Orofacial Cleft 2

For a phenotypic description and a discussion of genetic heterogeneity of nonsyndromic cleft lip/palate (CL/P), see 119530.

Ardinger et al. (1989) observed a significant association between 2 RFLPs at the transforming growth factor-alpha (TGFA; 190170) locus on chromosome 2p13 and the occurrence of clefting. The authors suggested that either the TGFA gene or DNA sequences adjacent to the locus contribute to the development of some cases of cleft lip with or without cleft palate in humans. However, in a study of 7 families with CL/P segregating in a dominant manner, the TGFA haplotype associations reported by Ardinger et al. (1989) were not seen, and in 1 family clefting did not cosegregate with TGFA, thus ruling out tight linkage (Hecht et al. (1990, 1991)).

In 96 unrelated patients with nonsyndromic CL/P, Chenevix-Trench et al. (1991) confirmed the existence of an excess frequency of the same TaqI allele found by Ardinger et al. (1989).

Vintiner et al. (1992) studied 8 families with cleft lip with or without cleft palate inherited in an apparently autosomal dominant manner and excluded linkage with TGFA.

In a study of 3 RFLPs at the TGFA locus in 60 unrelated British Caucasian subjects with nonsyndromic CL/P and 60 controls, Holder et al. (1992) found a highly significant association between the TaqI RFLP and the occurrence of clefting, and no significant association with the other 2 RFLPs.

Chenevix-Trench et al. (1992) extended their analysis of the TGFA TaqI RFLP to 2 other TGFA RFLPs and 7 other RFLPs at 5 candidate genes. Significant associations with the TGFA TaqI and BamHI RFLPs were confirmed. Of particular interest, in view of the known teratogenic role of retinoic acid, was a significant association with a PstI RFLP of RARA (180240) (P = 0.016, not corrected for multiple testing). The effect on risk of the A2 allele appeared to be additive; although the A2A2 homozygote only has an odds ratio of about 2 and recurrence risk to first-degree relatives of 1.06, because it is so common, it may account for as much as a third of the attributable risk of clefting. There was no evidence of interaction between the TGFA and RARA polymorphisms on risk, but jointly they appeared to account for almost half the attributable risk of clefting.

Sassani et al. (1993) and Shiang et al. (1993) likewise found a significant association between TGFA alleles and CL/P. Sequence analysis of the variants disclosed a cluster of 3 variable sites within 30 bp of each other in the 3-prime untranslated region previously associated with an antisense transcript in the TGFA gene (Shiang et al., 1993).

Farrall et al. (1993) attempted to resolve the apparent paradox concerning the role of TGFA in CL/P: the very strong support from population-based studies for TGFA as a susceptibility locus but the seeming exclusion of TGFA as a candidate locus by linkage studies in a series of multiplex CL/P families (Hecht et al., 1991). Farrall et al. (1993) stated that the genotypic association data show that TGFA is a susceptibility locus of modest effect which should be allowed for in the specification of genetic models for linkage analysis, and noted that analysis under an inappropriate model may result in erroneous exclusion of a candidate susceptibility locus.

Studies to determine whether women who smoke during early pregnancy are at increased risk of delivering infants with orofacial clefts have yielded conflicting results. In part, the inconclusive or contradictory findings result from inadequate study design. Using a large population-based case-control study, Shaw et al. (1996) investigated whether parental periconceptional cigarette smoking was associated with an increased risk for having offspring with orofacial clefts. They also investigated the influence of genetic variation at the TGFA locus on the relation between smoking and clefting. They found that risks associated with maternal smoking were most elevated for isolated CL/P and for isolated cleft palate when mothers smoked 20 or more cigarettes per day. Analyses controlling for the potential influence of other variables did not reveal substantially different results. Clefting risks were even greater for infants with the TGFA allele previously associated with clefting when the mothers smoked 20 or more cigarettes per day. These risks for white infants ranged from 3-fold to 11-fold across phenotypic groups. Paternal smoking was not associated with clefting among the offspring of nonsmoking mothers, and passive smoke exposures were associated with at most slightly increased risks. Shaw et al. (1996) concluded that this is an example of gene-environment interaction in the occurrence of orofacial clefting.

Mitchell (1996) analyzed published data on the association between nonsyndromic CL/P and genetic variation at the TGFA locus. She found evidence of significant heterogeneity in the TGFA allele frequencies between cases, but not controls, from different studies. Because of difficulties in identifying the source(s) of the observed heterogeneity, the evidence regarding an association remained inconclusive, in the opinion of Mitchell (1996).

A collection of 38 nonsyndromic orofacial cleft families of Italian ancestry from northeastern regions of Italy were studied by Pezzetti et al. (1998) for linkage to 2p13. The same collection of families had been used to demonstrate linkage to the 6p23 chromosome region, with the presence of genetic heterogeneity (Carinci et al., 1995; Scapoli et al., 1997). Evidence for genetic heterogeneity in this family set was apparent by both pairwise and multipoint linkage analyses. Moreover, lod scores greater than 3 were found for marker D2S378 on chromosome 2 when families linked to the 6p23 markers were analyzed. Taken together, these results indicated a role for the TGFA locus, or another gene physically close to it, and suggested an interaction between 2 different genes, OFC1 (119530) and OFC2, mapped in 6p23 and 2p13, respectively, in the development of the cleft.

Sull et al. (2009) genotyped 297 CL/P case-parent trios from 4 populations for 17 SNPs in the TGFA gene. When all trios were combined, 4 SNPs showed significant excess maternal transmission, 2 of which gave significant parent-of-origin likelihood ratio test values: rs3821261 (p = 0.004; imprinting odds ratio, 4.17) and rs3771475 (p = 0.027; imprinting odds ratio, 2.44). Haplotype analysis of these 2 SNPs also supported excess maternal transmission. Sull et al. (2009) suggested that TGFA influences risk of CL/P through unconventional means, with an apparent parent-of-origin effect involving excess maternal transmission and possible interaction with maternal exposures.