Morphogenesis and homology in arthropod limbs

American Zoologist, Jun 1999 by Williams, Terri A

The difficulty in comparing Artemia and Drosophila is a specific case of a more general problem mentioned above: our lack of understanding of how gene networks are modified during evolution. Thus, there is a tendency to use the presence of conserved genes as a short-hand for conserved morphogenetic units. For example, widespread conservation of engrailed throughout arthropods has indicated that A/P positioning is conserved. It has been suggested that branches simply arise from modulations in D/V patterning (Shubin et al., 1997). Specific hypotheses do not exist to account for how dorsoventral patterning molecules could be activated and deployed in a field of cells so as to both conform to a single model for gene regulation of limb development and account for known patterns of gene expression and morphogenesis (Williams and Nagy, 1996; Williams, 1998). A gap remains between our grasp of developmental genetics and morphological diversity.

One area of research that could fill this gap would be studies of both normal limb morphogenesis as well as experimental manipulations of morphogenesis. For example, it was originally predicted that limb development in arthropods would exhibit a fairly uniform sequence in which an initial unbranched limb bud would then sequentially give rise to other branches (Manton, 1977). However, this is demonstrably not true; the timing and expression of limb branches is quite diverse (Williams and Miller, 1996; Williams, 1998; Panganiban et al., 1995). Nonetheless, with further study, branching mechanisms could turn out to be more uniform than they appear. In vertebrate limb development, for example, it has been suggested that the process of chondrogenesis which produces highly diverse limb morphologies can be reduced to three basic patterns: de novo condensations, branching, and segmentation (Shubin and Alberch, 1986). On the other hand, morphogenetic mechanisms could be quite diverse in arthropods. It remains both unknown and unexplored whether the formation of articulated hard parts out of a continuous epidermal sheet necessarily involves mechanisms more diverse than the cartilage condensations involved in bone formation.

Studies of morphogenesis could be coupled with experimental studies designed to uncover conserved developmental mechanisms. For example, we do not yet know whether a single proximal/distal patterning axis exists in all arthropod limbs and, if so, whether some cells function as organizers for growth (like AER and Dll expressing cells). It is unknown what cellular behaviors form branches, analogous to the bifurcations that occur during cartilage condensation in vertebrates. Understanding the cellular basis of branching would provide a basis for predicting possible constraints on form analogous to the constraints on trifurcation based on the cellular dynamics of chondrogenesis (Oster et al., 1988). For example, one striking feature of development in branchiopod limbs is the large extent of ventral body wall that has been co-opted to make the limb buds. Do changes in the relative or absolute size of the limb field necessarily effect branching, as has been demonstrated experimentally in vertebrate limb (Alberch and Gale, 1985)?