A structural and biophysical characterisation of the ESCRT-0 complex

Abstract

The attenuation of cell surface signaling primarily occurs through the ubiquitination of activated receptors, and subsequent rapid endocytosis into cytosolic membrane bound transport carriers, including peripheral vesicles and endosomes. Here they are recognised by Endosomal Sorting Complexes Required for Transport (ESCRTs), which prime them for lysosomal degradation by sequestering them within intraluminal vesicles within endosomes. The ESCRT-0 complex, a heterodimer of Hepatocyte growth factor Receptor Substrate (HRS) and Signal Transduction Associated Molecule 2 (STAM2) is key to this process and provides the initial basis for recognition of ubiquitinated cargoes on the endosome surface. ESCRT-0 is poorly understood from a structural perspective, and the number of ubiquitin binding domains across the complex is debated. Both HRS and STAM2 are phosphorylated at a number of sites by unknown kinases downstream of epidermal growth factor receptor (EGFR) signaling, whilst HRS is also glycosylated in a mechanism proposed to regulate the interaction of the two proteins. Here, I generated a predicted structure for ESCRT-0 using AlphaFold Multimer. This model hinted at a hitherto unreported interaction between the two Vps27, HRS, STAM (VHS) domains present across HRS and STAM2. Although I was unable to detect the interaction in vitro using purified protein, I found evidence for this interaction in vivo using a split reporter assay. I showed that whilst the STAM2 VHS domain binds to monoubiquitin, the HRS VHS domain does not. I purified full length and C-terminal truncations of HRS and STAM2, before using cryoEM to investigate ESCRT-0 structure. The findings from this study suggested that ESCRT-0 is a flexible complex, with a well-defined previously identified core region. A central helix in the AlphaFold model of ESCRT-0 is not seen in the cryoEM model, perhaps due to a hinge region that provides a route to significant flexibility. Finally, I purify a kinase-active fragment of EGFR and ubiquitinate it in vitro. I investigate ESCRT-0 phosphorylation by a ubiquitinated EGFR fragment, finding that it is unlikely to phosphorylate ESCRT-0, and no priming effect occurs following addition of known ESCRT-0 kinases. This study has provided the basis for further investigation of ESCRT-0 by cryoEM, whilst also showing that EGFR does not phosphorylate ESCRT-0 in vitro. I established that the HRS VHS domain does not bind to ubiquitin and that the termini of HRS and STAM2 may combine to regulate ESCRT-0 function.

Date of Award	24 May 2024
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Philip Woodman (Main Supervisor) & Clair Baldock (Co Supervisor)