Whole exome sequencing of phenotypically extreme preterm infants identifies innate immune system-related biological pathways in retinopathy of prematurity

Abstract

Purpose : Retinopathy of prematurity (ROP) is a potentially blinding retinal neurovascular disease that affects premature infants. In addition to known clinical risk factors such as low birth weight (LBW) and early gestational age, there is a growing body of evidence supporting a genetic basis for ROP. However, it is still unclear which variants and/or genes are significantly associated with ROP. The purpose of this study is to identify novel, potentially pathogenic gene variants and biological pathways associated with ROP by analyzing whole exome sequencing data from preterm infants. To maximize variant discovery, we selected 100 subjects at phenotypic extremes (e.g. non-LBW infants with severe ROP, LBW infants with no ROP).

Methods : As part of a multicenter ROP cohort study, Imaging and Informatics for ROP (i-ROP) study, 100 non-Hispanic Caucasian preterm infants were included. All participants were subjected to whole exome sequencing on the BGISEQ-500 sequencer (BGI, Hong Kong). Candidate genes related to the pathogenesis of ROP were selected by literature review on human and animal studies (n=164) and an additional 99 genes related to the 164 genes were identified by predicting interaction networks with GeneMANIA. Within these genes, rare and low-frequency coding variants which were predicted to have functional damaging effects were subjected to gene set analysis using GSA-SNP.

Results : Mean BW of infants with type 1 or 2 ROP (n=58) and mild or no ROP (n=42) were 744g and 995g, respectively. We found 119 nonsynonymous SNPs predicted to be damaging, 3 stop-gain, 4 splice-site variants, 4 frameshift insertions or deletions (InDels), and 5 disruptive inframe InDels. The frequencies of subjects having the alternative A allele of SERPINE1 rs6090 and TLR2 rs5743704 were significantly higher in type 1 or 2 ROP than in mild or no ROP (10.3% versus 0% in both SNPs; P=0.038). Gene set analysis showed six Gene Ontology (GO) terms were significantly associated with ROP (q-value < 0.05). These six GO terms were mostly related to innate immune system, including “cellular response to lipopolysaccharide” and “regulation of interleukin-8 production.”

Conclusions : This study suggests that variants in innate immune system genes may be associated with ROP. These findings may contribute to better understanding about the genetics and pathophysiology of ROP.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.