Molecular Pathogenesis
Several genes and gene families including TWIST1, TCF12, ERF, FGFs, FGFRs, MSX2, ALX4, EFNB1, EFNA4, NELL1, RUNX2, BMPs, TGF-βs, SHH, IGFs, IGFRs, and IGFBPs regulate patency of the sutures of the calvarium, likely by interacting with one another. Clinically, Saethre-Chotzen syndrome (SCS) has phenotypic overlap with other craniosynostosis syndromes, particularly Muenke syndrome, caused by the p.Pro250Arg pathogenic variant in FGFR3 [Muenke et al 1997]. While the two genes lead clinically to the same primary malformation –premature fusion of the calvaria – it is not known if they lie in the same, parallel, or independent pathways during calvarial development.
Gene structure.
TWIST1 comprises two exons and one intron. The first exon contains an open reading frame encoding a 202-amino acid protein, followed by a 45-bp untranslated portion, a 536-bp intron, and a second untranslated exon (reference sequences NM_000474.3 and NP_000465.1).
Pathogenic variants. To date, more than 209 variants in TWIST1 have been reported to cause SCS, which results from functional haploinsufficiency of Twist-related protein 1, a basic helix-loop-helix (HLH) transcription factor. The majority of reported pathogenic variants are missense, nonsense, or frameshift (i.e., deletions/insertions/duplications/indels); however, a significant number of large deletion or chromosome rearrangements have also been reported [Gripp et al 2000, Cai et al 2003a, de Heer et al 2005, Kress et al 2006, Foo et al 2009, Roscioli et al 2013, Paumard-Hernández et al 2015, The Human Gene Mutation Database (registration required)]. All TWIST1 pathogenic variants cause functional haploinsufficiency.
All of the disease-associated variants are located within the coding region; no splice variants, intronic variants, or changes within the second exon have been reported. No apparent mutational "hot spot" has been identified.
Nonsense variants that preclude translation of the DNA binding domain and the HLH domain have been identified from the 5' end of the coding sequence to the end of the HLH motif.
Missense variants are clustered within the functional domains.
Four persons have been identified with pathogenic variants in the C-terminus, known as the TWIST box, a highly conserved region that binds and inhibits RUNX2 activation [
Kress et al 2006,
Seto et al 2007,
Peña et al 2010]. RUNX2 is considered the "master switch" for osteoblast differentiation and activity.
Functional haploinsufficiency of TWIST1, whether due to mutation in the DNA binding, HLH, or TWIST box domains, results in disinhibition of RUNX2 and enhances osteogenesis.
Normal gene product. The Twist-related protein 1 is a member of a large family of basic helix-loop-helix (bHLH) transcriptional regulators. The bHLH motif is identified by the following:
The basic domain that mediates specific DNA binding to a consensus hexanucleotide E-box (CANNTG)
The HLH domains containing two amphipathic helices that act as dimerization domains (dimerization is required for DNA binding)
A loop region that separates the two helices, spacing them appropriately for DNA binding and causing formation of a bipartite DNA-binding groove by the basic domain
Abnormal gene product.
TWIST1 pathogenic variants lead to haploinsufficiency [El Ghouzzi et al 2000]. Haploinsufficiency of Twist-related protein 1 changes the ratio of dimers and, therefore, the expression of downstream signaling molecules.
Nonsense and frameshift variants have been associated with disease.
Missense variants involving the helical domains lead to a loss of heterodimer formation that alters nuclear translocation.
In-frame insertion or missense variants within the loop domain alter dimer formation, but not the nuclear location of the protein.
These data suggest that protein degradation and altered subcellular localization account for the loss of functional Twist-related protein 1 from the abnormal allele in individuals with SCS. This model also supports the finding that the coronal sutures are predominantly fused in SCS, since these sutures have a higher level of gene expression of downstream activators, as shown in Twist-null/+ mice models [el Ghouzzi et al 1997, Bourgeois et al 1998, Carver et al 2002, Connerney et al 2008, Miraoui & Marie 2010].