Identification of 16q21 as a modifier locus for orofacial cleft phenotypes.

J.C. CARLSON; E.E. CASTILLA; C. PADILLA; G.L. WEHBY; J.C. MURRAY; E.J. LESLIE; A. PETRIN; I.M. ORIOLI; A.R. VIEIRA; E. FEINGOLD; J. STANDLEY; K. CHRISTENSEN; F. POLETTA; S.M. WEINBERG; M.L. MARAZITA

Orlando

Congreso; American Society of Human Genetics 67 Annual Meeting; 2017

American Society of Human Genetics

Orofacial clefts (OFCs) are common, complex birth defects with extremely heterogeneous phenotypic presentations. Two common subtypes ? cleft lip alone (CL) and cleft lip and palate (CLP) - are often combined for genetic analysis (i.e. cleft lip with or without cleft palate, CL/P). In such studies of CL/P, the underlying assumption is that the effect on risk of cleft may be identical for both CL and CLP subgroups. However, such investigations cannot separate variants for which the risk of OFC differs by subtype. Despite substantial data supporting a shared etiology between CL and CLP, recent evidence suggests that subtype-specific risk factors and/or genetic modifiers may distinguish these two phenotypes. Therefore, identification of genetic factors that act as modifiers of cleft subtypes is critical for understanding the substantial variability of OFCs between individuals. We performed genome-wide association scans for genetic modifiers by directly comparing 450 CL cases with 1692 CLP cases recruited from 13 countries around the world. Gene-based scans of low-frequency variants (MAF1%), a 16q21 locus was strongly associated with cleft type (p=5.6×10-8), which replicated in an independent sample of 360 CL and 725 CLP cases from Brazil, the Philippines, and Mongolia. Furthermore, the 16q21 alleles that increased CL risk were found at highest frequencies among individuals with a family history of CL compared to those with histories that included CLP (p=0.003). We further investigated the potential modifying behavior of the 16q21 locus on known CL/P risk loci with gene-gene interaction analyses and found significant interactions between the modifier locus and 8q21 (p=0.012) and 9q22 (p=0.023). In these interactions, the risk of CL increased only in the presence of both the CL/P risk allele and the 16q21 modifier; CLP risk was unchanged by the modifier. 16q21 contains several biologically plausible candidates including LINC00922, a long-non-coding RNA with putative gene regulatory function and CDH11, a craniofacially expressed protein that regulates extracellular matrix production. This study demonstrates the existence of common and low-frequency phenotypic modifiers for OFCs and further elucidates the complex genetic architecture of OFCs by identifying biologically plausible elements responsible for phenotypic heterogeneity.