Out on a limb

If babies could talk, they would no doubt testify that hands and feet are fascinating things. If they were in a particularly philosophical mood, they might — like many developmental biologists — wonder why there are five fingers on each hand, and how the parts of the limb are so neatly patterned and articulated. Limb development is a key research topic, which encompasses issues of cell differentiation, tissue patterning and signalling.

We owe much of our current understanding of limb morphogenesis to John Saunders and his colleagues, whose pioneering experiments in chick embryos led to the discovery of key signalling centres in the limb bud. In 1948, Saunders had shown that removal of a thickened layer of ectoderm that rims the distal tip of the limb bud — dubbed the apical ectodermal ridge (AER) — causes truncation of the limb. In a 1957 paper, Saunders, along with John Cairns and Mary Gasseling, grafted prospective thigh mesoderm from the chick leg bud to the apex of the wing bud, and they found that this mesoderm generated foot structures if it was placed in contact with the AER. However, if wing mesoderm was allowed to ingress between the grafted tissue and the AER, the grafted mesoderm retained its thigh identity.

Experiments such as these led to the suggestion that the AER is not only required for limb outgrowth, but it also provides signals that allow specific structures to form at different proximo–distal levels of the limb axis. Gail Martin and colleagues subsequently attributed the signalling activity of the AER to molecules of the fibroblast growth factor (FGF) family. It has become accepted that signals from the AER maintain a population of undifferentiated cells that give rise to successive skeletal elements of the proximo–distal limb pattern. However, recent findings from the laboratory of Clifford Tabin indicate that this pattern is not established progressively during limb-bud development, but instead results from the expansion of cell populations that are specified in the young limb bud.

Saunders was also interested in how the antero–posterior pattern of digits is generated in the limb. In 1968, Saunders and Gasseling identified a region at the posterior margin of the wing bud — later termed the zone of polarizing activity (ZPA) — which, when transplanted to an anterior position in the wing-bud margin, caused a mirror-image duplication of digits. In 1975, Cheryl Tickle, Dennis Summerbell and Lewis Wolpert proposed that the ZPA releases a diffusible morphogen, establishing a gradient such that the posterior-most digit arises closest to the source of the morphogen, and the more anterior digits emerge at sites with progressively lower morphogen concentrations. As retinoic acid (RA) can mimic the polarizing effect of the ZPA, it was initially thought that this might be the morphogen. However, Tabin's group showed that the ZPA signal is Sonic hedgehog, and that RA primarily induces the formation of an ectopic ZPA.

The concept of inductive interaction in morphogenesis that was formulated by Saunders et al. has proved to be remarkably robust, and it represents a fundamental model to elucidate the importance of signalling activity and tissue interactions for the development of complex structures.