Limb anomalies from evolutionary, developmental, and genetic perspectives.

Coming-on-land by vertebrates during the Devonian was preceded by a 100 million year history of evolution of fins from an early agnathan to a sarcopterygian state with proximal single stylopod bone and probable paired zeugopod bones. There is little disagreement about the homology [Owen, 1837: See Desmond, 1982; Owen, 1849 for a general discussion see Roth, 1988] of these three bones to the corresponding ones of present land vertebrates including those of birds and mammals; or, that the concept of homology in this context may safely be interpreted as meaning structural "identity" by virtue of descent from a common ancestor with a prototypic developmental plan irregardless of the corresponding innervating vertebral segments [qv Roth, 1988]. This extraordinarily conserved body plan in all four classes of tetrapods, including some 4500 living species of mammals, suggests early successful selection, adaptation, and emergence of developmental constraints assuring "proper" succession of proximo-distal epimorphic events and the structural and functional integrity of the autopod. The autopod is the most variable part of the tetrapod limb with humans, in contract to most other primates, retaining its most general form with little modification except for use of the thumb [Ankel-Simons, 1983]. There is also no question about the fact that during the later stages of blastogenesis the limb arises as a prepatterned single morphogenetic field from lateral plate (and somite) mesoderm and overlying ectoderm organizing in concert a single, orthotopic developmentally reactive system of ectoderm-covered mesodermal core with distal apical ectodermal ridge and posterior zone of polarizing activity. This assertion is based on two lines of evidence. First, experimental results [beginning with Harrison and Detweiler in 1918] recognized almost immediately as demonstrating not symmetrical, but "equipotential" fields with identical morphogenetic reaction potential in all vertebrates studied so far. One is tempted to say that these morphological results and interpretations have been, "triumphantly" confirmed by recent molecular work. Second, clinical insights beginning with thalidomide, and then drawing on the acrofacial dysostoses, the associations (VATER), and the discovery of the acrorenal polytopic field defect in humans, which found its explanation in the work of Lash and of Geduspan and Solursh (possibly involving a single molecule, namely, the insulin-like growth factor-I). It is evident that the gross morphological pattern set up in subsequent normal limb development is proximo-distally hierarchical (or at least sequential), and that the complex group of secondary (epimorphic) fields (perhaps as many as 33 as identified by analysis of mendelian mutations) is determined before cellular differentiation of the individual tissue components of the limb. The Anikin [1929] patterns of precartilage condensations, segmentations, and branchings in limb rudiments, while involving a specific type of cell (precartilage mesenchyme) in complex interaction with the extracellular matrix, must be looked at primarily as gross morphogenetic field events rather than as "fine" tissue differentiation sensu stricto. In view of the clinical evidence, the Shubin-Alberch-Oster model of (pre) cartilage events (condensation, segmentation, and branching), while universally valid as such, had best be regarded as events with morphogenetic potential rather than as invariable predictors of final structure.