association p-values for diabetic retinopathy. Among these, we identified FLCN as a susceptibility gene for diabetic retinopathy. Expression of FLCN in response to glucose is greater in individuals with diabetic retinopathy compared to diabetic individuals without retinopathy. Three large, independent cohorts of diabetic individuals revealed an enhanced association of FLCN eQTL to diabetic retinopathy. Mendelian randomization confirmed a direct positive effect of increased FLCN expression on retinopathy in diabetic individuals. Together, our studies integrating genetic association and gene expression implicate FLCN as a disease gene in diabetic retinopathy. We compared the differential response in gene expression to glucose for individuals with and without proliferative retinopathy. ‘Differential response’ in gene expression refers to gene expression comparisons that were determined by calculating the difference in response to glucose between groups. Specifically, we identified genes with a significant differential response in expression between diabetic individuals with and without proliferative diabetic retinopathy (RGpdr-ndr). We integrated these findings with results of a genome-wide association study meta-analysis of diabetic retinopathy and the results of a multi-tissue eQTL analysis to identify potential diabetic retinopathy susceptibility genes