The successful biogenesis – synthesis, delivery, and insertion into designated membranes – of membrane proteins is a crucial cellular process. One particular class of membrane proteins, tail-anchored (TA) proteins have a single transmembrane domain (TMD) that this located at their C-termini and are targeted to membranes post-translationally. Multiple pathways have been identified to target TA proteins to the ER membranes, but designated pathways for targeting TA proteins to the mitochondria remain elusive. The most well understood ER TA protein pathway is the Guided Entry of Tail-anchored proteins (GET) pathway, consisting of six (fungal) or seven (metazoans) proteins, SGTA, Get1-5, and Bag6 (metazoans only), has nearly been studied exclusively in Opisthokants (fungi and metazoans). Here we employed a combination of x-ray crystallography, cryo-electron microscopy, computational modeling, cellular biology, fluorescent imaging, and bioinformatics in order to understand the underlying factors that regulate the targeting of these TA proteins to their correct membranes. Our work reveals that ER-bound TA proteins tend to have a hydrophobic face whereas mitochondria-bound TA proteins contain a charge following their TMD. This finding corroborates our observation that the first component of the GET pathway to interact with TA proteins, SGTA, falls in a category of other hydrophobic segment binding domains, dubbed STI1-domains. Structures presented here demonstrate that the overall structure of Get3 is conserved in organisms as distant as Excavats and Opistokonts, and slight conformational changes in the ATPase allows the described chaperone cascade of the GET pathway to progress. Together these results refine the model for TA protein targeting to the ER membrane.
}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/VKH3-XX98, author = {Yun, Hyun Gi}, title = {Structural and Biochemical Studies of Enzymes in Bacterial Glycobiology}, school = {California Institute of Technology}, year = {2019}, doi = {10.7907/VKH3-XX98}, url = {https://resolver.caltech.edu/CaltechTHESIS:05292019-144516804}, abstract = {The speed that bacterial pathogens gain resistance to antibiotics is alarming. Designing new antibacterial agents is urgent, but it requires understanding their bacterial targets at the molecular level to achieve high specificity and potency. In this thesis, I discuss the structural and biochemical investigations of three potential protein targets for antibiotics. The first is a UDP-Glc/GlcNAc 4-epimerase, called Gne, from the human pathogen Campylobacter jejuni. This enzyme is the sole source of N-acetylgalactosamine (GalNAc) in C. jejuni, which is a common component in three major glycoconjugates decorating the cell surface and is critical for pathogenesis. The second target protein is an integral membrane protein, called MraY, which catalyzes the transfer of phospho-N-acetylmuramyl (MurNAc) pentapeptide to a lipid carrier, undecaprenyl phosphate (C55-P), producing Lipid I in the peptidoglycan biosynthesis pathway. In the following step, a peripheral protein called MurG catalyzes transferring N-acetylglucosamine (GlcNAc) to Lipid I and produces Lipid II, which provides the first building block of the peptidoglycan layer. Peptidoglycan is uniquely bacterial, with MraY and MurG both being essential for cell viability; therefore, they are attractive targets for the development of antibacterial agents and work toward their structures is presented. Finally, MraY from Escherichia coli is the target for the lysis protein E from phage ΦX174.Efforts toward elucidating the EcMraY-E complexstructure are demonstrated here. In total, this thesis provides important data toward a full mechanistic understanding of these important antibacterial targets.}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/Z9Q23XD3, author = {Marshall, Stephen Sandell}, title = {Improvement of Integral Membrane Protein Expression via Optimization of Simulated Integration Efficiency}, school = {California Institute of Technology}, year = {2018}, doi = {10.7907/Z9Q23XD3}, url = {https://resolver.caltech.edu/CaltechTHESIS:08292017-151537318}, abstract = {Integral membrane protein characterization is limited by the low levels of protein obtainable from heterologous overexpression in hosts such as Escherichia coli. Differences in the efficiencies of subdomains of the co-translational integration processes of membrane proteins into the membrane could explain the observed variation in the experimental expression of closely related homologs in E. coli. We have developed a method to predict and increase the expression of individual membrane proteins by optimizing the efficiency of their translocon-mediated integration into the membrane. The integration efficiency of each component of a membrane protein is calculated using a coarse-grained co-translational simulated integration model. The results of model simulations, experimental expression levels quantified by integral membrane protein-GFP fusion fluorescence, and a novel antibiotic survival test that reports on misintegration in vivo are applied to test the relationship between the integration efficiency of specific domains and experimental expression. Changes in simulated integration efficiencies due to sequence modifications agree with the effects on experimental expression in vivo. In the case of the TatC protein family, misintegration of the C-tail is found to be a major contributor to expression failure in E. coli. Beneficial sequence modifications that improve both simulated integration efficiency and experimental expression levels can be identified using the model. Preliminary evidence shows that simulated integration efficiency could potentially predict the effects of mutations on Haemophilus influenzae GlpG experimental expression in E. coli. The process described herein allows for the rational overexpression of integral membrane proteins through the identification and mitigation of inefficiencies in the underlying co-translational membrane integration process.
}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/Z91C1TXQ, author = {Mock, Jee Young}, title = {The Bag6 Complex: Biological Complexity through Modularity}, school = {California Institute of Technology}, year = {2017}, doi = {10.7907/Z91C1TXQ}, url = {https://resolver.caltech.edu/CaltechTHESIS:06052017-184553539}, abstract = {Proper synthesis and targeting of membrane proteins that contain hydrophobic transmembrane domains are mediated by chaperones and targeting factors. Tail-anchored (TA) proteins are a special class of membrane proteins that are characterized by a single carboxy (C) terminal helix that anchors them to biological membranes. Fungal Guided Entry of Tail-anchored protein (GET) pathway components, which include four soluble proteins—Sgt2, Get3, Get4, Get5—and two membrane bound receptors—Get1 and Get2—mediate TA biogenesis. These proteins maintain TA protein solubility in the aqueous cytosol and target TA to the endoplasmic reticulum. While most of the components are conserved in metazoans, one additional protein, Bag6, reorganizes the sorting complex from the heterotetrameric Get4-5 to the heterotrimeric Bag6-TRC35-Ubl4A. To understand the molecular architecture and mechanism of the Bag6 complex, we took a multidisciplinary approach that combines x-ray crystallography, biochemical reconstitution, and cell biology. Our studies demonstrate that the BAG domain of Bag6 is not a canonical BAG domain. Instead, main role of the Bag6 ‘mock’ BAG domain is to dimerize with Ubl4A. Furthermore, the truncated Bag6 complex defined in this study is sufficient to facilitate substrate transfer from SGTA to TRC40. Lastly, our results unequivocally establish TRC35 as a cytoplasmic retention factor for Bag6. These results provide structural, biochemical and cell biological bases for modular Bag6 function and regulation of nucleocytoplasmic distribution of Bag6 by TRC35.}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/Z95718ZD, author = {Gristick, Harry Benjamin}, title = {Investigating the Role of the GET3-GET4/GET5 Interaction During Tail-Anchor Protein Targeting}, school = {California Institute of Technology}, year = {2015}, doi = {10.7907/Z95718ZD}, url = {https://resolver.caltech.edu/CaltechTHESIS:06052015-162319516}, abstract = {The proper targeting of membrane proteins is essential to the viability of all cells. Tail-anchored (TA) proteins, defined as having a single transmembrane helix at their C-terminus, are post-translationally targeted to the endoplasmic reticulum (ER) membrane by the GET pathway (Guided Entry of TA proteins). In the yeast pathway, the handover of TA substrates is mediated by the heterotetrameric Get4/Get5 (Get4/5) complex, which tethers the co-chaperone Sgt2 to the central targeting factor, the Get3 ATPase. Although binding of Get4/5 to Get3 is critical for efficient TA targeting, the mechanisms by which Get4 regulates Get3 are unknown. To understand the molecular basis of Get4 function, we used a combination of structural biology, biochemistry, and cell biology. Get4/5 binds across the Get3 dimer interface, in an orientation only compatible with a closed Get3, providing insight into the role of nucleotide in complex formation. Additionally, this structure reveals two functionally distinct binding interfaces for anchoring and ATPase regulation, and loss of the regulatory interface leads to strong defects in vitro and in vivo. Additional crystal structures of the Get3-Get4/5 complex give rise to an alternate conformation, which represents an initial binding interaction mediated by electrostatics that facilitates the rate of subsequent inhibited complex formation. This interface is supported by an in-depth kinetic analysis of the Get3-Get4/5 interaction confirming the two-step complex formation. These results allow us to generate a refined model for Get4/5 function in TA targeting.}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/QVCV-8A76, author = {Chartron, Justin William}, title = {The Structure of a Transmembrane Protein Sorting Complex}, school = {California Institute of Technology}, year = {2013}, doi = {10.7907/QVCV-8A76}, url = {https://resolver.caltech.edu/CaltechTHESIS:10012012-155636249}, abstract = {The biogenesis of membrane proteins is an essential process in biology. It requires the protection of hydrophobic transmembrane domains from aggregation in the cytosol as well as targeting to the proper membrane. Tail-anchored (TA) proteins have a single transmembrane helix near their carboxyl termini and require a post-translational mechanism for targeting and insertion. In yeast, the Guided Entry of Tail-anchored proteins (GET) pathway delivers TA proteins to the endoplasmic reticulum (ER). A sorting complex comprising Get4, Get5, and Sgt2 load ER destined TA proteins onto the targeting factor Get3. X-ray crystallography, solution NMR, and small angle X-ray scattering were used to characterize this assembly. Get4 and Get5 form an extended adapter complex. Get4 maintains Get3 in a state competent to receive TA proteins. The N-terminus of Get5 tightly binds Get4, while the C-terminus of Get5 is a homodimerization domain, resulting in a heterotetrameric assembly. A ubiquitin-like domain within Get5 binds the heat-shock protein (HSP) co-chaperone Sgt2, providing a physical link between ER destined TA protein targeting and protein folding pathways. Sgt2 is also an extended homodimeric complex, and can directly bind four major classes of HSPs. The Get4/Get5/Sgt2 sorting complex is multivalent, flexible and the binding of individual components is transient. These results build a model for post-translational protein targeting in eukaryotes that is distinct from other pathways.}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, } @phdthesis{10.7907/AGES-WJ42, author = {Suloway, Christian J. M.}, title = {Structural Insights into Tail-Anchored Protein Targeting by Get3}, school = {California Institute of Technology}, year = {2012}, doi = {10.7907/AGES-WJ42}, url = {https://resolver.caltech.edu/CaltechTHESIS:02282012-151846967}, abstract = {Translocation of membrane proteins from the point of synthesis to their integration in the membrane is critical to the function of the cell. Tail-anchored (TA) proteins are an important class of membrane proteins with a single transmembrane domain (TMD) close to the carboxyl-terminus. They are defined topologically by having their amino-terminus in the cytosol and their carboxyl-terminus on the exterior side of the membrane. Since the TMD is sequestered by the ribosome during translation, co-translational translocation of TA proteins by the SRP-dependent pathway is not possible. The Guided Entry of Tail-anchored proteins (GET) pathway post-translationally targets TA proteins to the endoplasmic reticulum (ER) membrane. The conserved nucleotide hydrolase Get3 is the central protein in the pathway that specifically binds the TMD of TA proteins to chaperone them from a sorting complex of Get4, Get5, Sgt2 and other chaperones to an ER membrane receptor formed by Get1 and Get2. We have created a model for the mechanism of Get3 TA protein binding coupled to nucleotide state using X-ray crystallography, structural modeling and mutagenesis experiments. We then demonstrate expression, purification and crystallization of complexes of Get3 with TA proteins for structural studies. Finally, we present a crystal structure of a tetrameric archaeal Get3 homologue that forms a central hydrophobic chamber and is capable of binding TA proteins. Using small-angle X-ray scattering, the structure is comparable to a tetrameric fungal Get3 complex with TA protein, which is capable of TA protein membrane integration in vitro. This suggests a model in which a tetramer of Get3 binds TA proteins for delivery to the membrane.}, address = {1200 East California Boulevard, Pasadena, California 91125}, advisor = {Clemons, William M.}, }