The actin cytoskeleton has crucial roles in cell and tissue morphogenesis, in which it orders cellular space and transduces forces. New work in the October issue of Nature Cell Biology by Baum and Perrimon identifies the genes that control the polarized distribution of actin filaments in the Drosophila follicular epithelium.

Actin filaments are involved in processes as wide-ranging and diverse as cytokinesis, polarized intracellular transport, adhesion and migration. Therefore, the action of F-actin must be regulated in both a temporal and spatial manner to ensure 'the cap always fits' — in other words, to tailor the level and localization of actin to the required biological function. Many studies have identified the proteins that are needed for actin polymerization and depolymerization; for example, Enable (Ena) and VASP catalyse actin formation, whereas adenylyl-cyclase-associated proteins (CAP proteins) limit polymerization. However, little is known about the function of actin-regulatory proteins within an intact organism.

Baum and Perrimon used both genetics and cell biology in Drosophila melanogaster to study the spatial control of actin within the follicular epithelium. The follicle cells surround the germ-line cyst and, during oogenesis, undergo at least two significant shape changes. So, they are an ideal system in which to study the regulation of actin dynamics. By generating clones of cells that lack Ena, CAP (a Drosophila CAP protein homologue) or Abelson (Abl, a tyrosine kinase, which acts as a CAP-binding protein), Baum and Perrimon were able to study the role of each of these genes in modulating the spatial organization of the actin cytoskeleton within these cells. Clones of follicle cells which lack CAP maintain their epithelial polarity, but have increased levels of actin and defects in actin organization. (In the figure, CAP clones within the follicle cell epithelium, which are marked by the absence of green fluorescent protein (green), have increased levels of actin (pink) and Ena (blue) at the apical membrane.) Similarly, Abl clones also have subtle defects in actin organization. Ena clones, however, lose actin filaments.

Given that mammalian Abl binds CAP and antagonizes the function of Ena, Baum and Perrimon went on to determine whether CAP/Abl/Ena act together to control the level and localization of F-actin. Epistatic experiments allowed them to place Ena genetically downstream of CAP in regulating actin organization, as loss of CAP leads to a change in the localization of both Ena and Abl.

Therefore, it seems that the action of actin-regulatory proteins within an intact organism is very similar to the system previously observed in mammalian cells. CAP, in collaboration with Abl, inhibits actin polymerization, whereas Ena acts in the opposite way. This set of genes acts together to ensure the correct levels of actin are made and localized to their proper locations, and so they ensure the cap always fits.