A Footnote to the Evolution of Digits

Half a billion years ago, the first fourlegged land animal crawled out of the sea onto dry land. How did the limbs that creature crawled on evolve from the fins of its fishy ancestors? This question has long intrigued biologists. Fossil records suggest that tetrapod legs evolved step by step from fins, and comparative gene expression studies have provided some insights into how mutation and natural selection derived long limb bones from fin precursors. But the evolutionary path that lies between the structural elements (radials) of fish fins and the toes and fingers of tetrapod digits has remained obscured. Are tetrapod digits homologous to fish radials? Did the genetic capacity for digit differentiation exist in fish ancestors, or is it unique to tetrapods? In their recent PLOS Biology article, Denis Duboule, Joost M. Woltering, and colleagues shed new light on these questions from a comparative analysis of the regulatory mechanisms that control when and where certain members of the Hox gene family are turned on and off in zebrafish fins and mouse limbs. Two Hox gene clusters, HoxA and HoxD, are known to function in patterning the developing vertebrate limb. In land animals, but not in fish, HoxD has what is known as a ‘‘bimodal expression pattern,’’ meaning that one subset of Hoxd genes directs the development of the long bones on the proximal (body) side of the wrist or ankle, while another subset directs the development of the long bones on the distal side (i.e., the digits). This bimodal expression pattern is due to the preferential interaction of the 39 and 59 genes in these Hox clusters with flanking regions of DNA on their own side of the cluster. These flanking regions contain non-genic DNA in which are located so-called enhancers—DNA loci that control the transcription of genes. On the 39 side of the HoxA and Hox D clusters sit the proximal enhancers, which regulate the expression of proximal genes in the cluster in the proximal part of the limb, and on the 59 side are the distal enhancers, which regulate the expression of distal genes in the cluster, leading to the segregated pattern of 39 and 59 genes. Duboule and colleagues decided to find out whether the HoxA cluster shares this regulatory characteristic with HoxD, reasoning that if it does, that the bimodality arose before the Hox clusters duplicated, indicating also that the regulatory capacity to form digits was in place before land animals’ evolutionary emergence from the sea. Such a finding would lend strength to the argument for homology between digits and fin radials. By looking at the expression patterns of Hoxa genes in mouse embryos and by comparing them with the expression patterns of Hoxd genes, the researchers determined that the HoxA gene cluster does indeed exhibit biomodality, with one regulatory module directing the development of the digits and another orchestrating the development of the proximal segment of the limb. Although they noted some differences in the details of the bimodal expression of HoxA and HoxD genes, the researchers concluded that this pattern was common to both these Hox clusters and therefore predated the evolution of tetrapods from their fish ancestors. If both HoxA and HoxD clusters share the bimodality that is associated with the differential regulation of the proximal and