The conjugation of proteins with ubiquitin involves the sequential action of proteins known as E1 (ubiquitin activating enzyme), E2s or UBCs (ubiquitin conjugating enzymes), and E3s (ubiquitin protein ligases). We have previously determined that RING finger proteins are, in general, ubiquitin ligases and are now focused on understanding issues related to the mechanism of action and substrates of specific members of this family as well as identifying inhibitors. Our studies looking for inhibitors of Hdm2 have resulted in the identification of a family of compounds that inhibit Hdm2 E3 activity both in vitro and in cells (HLI98 family) and we have now identified a related small molecule that is highly soluble in water and exhibits greater potency than the HLI98 compounds. This small molecule, HLI373, selectively kills p53-expressing cancer cells in a manner similar to the original compounds. Additionally, our screens have also resulted in the identification of an inhibitor of the ubiquitin E1 inhibitor. This inhibitor markedly deceases ubiquitin-mediated degradation of substrates as well as non-proteolytic functions of the ubiquitin system including ligand-induced activation of Nf-kB and ligand-mediated ubiquitination of the epidermal growth factor receptor. It also results in increased levels of p53 and in selective cell death in transformed cells and may also serve as the basis for new therapeutics for cancer and other diseases in which perturbing functions of the ubiquitin system may be of beneficial. In this regard, multiple pharmaceutical companies are now pursuing the same targets. We are also carrying out structure-function and structural studies on the transmembrane pro-metastatic endoplasmic reticulum (ER)-associated degradation (ERAD) ubiquitin ligase gp78. We had previously identified a second E2 binding site on gp78 specific for the E2 Ube2G2, known as the Ube2G2 Binding Region (G2BR). We have found that the G2BR binds to a region of Ube2g2 distinct from where the RING finger binds this E2 ('backside' binding to the E2) and increases the affinity of Ube2g2 for the gp78 RING finger and therefore stimulates ubiquitination. We have determined that the means by which this allosteric activation occurs and have also identified a novel 'reverse' allosteric effect that allows Ube2G2 to dissociate after ubiquitin has been transferred. We had predicted that this is likely to be the first of multiple such binding sites on the backside of E2s and that these types of interaction may represent a general means by which E2-E3 function can be modulated. This turns out to be the case, as we have identified a second such binding site on the yeast Cue1p protein, which is integral to the two ERAD E3s of yeast. This region bears both similarities and differences in structure and function to the G2BR and demonstrates the heterogeneity of function that can be displayed by these backside binding sites. We have also identified an additional such site in the ubiquitin ligase AO7/RNF25, which has distinct functions and is also allowing for an exploration of the functional significance of mutations in the RING:E2 interface. Collectively these studies are leading to new insights into how RING finger proteins function together with E2s. Our work in this area has circled back around to the G2BR as we have identified another protein involved in ERAD that also has a G2BR-like domain and which also binds to Ube2G2. We are currently determining the similarities and differences between these two domains and assessing their relative roles in targeting specific proteins for degradation. In other studies on this topic, we have been interested in the molecular basis for the differential persistence of different serotypes of botulinum neurotoxin (Bo/NT). We have discovered that the ubiquitin ligase TRAF2 targets the shorter-lived Bo/NTE light chain, LCE, but not the very long-lived light chain of the most potent neurotoxin Bo/NTA, LCA. These findings provided the first evidence for a role for the ubiquitin system in botulinum neurotoxin. Our studies in this area have now moved on to an assessment of Bo/NTA. We have recently determined that LCA is targeted for degradation normally by the ubiquitin ligase HECT D2. We have also discovered that the longevity of this protein is a consequence of the action of a deubiquitinating enzyme (DUB) known as VCIP135. This DUB, about which little is known, stabilizes LCA by dynamically deubiquitinating this toxin. Thus, we have unmasked the basis for the most potent and persistent of toxins. We have also determined that VCIP135 binds to a C-terminal region of LCA distinct from where HECT D2 binds. It now becomes of great interest to consider means of shortening the persistence of LCA either by blocking VCIP135 binding or by inhibiting the DUB. Studies are being initiated that have this as a goal.
Showing the most recent 10 out of 21 publications
