Ubiquitin is a highly conserved eukaryotic protein of 76 amino acids that is added post-translationally to other proteins or to itself by a hierarchical cascade of enzymes (E1, E2, and E3) in distinct pathways. Often, a polyubiquitin chain is built onto a substrate to target it for degradation by the 26S proteasome. The selection of a ubiquitin conjugating enzyme (E2) for a particular ubiquitin ligase (E3) and the subsequent transfer of ubiquitin from the E2 onto the substrate lysine are two processes that remain to be better understood. It is known that E3s harbor a limited number of conserved sequence motifs that recruit specific E2 enzymes. One such motif is known as the RING domain, comprising of amino acids that coordinate two zinc atoms in a conserved structural fold. Based on the finding that RING domains from known or putative E3 partners of Ubc7, E2-25K and Cdc34 are able to activate an intrinsic ubiquitin to ubiquitin linkage activity in these three E2 enzymes, work in this thesis uses the premise that insights on E3-E2 interactions can be obtained by studying the interactions between an E2 enzyme with isolated RING domains.

A collection of human RING domains that belong to ∼100 E3s is available in our laboratory for a study toward the identification of RING domains that selectively activate Ubc7, E2-25K, or Cdc34. In work carried out in this thesis, eleven new Ubc7- and seven new E2-25K-activating RING sequences were identified. To understand the structure-function relationship of these E2-activating sequences, work in this thesis examined a RING-E2 fusion strategy to facilitate structural analysis of RING-E2 interactions. Independently of this thesis work, the gp78RING-Ubc7 fusion protein was used to obtain diffraction quality crystals from which a 2.2 Å resolution structure was obtained. Using the atomic coordinates of this structure, calculation and mutational analyses were carried out to assess selected contacts at the two-protein interface. Results from this analysis identified three residues in Ubc7 (R8, L66, and V113) that contribute importantly to the binding affinity, of indicating that positive interactions make a significant contribution to E2-E3 pairings. Thus, selective E2-E3 pairings can be dictated in part by the presence of specific residues in the sequence of these proteins that supply positive interactions. Based on this analysis, it can be concluded that certain amino acids cannot readily be accommodated into the gp78-Ubc7 interface, and a negative selection process may also contribute to the stringent pairing of an E2 with an E3.

Work in this thesis also examined the catalytic function of a set of residues in Ubc7. The transfer of ubiquitin from the thiolester-linked form in an E2~Ub complex to a lysine requires the latter to be in the deprotonated form. To examine residues in Ubc7 that may assist in this deprotonation step of the reaction, work in this thesis examined the pH dependence of Ubc7 activity, with the rationale that loss of this function in a Ubc7 mutant will lead to a pH profile that is dictated by the substrate lysine. This analysis indicates that H94 and E104 function together in this mechanism. This finding differs from that suggested for Ubc9, a SUMO-conjugating enzyme that belongs to the UBC family of proteins.

In conclusion, work carried out in this thesis led to the identification of a set of E2-activating RING domains that can be used to analyze how E3 and E2 enzymes are selectively paired. A study examining the interaction between the RING domain in gp78 and Ubc7 has provided initial structure-function insights. The work in this thesis also identified important catalytic residues in Ubc7. These results have led to further testable hypotheses on the selection of an E2 by an E3 and how the RING domain may allosterically activate an E2 enzyme.