Genetic analysis and functional assessment of a TGFBR2 variant in micrognathia and cleft palate

Abstract

Cleft lip and cleft palate are among the most common congenital anomalies and are the result of incomplete fusion of embryonic craniofacial processes or palatal shelves, respectively. We know that genetics play a large role in these anomalies but the list of known causal genes is far from complete. As part of a larger sequencing effort of patients with congenital craniofacial anomalies, we identified a rare candidate variant in transforming growth factor beta receptor 2 (TGFBR2). This variant alters a highly conserved amino acid and is predicted to be pathogenic by a number of metrics. The family history and population genetics suggest that this specific variant would be incompletely penetrant, but this gene has been convincingly implicated in craniofacial development. In order to test the hypothesis this might be a causal variant, we used genome editing to create the orthologous variant in a new mouse model. Surprisingly, Tgfbr2V387M mice did not exhibit craniofacial anomalies or have reduced survival, suggesting Tgfbr2V387M is not a causal variant for cleft palate/ micrognathia. The discrepancy between in silico predictions and mouse phenotypes highlights the complexity of translating human genetic findings to mouse models. We expect these findings will aid in interpretation of future variants seen in TGFBR2 from ongoing sequencing of patients with congenital craniofacial anomalies.

Fig 1. A novel TGFBR2 allele.

(a) TGFBR2 protein domain structure. Amino acid V387 is in an area of highly conserved sequence in the kinase domain.

Fig 2. A mouse model of Tgfbr2^V387M missense allele.

The amino acid and DNA sequence of Tgfbr2 shows high conservation between human (a) and mouse (b). (c) Sanger sequence of mice showing desired sequence change from G to A indicated by the red box in mutants. (d) PCR genotyping followed by restriction digest indicating ability to clearly differentiate wild-type, heterozygous, and homozygous mutant mice. (e) Survival of all mice are not significantly different than Mendelian expectations.

Fig 3. Craniofacial analysis of Tgfbr2^V387M mice.

(a-D) Skeletal preparations of animals at postnatal day (P) 60 from the lateral view (a,b) or focused on the palatal surface (c,d) do not appear any different between wild-type (a,c) and Tgfbr2 ^V387M/V387M homozygous mutants (b,d). (e,f) Histological analysis of wild-types (e) and mutants (f) at E18.5 also did not reveal any differences. Quantification of mandible length (g), skull length (h), skull width (i), ratio of mandible length to skull (j) or snout length (k) showed no difference between genotypes. Scale bars in a-f indicate 1 mm. Statistical values shown are from a one-way ANOVA.

Fig 4. Western blot analysis of TGF β signaling in Tgfbr2^V387M mice.

Western blot analysis with antibodies for phosphorylated SMAD2 (p-SMAD2), and SMAD2 antibodies. Quantification shows the ratio of pSMAD2 to total SMAD2 is not significantly different among genotypes. Statistical values shown are from one-way ANOVA.