Furrows (green) are not fully formed if recycling endosomes are disturbed (right).

Cytokinesis requires dramatic actin remodeling to produce the wall of actin and myosin that pinches apart the two daughters. Loads of membrane also have to be added to accommodate the increased surface area at the point of separation. On page 143, Riggs et al. show that these two processes may be coordinately accomplished by recycling endosomes (REs)—vesicles that take in and then return plasma membrane components via a centrosome-targeted pathway.

The group had previously found that a centrosome-associated protein called Nuf is necessary for actin remodeling during the cytokinetic-like furrow formation of multinucleated fly embryos—a process that prevents individual dividing nuclei from bumping into each other. Centrosomes organize microtubules but are not known to coordinate actin polymerization. So when Nuf turned out to be a homologue of Arfophilin, a mammalian GTPase effector found on REs, the authors guessed that this vesicle association, rather than centrosome association, was relevant to organizing furrows. Riggs et al. now show that indeed Nuf is part of the RE pathway.

Nuf influences REs by binding to the small GTPase Rab11 and localizing it to REs. Mutation of either Rab11 or Nuf inhibits both membrane recruitment and actin remodeling at early stages of furrow formation. The group suggests two models to explain RE involvement in furrow formation and, by extension, cytokinesis. Vesicles budding from the plasma membrane might grab hold of pieces of actin at the cortex, thus bringing both membrane and actin to the furrow. Alternatively, REs might harbor actin-organizing or -polymerizing activities such as Rac1. If so, REs could be involved in actin remodeling in other processes, including phagocytosis. The results also suggest that other unexplained functions attributed to centrosomes may instead be related to REs. ▪