A phenotypic and genetic characterization of the cell adhesion molecules echinoid and friend-of-echinoid in the directed cell movements of ommatidial rotation during Drosophila eye development.

摘要

Correct development of multicellular organisms relies on the precise patterning of cells, which must respond to and interpret specific cues that instruct the cells to differentiate and often undergo directed cell movements and rearrangements to give rise to functional tissues and organs. Differential adhesion between the stationary and mobile cells permits and promotes these cellular movements, effecting patterning of cells and tissues. During Drosophila eye development, groups of cells, the ommatidial precursors, undergo a 90° rotational movement within a matrix of stationary cells, providing the cell motility readout of tissue polarity. The mechanisms that regulate ommatidial rotation are not well understood.;In order to better understand how ommatidia coordinate cell signaling and cell adhesion to regulate the directed cell movement of ommatidial rotation, I investigated the roles of two cell adhesion molecules, Echinoid (Ed) and Friend-of-Echinoid (Fred), in this process. Initially, I characterized the misrotation phenotypes resulting from loss-of-function mutations in these two genes, and used a genetic approach to ascertain that they function during larval development and cooperate to regulate rotation.;To understand the underlying mechanism by which ed and fred regulate rotation, I performed a row-by-row analysis of Ed and Fred protein localization during ommatidial rotation, and found that these proteins localize in patterns that are consistent with an affect on cell-cell adhesion. This observation led to the hypothesis that different levels of Ed or Fred in rotating vs. nonrotating cells provide a permissive environment for cell movement at the beginning of ommatidial rotation. Beginning midway through ommatidial rotation, equalizing levels of these proteins in the ommatidial cells and the interommatidial cells leads to a restrictive environment, thus slowing ommatidial rotation. In support of this hypothesis, I demonstrate that manipulating levels of these proteins and interfering with the establishment of the early permissive environment slows ommatidial rotation.;My work also provides evidence that Ed and Fred may regulate signaling in the slow phase of ommatidial rotation. Mosaic analysis identified a requirement for ed and fred in photoreceptors R1, R6, R7 and the cone cells for proper ommatidial rotation. In addition, I used a genetic approach to identify potential interactors of ed and fred in rotation, and found that both genes interact with two downstream effectors of Egf signaling: the Mapk/Pnt transcriptional output and the Cno cytoskeletal/junctional output. Furthermore, my analysis of the cno loss-of-function phenotype provides the first indication that Cno inhibits ommatidial rotation.;Egf signaling promotes ommatidial rotation, although the underlying mechanism is unclear. I hypothesize that Egfr signaling promotes ommatidial rotation by inhibiting Cno activity in the ommatidial cells. As ommatidial rotation slows, Ed and Fred cooperate to regulate the Egf receptor in R1, R6, R7 and the cone cells, and increased inhibition of the Egf receptor as Ed levels rise leads to an increase in Cno activity and the cessation of ommatidial rotation.;Using a genetic approach, I also identified the tissue polarity genes as interactors of ed and fred in rotation. Intriguingly, ed and fred specifically modify different subsets of the TP genes. Mosaic analysis of the tissue polarity gene strabismus (stbm) identified a requirement for stbm in photoreceptor R7, thus providing the first indication of a role for a tissue polarity gene outside of photoreceptors R3 and R4 to regulate some aspect of tissue polarity.