Search

The Drosophila compound eye is formed by selective recruitment of undifferentiated cells into clusters called ommatidia during late larval and early pupal development. Ommatidia at the edge of the eye, which often lack the full complement of photoreceptors and support cells, undergo apoptosis during mid-pupation. We have found that this cell death is triggered by the secreted glycoprotein Wingless, which activates its own expression in peripheral ommatidia via a positive feedback loop. Wingless signaling elevates the expression of the pro-apoptotic factors head involution defective, grim and reaper, which are required for ommatidial elimination. We estimate that approximately 6-8% of the total photoreceptor pool in each eye is removed by this mechanism. In addition, we show that the retinal apoptosis previously reported in apc1 mutants occurs at the same time as the peripheral ommatidial cell death and also depends on head involution defective, grim and reaper. We consider the implications of these findings for eye development and function in Drosophila and other organisms.

Wingless directs many developmental processes in Drosophila by regulating expression of specific target genes through a conserved signaling pathway. Although many nuclear factors have been implicated in mediating Wingless-induced transcription, the mechanism of how Wingless regulates different targets in different tissues remains poorly understood. We report here that the split ends gene is required for Wingless signaling in the eye, wing and leg imaginal discs. Expression of a dominant-negative version of split ends resulted in more dramatic reductions in Wingless signaling than split ends-null alleles, suggesting that it may have a redundant partner. However, removal of split ends or expression of the dominant-negative had no effect on several Wingless signaling readouts in the embryo. The expression pattern of Split ends cannot explain this tissue-specific requirement, as the protein is predominantly nuclear and present throughout embryogenesis and larval tissues. Consistent with its nuclear location, the split ends dominant-negative acts downstream of Armadillo stabilization. Our data indicate that Split ends is an important positive regulator of Wingless signaling in larval tissues. However, it has no detectable role in the embryonic Wingless pathway, suggesting that it is a tissue or promoter-specific factor.