Retinoic acid-induced asymmetric craniofacial growth and cleft palate in the TO mouse fetus.

The etiology and pathogenetic mechanisms of cleft palate (CP) are rather uncertain. Both genetic and environmental factors are known to cause failure of horizontalization and/or failure of fusion of the palatal shelves resulting in CP. Retinoic acid (RA)-induced CP in the mouse is reported to exhibit two peaks of incidence separated by a less sensitive window. The morphologic bases of the differential sensitivity are not known. The objectives of this study were to determine whether the TO mouse had similar peaks of sensitivity to RA-induced CP, and if it did, to evaluate the morphologic and histologic bases of CP induced at an early [Gestation Day (GD) 8] and at a late (GD 12) stage of embryonic development. Single doses of all-trans-RA were administered to groups of mice on one of GD 8 to 15. On GD 18, fetuses were evaluated for the presence of CP, and the developmental stage of the palatal shelves was determined. All doses of RA were found to induce a high incidence of CP in the GD 8 to 13 treatment groups. GD 14 and 15 were not susceptible. There were no stage-dependent peaks or less sensitive windows, indicating that RA-induced CP in this strain is a continuum from GD 8 through 13. Morphologically clefting in the GD 8-RA treatment group was characterized by extreme hypoplasia (65% to 100%, depending on the dose) or agenesis (35% in the 200 mg/kg group) of the palatal shelves and associated with astomia, microstomia, aglossia, microglossia, and micrognathia with fusion of mandible, maxilla, and zygoma. Treatment on subsequent days of gestation resulted in CP with the shelves reaching progressively higher levels of maturity in terms of developmental staging. There was no case of CP with horizontalized shelves apposing but failing to fuse with each other. The facial skeleton of GD 12-RA group was hypoplastic but not malformed. Reduction in all dimensions of the cranium and mandible was highly significant (P < 0.001) in the GD 8-RA group, whereas there was a clear imbalance between the vertical growth and that in other directions in the GD 12-RA group. The CR length, head and body weights, and the protein content of heads of GD 8-RA-treated embryos were significantly reduced. Histologic studies showed that both the intrinsic and extrinsic muscles of the tongue and face, growth of the Meckel's cartilage, and ossification of the mandible were severely affected in the GD 8 treatment group, whereas these tissues were only moderately affected in the embryos of the GD 12-RA group. However, the quality of cytodifferentiation of the muscles was not affected in either group. These data provide evidence for the susceptibility continuum of CP in this strain. They also indicate that agenesis and hypoplasia of the palatal shelves and primordia of craniofacial skeleton and musculature contribute to CP, the relative involvement of the components depending on the stage of drug administration. In the absence of pronounced cell death, it appears that RA possibly produces its deleterious effects on the precursors of craniofacial primordia, such as the neural crest, by misexpression of developmentally important genes.