Genetic and pathologic aspects of retinoic acid-induced limb malformations in the mouse.

Because all-trans retinoic acid (atRA) is teratogenic in all species tested and many of the specific defects induced are common across the phylogenetic spectrum, it would be logical to predict that murine strain differences in teratology to this agent are minimal. However, for specific defects, strain susceptibilities are vastly different. Studies with atRA have shown stark differences between C57BL/6 and SWV mouse strains in susceptibility to postaxial forelimb ectrodactyly and ectopic hindlimb formation, with the C57 strain being more susceptible for both defects. Various approaches were used to determine why these strains differ in susceptibility, but the mechanisms remain unknown. Hindlimb duplications were hypothesized to be caused by the formation of ectopic posterior body axes. For forelimb ectrodactyly, a locus on chromosome 11, Rafar, has linkage to the strain difference, and mRNA localization has shown that specific genes (Fgf8, Dlx3, Bmp4, and Sp8) in the postaxial preAER (prior to formation of the apical ectodermal ridge) of the developing limb bud (the site of the defect) were downregulated hours after atRA administration more in the susceptible C57 than in the SWV strain. Because both atRA and divalent cadmium induce postaxial forelimb ectrodactyly (right-sided predominance) at a high rate in C57BL/6 and low in the SWV strain, there is debate as to whether they share a common mechanism. These teratogens cause a greater-than-additive level of forelimb ectrodactyly when coadministered at low doses, but cadmium does not induce ectopic hindlimb formation. The hypothesis is that these agents have separate molecular pathologic pathways that converge to perturb a common anatomic structure.