ABSTRACT
Robinow syndrome (RS) is a rare disease caused by mutations in seven WNT pathway genes. Features include craniofacial widening and jaw hypoplasia. We used the chicken embryo to test two autosomal dominant RS (ADRS) missense FZD2 variants on the frontonasal mass, the affected region in RS. The wild-type (wt) and variant hFZD2 inhibited beak ossification. The bone hypoplasia was possibly mediated by decreased levels of WNT and BMP pathway genes. In primary cultures, hFZD2 variants inhibited chondrogenesis, increased nuclear shuttling of β-catenin and increased expression of TWIST1, both known to suppress chondrogenesis. In luciferase reporter assays, proteins coding for 1301G>T and 425C>T FZD2 variants weakly activated canonical WNT reporter and dominantly interfered with wtFZD2. In the JNK-PCP WNT pathway luciferase assay, only the 425C>T showed a loss-of-function. The 1301G>T variant presumably acts through a JNK-independent pathway. This is the first study to demonstrate that the ADRS-FZD2 missense variants cause craniofacial and WNT signaling defects. Frontonasal mass width is increased by both hFZD2 variants which sheds light on the ontogeny of the broad facial features seen in individuals with RS.
Summary Statement Gain-of-function studies on FZD2 missense variants associated with Robinow syndrome led to increased facial width, altered Wnt signaling and inhibition of beak skeletogenesis in chicken embryos.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Author email addresses: Shruti.tophkhane{at}ubc.ca Kfu{at}dentistry.ubc.ca everheye{at}sfu.ca Richman{at}dentistry.ubc.ca
1.Author added: Esther M. Verheyen 2.New title: Craniofacial studies in chicken embryos confirm the pathogenicity of Frizzled2 variants associated with Robinow syndrome 3.Other edits 4.FZD2 variants are referred to as nucleotide change 425C>T, 1301G>T instead of amino acid change P142L, G434V, respectively. This applies to the MS, figures, and tables 5.Added stage 28 frontonasal mass width and TUNEL quantification to Figure 2, updated summary figure for signaling clarity, included skeletal analysis contingency graphs in supplementary images, and specified error bars as standard deviation in figure legends. 6.Sample size Methods updated to clarify sample sizes for all experiments and pooled biological replicates for qRT PCR. Edits detail unknown initial RCAS virus titer and its replication competence, leading to variable viral spread. Analysis was always limited to injected side. 7.Further emphasized significance of the 425C>T variant losing interactions with the WNT5A ligand and ROR2 coreceptor. 8.Removed in vivo nuclear β catenin data from Figure 2 and STF assay from figure 5 due to potential confusion. 9.Clarified similarities and differences between wild type and mutant FZD2 Experiments revealed non selective effects of mutant FZD2 indicating a general gain of function, shown in ossification in vivo, qRT PCR, TUNEL, and BrdU assays. Conversely, selective effects of mutant FZD2 were evident in luciferase assays (STF, ATF, SOX9 activity) and micromass cultures (thickness, cartilage amount, TWIST1, CTNNB1 expression). Evidence for variant function changes was strongest in studies using primary facial mesenchyme cells, particularly through luciferase assays and micromass cultures. Epithelium removal isolated variant effects on mesenchyme, not observed in vivo. 10.Clarified that variants inhibit frontonasal mass narrowing via distinct mechanisms: 425C>T variant is inactive in JNK PCP pathway, while 1301G>T variant inhibits RhoA ROCK in the second PCP pathway. Additionally, this study is the first to associate the BMP pathway with ADRS phenotypes.