The ULK3 Kinase Regulates the ESCRT pathway in the Abscission Checkpoint

The endosomal sorting complex required for transport (ESCRT) pathway mediates important membrane fission events, including the physical separation of daughter cells during the abscission step of cytokinesis. Late acting proteins in this pathway, called ESCRT‐III subunits, are recruited to the intercellular bridge that connects two daughter cells where they appear to polymerize into spiraling membrane‐bound filaments that draw membranes together and promote fission. Abscission is highly regulated; if a dividing cell cuts its membrane prematurely (e.g., in the presence of unsegregated DNA) or fails to divide entirely, the daughters may become aneuploid. Progression through abscission is regulated by the abscission checkpoint; a genome protection mechanism that functions through the Aurora B kinase and the ESCRT‐III subunit CHMP4C to delay abscission until upstream mitotic events have been resolved. Checkpoint inputs include lagging chromosomes within the intercellular bridge, aberrant nuclear pore assembly, and mechanical tension at the intercellular bridge. Our goal is to understand how ESCRT‐mediated abscission is coordinated with these other steps in the mitotic program. We recently identified a second kinase, ULK3, that functions downstream of Aurora B in the abscission checkpoint. ULK3 contains tandem C‐terminal M icrotubule I nteracting and T rafficking (MIT) domains, which serve as docking sites for ESCRT‐III proteins. Notably, ULK3 binds and phosphorylates a subset of ESCRT‐III proteins, including CHMP4C and IST1. We have determined the crystal structure of the second ULK3 MIT domain in complex with its IST1 binding site, and characterized the unusually tight binding interaction. We further find that ULK3 phosphorylation provides a reversible inhibitory signal that stalls abscission until mitotic errors can be resolved; phospho‐mimetic IST1 fails to support cytokinesis, and non‐phosphorylateable IST1 cannot support a functional abscission checkpoint. We are now mapping ULK3 phosphorylation sites on other ESCRT‐III subunits, with the goal of identifying a consensus site for ULK3 substrates. We are also testing the mechanistic consequences of ULK3 phosphorylation on ESCRT‐III function using in vitro ESCRT‐III assembly assays and cytokinesis assays in cells. Preliminary studies suggest that ULK3 phosphorylation inhibits the ability of ESCRT‐III proteins to form the filaments that constrict the midbody. Thus, we are learning how ULK3 phosphorylation of ESCRT‐III proteins inhibits the abscission checkpoint and laying the groundwork for identifying additional ULK3 substrates. Support or Funding Information This work was supported in part by an American Cancer Society Postdoctoral Fellowship to DMW (“ULK3 regulation of the ESCRT pathway in abscission”, PF‐14‐102‐01‐CSM) and NIH GM112080 to WIS