Rethinking the proximodistal axis of the vertebrate limb in the molecular era

Abstract

The development of the vertebrate limb has long served as a paradigm for understanding the fundamental processes by which an undifferentiated field of cells gains spatial pattern and undergoes coordinated differentiation to produce the exquisitely complex structures that characterize our functional anatomy. The limb bud emerges from the flank as a mound of seemingly homogeneous mesenchymal cells within an ectodermal jacket. Yet within days this mass of cells gives rise to skeletal elements, muscles, and connective tissues organized in the form of a mature limb. Looking at one’s own arm or leg, this organization can immediately be appreciated along three different axes: The back of the hand (dorsal) is different from the palm (ventral), the thumb (anterior) is distinct from the little finger (posterior), and the upper arm (proximal) is different from the lower arm and the hand (distal). The process by which differences along the proximodistal axis are established has been particularly controversial, with competing models proposed to explain the outcome of both classical and genetic manipulations. However, if one examines the two major models that have been previously proposed, neither of them is tenable in the context of our current knowledge of gene activity in the developing limb. Proximodistal patterning therefore needs to be placed into a new framework based directly on the molecular data. The first key insight into the process of limb patterning came almost 60 yr ago when John Saunders discovered that the apical ectodermal ridge (AER), a thickened ridge of ectodermal cells that runs along the anterior– posterior axis of the distal limb bud (equivalent to a ridge running along the distal edge of the hand from which the finger tips will eventually form) is necessary for the successful outgrowth of the limb along the proximodistal axis (Saunders 1948). To explain how different structures arise at different proximodistal levels as the limb bud grows out, a model was subsequently proposed based on progressive specification of increasingly distal cell fates in a domain ∼300 µm deep directly below the AER (Summerbell et al. 1973). According to this view, under the influence of the AER, cells are maintained in a so-called progress zone. These cells continuously acquire ever more distal positional information through the influence of an internal clock that is kept active as long as the cells receive signaling from the overlying AER. This process gives the cells a distal fate proportional to the length of time the cells remain in the progress zone. When cells move out of the range of AER signaling— i.e., out of the progress zone—the clock stops and their proximodistal fate therefore becomes fixed. This happens continuously, as all the cells in the progress zone—indeed, all of the cells of the limb bud—are dividing. As more cells are produced, only the members of the population closest to