Using a combination of genetic or proteomic screens, biochemical reconstitution, and small molecule discovery, we are aspiring to discover, and mechanistically understand, signaling pathways that guide human cell fate decisions. Our work often starts with investigating the role of ubiquitin-dependent modifications in development, but has branched out to dissect the consequences of protein misfolding and aggregation, aberrant complex formation, or defective redox regulation. Our multidisciplinary approach is made possible by close collaborations with groups in the exciting Bay Area, including at UC Berkeley, UCSF, Stanford and Genentech. We have successfully translated findings from our work into new approaches to chemotherapy, spearheaded by Nurix Therapeutics. It is one of our most important goals to accomplish the same for the treatment of neurodegenerative diseases.
Projects in the lab are driven by the interests of students and postdocs. Postdocs leaving the lab take all their projects with them. Current projects include, but are not limited to, the three examples described below:
We recently discovered the core components of the reductive stress response (Manford et al., Cell 2020). This protective system detects the pervasive absence of reactive oxygen species (ROS), a sign of an inactive mitochondrial electron transport chain, and in turn triggers the proteasomal degradation of a mitochondrial gatekeeper to jumpstart oxidative phosphorylation. The reductive stress response therefore controls production of ATP and ROS and is therefore a key player in maintaining physiological energy and ROS states. Our work has opened many exciting questions that we are actively pursuing. We are specifically interested in:
While most quality control pathways target misfolded proteins, we recently found that cells can also recognize protein complexes that contain inappropriate subunits. Our first example of such quality control of protein complex composition was centered of the E3 ubiquitin ligase SCFFBXL17, which can selective degrade inactive heterodimers of BTB domain containing proteins, while it leaves active BTB homodimers intact (Mena et al., Science 2018). Structural studies showed that SCFFBXL17 recognizes the shape and complementarity of BTB domains, thus using distinct modes of substrate recognition than most known E3 ligases (Mena et al., Nature 2020). We named this pathway dimerization quality control (DQC). Interestingly, loss of DQC leads to specific phenotypes in the central and peripheral nervous system, underscoring a critical role in human development. We are now interested in asking:
While dissecting the molecular mechanism of ubiquitin chain formation, we discovered functions for the first atypical ubiquitin chain, i.e. K11-linked conjugates, and for the first heterotypic chain, i.e. K11/K48-branched conjugates (Jin et al., Cell 2008; Meyer et al., Cell 2014). Biochemical work showed that K11/K48-branched chains are proteasomal priority signals reserved for the most difficult proteolytic targets. Indeed, by developing bispecific antibodies against K11/K48-branched chains, which we accomplished together with Marissa Matsumoto and Vishva Dixit at Genentech, we found that K11/K48-branched ubiquitin chains are required for rapid elimination of aggregation-prone nascent polypeptides (Yau et al., Cell 2017). We also found that these conjugates are assembled by specific E3 ligases, i.e. UBR4 and UBR5, and recognized by dedicated effectors, including the ubiquitin-selective segregase p97/VCP, whose function we described many years ago (Rape et al., Cell 2001). All effectors of K11/K48-branched ubiquitin chains are mutated in familial forms of neurodegenerative diseases. Thus, we are now very interested in asking: