"Elements of the Ubiquitin Ligase Doa10 Regulating Protein Ubiquitylati" by Adrian Mehrtash

Date of Award

Fall 2022

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Molecular, Cellular, and Developmental Biology

First Advisor

Hochstrasser, Mark

Abstract

In endoplasmic reticulum-associated degradation (ERAD), misfolded proteins at the endoplasmic reticulum (ER) are degraded by the proteasome after undergoing ubiquitylation. In yeast, Doa10 (MARCHF6/TEB4 in humans) is a conserved membrane-embedded E3 ubiquitin ligase that functions with the E2 ubiquitin-conjugating enzymes Ubc6 and Ubc7. These E2s use a sequential E2 mechanism where Ubc6 attaches a single ubiquitin molecule to a lysine, serine, or threonine residue of a substrate that is subsequently elongated by Ubc7 to form a poly-ubiquitin chain. How the Doa10 complex mediates non-canonical and sequential poly-ubiquitylation of its substrates is unclear. I determine the conserved 16-residue C-terminal element (CTE) of Doa10 is required for E3-mediated activity of Ubc6. Therefore, only Doa10 substrates that undergo ubiquitylation through an alternative mechanism are degraded in CTE mutant cells. Protein structure prediction by AlphaFold2 suggests the CTE interacts with the catalytic RING-CH domain of Doa10, implying a direct role in substrate ubiquitylation. Through truncation analysis, I define the minimal E2-binding region of Doa10, which includes 9 of its 14 transmembrane helices. Structural predictions suggest Doa10 forms an ERAD protein channel and E2-binding occurs within the minimal cofactor-binding region defined here. In particular, Cue1 (the Ubc7 activator) and Ubc6 are predicted to interact outside of the Doa0 channel near TM7, TM9, and Loop 8. Interestingly, Loop 8 of Doa10 forms a unique globular domain at the cytosolic/ER membrane surface, where it interacts extensively with Cue1 and Ubc6. The formation of a large protein channel by the 14 TMs of Doa10 is consistent with this E3 having a direct role in retrotranslocation (membrane extraction) of its substrates. The predicted protein channel is lined by the conserved TD domain, which forms a cave-like feature, and likely represents the substrate-binding site of Doa10. Charged residues within the cave-like feature are highly conserved and mutagenesis studies found this region is sensitive to mutation. While Doa10 likely contributes to the retrotranslocation of membrane substrates, I provide evidence that suggests membrane substrates can be retrotranslocated independently of the ERAD E3 ligase complexes and the Derlin Dfm1. Collectively; these proteins represent the known retrotranslocation pathways in yeast. Therefore, I predict the existence of an unidentified ERAD E3-independent retrotranslocation pathway. Overall, these results provide mechanistic insight into how Doa10, and potentially other ERAD ligase complexes, interacts with its cognate E2 enzymes and mediates substrate turnover.

Share

COinS