The human GID complex engages two independent modules for substrate recruitment

Abstract

The human GID (hGID) complex is an evolutionary conserved E3 ubiquitin ligase regulating diverse biological processes including glucose metabolism and cell cycle progression. However, the biochemical function and substrate recognition of the multi-subunit complex remains poorly understood. While the yeast GID complex recognizes Pro/N-end rule substrates via yeast Gid4, the human GID complex requires a WDR26/Gid7-dependent module to trigger proteasomal degradation of mammalian HBP1. Here, using biochemical assays, crosslinking-mass spectrometry and cryo-electron microscopy, we show that hGID unexpectedly engages two distinct modules for substrate recruitment, dependent on either WDR26 or GID4. WDR26 together with RanBP9 cooperate to ubiquitinate HBP1 in vitro, while GID4 is dispensable for this reaction. In contrast, GID4 functions as an adaptor for the substrate ZMYND19, which surprisingly lacks a Pro/N-end rule degron. GID4 substrate binding and ligase activity is regulated by ARMC8 alpha, while the shorter ARMC8 beta; isoform assembles into a stable hGID complex that is unable to recruit GID4. Cryo-EM reconstructions of these hGID complexes reveal the localization of WDR26 within a ring-like, tetrameric architecture and suggest that GID4 and WDR26/Gid7 utilize different, non-overlapping binding sites. Together, these data advance our mechanistic understanding of how the hGID complex recruits cognate substrates and provide insights into the regulation of its ligase activity.

A chemical probe to modulate human GID4 Pro/N-degron interactions

“…PFI-7, however, facilitated the identification of proteins, or their interactors, that bind directly via the GID4 substrate binding pocket. Interactors that were lost upon PFI-7 treatment included RNA helicases DDX17, DDX21, and DDX50 that have previously been reported to associate with the CTLH complex 16,18,34,36 , and that we now show are GID4-dependent interactors. Indeed, many of the direct GID4 interactors are nucleolar proteins including DDX21 39 , DDX50 40 , and KIN 41 , among others.…”

“…The first human CTLH substrate to be identified was HMG box protein 1 (HBP1) 18 , a transcription factor responsible for inhibiting expression of pro-proliferative cell cycle regulators. HBP1 is ubiquitylated and degraded independently of GID4 16,18 , suggesting degradation is mediated via an alternative recognition subunit in CTLH other than GID4. Since then, a handful more degradative substrates of the complex have been suggested including BICC1 19 , MKLN1 20 , LMNB2 21 , RAF1 22 , and ZMYND19 16 .…”

Towards a structurally resolved human protein interaction network

“…A complete list of all datasets is available from the supplementary data. A subset of Cross-link data was collected from [34][35][36][37][38][39][40][41][42][43][44] , filtered for peptides unique to only one protein sequence. A cross link was considered validated by the structure if the distance between the epsilon amino groups on the side-chains of the relevant pair of lysine residues were within 32Å.…”

Towards a structurally resolved human protein interaction network

“…A subset of crosslink data was collected from refs. [33][34][35][36][37][38][39][40][41][42][43] , and filtered for peptides unique to only one protein sequence. A crosslink was considered validated by the structure if the distance between the epsilon amino groups on the side-chains of the relevant pair of lysine residues was within 32 Å.…”