Craniofacial malformations, often associated with syndromes, are prevalent birth defects. Emerging evidence underscores the importance of m⁶A modifications in various bioprocesses such as stem cell differentiation, tissue development, and tumorigenesis. Here, in vivo, experiments with zebrafish models revealed that mettl3-knockdown embryos at 144 h postfertilization exhibited aberrant craniofacial features, including altered mouth opening, jaw dimensions, ethmoid plate, tooth formation and hypoactive behavior. Similarly, low METTL3 expression inhibited the proliferation and migration of BMSCs, HEPM cells, and DPSCs. Loss of METTL3 led to reduced mRNA m⁶A methylation and PSEN1 expression, impacting craniofacial phenotypes. Co-injection of mettl3 or psen1 mRNA rescued the level of Sox10 fusion protein, promoted voluntary movement, and mitigated abnormal craniofacial phenotypes induced by mettl3 knockdown in zebrafish. Mechanistically, YTHDF1 enhanced the mRNA stability of m⁶A-modified PSEN1, while decreased METTL3-mediated m⁶A methylation hindered β-catenin binding to PSEN1, suppressing Wnt/β-catenin signaling. Pharmacological activation of the Wnt/β-catenin pathway partially alleviated the phenotypes of mettl3 morphant and reversed the decreases in cell proliferation and migration induced by METTL3 silencing. This study elucidates the pivotal role of METTL3 in craniofacial development via the METTL3/YTHDF1/PSEN1/β-catenin signaling axis.

Skeletal Class III malocclusion with maxillary deficiency is a severe maxillofacial disease with unclear pathogenic mechanisms. We recruited a Han Chinese family who was clinically diagnosed with skeletal Class III malocclusion and maxillary deficiency. Using whole exome sequencing, a missense variant in ADAMTS2 (NM_014244: c.3506G>T: p.G1169V) was identified and predicted as deleterious by in silico tools. We also found ADAMTS2 variants associated with deficient maxillary development in a cohort. ADAMTS2 expression in HEK293 cells showed significant decrease due to the variant, which was also consistent in dental pulp stem cells from the proband and a healthy control. In the adamts2-knockdown zebrafish model, the length and width of the ethmoid plate, as well as the length of the palatoquadrate became significantly shorter than the control group (p < 0.001), while there was no significant difference in the length and width of the mandible. The expression of Sox3, which was required in early embryonic craniofacial development, was significantly downregulated in the adamts2-knockdown zebrafish embryos. Bioinformatic and cellular studies showed that the decreased expression of ADAMTS2 may inhibit downstream ErbB signaling pathway transduction and restrain subsequent osteogenesis in human adult mesenchymal stromal cells. Collectively, these data showed that ADAMTS2 (c.3506G>T: p.G1169V) may confer susceptibility to risk of skeletal Class III malocclusion with maxillary deficiency.