Autophagosome closure by the ESCRT machinery

Project Summary/Abstract The goal of this project is to address a fundamental gap in knowledge on how phagophores are closed to form double membrane autophagosomes. During macroautophagy (hereafter autophagy), crescent-shaped phagophores elongate around cytoplasmic material and seal to generate double-membrane autophagosomes that fuse with lysosomes for cargo degradation. As the phagophore rim narrows, the membranes must undergo fission to separate the inner and outer membranes in a process that bears resemblance to endosomal sorting complexes required for transport (ESCRT)-mediated membrane scission. Using our elegant HaloTag-LC3 autophagosome completion assay, we provided the first experimental evidence for the ESCRT machinery in mammalian phagophore closure and identified the ESCRT-I subunit VPS37A as critical factor for the recruitment of downstream ESCRTs to the phagophore. Notably, we found that the N-terminal putative ubiquitin E2 variant (PUEV) domain of VPS37A is uniquely required for autophagosome closure but is dispensable for other ESCRT-mediated membrane abscission processes, including endosome receptor sorting and cytokinesis. Compartment-specific targeting factors initiate the sequential recruitment of the four ESCRT complexes (ESCRT-I, -II, -III and VPS4) to the membrane scission site. While the phagophore-specific targeting factors for the VPS37A-containing ESCRT-I complex are unknown, our preliminary study has revealed that VPS37A PUEV interacts with highly curved membranes containing anionic lipids and lipid packing defects, which are all features of the phagophore rim. We hypothesize that the PUEV selectively interacts with highly curved phagophore membranes to target ESCRT-I to the phagophore. Furthermore, our preliminary work has revealed that ESCRT recruitment to phagophores requires protein ubiquitylation and the LC3/GABARAP conjugation machinery, leading us to believe that the stabilization of membrane-associated ESCRT-I requires additional interactions with phagophore-associated ubiquitylated cargo. We are in an ideal position to test our hypotheses in the following Specific Aims: (1) to determine how VPS37A targets ESCRT-I to phagophores and directs the assembly of downstream ESCRTs during autophagosome biogenesis; (2) to identify phagophore-specific targeting factors for ESCRT-I during autophagy. As autophagy is involved in numerous physiological and pathological processes, these studies will have far-reaching implications for human health and disease.