The development of the human lip and palate is a delicate process, requiring tissues from both sides of the face to migrate and meet neatly in the middle. It is therefore not surprising that cleft lip and palate, in which this process goes awry, is relatively common — it affects 1 in 700 newborns — and can be caused by either genetic or environmental influences. In an inventive and thorough study, Kondo and colleagues now identify the causative gene of two common cleft lip or palate disorders.

The two syndromes in question — Van der Woude (VWS) and popliteal pterygium (PPS) — are characterized by cleft lip and palate, with PPS individuals also sufferring from skin and genital abnormalities. It had been suspected that the two disorders are allelic, and that they mapped somewhere in the 1q32–41 interval. This was confirmed when the authors sequenced the candidate region in a pair of identical twins that were discordant for the VWS phenotype. The only genetic difference was a nonsense mutation in the interferon regulatory factor 6 ( IRF6 ) gene, which belongs to a family of nine transcription factors. IRF6 mutations were also found in 57 other families with either VWS or PPS; the causative role of IRF6 in VWS and PPS was further supported by the high expression of IRF6 in the tissues that are affected in the two disorders.

The two syndromes might be allelic but they are characterized by a different spectrum of mutations in the functional regions of IRF6: VWS-associated mutations map to both the DNA-binding and dimerization domains, whereas lesions in PPS patients affect only DNA-binding residues. These lesions could explain the dominant nature of VWS and PPS: the common, protein-truncating mutations seen in VWS are probably haploinsufficient null alleles, whereas the PPS mutant proteins have a dominant negative effect, as they retain the ability to bind protein partners but not DNA.

This work provides proof of principle that identical twins can be used to pinpoint disease genes, an approach that removes the complicating presence of neutral SNPs. Nonetheless, mutations in IRF6 don't explain everything, as phenotypic variation in VWS and PPS patients betrays the presence of modifier genes and/or random effects.