The TET-Sall4-BMP regulatory axis controls craniofacial cartilage development

Craniofacial microsomia (CFM) is a congenital defect that usually results from aberrant development of embryonic pharyngeal arches. However, the molecular basis of CFM pathogenesis is largely unknown. Here, we employ the zebrafish model to investigate mechanisms of CFM pathogenesis. In early embryos, tet2 and tet3 are essential for pharyngeal cartilage development. Single-cell RNA sequencing reveals that loss of Tet2/3 impairs chondrocyte differentiation due to insufficient BMP signaling. Moreover, biochemical and genetic evidence reveals that the sequence-specific 5mC/5hmC-binding protein, Sall4, binds the promoter of bmp4 to activate bmp4 expression and control pharyngeal cartilage development. Mechanistically, Sall4 directs co-phase separation of Tet2/3 with Sall4 to form condensates that mediate 5mC oxidation on the bmp4 promoter, thereby promoting bmp4 expression and enabling sufficient BMP signaling. These findings suggest the TET-BMP-Sall4 regulatory axis is critical for pharyngeal cartilage development. Collectively, our study provides insights into understanding craniofacial development and CFM pathogenesis. 10× library construction and sequencing Single-cell suspensions were processed for library construction using a Chromium Next GEM Single Cell 30 Reagent Kit v3 following the manufacturer’s instructions. Libraries were then sequenced on the Illumina Novaseq 6000 platform. Bulk RNA-seq library construction and sequencing Total RNAs were purified using TRIzol reagent (Life Technologies, 343702). Libraries were constructed using a VAHTS Universal V8 RNA-seq Library Prep Kit for MGI (Vazyme) according to the manufacturer’s instructions and then sequenced on the MGISEQ-2000 platform. 5hmC hMeDIP-seq library construction and sequencing DNA was purified from tissues according to the manufacturer’s instructions. 5hmC DNA immunoprecipitation (hMeDIP) was performed as previously described. Briefly, genomic DNA was digested with proteinase K at 55°C overnight and then with RNase A at 37°C for 1 h. Genomic DNA was purified by phenol/chloroform extraction and recovered by ethanol precipitation. Purified genomic DNA was sonicated to an average size of 200–400 bp and heat-denatured (95°C, 10 min). Then, 5 μg of fragmented genomic DNA was diluted in 500 μL DNA-IP buffer [10 mM sodium phosphate (pH 7.0), 140 mM NaCl, 0.05% Triton X-100] and pre-cleared with 50 μL protein A/G beads at 4°C for 3 h. Five percent of the pre-cleared genomic DNA was saved as an input sample. The DNA-IP sample was incubated with 5 μg 5hmC antibody (Active Motif, 39769) at 4°C overnight and then with 50 μL protein A/G beads for 4 h at 4°C. Three washes for a total of 30 min were then performed to remove non-specifically bound antibodies. The beads were then treated with 200 μL TE buffer (pH 8.0) plus proteinase K for at least 3 h at 1000 rpm, 55°C in a thermomixer. The DNA was then purified by phenol/chloroform extraction followed by ethanol precipitation. Purified DNA was dissolved in ultrapure H2O, and libraries were constructed using a VAHTS Universal DNA Library Prep Kit for MGI (Vazyme) according to the manufacturer’s instructions. Libraries were then sequenced on the MGISEQ-2000 platform.