The team set out to investigate the migration function of these enzymes further by making knockouts. “That was about a year and a half ago,” says Devreotes. “We still haven't got around to studying migration.”
Instead they immediately came across a striking cytokinesis defect in the mutants, which couldn't regulate PI(3,4,5)P3 levels because of a lack of both PTEN and two PI3 kinases. Cells continued to grow but cytokinesis often failed, especially for cells growing in suspension. In wild-type cells undergoing cytokinesis, by contrast, the kinases and PI(3,4,5)P3 were concentrated at the poles of cells and PTEN was at the furrow. This localization was blocked if spindle formation was prevented.
Before cytokinesis, cells round up. This coincided in wild-type cells with the arrival of PTEN at the plasma membrane all around the cell. “That would tend to erase the polarity from migration,” says Devreotes. The spindle then may reestablish two opposing polarities, setting the two ends of the cell off in opposite directions, and simultaneously instructing the middle of the cell to contract. Dictyostelium already has a reputation for cytokinesis that relies on daughter cells pulling themselves away from each other, so it will be important to establish whether other organisms use similar mechanisms.