Abstract
Purpose :
Recent genome wide association studies have identified 305 common genetic variants robustly associated with primary open angle glaucoma (POAG) and primarily located in noncoding regions. While noncoding variants can contribute to disease by regulating gene expression, coding variants disrupting protein function may also contribute to complex disease such as is known for age-related macular degeneration. The purpose of this study is to complete an exploratory analysis to assess the disease contribution of protein disrupting variants impacting genes located in genomic regions associated with POAG.
Methods :
2606 cases and 2606 controls from the NEIGHBORHOOD NEIGHBOR and MEE cohorts were genotyped using the HumanExome BeadChip array. After quality control variants affecting protein structure (missense, nonsense, splice site), and with at least moderate deleterious effects (CADD >10), and in genes located within genomic regions associated with POAG (3617 genes), were selected for association analysis using Fisher exact test and gene burden testing (N= 36,737 variants). The majority (83%) of variants were low frequency or rare (minor allele frequency less than 0.5%), with more common variants included in the analysis if they were protein disrupting.
Results :
Two common coding variants were significantly associated with POAG at study-wide threshold (P<1.36E-6), CAV2 (Q117E, P= 2.84E-7) and SIX6 (H141N, P=1.58E-10), and 862 variants demonstrated at least nominal evidence for association (P< 0.05). Of these, two or more variants with the same direction of effect were found in each of 39 genes. Gene burden testing for these 39 genes identified significant association (P<9.6E-4) for SLC44A4, SIX6, MUC2 and SSPN and interesting association (P<0.01) for APC2, COL11A1, LYPLAL1, IRPKB, and KCNB1. Among the group of genes with at least two nominally associated variants are genes involved in neurodegeneration (ATXN2, ATG2A), ocular development (SIX6, CDK133, ZHX2) and extracellular matrix function (CHI3L1, COL11A1, COL24A1, FNDC1).
Conclusions :
These results support the hypothesis that coding variants can contribute to complex disease pathogenesis and that some causal genes within GWAS loci may impact the disease through protein disruption. Larger sample sizes are needed for further evaluation of these protein coding variants.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.