|Home | About | Journals | Submit | Contact Us | Français|
In a recent issue of Molecular Cell, Das et al. (2009) show that the G2BR domain of gp78, a RING-family E3 ubiquitin ligase, binds the E2 Ube2g2 and induces conformational changes within Ube2g2 to allosterically enhance its interaction with cognate E3 RING domains.
In the ubiquitin-Proteasome system, E1-E2-E3 enzymatic cascades transfer ubiquitin (Ub) onto protein substrates. First, after an ATP-consuming step, an E1 forms a thiolester bond with Ub, and promotes Ub transfer to the catalytic cysteine of an E2. A Ub-loaded E2 assembles with an E3 for substrate ligation to Ub or polyUb chains, which either trigger degradation by the 26S Proteasome or affect other functions such as subcellular localization. E3s for Ub often contain one of two structurally- and mechanistically-distinct classes of E2-binding/“catalytic” domains: HECT and RING (Deshaies and Joazeiro, 2009). HECT E3s contain a catalytic cysteine that receives Ub from an E2 to form a transient thiolester-linked E3~Ub intermediate, prior to Ub transfer to a substrate. By contrast, RING E3s bind a thiolester-linked E2~Ub complex, and promote Ub transfer from an E2 to a substrate. In many cases, an E2 is thought to cycle between binding E1 for loading with Ub and binding E3 for Ub transfer. Thus, E2 interactions with both HECT and RING E3s are often fleeting (Deshaies and Joazeiro, 2009).
gp78 is one RING E3 that forms a particularly stable complex with its cognate E2, Ube2g2 (Chen et al., 2006). A membrane glycoprotein, gp78 is also known as the tumor autocrine motility factor receptor, and has been functionally associated with cell migration, tumor invasion, and metastasis. gp78 is primarily localized to the endoplasmic reticulum (ER) to regulate ER associated degradation (ERAD) by promoting destruction of misfolded proteins (Chen et al., 2006). The 298-residue gp78 N-terminal domain contains at least five membrane-spanning helices that anchor the protein in ER membranes. The C-terminal cytoplasmic tail has at least five protein-protein interaction motifs, including RING, self-association, and CUE (coupling of Ub conjugation to ERAD) domains, the “G2BR” Ube2g2-binding region, and a C-terminal binding site for p97 (Chen et al., 2006; Li et al., 2009). gp78 functions with the E2 Ube2g2 to generate Lys48-linked polyUb chains in an intriguing manner: rather than Ub molecules being transferred one-at-at-time from the Ube2g2 catalytic cysteine to a substrate, gp78 contains an oligomerization sequence that is required for the gp78 RING to promote Ub transfer between Ube2g2 molecules such that a polyUb chain synthesized on Ube2g2 is transferred en bloc (Li et al., 2009 and references therein; also data reviewed in Deshaies and Joazeiro, 2009). Notably, formation of these Lys48-liked polyUb chains requires an extended loop, referred to as the “L2-”, “acidic-” or “β4α2-“ loop that for other E2s has been shown to associate with RING domains (Zheng et al., 2000) and influence processivity and specificity of Lys48-linked polyUb chain synthesis (Petroski and Deshaies, 2005; also data reviewed in Deshaies and Joazeiro, 2009).
Besides the widely recognized E2-E3 interaction between Ube2g2 and the gp78 RING domain, a second E2-binding site (G2BR) of gp78 was reported to interact with Ube2g2 (Chen et al., 2006). Now, the groups of Brunger and Ye, and of Byrd and Weissman have reported the crystal structure of the gp78 G2BR-Ube2g2 complex (Das et al., 2009; Li et al., 2009): G2BR forms a helix that binds the Ube2g2 “backside” (Brzovic et al., 2006) opposite from the catalytic cysteine and surrounding loops, and distal from the RING domain binding site (Zheng et al., 2000). Unexpectedly, Das et al. found that the G2BR not only recruits Ube2g2, but also induces conformational changes in the β4α2- and α2α3- loops that surround the Ube2g2 catalytic cysteine far from the G2BR-interacting surface (Figure 1). Das et al. further showed that G2BR binding influences Ube2g2 activity in at least two ways. First, the conformational changes in the β4α2- and α2α3- loops decrease accessibility around the Ube2g2 active site (Das et al., 2009). Indeed, G2BR-binding hindered E1-mediated loading of Ub onto the Ube2g2 catalytic cysteine. Furthermore, this inhibitory effect of G2BR depended on the β4α2-loop, as its deletion partially restored Ub loading in the presence of G2BR (Das et al., 2009).
The second and perhaps most striking effect of G2BR binding is its impact on Ube2g2 association with the gp78 RING domain. The G2BR interaction results in nearly 50-fold increased affinity between Ube2g2 and the gp78 RING domain (Das et al., 2009). This is not simply an effect of raising effective concentration by gp78 recruiting Ube2g2 through multiple sites: remarkably, providing the G2BR as a separate peptide in trans is sufficient to increase Ube2g2 binding to the gp78 RING domain, and polyubiquitination (Das et al., 2009). Binding to the G2BR appears to generally enhance Ube2g2 interaction with partner RING domains, because the G2BR peptide also enhanced polyubiquitination by RING domain containing fragments of other Ube2g2 partners, such as HsHRD1 and Trc8 (Das et al., 2009).
How might G2BR-binding mediate long-distance conformational changes in Ube2g2 to enhance binding to RING domains? It is tempting to speculate that clusters of energetically coupled side-chains sense G2BR binding on the Ube2g2 “backside”, and translate this information into changes across the Ube2g2 structure (Ozkan et al., 2005). While future structural studies of Ube2g2 dually-engaged by the gp78 G2BR and RING will be required to fully understand the allosteric regulation, the β4α2-loop corresponds to a RING-binding region (Deshaies and Joazeiro, 2009; Zheng et al., 2000) and is thus an attractive candidate for playing a role in G2BR-mediated enhanced E2-RING affinity.
It is intriguing that the G2BR has opposite effects on E1-mediated Ub loading of Ube2g2, and subsequent E3 RING domain binding/Ub unloading. At present, it is unclear how the G2BR-mediated decrease in forming a Ube2g2~Ub thiolester intermediate might influence flux through the E1-E2-E3 cascade. Nonetheless, it is clear that G2BR-binding to the Ube2g2 backside has pleiotropic effects on both protein-protein interactions and the environment of the E2 active site.
The finding that the G2BR binding to the Ube2g2 backside influences numerous E2 functions at a distance may serve as a paradigm for understanding regulation of other E2s. Many E2s have additional partners besides E1 and HECT and/or RING domains. It is possible that some of these may function in a manner analogous to G2BR, to allosterically regulate E2 functions. For example, in addition to a RING domain, the E3 Ubr1 has a “basic residue region” essential for high-affinity interaction with the E2 Ubc2 (Xie and Varshavsky, 1999). Another type of regulation is observed for the HECT E3 Smurf2: a distinct protein, Smad7, promotes binding of the E2 UbcH7 to the Smurf2 HECT domain (Ogunjimi et al., 2005). A different interaction important for polyubiquitination is observed for E2s in the UbcH4/5-family, which bind noncovalently to Ub via their “backside” surface corresponding to where Ube2g2 binds the G2BR (Brzovic et al., 2006). The precise mechanisms by which these and other interactions alter functions of E2s remain incompletely understood. The existence of fascinating allosteric mechanisms underlying these and/or other E2 interactions may be foreshadowed by the unexpected G2BR-mediated allosteric regulation of Ube2g2 (Das et al., 2009).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.