Patterning of the rib cage is complex, not least because it derives from two different tissues - the somitic mesoderm (which forms the vertebrae and ribs) and the lateral plate mesoderm (which forms the sternum). On , Deneen Wellik and colleagues shed light on the `Hox code' that orchestrates development of the thoracic skeleton. They made mutant mice lacking whole groups of paralogous Hox genes - Hox5, Hox6 and Hox9 - which had dramatic effects on rib cage morphology. Surprisingly, the phenotypes do not fit with the simplest version of the model of `posterior prevalence', in which posterior groups of Hox genes are functionally dominant over the next-most anterior group. Consistent with Hox genes in other tissues, the phenotypes in the somite-derived skeleton do have consistent colinearity - the genomic distribution of the genes is reflected in their spatial and temporal expression and function during development. However, no such colinearity is seen in the sternum. The authors conclude that Hox patterning of the thoracic skeleton occurs through different mechanisms depending on the mesodermal tissue from which it derives.
