Nonsyndromic Orofacial Clefts – Identifying Putative Disease Causing Genes by CNV Analysis of Whole Exome Sequencing Data

Selina Hölzel1, Nina Ishorst1, Carola Greve1, Franziska Degenhardt1, Dmitriy Drichel2, Carlo Maj3, Michael Nothnagel2, Jayne Y. Hehir-Kwa4, Joris A. Veltman4, Teresa Kruse5, Carine Carels6, Iris van Rooij7, Kerstin U. Ludwig1, Elisabeth Mangold1

1 Institute of Human Genetics, University of Bonn, School of Medicine & University Hospital Bonn, Bonn, Germany
2 Cologne Center for Genomics, University of Cologne, Cologne, Germany
3 Institute for Genomic Statistics and Bioinformatics, University of Bonn, School of Medicine and University Hospital Bonn, Bonn, Germany
4 Department of Human Genetics, Donders Institute, Radboud University Medical Center, Nijmegen, The Netherlands
5 Department of Orthodontics, University of Cologne, Cologne, Germany
6 Orthodontics, University Hospitals KU Leuven, Leuven, Belgium
7 Department for Health Evidence, Radboud Institute for Health Sciences, Radboud university medical center, Nijmegen, The Netherlands

Nonsyndromic cleft lip with/without palate (nsCL/P) and cleft palate only (nsCPO) are frequent congenital malformations with high heritability estimates. Common risk loci explain only some of their heritability. Rare highly penetrant copy number variations (CNVs) encompassing possible clefting susceptibility genes might partially explain the missing heritability. Aim of this study was to identify such genes by detecting de novo CNVs from whole-exome sequencing (WES) trio data.

CNVs were called from WES data of 50 nsCL/P and 43 nsCPO Central European patient/parent trios using XHMM, CoNIFER and EXCAVATOR2. Defining XHMM as primary algorithm we established a filtering pipeline considering sample wide frequency, de novo status, calling intersection (CoNIFER and/or EXCAVATOR2), annotation to RefSeq-genes, overlap with regions of segmental duplications and population frequency. CNVs were validated using qPCR.

Three interesting de novo CNVs were identified: A 86kb deletion on Chr8 in an nsCL/P patient, a 16Mb deletion on Chr1 in an nsCPO patient, and an 877kb duplication on Chr2 in an nsCPO patient. Interesting candidate genes at these loci are LHX8, DLX1 and DLX2, all suggested to contribute to craniofacial development: Targeted mutations of Lhx8 in mice lead to an isolated cleft palate. In embryonic mice Dlx1 and Dlx2 are expressed in both the maxillary and the mandibular component of the first pharyngeal arches, and Dlx1/2-/- mutant mice exhibit a cleft palate.

Our results indicate that rare highly penetrant CNVs can contribute nsCL/P and nsCPO risk. To achieve more evidence for our findings, replication studies in larger patient/control cohorts are essential.