The determinants of chain type specificity and the mechanism of polyubiquitination by HECT E3s

Ubiquitination is a post-translational modification that can take several forms. Some proteins are modified with a single ubiquitin molecule, while others are modified with polyubiquitin chains. Each type of ubiquitination is thought to have distinct biological functions. The best-characterized types of ubiquitin modification are K48-linked polyubiquitination, which serves as a signal for proteasomal degradation and K63-linked polyubiquitination, which has non-proteolytic functions such in DNA repair, signaling, and endocytosis. HECT ubiquitin ligases (HECT E3s) form a class of E3s, defined by a C terminal catalytic domain. Several lines of evidence suggested that the HECT E3s assemble a polyubiquitin chain in a sequential manner with one molecule of ubiquitin at a time being conjugated to the distal ubiquitin of the chain. In the process of chain elongation, not all HECT E3s target a common internal lysine of ubiquitin, leading to diversification of chain type specificity in HECT E3s. For example, yeast Rsp5 forms K63 chains, while human E6AP forms K48 chains. Two important mechanistic questions were addressed in my work: 1) what are the determinants of chain type specificity of HECT E3s, and 2) what allows the distal ubiquitin of a chain to be continuously oriented near the active site of the HECT domain in the course of a sequential polyubiquitination reaction?
I have determined that the chain type specificity of Rsp5 is a function solely of the HECT domain. Further, through the generation of chimeric HECT E3s, I demonstrated that chain type specificity determinants are located within the last 60 amino acids of the C lobe of the HECT domain. To address the second question, we solved the structure of Rsp5 HECT domain in complex with non-covalently bound ubiquitin in collaboration with Jue Chen’s laboratory (Purdue University). From the structure, we found that the N lobe of the HECT domain binds ubiquitin in a manner distinct from other known ubiquitin binding domains, and I have shown that Rsp5 proteins defective for ubiquitin binding are defective for chain elongation. We hypothesize that the ubiquitin binding site functions in the recruitment of the distal ubiquitin of polyubiquitin chain for efficient polyubiquitination.