Cleft lip and/or palate (CL/P) is a group of congenital birth defect caused by the failure of the lip and/or palate to properly fuse during facial development. This defect occurs in approximately 1:700 live births and is the most second most common developmental defect. Twin studies and evaluation of family history reveals that risk for CL/P is influenced by genetics. However, to date less than half of the heritable risk for CL/P has been ascribed to specific genes.
To identify new genes involved in CL/P, our colleagues, Dr. Manak and Dr. Murray, screened DNA from CL/P patients for rare copy number variants. A single copy deletion of Isthmin1 (ISM1) was identified in this screen. To test the hypothesis that this deletion contributed to the pathogenesis of CL/P we conducted functional tests of the zebrafish ortholog, isthmin1. The results indicated that Ism1 is necessary for development of the face in zebrafish, supporting the hypothesis.
Together with Dr. Bassuk, we applied a similar approach to another structural birth defect, spina bifida. Dr. Manak discovered a de novo single copy deletion encompassing Glypican 5 (GPC5) and part of Glypican 6 (GPC6) in a spina bifida patient. Our functional tests in zebrafish support the notion that mutations in GPC5 cause spina bifida in some patients.
To identify additional CL/P loci, we investigated two putative transcriptional targets of Interferon Regulatory Factor 6 (IRF6), a transcription factor encoded by a gene which is mutated in the majority of patients with Van der Woude orofacial clefting syndrome. The Cornell lab found evidence that Irf6 regulates expression of grhl3 and klf4 in zebrafish periderm. Partly in response to our findings, Jeff Murray's group sequenced the coding region GRHL3 gene in Van der Woude patients lacking IRF6 mutations and found 8 different coding mutations in GRHL3. We tested 5 of theses GRHL3 mutations in zebrafish-based functional studies, and found the 5 patient-derived GRHL3 variants had dominant negative effects. We conclude that GRHL3 is indeed a CL/P locus. Because all the patient derived variants were dominant negative (as opposed to null), and because such variants would be expected to block the function of the other copy of GRHL3 as well as other family members (GRHL1 and GRHl2), which are also cleft candidate genes.
Multiple KLF4 coding mutations were also detected in patients with CL/P. We tested one of them but did not detect evidence of disruption in protein function. We conclude that this mutation does not seem to affect the function of KLF4, even though it was predicted to be damaging. This enforces the idea that conclusions from in silico studies should be examined in vivo.
Finally, the protein structure of Irf6 has been examined to identify important residues within the C-terminus of the protein. Using constructs built by Dr. Mankad, we tested 5 different phosphor-mimetic amino acid substitutions. Of the variety of constructs, only Irf6 S447D in zebrafish Irf6 was able to cause ectopic expression of downstream Irf6 targets. This predicts that phosphorylation at S447 is sufficient to activate Irf6.
All of these studies have expanded our understanding of the genetics behind CL/P, either by discovering new loci in human patients and testing them in Danio rerio, or from finding downstream targets of Irf6 in zebrafish, sequencing human patients for mutations in those genes, and then testing the functional changes of those variants. From our work with zebrafish, we can determine additional components of the IRF6 regulatory network.