Structural Insights into WD-Repeat 48 Activation of Ubiquitin-Specific Protease 46

Abstract

Protein ubiquitination patterns are an important component of cellular signaling. The WD-repeat protein WDR48 (USP1-associated factor UAF-1) stimulates activity of ubiquitin-specific proteases USP1, USP12, and USP46. To understand how WDR48 exerts its effect on the USP scaffold, we determined structures of the ternary WDR48:USP46:ubiquitin complex. WDR48 interacts with the USP46 fingers subdomain via a relatively small, highly polar surface on the top center of the WDR48 β propeller. In addition, WDR48 has a novel ancillary domain and a C-terminal SUMO-like domain encircling the USP46-bound ubiquitin. Mutation of residues involved in the WDR48:USP46 interaction abrogated both binding and deubiquitinase activity of the complex. An analogous mutation in USP1 similarly blocked WDR48-dependent activation. Our data suggest a possible mechanism of deubiquitinase stimulation via stabilization and prolonged residence time of substrate. The unprecedented mode of interaction between the USP fingers domain and the WD-repeat β propeller serves as a prototypical example for this family of deubiquitinases.

Allosteric Activation of Ubiquitin-Specific Proteases by β-Propeller Proteins UAF1 and WDR20

“…Similar to most USPs of known structures, the active site of USP12 is in a catalytically competent conformation (Figure S1B), suggesting that its enzymatic activation does not involve major reorganization of the catalytic triad. Superposition analysis of USP12 with several Ub-bound USP structures, including Ub~USP46 and Ub~USP14, reveals a partially closed catalytic cleft of the free enzyme, which is constricted by two previously highlighted USP Palm domain surface loops, Blocking Loop (BL) 1 and 2 (Figure 1B, S1C, and S1D) (Hu et al, 2005; Yin et al, 2015). To accommodate the Ub tail, these two loops would have to undergo conformational changes to make room for the substrate.…”

“…Few structural differences were previously detected in the post-reaction form of USP46 upon UAF1 binding (Yin et al, 2015). By contrast, a comparison between the free and UAF1-bound USP12 structures without a suicidal Ub unveils a concatenation of conformational differences spreading across the protein.…”

“…The enzyme adopts a canonical right hand-like USP fold, which consists of three subdomains, Fingers, Thumb, and Palm (Figure 1A and S1A). Previous studies have shown that the Fingers domain of USPs is responsible for cradling the globular domain of ubiquitin (Ub), whereas the Thumb and Palm domains flank a long and narrow catalytic cleft, recognizing the extended Ub tail and presenting its C-terminus to the active site cysteine (Hu et al, 2002; Yin et al, 2015). Similar to most USPs of known structures, the active site of USP12 is in a catalytically competent conformation (Figure S1B), suggesting that its enzymatic activation does not involve major reorganization of the catalytic triad.…”

“…In other known USP structures, the extended Fingers domain is defined by a four-stranded antiparallel β-sheet, which is frequently stabilized by a zinc ion coordinated at the fingertips region (Figure 1C) (Avvakumov et al, 2006; Ernst et al, 2013; Hu et al, 2002; Köhler et al, 2010; Lei et al, 2014; Ratia et al, 2006; Renatus et al, 2006; Samara et al, 2010; Ye et al, 2011; Yin et al, 2015). Surprisingly, in one copy of USP12, no clear electron density is present for part of the sequence at the outer edge of the Fingers domain, which we name the Pinky Finger (Figure 1A and 1C).…”

“…But question remains as to whether these substrate-induced changes are involved in USP activation by non-substrate regulatory partners. In a recent report, the crystal structures of free and UAF1-complexed USP46 have been determined both with a conjugated suicidal ubiquitin substrate (Yin et al, 2015). However, a structural comparison of the enzyme in these two post-reaction forms revealed few conformational differences, leaving the mechanism of USP46 activation by UAF1 elusive.…”