Genome-wide mapping of expression in WT and p63 mutant mouse mandibular prominences at E10-E13. Mus musculus

Despite even large phenotypic differences among vertebrate groups, dentitions and jaws fit and function together, yet the genetic processes that orchestrate cranial and dental morphogenesis remain poorly understood. In the p63-/- mouse mutant, teeth but not jaws fail to form. This edentate mouse is a model with which to tease out genes important for odontogenesis but not jaw morphogenesis, and which may thus allow dentitions to change during development and evolution without necessarily affecting the jaw skeleton. With the working hypothesis that tooth and jaw development are autonomously controlled by discreet gene regulatory networks, we probed for genes crucial for tooth development only. Using gene expression microarray assays validated by quantitative reverse-transcription PCR, we contrasted expression in mandibular prominences at embryonic days (E) 10-13 among mice with normal lower jaw development and either normal (p63+/-, p63+/+) or arrested (p63-/-) tooth development. We predicted that expression of a suite of genes specific to odontogenesis would differ in the edentate mice. The p63-/- mice showed significantly different expression (fold change ≥1.5, ≤-1.5; p≤0.05) of several genes, some of which are already reported to help regulate odontogenesis (e.g., p63, Osr2, Cldn3/4) and/or to be targets of p63 (e.g., Jag1/2, Fgfr2), others of which have no previously reported roles in odontogenesis or the p63 pathway (e.g, Fermt1, Cbln1, Pltp, Cxcl14, Krt8, and additional keratin and claudin family members). As expected, from E10-E13 few genes known to regulate mandible morphogenesis differed in expression between mouse strains. Thus this study links for the first time several genes to odontogenesis and/or the p63 signaling network. We propose that these genes act in a novel odontogenic network that is exclusive of lower jaw morphogenesis, and posit that this network evolved in oral, not pharyngeal, teeth. Overall design: All experiments used mouse tissues from the B6.129S7-Trp63tm2Brd/J (generated by Mills et al., (1999)) and C57BL6/J strains (The Jackson Laboratory/JAX). We refer to mice with both copies of the mutant p63 allele as homozygote or p63-/-. The p63+/- samples represent B6.129S7-Trp63tm2Brd/J and p63-/- represent embryos from crosses of B6.129S7-Trp63tm2Brd/J x B6.129S7-Trp63tm2Brd/J mice. For each stage, p63-/- and p63+/- littermates were collected to minimize between-litter variation within each genotype. Also for each stage, three biological replicates (i.e., three embryos, each from a different litter) were assayed per genotype. Samples were not pooled between litters.