Evolutionarily recent, exonic mutations are strong genetic risk factors for endometrosis

Abstract

Most genetic studies of endometriosis have searched for relatively common “ancestral” genetic markers, even though most diseases with decreased reproductive fitness are more likely to be impacted by recent (rare) and even de novo mutations. Whole genome/exome sequencing has now been published on nearly a million human subjects. Collectively these studies have identified thousands of rare exome variants (seen in < 1% of the general population) predicted to affect protein function. We suspect that many of these mutations contribute to reproductive disorders. We tested 27,924 rare exonic variants with a frequency < 0.01 among controls and with a minimum of 3 heterozygous individuals in both our case and control populations. non-synonymous protein-damaging exome variants for association with endometriosis. 1,537 Caucasian endometriosis patients were genotyped and compared with published population data (n>50,000). All endometriosis subjects had surgically proven disease. All DNA samples were tested using the Infinium HumanExome BeadChip (Illumina, San Diego). The significance of association was calculated using Fisher’s Exact Test. The in-silico annotation of variant function was predicted using polyphen 2 database. 21 rare exonic variants were associated with endometriosis (p<5×10-8) with a mean odds ratio of 6.7. Genes affected were: AP1M2, BTNL2, C8orf74, CLEC4F, CR1L, FAM71E2, FRAS1, GJB2, HYAL2, IKBKE, IQCH, KRTAP19-8, PMFBP1, PPM1E, RAPSN, RHBDD1, SLC24A5, SLC38A6, STAM2, THOC5, none of which have previously been associated with endometriosis. The high odds ratio suggests that these markers are stronger risk factors than the single nucleotide polymorphism (SNP) associations published to date. Nineteen of the variants are non-synonymous and 6 of the variants are reported by PolyPhen2 as “possibly damaging”. Although some of the implicated genes are involved in inflammation, macrophage activation, adhesion, and receptor and signaling activity, most variations point to proteins not previously studied in the context of endometriosis. Rare protein altering mutations of large effect contribute to the genetic risk for endometriosis. Finding of novel genes not previously implicated in endometriosis will help expand the functional understanding of the pathogenesis of endometriosis.