A de novo MSX1 frameshift, c.299delC, in a 6-year-old Japanese boy with nonsyndromic tooth agenesis; the C-terminal loss-of-function region and an uncommon canine involvement

What was tested

Sano and colleagues at Aichi Gakuin University, Tsurumi University and the Institute for Developmental Research, Aichi Developmental Disability Center, ran a genetic diagnosis on a single patient: a 6-year-old Japanese boy with congenital tooth agenesis, born to nonconsanguineous parents, with no family history of cleft lip or palate or ectodermal disorders (DOI 10.1038/s41439-026-00343-5, Human Genome Variation, volume 13, article 10, published online 2 April 2026). Saliva DNA from the boy and both parents was analyzed by trio-based whole-exome sequencing (Agilent SureSelect Human All Exon, Illumina NovaSeq 6000, 150-base-pair paired-end reads), with Sanger sequencing confirmation in all three. The study was approved by the IRBs of the three institutions, funded by the Japan Agency for Medical Research and Development (JP17nk0101334 and JP20ek0109397).

What they found

The trio screen returned one predicted loss-of-function variant in a known tooth-agenesis gene: a heterozygous de novo frameshift in MSX1, NM_002448.3:c.299delC in exon 1. The deletion shifts the reading frame and produces a truncated protein, p.(Pro100Argfs*60), with an aberrant 60-amino-acid C-terminal tail that is expected to lose function. The variant is absent from gnomAD, ExAC, dbSNP, the 1000 Genomes Project and the NHLBI Exome Variant Server, and the authors classify it as pathogenic under ACMG, AMP and CAP guidelines on criteria PVS1 (null variant) and PS2 (confirmed de novo). The truncation is predicted to disrupt both the homeodomain and the C-terminal MH6 (PIAS) domain, regions tied to DNA binding, nuclear localization and transcriptional regulation; the paper notes prior evidence that MSX1 haploinsufficiency decreases BMP4 and LEF1 signaling in the interdental lobe during tooth development, so the variant is read as failing to transmit sufficient signaling for normal tooth germ development. Phenotypically, the boy’s agenesis includes a missing canine, which the authors flag as uncommon: in WNT10A and LRP6 carriers premolars and molars are affected more often, and canine agenesis in MSX1 disease had been documented only sporadically, for example in one individual in a 2024 Chinese cohort (Ding et al.).

What it does not show

This is a one-patient case report with a candidate-gene design, not a cohort study: a single trio with one pathogenic-classified variant cannot establish how often c.299delC or similar C-terminal MSX1 variants occur, or their penetrance. The functional effect of p.(Pro100Argfs*60) is predicted, not tested here: no cellular assay, no edited model, no rescue experiment appears in this paper; the loss-of-function and BMP4/LEF1 mechanism claims rest on prior literature the authors cite. The retrieved text does not state how many teeth the boy is missing or their full distribution, so we report none, and the hypodontia-versus-oligodontia classification of this case is left open on the record we read. The classification as nonsyndromic reflects the absence of clefting or ectodermal findings in this one child; MSX1 variants are also associated with syndromic presentations, so the phenotypic boundary is a statement about this case, not the gene. And a de novo dominant frameshift in one child says nothing about population risk or about treatment.

Where it sits in the field

For the bioengineered tooth germ program, this is a human data point on the control layer the program eventually has to command: which genes decide that a tooth germ initiates at all. The paper itself counts more than 350 genes involved in tooth development, and MSX1 sits in the epithelial-mesenchymal signaling network that the germ-engineering route depends on for initiating and patterning a germ. It also sharpens a contrast the site tracks across programs: loss of Sostdc1, which encodes the USAG-1 protein, produces supernumerary teeth in mice, while loss of function of MSX1 in this boy means teeth that never form. Tooth number in mammals is set by the balance of that network, and the same biology that removes a tooth when one dosage-sensitive gene breaks is the biology that induced-extra-tooth programs, anti-USAG-1 included, are trying to steer rather than repair. The new variant adds to a short list of reported MSX1 tooth-agenesis alleles and extends the phenotype to canine involvement; it does not change any program’s readiness tier. The current field assessment stands at /field/.

Where we differ from the coverage

We found no press or popular coverage of this paper to differ from. Against the paper’s own framing, we hold back one inference: the abstract’s conclusion that the case “supports the utility of genomic testing for molecular diagnosis” is fair as far as it goes, but from a single patient the only diagnosis utility demonstrated is for this child and his recurrence-risk counseling; broader utility claims need the cohort data this design does not supply. We also note for the record that the pathogenic classification is guideline-based and prediction-supported, not functionally demonstrated, which matters when the variant is held up as evidence for what the MSX1 C terminus does.

Provenance: grounded in the article’s full main text as published by the journal and in the Crossref record for DOI 10.1038/s41439-026-00343-5, both retrieved 20 October 2026; every number and classification above is taken from those sources. Method and sourcing standard at /method/.