Reinventing a Common Strategy for Patterning the Eye

The vertebrate eye develops from apposed neural (op-tic vesicle) and surface (lens placode) ectoderm and its early development involves reciprocal inductive processes between these two tissues, both of which ex-Department of Biology press and require Pax6. In flies, a single ectodermal New York University epithelium forms both the retinal and lens components 1009 Main Bldg., 100 Washington Square East of the eye as well as the antenna. Pax6 expression is New York, New York 10003 activated by the embryonic patterning system and is † Skirball Institute of Biomolecular Medicine and then restricted to the eye-antennal primordium as well Department of Cell Biology as to subsets of cells in the nervous system. Although fly NYU School of Medicine and vertebrate eyes appear to develop very differently, it 540 First Avenue is possible that the greater emphasis on induction in New York, New York 10016 vertebrate eye development results from the use of transplantation methods rather than the genetic techniques used in flies. Master regulatory genes have been defined by their abil-In a recent issue of this journal, Kumar and Moses ity to trigger the complete development of an organ or (2001) reported that early determination of the Drosoph-appendage, such as an eye or a wing. One of the most ila eye may also be the result of an inductive process, striking examples is Pax6, a gene whose function is as the choice between eye vs. antennal fates is con-required to produce eyes in most species and whose trolled by antagonistic interactions between the Notch expression is sufficient to induce ectopic eye develop-(N) and epidermal growth factor receptor (EGFR) path-ment in flies and in frogs (Chow et al.,1999; Gehring ways (Figure 1). EGFR activity suppresses Pax6 expres-and Ikeo, 1999). The extreme degree of conservation of sion in the antennal disc through an as yet unknown function for this factor in species with clearly different mechanism. This restricts to the eye disc the activation types of eyes was surprising, given the widely accepted of the conserved cascade of downstream " eye determi-model that eyes have appeared independently many nation genes " (Desplan, 1997) at the second instar larval times throughout evolution (Salvini-Plawen and Mayr, stage, i.e., relatively late in development. The N pathway 1977). Now, recent reports further challenge our views antagonizes EGFR and allows eye development to pro-on how eyes have evolved by identifying common differ-ceed. This work is consistent with the …