Protein-based life faces constant and dynamic stress caused by protein mis-folding. This can cause a variety of potentially lethal alterations in cell physiology that underlie many important clinical maladies. Eukaryotic cells have evolved a variety of quality control mechanisms to ensure toxic proteins are removed. The Ubiquitin Proteasome System (UPS) is the principal mechanism of protein degradation in eukaryotes that utilizes ubiquitin ligases E1, E2 and E3 to tag misfolded proteins with polyubiquitin chains. Of these enzymes, E3 ligases determine the specificity in identifying misfolded substrates. We have discovered that the highly conserved E3 ubiquitin ligase Ubr1 works together with cellular chaperones to detect and destroy misfolded proteins by selectively ubiquitinating them for degradation. Ubr1 is highly conserved in all eukaryotes and well-studied in other capacities, but, little is known about its role in quality control. We have employed the yeast model misfolded substrate ssCPY*-GFP in a high throughput array based screen Systematic Genotype-Phenotype Array (SGPA) to uncover other genes involved in cellular quality control, including those that may work in the Ubr1/San1 pathway, and those that work in distinct branches of this process. From our screen, we discovered a set of genes related to translational modification and ubiquitin cleavage that appear to have unanticipated roles in protein quality control. The genes revealed in these studies are, to a large extent, conserved across all eukaryotes, so the knowledge we gain of their roles in quality control will impact future studies and hold clinical relevance.

As a part of protein quality control (PQC), the ubiquitin-proteosome system (UPS) ubiquitinates and degrades aberrant proteins. In this role, the UPS must identify damaged, orphaned, and misfolded species amidst a proteome-worth of normal, folded proteins. This exquisite specificity is achieved by PQC E3 ligases, each of which recognizes a broad range of PQC substrates. Alongside efforts to understand how PQC ligases effect selectivity, there has been sustained interest in PQC substrates and their distinguishing structural characteristics. Perhaps the best studied substrates are disease causative mutants, such as the most common cause of cystic fibrosis, CFTRΔF508. Numerous high-throughput screens have also aimed to identify “degrons,” discrete amino-acid sequences that elicit degradation. Despite these efforts, isolating and characterizing PQC substrates remains an exigent mode of inquiry in the field. Here, we present insights gained from PQC substrates both new and old. An initial study underlines the benefits of integrating transgenes onto yeast chromosomes: Stable integration of GFP-tagged proteins allows rapid and reliable quantitation of protein steady-state levels, a crucial indicator of substrate (in)stability. A second study demonstrates the value of one model, misfolded substrate, Sec61-2-GFP. Sec61-2-GFP allowed us to monitor each step of inner-nuclear-membrane-associated degradation (INMAD), from ubiquitination to retrotranslocation of the full-length protein. Sec61-2-GFP was also lethal in the combined absence of ERAD and INMAD, further evincing the elucidative power of appropriate model substrates. These studies and others justify the “Structure-Misfunction” screen, a platform to isolate minimally misfolded versions of cytosolic proteins. Tellingly, the screen seems to have uncovered a novel PQC pathway in the Saccharomyces cerevisiae cytosol. Structure-Misfunction analysis also provided insight into misfolding itself. Our studies show that different destabilizing point mutations within one domain can lead to entirely distinct PQC outcomes. These data suggest that minimal misfolding can cause “local” as opposed to “global” unfolding in vivo. The screen also identified mutants of chorismate mutase that can be stabilized by the allosteric effector tryptophan, a striking example of chemical chaperoning. Thus, the screen is a simple genetic approach to uncovering novel features of cell and structural biology.

As the rate of Type II Diabetes Mellitus (T2DM) develops not only in adults but in youth, many other health related problems are starting to arise in association. A major complication related to T2DM is cardiovascular disease, which female diabetic patients have been found to have a worsened outcome compared to male diabetic patients. To get a better understanding of the sex difference in cardiovascular disease and other complications associated with T2DM, mitochondrial function was investigated to find if it was an underlying cause and if there was a difference between males and females. Through an obesity driven T2DM model, young female and male mice were fed a 45% kcal fat diet (HFD) to induce diabetes and a 10% kcal fat diet (LFD) as control. Males on HFD significantly increased weight with a more profound impairment to glucose tolerance, while females on HFD slowly increased weight, developed ulcerative dermatitis (UD), and did not have as profound impairment to glucose tolerance. There was also a statistically significant three-way interaction between mitochondrial function, gender and diet in the hippocampus and kidney. A two-way interaction between mitochondrial function and diet was found to be statistically significant in the heart, but a statistically significant interaction between mitochondrial function and gender was found in the hippocampus, liver and pancreas.

Protein quality control is an essential set of processes which allow cells to detect and mange misfolded proteins. Deficits in protein quality control increase with aging and are associated with disease, from inborn disorders to neurodegeneration and cancer. One arm of quality control is endoplasmic reticulum (ER) associated degradation (ERAD) a pathway which identifies and ubiquitinates ER and secretory pathway proteins. The conserved E3 ligase Hrd1 catalyzes the ubiquitination of a variety of misfolded substrates, but was discovered for the regulated degradation of HMG-CoA reductase, a critical sterol pathway enzyme, in yeast. The yeast isozyme Hmg2 is degraded when flux through the sterol pathway is high and stabilized when levels of pathway molecules fall. Work in our laboratory found that under degradation conditions, Hmg2 takes on some of the features of a misfolded protein, including becoming more susceptible to proteolysis in vitro. Because of the clinical importance of sterols and protein quality control in general, and because of questions about how quality control pathways distinguish misfolded proteins from the structural diversity in the cell, we are interested in learning more about how Hmg2 transitions between its normal and misfolded states. In this work, we found that the isoprenoid GGPP causes direct, reversible misfolding of Hmg2. This action is potent, specific for GGPP’s structure, and antagonizable with a similar molecule. In response to GGPP, Hmg2 takes on the features of a misfolded protein. Its structure becomes more accessible to proteases and detergents, it is degraded by ERAD, and these effects are reversible by treatment with chemical chaperones. Hmg2 is a multimer, and its multimerization is unchanged by GGPP treatment. These actions require the sterol sensing domain (SSD), a domain found in many human disease proteins. SSD mutations prevent Hmg2 misfolding in cis and even in wild type Hmg2 in trans. Taking these findings together, we liken Hmg2 misfolding to allostery, with an output of misfolding.

A recently discovered class of endogenous lipids called fatty acid esters of hydroxy fatty acids (FAHFAs) display potent anti-diabetic and anti-inflammatory effects. This leads to the question of how these FAHFAs are regulated in vivo.

My thesis work focused on understanding how FAHFAs are biochemically degraded in an animal model. Prior to my work, androgen induced gene 1 (AIG1) and androgen-dependent TFPI regulating protein (ADTRP) were discovered as FAHFA hydrolases and I endeavored to identify Drosophila melanogaster orthologs to provide an additional model to study FAHFA regulation. First, I confirmed that FAHFAs are present in flies, suggesting this is a well conserved lipid class. Conservation analysis revealed two putative FAHFA hydrolases, CG3625b and CG11601, in the Drosophila melanogaster genome. Like AIG1 and ADTRP, CG3625b and CG11601 have a conserved active site threonine, and testing the biochemical activity of these proteins demonstrated that CG3625b and CG11601 are FAHFA hydrolases. The results show that D. melanogaster can be used as a possible animal model to further elucidate how FAHFAs are physiologically regulated.

Parkinson’s Disease (PD) is a neurodegenerative movement disorder that affects a high percentage of the elderly population over 60. Autosomal dominant point mutations in the Leucine-Rich Repeat Kinase (LRRK2) gene mark the most common cause of familial PD. The main features of LRRK2 that drive PD include but are not limited to its kinase activity and binding partners. However, cellular functions and interactors of PD-driving and wild-type LRRK2 are only partially defined. Therefore, characterizing LRRK2 binding partners becomes one of the top priorities in learning about the disease. Chapter 1 provides a detailed background on LRRK2, its known functions, and interactors. Chapter 2 demonstrates Rab38 as a novel kinase regulator of LRRK2 under healthy and pathologic physiological conditions with LRRK2 PD mutants in murine melanocytes. Moreover, Rab38 also mediates membrane association and cellular localization of LRRK2. Rab38 and its more widely studied homolog Rab32 have been treated as proteins of the same functions; my experiments elucidate more distinct functions of Rab38 and Rab32 in murine melanocytes. Chapter 3 delineates a well-known LRRK2 differential binding partner, 14-3-3, while using it to calibrate an affinity purification mass-spectrometry (AP-MS) system for novel LRRK2 interactor screening. CLIMP63 is believed to be a novel binding partner of LRRK2 in our first LFQ (label-free quantification) MS screen. While investigating the binding between CLIMP63 and LRRK2 WT and R1441G, I found that exogenous Myc-CLIMP63’s binding to LRRK2 is mediated by LRRK2 residues independent from S910/935, which calls for further investigation into its precise binding sites.