Glucocorticoid receptor binding sites are enriched for genetic associations with intraocular pressure variation

Abstract

Purpose : Glucocorticoids (GC) are a class of steroid hormones with strong anti-inflammatory effects used to treat various conditions, including ocular disorders. GCs regulate gene expression by binding to glucocorticoid receptors (GR) in the cytoplasm, which in turn translocate to the nucleus and either bind directly to glucocorticoid response elements (GREs) throughout the genome or indirectly via tethering to other transcription factors. GCs elevate intraocular pressure (IOP) in a large proportion of patients, which can lead to optic nerve damage and vision loss. There is long-standing evidence that genetic factors influence the development of GC-induced ocular hypertension; however, the genetic basis remains unknown. Here we test the hypothesis that common variants in GR binding sites (GBS) affect IOP levels.

Methods : We assessed whether common variants (minor allele frequency, MAF >1%) that fall in direct or tethered GBS regions were enriched for genetic associations (P<0.05) with IOP compared to null variants using QTLEnrich, which corrects for linkage disequilibrium (LD), MAF, and distance to gene transcription start sites. We analyzed 10,900 direct GBS and 16,100 tethered GBS, identified with ChIP-Seqin alveolar basal epithelial cells in response to GC treatment (Vockley et al., 2016). Summary statistics from the UK Biobank IOP genome-wide association study of ~140,000 individuals were used. We further tested whether any of the 112 known IOP loci contained a GBS. If so, we checked whether variants in the GBS were associated with gene expression changes by inspecting expression quantitative trait loci (eQTLs) from 49 human tissues in GTEx.

Results : We found significant enrichment of IOP associations among variants in both direct and tethered GBS regions (P<1E-05, adjusted fold-enrichment of 2.1 and 2, respectively), proposing hundreds of new genetic associations in GREs that may affect IOP levels. Of the 112 known IOP loci, 14 (12.5%) and 33 (29.5%) loci are in LD (r² > 0.8 and r² > 0.5, respectively) with at least one GBS. The majority of these GBS are significant eQTLs in one or more tissues, including artery, suggesting possible target genes through which the GBS may be mediating their effect on IOP.

Conclusions : Our work proposes that genetic variation in GBS significantly contributes to IOP variation, and may play a role in GC-induced ocular hypertension.

This is a 2020 ARVO Annual Meeting abstract.