{"http:\/\/dx.doi.org\/10.14288\/1.0389991":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Medicine, Faculty of","type":"literal","lang":"en"},{"value":"Biochemistry and Molecular Biology, Department of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeCampus":[{"value":"UBCV","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"Young, John William","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2024-04-30T07:00:00Z","type":"literal","lang":"en"},{"value":"2020","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#relatedDegree":[{"value":"Doctor of Philosophy - PhD","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeGrantor":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"Many periplasmic and extracellular Escherichia coli (E. coli) proteins are transported across the inner bacterial membrane through the highly conserved heterotrimeric SecYEG protein-conducting channel. During post-translational translocation, polypeptide substrates are driven across the membrane through SecYEG by the ATPase SecA, which binds to SecYEG and couples nucleotide hydrolysis to polypeptide movement. In the first part of this thesis, we study the dynamics of SecYEG-SecA interactions. We show that SecA is a highly dynamic enzyme, repeatedly binding to and dissociating from SecYEG during substrate translocation. Using two model Sec-dependent protein substrates, we show that the importance of these dynamics for efficient translocation depends on the length of the translocating protein substrate. In the second part of this thesis, we turn to quantitative proteomics to identify novel interactors of the SecYEG complex. Previous studies have identified and validated a series of membrane embedded interactors of SecYEG using classical detergent-based methods. However, it is possible that other important interactors of the Sec translocon may exist which have not yet been identified by detergent-based proteomic methods - the difficulties of using detergent-based methods to identify and characterize transient interactors of membrane proteins and complexes are well-documented. Here, we employ the peptidisc - a \"one-size-fits-all\" membrane mimetic - to identify and characterize potentially novel interactors of the Sec translocon in detergent-free conditions. One of the most notable interactions identified in this work is a super-complex between the Sec translocon and the outer membrane embedded Bam complex, which is required for insertion of outer membrane proteins (OMPs). This observation is particularly astonishing and has implications for our understanding of outer membrane protein biogenesis. Finally, we develop a functionalized variant of the peptidisc scaffold and demonstrate its utility for isolation of the membrane proteome.  As a simple case study, we employ the functionalized peptidisc scaffold to survey changes in the membrane proteome caused by altered gene expression. Potential future applications of the peptidisc membrane mimetic in the fields of membrane protein biochemistry and membrane proteomics will also be discussed.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/74194?expand=metadata","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"   EXPLORING THE INTERACTOME OF THE BACTERIAL SEC TRANSLOCON  by  John William Young  B.Sc., The University of British Columbia, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2020  \u00a9 John William Young, 2020  ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: Exploring the Interactome of the Bacterial Sec Translocon  submitted by John William Young in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biochemistry and Molecular Biology  Examining Committee: Dr. Franck Duong, Biochemistry and Molecular Biology Supervisor  Dr. Robert Molday, Biochemistry and Molecular Biology Supervisory Committee Member  Dr. Leonard Foster, Biochemistry and Molecular Biology Supervisory Committee Member Dr. Thibault Mayor, Biochemistry and Molecular Biology University Examiner Dr. Christopher Loewen, Cellular and Physiological Sciences University Examiner   iii  Abstract Many periplasmic and extracellular Escherichia coli (E. coli) proteins are transported across the inner bacterial membrane through the highly conserved heterotrimeric SecYEG protein-conducting channel. During post-translational translocation, polypeptide substrates are driven across the membrane through SecYEG by the ATPase SecA, which binds to SecYEG and couples nucleotide hydrolysis to polypeptide movement. In the first part of this thesis, we study the dynamics of SecYEG-SecA interactions. We show that SecA is a highly dynamic enzyme, repeatedly binding to and dissociating from SecYEG during substrate translocation. Using two model Sec-dependent protein substrates, we show that the importance of these dynamics for efficient translocation depends on the length of the translocating protein substrate. In the second part of this thesis, we turn to quantitative proteomics to identify novel interactors of the SecYEG complex. Previous studies have identified and validated a series of membrane embedded interactors of SecYEG using classical detergent-based methods. However, it is possible that other important interactors of the Sec translocon may exist which have not yet been identified by detergent-based proteomic methods - the difficulties of using detergent-based methods to identify and characterize transient interactors of membrane proteins and complexes are well-documented. Here, we employ the peptidisc - a \"one-size-fits-all\" membrane mimetic - to identify and characterize potentially novel interactors of the Sec translocon in detergent-free conditions. One of the most notable interactions identified in this work is a super-complex between the Sec translocon and the outer membrane embedded Bam complex, which is required for insertion of outer membrane proteins (OMPs). This observation is particularly astonishing and has implications for our understanding of outer membrane protein biogenesis. Finally, we develop a functionalized variant of the peptidisc scaffold and demonstrate its utility for isolation iv  of the membrane proteome.  As a simple case study, we employ the functionalized peptidisc scaffold to survey changes in the membrane proteome caused by altered gene expression. Potential future applications of the peptidisc membrane mimetic in the fields of membrane protein biochemistry and membrane proteomics will also be discussed.   v  Lay Summary  Membrane proteins are a highly relevant class of biological molecules and comprise ~60% of current drug targets. Membrane proteins control many of the essential processes of life, including the process of protein transport across cellular membranes - otherwise known as \"protein translocation\". In bacteria, protein translocation is mediated by the SecYEG membrane protein complex. Because protein translocation is so well conserved across evolution, the bacterial model has been widely used as a simple model to thoroughly study the mechanism of protein transport.  This thesis provides additional insights into bacterial protein transport mechanisms. Additionally, we use a technique called \"quantitative proteomics\" to search for novel and uncharacterized membrane protein interactors of the SecYEG complex which may modulate\/influence its protein transport activity. Finally, more generally, we present a new tool for the isolation and characterization of membrane proteins.    vi  Preface I wrote Chapter 1 with input from my supervisor, Dr. Duong. Permission was obtained from the respective journals to reuse figures and is indicated where applicable. Figure 14 was prepared by Harveer Singh Dhupar.  A version of Chapter 2 has been published:  Young, J., and F. Duong. 2019. \"Investigating the stability of the SecA-SecYEG complex during protein translocation across the bacterial membrane.\"  J Biol Chem 294 (10):3577-3587. doi: 10.1074\/jbc.RA118.006447. I performed all the experiments and the downstream data analysis. I wrote the manuscript with input from Dr. Duong.  A portion of Chapter 3 has been published and adapted for use in this thesis: Carlson, M.L., Stacey, R.G., Young, J.W., I.S. Wason, Z. Zhao, D.G. Rattray, N. Scott, C.H. Kerr, M. Babu, L.J. Foster, and F.V.H. Duong. 2019. \"Profiling the E. coli membrane interactome captured in peptidisc libraries.\" Elife 8. doi: 10.7554\/eLife.46615. An initial draft of this publication was written by Michael Carlson and R. Greg Stacey with input from all other authors. The final version of the manuscript was written by myself and Dr. Duong. All authors approved the final version of the manuscript. I designed and performed all experiments in Chapter 3. Zhiyu Zhao assisted with SILAC labeling and protein expression. James Saville assisted with cloning of BamA. David G. Rattray performed digestion and STAGE tipping of mass spectrometry samples. David G. Rattray also performed analysis of the mass spectrometry data using MaxQuant.  vii  A version of Chapter 4 has been accepted for publication in the Journal of Proteome Research: \"Young, J.W., Wason, I.S., Zhao, Z., D.G. Rattray, L.J. Foster, and F.V.H. Duong. 2020             \"His-tagged Peptidiscs Enable Affinity Purification of the Membrane Proteome for Downstream Mass Spectrometry Analysis\" (Submitted January 17th, 2020; accepted April 17th, 2020). I wrote the manuscript with input from Dr. Duong. All authors approved the final version of the manuscript. I designed and performed all experiments in Chapter 4. I analyzed the experimental data and prepared all figures and tables with input from Irvinder S. Wason and Zhiyu Zhao. David G. Rattray performed digestion and STAGE tipping of mass spectrometry samples. David G. Rattray also performed MaxQuant analysis of the mass spectrometry data.   I wrote Chapter 5 with input from my supervisor, Dr. Duong. viii  Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ....................................................................................................................... viii List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiii List of Abbreviations and Terms .............................................................................................. xiv Acknowledgements ................................................................................................................... xvii Dedication ................................................................................................................................. xviii Chapter 1: Introduction ................................................................................................................1 1.1 Overview of Sec-mediated protein translocation ............................................................ 1 1.2 The structure of the SecYEG complex ........................................................................... 5 1.3 Properties of pre-protein substrates ................................................................................ 7 1.4 The structure and mechanism of the SecA ATPase ...................................................... 10 1.5 Ancillary subunits of the Sec translocon ...................................................................... 17 1.5.1 SecDF .................................................................................................................... 18 1.5.2 YidC ...................................................................................................................... 22 1.5.3 The holo-translocon - SecYEG-SecDF-YidC ....................................................... 24 1.5.4 YfgM and PpiD ..................................................................................................... 25 1.6 Overview of methods for identifying transient interactors of membrane proteins ....... 28 1.6.1 Blue Native PAGE ................................................................................................ 29 ix  1.6.2 Affinity Purification Mass Spectrometry (AP\/MS) .............................................. 31 1.6.3 The Peptidisc Library ............................................................................................ 35 1.6.4 BioID..................................................................................................................... 38 1.6.5 Native Mass Spectrometry (nMS) ........................................................................ 40 1.7 Overview of Objectives ................................................................................................ 42 Chapter 2: Investigating the stability of the SecA-SecYEG complex during protein translocation across the bacterial membrane ............................................................................45 2.1 Introduction ................................................................................................................... 45 2.2 Materials and Methods .................................................................................................. 48 2.2.1 Protein Production and Purification ...................................................................... 48 2.2.2 Fluorescent Labeling ............................................................................................. 51 2.2.3 SecY-SecA cross-linking ...................................................................................... 52 2.2.4 Proteoliposome reconstitutions ............................................................................. 53 2.2.5 Protein translocation assays .................................................................................. 54 2.2.6 Analysis of Translocation Data ............................................................................. 54 2.3 Results ........................................................................................................................... 57 2.3.1 Effect of a SecA mutation which stabilizes SecA-SecYEG interactions ............. 57 2.3.2 PrlD is able to out-compete wild-type SecA at SecYEG ...................................... 59 2.3.3 Covalent cross-linking of SecY-SecA mimics the effect of the PrlD mutation ... 62 2.3.4 The importance of SecA-SecY processivity varies between preprotein substrates 65 2.3.5 Influence of substrate leader peptide and mature domain on SecA ...................... 67 2.3.6 Influence of substrate length on SecA processivity .............................................. 71 x  2.4 Discussion ..................................................................................................................... 75 Chapter 3: Utilizing the peptidisc membrane mimetic to identify novel transient interactors of the E. coli SecY complex .........................................................................................................78 3.1 Introduction ................................................................................................................... 78 3.2 Materials and Methods .................................................................................................. 83 3.2.1 Reagents ................................................................................................................ 83 3.2.2 Plasmids ................................................................................................................ 83 3.2.3 Expression of target proteins in SILAC labeling conditions ................................ 84 3.2.4 Digestion of protein samples ................................................................................ 85 3.2.5 Liquid chromatography and mass spectrometry analysis ..................................... 86 3.2.6 Validation of the HTL-YfgM-PpiD (HMD) interaction ....................................... 88 3.3 Results ........................................................................................................................... 89 3.3.1 Experimental validation of binary interactions by affinity purification mass spectrometry (AP\/MS) .......................................................................................................... 89 3.3.2 Identification of Bam complex interactors ........................................................... 92 3.3.3 Identification of YidC interactors ......................................................................... 95 3.3.4 Identification of SecDF interactors ....................................................................... 97 3.3.5 Experimental validation of the HMD complex ..................................................... 99 3.4 Discussion ................................................................................................................... 102 Chapter 4: His-tagged Peptidiscs Enable Affinity Purification of the Membrane Proteome for Downstream Mass Spectrometry Analysis ........................................................................105 4.1 Introduction ................................................................................................................. 105 4.2 Materials and Methods ................................................................................................ 110 xi  4.2.1 Reagents and Plasmids ........................................................................................ 110 4.2.2 Preparation of native E. coli membranes ............................................................ 110 4.2.3 Expression and purification of MsbA ................................................................. 111 4.2.4 On-gradient reconstitution of MsbA ................................................................... 112 4.2.5 Depletion of SecDFyajC ..................................................................................... 112 4.2.6 Preparation of peptidisc libraries ........................................................................ 113 4.2.7 Protein digestion and LC-MS\/MS analysis......................................................... 114 4.2.8 Liquid chromatography and mass spectrometry analysis ................................... 115 4.2.9 Analysis of Mass Spectrometry Data .................................................................. 116 4.3 Results ......................................................................................................................... 117 4.3.1 Protein reconstitution with His-tagged peptidisc peptides .................................. 117 4.3.2 Proteomic analysis of the affinity purified peptidisc library .............................. 120 4.3.3 Effect of SecDFyajC depletion on the E. coli membrane proteome ................... 126 4.4 Discussion ................................................................................................................... 132 Chapter 5: Conclusions and Future Directions ......................................................................135 Bibliography ...............................................................................................................................144  xii  List of Tables  Table 2.1 List of constructs employed in this study ................................................................. 56 Table 4.1 Depletion of soluble proteins following library purification over Ni-NTA ......... 124 Table 4.2 Enrichment of membrane proteins following library purification over Ni-NTA..................................................................................................................................................... 125 Table 4.3 Comparison of protein intensities between the DF-\/DF+ starting and purified libraries. ..................................................................................................................................... 131  xiii  List of Figures  Figure 1.1 Schematic of co-translational protein translocation. ............................................... 3 Figure 1.2 Schematic of post-translational protein translocation. ........................................... 4 Figure 1.3 Crystal structure of the SecYEG complex (PDB code: 1RH5). .............................. 6 Figure 1.4 Formation of \"folding intermediate\" states prior to translocation. ...................... 8 Figure 1.5 A schematic of substrate targeting to the Sec translocon. .................................... 10 Figure 1.6 Structure of SecA bound to the SecYEG complex (PDB code: 3DIN) ................ 12 Figure 1.7 Mechanisms of SecA-mediated protein translocation ........................................... 16 Figure 1.8 Crystal structures show conformational changes in SecDF ................................. 20 Figure 1.9 A working model of the PMF-driven translocation enhancement by SecDF ..... 22 Figure 1.10 Crystal structure of YidC ...................................................................................... 23 Figure 1.11 Experimentally determined topologies of YfgM and PpiD ................................. 26 Figure 1.12 Native PAGE analysis of the E. coli membrane proteome ................................. 30 Figure 1.13 Schematic of the classical detergent-based AP\/MS workflow ............................ 32 Figure 1.14 Incorporation of SILAC labeling into the AP\/MS workflow ............................. 34 Figure 1.15 Schematic of the peptidisc library workflow ....................................................... 37 Figure 1.16 Schematic of the BioID experimental workflow .................................................. 39 Figure 2.1 Translocation activity of PrlD23 with the substrate PhoA ................................... 59 Figure 2.2 Competition assay(s) between PrlD SecA and wild-type SecA ............................. 62 Figure 2.3 Translocation activity of a covalently linked SecA-SecYEG complex ................. 64 Figure 2.4 Translocation activity of PrlD23 SecA with the substrate proOmpA ................. 66 Figure 2.5 Effect of the substrate leader peptide on SecA processivity ................................. 68 Figure 2.6 Effect of substrate mature domain on SecA processivity ..................................... 70 Figure 2.7 Effect of substrate length on SecA processivity ..................................................... 72 Figure 2.8A proPhoA truncation mutant is no longer dependent on SecA processivity ...... 74 Figure 3.1 Preservation of labile interactors of the Sec translocon in peptidisc ................... 80 Figure 3.2 The peptidisc AP\/MS workflow ............................................................................... 90 Figure 3.3 Identification of SecYEG interactors by AP\/MS. .................................................. 91 Figure 3.4 Identification of BamA interactors by AP\/MS. ...................................................... 94 Figure 3.5 Identification of YidC interactors by AP\/MS. ....................................................... 96 Figure 3.6 Identification of SecDF interactors by AP\/MS. ..................................................... 98 Figure 3.7 Experimental validation of the HTL-YfgM-PpiD (HMD) complex ................... 101 Figure 4.1 Overview of the functionalized peptidisc workflow. ........................................... 109 Figure 4.2 Reconstitution efficiency of MsbA with the functionalized peptidisc scaffold. . 119 Figure 4.3 The his-tagged peptidisc scaffold enables purification of the E. coli membrane proteome. ................................................................................................................................... 122 Figure 4.4 Global changes in peptide intensity following library purification. .................. 123 Figure 4.5 Proteomic analysis of a SecDFyajC depletion strain ........................................... 129 Figure 4.6 Global changes in peptide intensities after library purification. ........................ 130  xiv  List of Abbreviations and Terms  AP\/MS: Affinity purification followed by mass spectrometry analysis ATP: Adenosine triphosphate BL21: E. coli strain suitable for protein over-expression CV: Column volumes DDM: n-Dodecyl \u03b2-D-maltoside, a non-ionic detergent DTT: Dithiothreitol, a reducing agent E. coli: Escherichia coli EM: Electron microscopy HTL: The bacterial holo-translocon. Contains one copy each of SecYEG, SecDF and YidC. HMD: An expanded variant of the HTL complex. Contains one copy each of SecYEG, SecDF, YidC, YfgM and PpiD. IPTG: Isopropyl-1-thio-\u03b2-D-galactopyranoside IMV: Inverted (inside out) inner membrane vesicle KM9: An E. coli strain in which the F1-F0 ATP synthase genes are knocked out LC-MS\/MS: Liquid Chromatography-Tandem Mass Spectrometry nMS: Native Mass Spectrometry Ni-NTA: A nickel-charged affinity resin used to purify recombinant proteins containing a poly-histidine (His) tag. PAGE: poly-acrylamide gel electrophoresis  xv  Peptidisc: A newly developed membrane mimetic. Membrane proteins in peptidiscs are encapsulated by multiple copies of a 37 amino acid synthetic polypeptide termed the \"peptidisc peptide\". The peptide scaffold adopts to the conformation and shape of the reconstituted membrane protein target. PMF: an electrochemical gradient across the inner membrane of E. coli. The periplasmic side of the membrane is acidic relative to the cytoplasmic side. PpiD: Peptidyl-prolyl cis-trans isomerase D. A putative periplasmic chaperone anchored to the E. coli inner membrane by an N-terminal transmembrane segment. A known interactor of the SecYEG complex (see below). SecDF:  A heterodimeric complex embedded in the E. coli inner membrane which interacts with the SecYEG complex (see below). Forms a proton channel across the membrane. Harnesses the energy from proton movement to accelerate the rate of protein translocation through SecYEG. SecYEG: A heterotrimeric membrane-embedded complex composed of one copy each of SecY, SecE, and SecG. Forms a protein-conducting channel across the inner membrane of E. coli. SDS: sodium dodecyl sulfate SMALPs: styrene maleic acid lipoparticle, polymer lipid particle formed by solubilization of a lipid bilayer by addition of excess styrene maleic acid polymer. These particles contain polymer, lipids and membrane protein(s). SILAC: Stable Isotope Labeling by Amino Acids in Cell Culture. TEMED: N,N,N\u2032,N\u2032-Tetramethylethylenediamine TSG: Tris-Salt-Glycerol buffer. Typically contains 50 mM Tris (pH 8), 50 mM NaCl, 10% Glycerol, unless otherwise stated.  xvi  YfgM: A putative periplasmic chaperone anchored to the E. coli inner membrane by an N-terminal transmembrane segment. A known interactor of PpiD and the SecYEG complex. YibN: An E. coli inner membrane protein of unknown function. YidC: A membrane insertase. Resides in the inner bacterial membrane and inserts nascent membrane proteins into the lipid bilayer. Functions either independently or together with SecYEG.   xvii  Acknowledgements I thank my supervisor Dr. Franck Duong for his mentorship and guidance over the last six and a half years, and for allowing me to work on such an exciting and fascinating research project. The training I've received here in the fundamentals of membrane protein biochemistry and scientific writing will be invaluable as I progress in my scientific career. I also appreciate Franck's valuable advice on what steps to take (and which ones not to take) as I've mapped out my future career path. I also want to thank my supervisory committee - Drs. Robert Molday and John Smit - for valuable suggestions and input on my research as I've progressed through my PhD.   In my time in the Duong lab, I've been privileged to work with many exceptional colleagues. I especially want to thank Irvinder Wason, James Saville, Harveer Dhupar, Zhiyu (Katherine) Zhao, Kate Radford and Michael Carlson for stimulating discussions on my research, and for creating a stimulating and enjoyable working environment.   I thank Drs. Filip van Petegem, Leonard Foster, Sheila Teves, and Huan Bao for valuable discussions and career advice - particularly towards the end of my PhD. I thank Doris Metcalf for valuable assistance with administrative matters during my degree.  I thank my parents for inspiring me to pursue excellence in all that I do, and for encouraging me to pursue academic research. Lastly, I thank my spouse Renee for all of her emotional support and encouragement during this degree.  xviii  Dedication  To my family, who made all of this possible.     1  Chapter 1: Introduction  1.1 Overview of Sec-mediated protein translocation  Protein transport across membranes - or \"protein translocation\" - is a highly conserved process that is fundamental to life. This process is largely mediated by the general Secretory - or \"Sec\" pathway (Park and Rapoport 2012, Rapoport, Li, and Park 2017, Komarudin and Driessen 2019, Tsirigotaki et al. 2017). The Sec pathway has been extensively studied with a variety of genetic, proteomic and biochemical techniques using the gram-negative bacteria Escherichia coli as a model system. The core of the bacterial Sec translocon is the membrane-embedded SecYEG complex, which forms a protein-conducting channel across the inner bacterial membrane and is essential for the translocation of proteins either into or across the cell membrane (Tsirigotaki et al. 2017, Komarudin and Driessen 2019). The core protein-conducting channel is formed by the SecY subunit. The SecE and SecG subunits stabilize the SecY subunit and help to recruit multiple different cytosolic interactors which bind to the Sec translocon - most notably the ribosome and the motor ATPase SecA.   The Sec translocon forms a passive conduit for protein translocation. The energy to drive translocation is provided by cytosolic interacting partners (Crane and Randall 2017, Park and Rapoport 2012, Rapoport, Li, and Park 2017, Tsirigotaki et al. 2017). At the beginning of the translocation reaction, proteins are directed towards the Sec channel for export by a cleavable hydrophobic N-terminal signal peptide. Translocation proceeds by one of two possible     2  mechanisms - either co-translationally or post-translationally. In bacteria, the co-translational pathway is typically used by nascent inner membrane proteins while the post-translational pathway is followed by most secretory proteins and by nascent outer membrane proteins (Collinson, Corey, and Allen 2015, Rapoport, Li, and Park 2017).   During co-translational translocation, translating ribosomes bearing a nascent polypeptide chain are targeted to the Sec translocon by the signal recognition particle (SRP) and the SRP receptor FtsY (Figure 1.1) (Park and Rapoport 2012, Rapoport, Li, and Park 2017, Tsirigotaki et al. 2017). The nascent membrane protein is transferred from the ribosome into the Sec channel as it is being synthesized. Hydrophobic transmembrane helices then diffuse out of the Sec translocon into the interior of the lipid bilayer (Park and Rapoport 2012, Rapoport, Li, and Park 2017, Tsirigotaki et al. 2017).                3    Figure 1.1 Schematic of co-translational protein translocation.   A nascent polypeptide (hydrophobic transmembrane segments in red) is recognized by SRP and is targeted to the Sec translocon by SRP and the SRP receptor. The nascent membrane protein is inserted into the lipid bilayer via the Sec translocon as its synthesis is taking place. The newly synthesized membrane protein is then released into the lipid bilayer. Adapted with permission from (Pohlschr\u00f6der et al. 2005)    ANRV253-MI59-05 ARI 4 August 2005 11:9unfolding the protein and maintaining itin an unfolded state for Sec secretion mayplace an excessive demand on intracellularresources, such as chaperones and nucleosidetriphosphates. Such rapid folding kinetics oranomalous features of the protein\u2019s primarystructure may render these substrates in-compatible with Sec transport. For example,haloarchaea, including the above-mentionedH. volcanii, are organisms that maintain a highinternal salt concentration in order to balancethe high environmental salt levels. Such highsalt may significantly alter folding kineticsor require proteins to fold in the chaperone-rich cytoplasm (86). Hence, exporting themajority of secretory proteins via the Tatpathway may have been an evolutionaryadaptation to ensure proper protein foldingin near-saturating salt.IMPLICATIONS FOR THEEVOLUTION OF PROTEINTRANSLOCATIONWith the existence of self-replicating molec-ular species came a major evolutionary ad-vancement: the ordered segregation of suchmolecules from the surrounding milieuby virtue of lipids. In contrast to otherbiomolecules, there are currently no convinc-ing scenarios for an abiotic origin of long-chain fatty acids or isoprenes that are essen-tial components of rigid lipid membranes (17,101). As there is no evidence for the ability ofFigure 3A theoretical timeline of protein translocation evolution. The earliest mechanism of proteintranslocation likely involved the spontaneous insertion of membrane proteins. It is possible that thisprocess was almost completely replaced by dedicated protein translocation pathways, such as a YidC-likepathway or the ancient Sec pathway. Of these two mechanisms, it is unclear which evolved first. Theadvent of cytoplasmic factors that facilitated membrane targeting of extracytoplasmic proteins greatlyincreased the efficiency of membrane protein insertion and later protein translocation acrosshydrophobic membranes. Of the pathways discussed in this review, the Tat pathway is likely the mostrecently developed protein translocation pathway.102 Pohlschro\u00a8der et al.SRP receptor    4  During post-translational translocation, nascent secretory proteins are fully synthesized by ribosomes within the cytosol before being delivered to the motor ATPase SecA, which subsequently binds to SecYEG to drive protein export through the Sec channel via a succession of ATP-dependent conformational changes (Figure 1.2) (Rapoport, Li, and Park 2017, Chatzi et al. 2014, Tsirigotaki et al. 2017).    Figure 1.2 Schematic of post-translational protein translocation. The preprotein substrate (export signal sequence in red) is recognized by chaperones such as SecB, which direct the substrate towards the cytosolic motor ATPase SecA. SecA subsequently binds to the membrane-embedded SecYEG complex (Step 1) and drives the preprotein across the membrane via a series of ATP-dependent conformational changes (Step 2). Important structural features of SecA are highlighted in colours and will be described below. After the translocation process is complete, the substrate is released into the periplasm, and SecA dissociates from the SecY channel (Step 3). Reproduced with permission from (Park and Rapoport 2012).  BB41CH02-Rapoport ARI 11 April 2012 8:14ExtracellularspaceCytosolPPXDTwo-helixfingerNBD1\/2ClampSecY channelSecY channelTranslocatingpolypeptide(model)Figure 5Structure of the Thermotoga maritima SecA-SecY complex. A hypothetical translocating polypeptide chain isshown in blue. The clamp formed by rotation of the PPXD (red ) positions the polypeptide over the SecYpore. The two-helix finger ( green) contacts the polypeptide with its tip. The nucleotide bound between theNBDs is shown in ball presentation. Abbreviations: PPXD, polypeptide-cross-linking domain;NBD, nucleotide-binding domain.PeriplasmCytosolATP ADP + PiSecBSecASecYPreproteinTwo-helixfingerPPXDSignal sequence+ ATPClamp1 32Figure 6Model of SecA-mediated posttranslational translocation in bacteria. The scheme shows the postulated stepsin the posttranslational translocation of a secretory protein. Step 1: SecA binds to a completed polypeptidechain and inserts it into the SecY channel. The cytosolic chaperone SecB is released during this process.Step 2: During repeated ATP hydrolysis cycles, movements of the two-helix finger push the polypeptide intothe channel. The clamp might hold the polypeptide chain while the two-helix finger resets to grab the nextsegment of the substrate. Step 3: After translocation is terminated, SecA is released from SecY. Abbreviation:PPXD, polypeptide-cross-linking domain.30 Park \u00b7 RapoportAnnu. Rev. Biophys. 2012.41:21-40. Downloaded from www.annualreviews.org Access provided by University of British Columbia on 01\/03\/15. For personal use only.    5  1.2 The structure of the SecYEG complex  Several high-resolution structures - determined by both x-ray crystallography and by electron cryo-microscopy (Cryo-EM) - have revealed the architecture of SecYEG (Rapoport, Li, and Park 2017, Li et al. 2016, Park and Rapoport 2012, Collinson, Corey, and Allen 2015, Ma et al. 2019, Tanaka et al. 2015, Crane and Randall 2017, Tsirigotaki et al. 2017). The channel consists of three membrane-bound subunits: the SecY subunit forms the channel pore and consists of ten symmetrically arranged transmembrane (TM) alpha-helices (Figure 1.3). Helices 1-5 form the N-terminal half of SecY, while helices 6-10 form the C-terminal half. The SecY channel narrows in the middle of the lipid bilayer and is gated by a pore-ring motif which consists of 6 conserved Isoleucine residues (Figure 1.3). Previous studies have found that these residues are important for preventing ion leakage through the channel pore (Dalal and Duong 2009, Dalal, Bao, and Duong 2010, Park and Rapoport 2011). The pore ring is further supplemented by a flexible plug helix, which lies on the periplasmic side of the pore ring (Figure 1.3). The plug helix also plays a role in preventing ion leakage through the Sec channel (Dalal and Duong 2009, Park and Rapoport 2011, Tanaka et al. 2015).  Previous work from our laboratory has shown that the SecY plug domain also influences rates of protein translocation (Tam et al. 2005, Maillard et al. 2007). A further important structural motif is the lateral gate, an opening between TMs 2 and 7 of SecY that faces the hydrophobic interior of the lipid bilayer (Figure 1.3). The lateral gate serves as an exit for TMs of translocating polypeptides to diffuse into the lipid bilayer (Tanaka et al. 2015, Park and Rapoport 2012, Rapoport, Li, and Park 2017).        6  The SecE subunit contains two transmembrane helices which wrap around SecY, providing vital stability to the channel pore (Figure 1.3) (Collinson, Corey, and Allen 2015, Crane and Randall 2017, Park and Rapoport 2012, Rapoport, Li, and Park 2017).  The SecG subunit has two transmembrane helices which are more peripherally associated with SecY (Rapoport, Li, and Park 2017, Park and Rapoport 2012, Crane and Randall 2017, Tanaka et al. 2015).    Figure 1.3 Crystal structure of the SecYEG complex (PDB code: 1RH5).  (A) The structure of SecYEG viewed from the cytosol. The N- and C-terminal halves of the SecY subunit are depicted in blue and red, respectively. The six residues comprising the pore ring are shown in green and the plug helix is shown in yellow. The position of the lateral gate (between TMs 2 and 7) is indicated. The SecE subunit is shown in beige and the SecG subunit is shown in purple. (B) A cutaway side view of a space-filling model of the SecYEG complex viewed in the membrane. Colour coding as in (A). Adapted with permission from (Rapoport, Li, and Park 2017).    CB33CH01-Rapoport ARI 11 May 2017 13:46a b\u03b3-Subunit (SecE)\u03b2-SubunitTM8TM7TM2TM3Pore ringPlug90\u02daCytosolMembranePore ringPlugLateralgateFigure 1Crystal structure of the idle SecY channel from Methanocaldococcus jannaschii (PDB code 1RH5). (a) Viewfrom the cytosol. The N- and C-terminal halves of the \u03b1-subunit (SecY) are shown in blue and red,respectively, the \u03b2-subunit in purple, and the \u03b3-subunit (SecE) in beige. The plug domain is in yellow, andthe pore ring residues are shown as green sticks and balls. Transmembrane (TM) segments forming thelateral gate are labeled. (b) Cutaway side view of a space-filling model of the channel in the membrane.\u03b3-subunits (Sec61\u03b2 and Sec61\u03b3 in eukaryotes and SecG and SecE in bacteria). Much of thecurrent mechanistic understanding of protein translocation originates from the crystal structureof an archaeal SecY complex (from Methanocaldococcus jannaschii) (van den Berg et al. 2004).The structure, which corresponds to the idle (closed) state, showed that the \u03b1-subunit (SecY)is divided into N- and C-terminal halves, TM segments 1\u20135 and 6\u201310, respectively, which arepseudosymmetrical and surround a central pore (Figure 1a). The two halves are linked by a loopbetween TM5 and TM6 on the extracellular side. The \u03b3-subunit contains an amphipathic helixthat lies flat on the cytosolic surface and a TM segment that diagonally crosses the membrane andkeeps the two halves of the \u03b1-subunits together. The \u03b2-subunit makes only few contacts with the\u03b1-subunit, which may explain why it is dispensable for the function of the channel. Viewed fromthe side, the channel pore has an hourglass shape with a constriction in the center of themembrane(Figure 1b). The cytosolic cavity is empty, whereas the extracellular cavity is occupied by a domaintermed the plug. At the constriction of the pore is a ring of six aliphatic amino acids that projecttheir hydrophobic side chains radially inward. In Escherichia coli, all six pore ring residues areisoleucines. The channel has a lateral gate that is bordered by segments of TM2 and TM3 on oneside of the interface and by segments of TM7 and TM8 on the other side of the interface.Crystal structures showed that complexes from Thermotoga maritima, Aquifex aeolicus, Thermusthermophilus, Pyrococcus furiosus, and Geobacillus thermodenitrificans have the same architecture asthe M. jannaschii complex (Egea & Stroud 2010, Li et al. 2016, Tanaka et al. 2015, Tsukazakiet al. 2008, Zimmer et al. 2008). Cryo-EM structures of ribosome-bound SecY\/Sec61 complexesconfirm that the channel architecture is universally conserved up to mammals ( Jomaa et al. 2016,Voorhees & Hegde 2016a, Voorhees et al. 2014). One difference is that the bacterial \u03b2-subunits(SecG) have an additional TM segment that precedes the one that is common to all SecY\/Sec61complexes (Zimmer et al. 2008).The only truly idle channel structure is that of the M. jannaschii complex (van den Berg et al.2004). In this case, the channel is closed both across the membrane and laterally toward the lipidwww.annualreviews.org \u2022 Structural and Mechanistic Insights into Protein Translocation 1.3Changes may still occur before final publication online and in printAnnu. Rev. Cell Dev. Biol. 2017.33. Downloaded from www.annualreviews.org Access provided by University of British Columbia on 07\/11\/17. For personal use only.             A B    SecESecGSecY    7  1.3 Properties of pre-protein substrates  How do nascent secretory proteins remain unfolded yet soluble before being directed towards the translocon and then driven across the membrane? Recent work by the Economou group has shown that - before being targeted to the Sec machinery - preprotein substrates adopt loosely folded states in the cytosol, rather than remaining fully unfolded. These loosely folded states are termed \"folding intermediates\", and are surprisingly stable over time (Figure 1.4) (Tsirigotaki et al. 2018, Sardis et al. 2017, Chatzi et al. 2017). Furthermore, these \"folding intermediates\" are vital for maintaining preprotein substrates in a non-aggregated, translocation competent state (Tsirigotaki et al. 2018).       8   Figure 1.4 Formation of \"folding intermediate\" states prior to translocation.  After emerging from the ribosome, preprotein substrates adopt loosely folded \"folding intermediate\" states, rather than remaining fully unfolded. These \"folding intermediates\" are critical to maintaining the substrate in a non-aggregated, translocation-competent state. These loosely folded preproteins can be recognized by chaperones and delivered to the translocon (yellow cube). Alternatively, the loosely folded substrate can be targeted to the translocon independently of chaperones. Reproduced with permission from (Tsirigotaki et al. 2018).     aggregation-promotion is due to an inherent signal peptideproperty.To determine how signal peptides promotemature domain ag-gregation, we used proPpiA as amodel.We diluted proPpiA fromchaotrope to aqueous solution and determined its folding ki-netics by HDX-MS, either directly (total: dotted line; Figure 7A,left) or after an ultra-centrifugation step to remove aggregatingspecies (soluble: solid line; right). proPpiA transited to folding in-termediates with similar kinetics to those seen for the maturedomain PpiA (Figure 5B), and hence the signal peptide had nomajor role in this process. However, in contrast to the foldingintermediates seen with PpiA (Figure 5B, right), part of theunfolded ensemble in proPpiA was aggregation prone (we referto this as \u2018\u2018Ioff\u2019\u2019) and was selectively sedimented and removedby centrifugation (Figure 7A, compare dotted line, left with solidline, right). We refer to the population remaining after ultra-centri-fugation as being on pathway for folding or \u2018\u2018Ion\u2019\u2019 as it proceededquantitatively with acquiring a folded state after 15 min of incu-bation (Figure 7A, right). This population displayed broader dis-tributions suggestive of more complex folding intermediateensembles.Apparently, by preventing rapid Ion to F transitions, signalpeptides stabilize non-folded intermediate ensembles and thisallows shifting of the equilibrium to undesired states, i.e., Ion con-version to hydrophobic residue exposing Ioff intermediates thatcan go off the folding pathway toward aggregation (Figure 7B).Folding intermediates of various proteins may differ in their ag-gregation propensities and abundance.DISCUSSIONPost-translationally secreted proteins must remain soluble andtemporarily unfolded during cytoplasmic transit, targeting, andsecretion before folding resumes at the trans side of the mem-brane. Under optimal lab culture conditions in rich media in vivothis process can be completed in minutes (e.g., proMBP,proOmpA) (Ryan and Bassford, 1985; Tanji et al., 1991). Typicalfolders, e.g., cytoplasmic proteins, acquire native states in mi-croseconds-seconds (Mayor et al., 2003). Post-translational pre-protein targeting is a slower process.We now show by in vitro dissection, that fast folding is not awidespread characteristic of the secretome. Instead, mostsecretory proteins maintain loose 3D states for long periods oftime (Figures 3C and 3D). These kinetically trapped states ofsecretory proteins are \u2018\u2018on pathway\u2019\u2019 for targeting (Figures 1Eand 2E) and folding (PpiA, Figure 5B; PhoA, Figure S1G). This un-usual protein behavior is an inherent mature domain property(Figures 3C\u20133F). Several primary sequence features, such asABCFigure 7. Signal Peptide-Induce Stabilization of Intermediates In-curs an A gregatio \u2018\u2018Penalty\u2019\u2019(A) Folding kinetics (time; left) of the aggregation-prone proPpiA (as in Fig-ure 5B). The total protein (dotted line, left; no-centrifugation prior to isotopelabeling) is compared with the fraction remaining soluble (solid line, right; su-pernatant after high-speed centrifugation; experimental conditions permittedcentrifugation only after 1 min); n = 2. U, unfolded; F, folded; Ion\/Ioff, on\/off-pathway folding intermediates. Charge state: 24+. Intensities for the solubleand total populations have been normalized to the main peak of their individualMS spectra, and are therefore not directly comparable for quantification ofpopulations between the two sets.(B) Proposed model for the signal peptide effect on the folding kinetics andaggregation of mature domain folders. Mature domains collapse into anensemble of intermediates (Ion) that fold to native states (F). When present, thesignal peptide delays Ion to F transition and hence favors Ion to Ioff equilibria(i.e., aggregation-prone transitions).(C) Model of stochastic preprotein targeting to the Sec translocase (yellow; seeDiscussion for details). Green, signal peptide; orange, mature domain. Route I,chaperone-unassisted targeting-due to the inherentmature domain propertiesand\/or signal peptides; route II, chaperone-assisted targeting.Structure 26, 695\u2013707, May 1, 2018 703          9  The Economou group also identified a surprising role of the preprotein mature domain in targeting putative substrates to the Sec machinery. While the role of the substrate export signal sequence - or \"leader peptide\" has been thoroughly characterized, the role of the substrate mature domain remained largely unexplored. By examining the sequence of a well-characterized model secretory protein, Economou and colleagues identified a number of short, mildly hydrophobic sequences within the substrate mature domain. These short sequences were termed \"Mature Domain Targeting Signals\", or MTSs  (Chatzi et al. 2017, Tsirigotaki et al. 2018). They further showed that these sequences are essential for efficient protein translocation, and that they interact directly with the motor ATPase SecA (Figure 1.5).  A bioinformatic comparison of the sequences of multiple different E. coli secretory proteins reveal that these MTSs are widely prevalent across the E. coli secretome and represent a previously unexplored role of the preprotein mature domain in targeting substrates to the Sec translocon (Chatzi et al. 2017).       10   Figure 1.5 A schematic of substrate targeting to the Sec translocon.  The preprotein substrate's signal peptide is shown in green; the mature domain targeting signal (MTS) is in orange. For simplicity, only one MTS is shown here. The binding sites on SecA for the signal peptide and the MTS are indicated. Reproduced with permission from (Sardis et al. 2017).  1.4 The structure and mechanism of the SecA ATPase  The structure of the SecA ATPase - both on its own and in complex with SecYEG - has been thoroughly examined in the literature. Cytosolic SecA is known to self-associate into dimers and has been crystallized in multiple different dimeric conformations (Tsirigotaki et al. 2017, Gouridis et al. 2013). This lack of a consistent dimeric interface suggests that the dimeric conformation of cytosolic SecA is somewhat labile (Gouridis et al. 2013, Tsirigotaki et al. 2017, Chatzi et al. 2014).   RESULTSPreproteins Bind Synergistically on SecA by BridgingTwo Independent Binding SitesTo dissect how preproteins bind and engage the translocase wedetermined the affinities (Kd) of different proPhoA moieties (Fig-ure 1B). The bivalent ligand proPhoA binds to the SecYEG\/Atranslocase (embedded in inverted inner membrane vesicles[IMVs]) with a higher affinity (row 1) and a gain in negative DG(right), compared with either one of the component parts, signalpeptide (row 2) and mature domain (row 3).Multivalent ligands can display either allosteric or configura-tional cooperativity; negative or positive (Whitty, 2008). If a signalpeptide is added in trans, in excess (20 mM) it binds and reducesthe activation energy of SecA (Gouridis et al., 2009) but does notenhance mature domain binding (row 4). Likewise, excess ofmature domain does not improve signal peptide affinity (datanot shown). Therefore, the two sites are distinct and independentFigure 1. Cooperative Binding of proPhoA on the Translocase(A) Schematic representation of the Sec translocase. Signal peptide (green), mature domain targeting signals (MTS; orange) and their binding sites on SecA areindicated.(B) Equilibrium dissociation constants (Kd) and free energy of association (DG; right) of the indicated proPhoA and mutant derivatives (left) for the wild-type (WT;rows 1\u20139 and 12\u201319), SecYEG-SecA(IL) (Gelis et al., 2007) (IL; row 10) and SecYEG-SecA(LCt) (Chatzi et al., 2017) (LCt; row 11) translocase. SP, signal peptide(green); MD, mature domain (gray); cyan asterisk, site of the 8 residue M1mutation; only MTS 1 and 2 are shown (orange); linker, early mature domain region thatlinks signal peptide and mature domain. NM, non-measurable; NA, non-applicable. n = 3\u20139; values represent mean \u00b1 SEM.(C\u2013E) Synergy factor (C), translocase triggering (D), and in vitro preprotein translocation (E) of the indicated proPhoA derivatives. (C) The synergy factor (Fsyn = Kdmature domain\/Kd preprotein; cyan asterisk, site of the 8 residue M1 mutation. Kd values in (B) describe the occupancy of the SecA receptor when both signalpeptides and MTS are present. Fsyn > 1: optimal interaction with the receptor; <1: sub-optimal; = 1: non-measurable synergy. NA, non-applicable. Valuesrepresent mean \u00b1 SEM. (D) Activation energy of wild type SecA (Ea) was determined in the presence of wild-type SecYEG-IMVs and the indicated proPhoAderivatives. The Ea of SecA (alone or bound on SecYEG-IMVs Ea = 180 kJ mol!1; indicated by a red arrow on the x axis) lowers significantly upon signal peptidebinding and is indicative of translocase triggering (Gouridis et al., 2009). n = 6\u20138; values represent mean \u00b1 SEM. (E) In vitro translocation of the indicatedpreproteins using wild-type SecA-SecYEG translocase, under the same conditions; values were expressed as the percentage of the translocated proPhoA. n = 3;values represent mean \u00b1 SEM.See also Figure S1; Table S1.Structure 25, 1056\u20131067, July 5, 2017 1057             11   Upon binding to SecYEG, however, the consensus in the literature is that the SecA dimer dissociates, leaving only a SecA monomer bound to the translocon (Alami et al. 2007, Dalal et al. 2012). The structure of SecA in complex with SecYEG has been determined, revealing one protomer of SecA bound to one protomer of SecYEG, suggesting that this is the active form of the translocon (Figure 1.6) (Collinson, Corey, and Allen 2015, Whitehouse et al. 2012, Li et al. 2016, Ma et al. 2019).    Several structures of SecA in complex with SecYEG - both in the presence and absence of a preprotein substrate - have been presented in the literature (Zimmer, Nam, and Rapoport 2008, Li et al. 2016, Ma et al. 2019). These structures have shed valuable insight into the mechanism of the translocation reaction. SecA is a large, multi-domain protein, with two conserved nucleotide binding domains (NBDs) which move relative to each other during the ATP hydrolysis cycle. An additional functionally important motif is the two-helix finger (2-HF) - two helices connected by a loop - which interact directly  with the mouth of the SecY channel as well as with a translocating substrate (Figure 1.6) (Zimmer, Nam, and Rapoport 2008, Whitehouse et al. 2012, Bauer et al. 2014, Bauer and Rapoport 2009).  The translocating substrate is held in position above the channel opening by the SecA \"clamp\", which is formed between the SecA polypeptide crosslinking domain (PPXD) and one of the NBDs (Figure 1.6) (Bauer et al. 2014, Catipovic et al. 2019, Zimmer, Nam, and Rapoport 2008). When a polypeptide substrate inserts into the SecY channel, it promotes widening of the SecY lateral gate as well as displacement of the SecY plug     12  domain. These changes in the conformation of the SecY channel open up a pathway for the translocating polypeptide (Li et al. 2016, Ma et al. 2019, Rapoport, Li, and Park 2017).  Figure 1.6 Structure of SecA bound to the SecYEG complex (PDB code: 3DIN)  The crystal structure reveals a single copy of the SecA ATPase bound onto the SecYEG protein conducting channel. A hypothetical preprotein substrate is shown in blue; the two-helix finger is shown in green, and the preprotein crosslinking domain (PPXD) is shown in red. Reproduced with permission from (Park and Rapoport 2012).    Despite these advances, several aspects of the mechanism and dymanics of the translocation reaction remain controversial (Bauer et al. 2014, Allen et al. 2016, Fessl et al. 2018, Catipovic et al. 2019, Corey et al. 2019). Specifically, the extent of movement of the SecA 2-HF and the BB41CH02-Rapoport ARI 11 April 2012 8:14ExtracellularspaceCytosolPPXDTwo-helixfingerNBD1\/2ClampSecY channelSecY channelTranslocatingpolypeptide(model)Figure 5Structure of the Thermotoga maritima SecA-SecY complex. A hypothetical translocating polypeptide chain isshown in blue. Th clamp or ed by rot ti n of the PPXD (red ) positions the polypeptide over the SecYpore. The two-helix finger ( green) contacts the polypeptide with its tip. The nucleotide bound between theNBDs is shown in ball presentation. Abbreviations: PPXD, polypeptide-cross-linking domain;NBD, nucleotide-binding domain.PeriplasmCytosolATP ADP + PiSecBSecASecYPreproteinTwo-helixfingerPPXDSignal sequence+ ATPClamp1 32Figure 6Model of SecA-mediated posttranslational translocation in bacteria. The scheme shows the postulated stepsin the posttranslational translocation of a secretory protein. Step 1: SecA binds to a completed polypeptidechain and inserts it into the SecY channel. The cytosolic chaperone SecB is released during this process.Step 2: During repeated ATP hydrolysis cycles, movements of the two-helix finger push the polypeptide intothe channel. The clamp might hold the polypeptide chain while the two-helix finger resets to grab the nextsegment of the substrate. Step 3: After translocation is terminated, SecA is released from SecY. Abbreviation:PPXD, polypeptide-cross-linking domain.30 Park \u00b7 RapoportAnnu. Rev. Biophys. 2012.41:21-40. Downloaded from www.annualreviews.org Access provided by University of British Columbia on 01\/03\/15. For personal use only.    13  importance of these movements during the translocation reaction remain unclear. Early studies on the mechanism of SecA-mediated translocation using the model preprotein substrate proOmpA postulated that protein translocation occurs by a power-stroke mechanism (Figure 1.7).  In this model, ATP binding to SecA leads to conformational changes which actively pushes a portion of the translocating preprotein into the SecY channel. This \"pushing\" action is likely mediated by the 2-HF. ATP hydrolysis allows the conformation of SecA to \"reset\" itself before the next power-stroke event (Figure 1.7) (Schiebel et al. 1991, Sato et al. 1997, Uchida, Mori, and Mizushima 1995). These studies postulated that translocation occurs in well-defined steps of 20-30 amino acids, and that the precise size of each discrete step may vary depending on the identity of the substrate undergoing transport (Sato et al. 1997, Uchida, Mori, and Mizushima 1995).  While this purely power-stroked based model of protein translocation is straightforward and compelling, it has been largely refuted in subsequent studies. A recent series of studies from the Rapoport lab report that ATP binding to SecA does indeed trigger a \u201cpower stroke\u201d, resulting in forward movement of a polypeptide segment into the mouth of the SecYEG channel (Figure 1.7) (Bauer et al. 2014, Catipovic et al. 2019). Forward polypeptide movement is mediated by the 2-HF.  The \"power-stroke\" event leads to a large movement of the 2-HF, causing it to push the translocating polypeptide deep into the SecY channel (Bauer et al. 2014, Catipovic et al. 2019). Such large-scale movements of the 2-HF are supported by both crosslinking and by single-molecule FRET observations (Whitehouse et al. 2012, Catipovic et al. 2019). Following ATP hydrolysis, - when ADP is still bound to the enzyme - SecA adopts a different conformation, \"resetting\" the 2-HF and allowing the polypeptide to slide across the SecYEG channel until the     14  next ATP binding event (Figure 1.7) (Bauer et al. 2014, Erlandson et al. 2008). In this model, SecA exists mainly in an ADP-bound state when translocation is taking place (Bauer et al. 2014, Ding, Mukerji, and Oliver 2003). Importantly, this model relies mostly on passive diffusion of the substrate through the SecY channel, rather than on active power-stroke events pushing the substrate forward (Figure 1.7) (Bauer et al. 2014, Catipovic et al. 2019). It is not difficult to envision that translocation occurring by a mixed power-stroke\/diffusion model could be much more efficient than translocation occurring by a purely power-stroke based model.   The mixed power-stroke\/diffusion model implies that mobility of the SecA 2-HF is necessary for efficient translocation to take place. The Collinson research group found, however, that covalently crosslinking the 2-HF near the mouth of the SecY channel does not appear to impede the translocation reaction (Whitehouse et al. 2012). Thus, a second possible model that has gained traction in the literature which does not require a \"power stroke\" (Allen et al. 2016). Recent work from the Collinson research group suggests that SecA rather operates as a \"Brownian ratchet\", allowing substrates to passively diffuse within the SecY channel while the enzyme is in an ADP-state (Figure 1.7). When a steric blockage is reached due to a stretch of bulky or aromatic residue within the substrate, ATP binding to SecA leads to a modest conformational change in the SecA 2-HF, which temporarily widens the SecY channel, thus allowing the blockage to pass (Allen et al. 2016, Fessl et al. 2018). This second model is almost exclusively diffusion-based and does not rely on an active \"power-stroke\" at all (Figure 1.7). In both models, transport largely relies on passive sliding of substrates through the SecY channel without specific sequence recognition. Thus, both models can potentially explain why the translocon is able to handle such a wide variety     15  of proteins, each with a different sequence (Bauer et al. 2014, Allen et al. 2016, Fessl et al. 2018, Cranford-Smith and Huber 2018).   Downloaded from www.asmscience.org byIP:  128.189.231.76On: Wed, 11 Dec 2019 18:44:54ASMscience.org\/MicrobiolSpectrum 7SecA-Mediated Protein Translocation through the SecYEG ChannelADownloaded from www.asmscience.org byIP:  128.189.231.76On: Wed, 11 Dec 2019 18:44:54ASMscience.org\/MicrobiolSpectrum 7SecA-Mediated Protein Translocation through the SecYEG ChannelBDownloaded from www.asmscience.org byIP:  128.189.231.76On: Wed, 11 Dec 2019 18:44:54ASMscience.org\/MicrobiolSpectrum 7SecA-Mediated Protein Translocation through the SecYEG ChannelC    16  Figure 1.7 Mechanisms of SecA-mediated protein translocation  (A) The Power-Stroke model. After binding onto SecYEG, ATP binding to SecA leads to conformational changes in the two-helix finger which actively \"push\" a portion of the preprotein substrate into the SecYEG channel. After ATP hydrolysis, the finger \"resets\" in preparation for another ATP binding event. (B) The Mixed Power-Stroke\/Diffusion model. As in (A), ATP binding to SecA causes the two-helix finger to push a portion of the preprotein substrate into the SecYEG channel. After ATP hydrolysis, the finger resets, allowing the substrate to diffuse passively forward into the channel. (C) The Brownian Ratchet model. SecA largely allows the substrate to passively diffuse through the SecY channel in this model, without any active, ATP-dependent \"pushing\". Adapted from (Komarudin and Driessen 2019). \u00a9 American Society for Microbiology. Used with permission. No further reproduction or distribution is permitted without the prior written permission of American Society for Microbiology.    Since both models described above largely rely on passive diffusion of the substrate to drive translocation, another key mechanistic question arises: how does the Sec machinery ensure that diffusion of the translocating polypeptide occurs mainly in the forward direction, rather than resulting in non-productive \"back-sliding\" of the substrate? A recent study by Collinson and colleagues has shed light on this important question. They demonstrate that while substrates are largely devoid of secondary structure while they undergo transit through the Sec translocon, the environment within the periplasmic side of the SecY channel favours formation of substrate secondary structure as the substrate emerges out of the channel (Corey et al. 2019). The local environment within the cytosolic side of the SecY channel, by contrast, disfavours secondary structure formation (Corey et al. 2019). Since substrates which possess stable secondary structure cannot efficiently pass through the SecY channel, it appears that formation of substrate secondary structure on the periplasmic side of the SecY channel is an effective way for the system to prevent non-productive back-sliding of a polypeptide substrate (Corey et al. 2019, Ahdash et al. 2019).      17  The recent studies described above have provided a significant advance in terms of our understanding of the mechanism(s) of Sec-mediated protein translocation. Both models are based on compelling experimental evidence, but which is more correct? It is possible that SecA is capable of acting by either a purely diffusional \"Brownian Ratchet\" mechanism, or a mixed power-stroke\/diffusion mechanism. The mechanism used by SecA to drive transport may vary depending on the nature of the preprotein substrate.  1.5 Ancillary subunits of the Sec translocon  The E. coli Sec translocon has a number of well-characterized membrane-bound ancillary subunits. These additional subunits associate transiently with the core translocon and augment the activity of the core translocon during certain steps of the translocation reaction. The best characterized Sec ancillary subunits are the SecDF sub-complex, the membrane insertase YidC and the membrane-bound periplasmic chaperones YfgM and PpiD (Rapoport, Li, and Park 2017, Petriman et al. 2018, Crane and Randall 2017, Maddalo et al. 2011, G\u00f6tzke et al. 2014, Schulze et al. 2014). I will discuss each of these in turn. Our mechanistic understanding of the interactions between SecYEG and its ancillary subunits lags significantly behind our understanding of the interactions between SecYEG and its soluble interactors. This is primarily due to the difficulty of isolating sufficient quantities of stable, homogenous multi-subunit membrane protein complexes that are amenable to biochemical and high-resolution structural analysis. In the sections below, I will describe the current state of knowledge on known ancillary subunits of the Sec translocon.      18  1.5.1 SecDF  A physical interaction between SecDF and SecYEG were first identified in the early 1990s (Duong and Wickner 1997a). This landmark finding substantiated an already compelling body of genetic and biochemical data suggesting that SecDF is directly involved in Sec-dependent protein translocation across the inner bacterial membrane (Duong and Wickner 1997b, Riggs, Derman, and Beckwith 1988, Economou et al. 1995, Pogliano and Beckwith 1994, Rollo and Oliver 1988).   The importance of SecDF for Sec-dependent protein translocation is underscored by the observation that cells lacking SecDF display strong transport defects and have reduced viability (Duong and Wickner 1997b, Kato, Nishiyama, and Tokuda 2003, Pogliano and Beckwith 1994). Furthermore, cells harbouring loss-of-function mutations in either SecD or SecF exhibit heightened expression levels of the motor ATPase SecA (Riggs, Derman, and Beckwith 1988, Rollo and Oliver 1988). The effect of SecDF depletion on cell viability and on translocation activity becomes more pronounced when the SecG subunit of the core translocon is also deleted (Kato, Nishiyama, and Tokuda 2003). Following these initial genetic observations suggesting a SecYEG-SecDF interaction, biochemical studies revealed that SecDF associates transiently with SecYEG and harnesses the proton motive force across the inner membrane to accelerate the rate of protein translocation -  particularly at the later stages of the translocation reaction (Tsukazaki et al. 2011, Duong and Wickner 1997b, a). At these later stages, SecDF may be able to drive translocation of preprotein substrates independently of either ATP or SecA (Tsukazaki et al. 2011, Duong and Wickner 1997b).     19   Crystallographic studies on the SecDF complex have revealed that it contains 12 transmembrane helices. Both SecD and SecF possess large periplasmic domains. The SecD periplasmic domain is termed P1; the SecF periplasmic domain is termed P4 (Figure 1.8) (Tsukazaki et al. 2011, Tsukazaki 2018, Furukawa et al. 2017). By varying crystallographic conditions, the authors were able to capture SecDF in distinct conformational states, suggesting that the complex undergoes large conformational changes during protein translocation (Figure 1.8A) (Furukawa et al. 2018, Furukawa et al. 2017, Tsukazaki et al. 2011, Tsukazaki 2018). These states have been named \"F\"(\"Facing\") and \"I\" (\"Intermediate\"), depending on the position of the large periplasmic P1 domain of SecD. In the \"F\" conformation, the P1 domain is close to the membrane surface; in the \"I\" conformation, the P1 domain is rotated by 120 degrees, and is extended more into the periplasm (Tsukazaki et al. 2011, Tsukazaki 2018, Furukawa et al. 2018).   Electrophysiology experiments reveal that SecDF is a bona fide proton channel, allowing protons to flow into the cytoplasm from the more acidic periplasm (Tsukazaki et al. 2011). This important observation is substantiated by the crystal structure of SecDF, which reveals a proton channel within the transmembrane portion of SecDF (Figure 1.8) (Tsukazaki 2018, Furukawa et al. 2017). Mutation of conserved charged residues within the channel abolish the activity of SecDF, underscoring the importance of this channel for SecDF function (Tsukazaki et al. 2011, Furukawa et al. 2017).        20   Figure 1.8 Crystal structures show conformational changes in SecDF (A) The structures of SecDF are shown in the F (left) and I (right) forms. The TM region of SecD is shown in red; the TM region of SecF is shown in blue. The conformational changes in the P1 domain of SecD are clearly evident between the F form and I form (PDB code: 3AQP). (B) The structure of SecDF in the I form. The proton channel through the transmembrane domain of SecDF is shown in black (PDB code: 5XAP). A hydrophobic groove in the P1 domain which interacts with emerging preprotein substrates is shown in blue. Reproduced with permission from ((Tsukazaki et al. 2011, Furukawa et al. 2017). SecDF variants (described below) did not support the PMF-dependentcompletion of pro-OmpA translocation (Fig. 2c).To gain structural insights into the PMF-dependence of SecDFfunction, we compared its structure with that of AcrB20, an RNDsuperfamily proton\/multi-drug antiporter16 (Supplementary Fig. 5).AcrB forms a homotrimer, whereas SecDF is monomeric. Althoughthe TM segments of SecDF and AcrB share low sequence identity of15% (Supplementary Fig. 6), the structures of their TM regions aresimilar, yielding a root mean square deviation of ,2.7 A\u02da for the Caatoms of the TM helices. By contrast, the structures and functions ofthe periplasmic regions are different between SecDF and AcrB. TheTM region of AcrB is thought to participate in proton transport andcontains several conserved, charged residues important for drugexport activity, such as Asp 407, Asp 408, Lys 940 and Arg 97121(Supplementary Fig. 5e). Asp 407, Thr 978 and Arg 971 in AcrB havestructural counterparts in TtSecDF: Asp 340, Thr 675 and Arg 671respectively (Fig. 3a and Supplementary Fig. 5e). The conservedSecDF residues are clustered at the TM interfaces between SecD andSecF and in the periplasmic base region underneath the head(Supplementary Fig. 7). We also note that Asp 637 is a highly con-served, membrane-embedded charged residue. Complementationtests indicated that the Asp519Asn mutation in EcSecD as well asthe Asp213Asn and Arg247Met mutations in EcSecF (Fig. 3a) com-pletely abolished SecDF activity and conferred some dominant-negative phenotypes (Fig. 3b). Thus, these conserved charged residuesare crucial for SecDF function, consistent with the hypothesis thatSecDF conducts protons through the conserved TM region.The halophilic marine bacteria Vibrio spp. use a sodium ion (Na1)gradient instead of PMF for some cellular processes22. Vibrio algino-lyticus has two sets of secDF genes encoding SecDF-1 and SecDF-2complexes respectively. When VaSecDF-1 fromV. alginolyticus 138-2was expressed in the E. coli secD1 (Cs) mutant, the addition of NaCl,PeriplasmCytoplasmTM7\u201312TM1\u20136P1(head)P1(base)P412345678910 1112HingeSecD region SecF regionaTM7\u201312TM1\u20136P1(head)P1(base)P4b123 456789101112*cd eP4P1(base)P1(head)HingeTM7\u201312TM1\u20136P1(head)P1(base)P4TM7\u201312TM1\u20136P1(head)P1(base) P4PeriplasmPeriplasmCytoplasmCytoplasmf gI formF formFigure 1 | Structures of T. thermophilus SecDF.a, b, The crystal structure of full-length SecDFviewed from the membrane side (a) and theperiplasmic side (b). c, TtSecDF cross-sectioned atthe middle of the TM, viewed from the periplasm.The asterisk indicates the pseudo-symmetrical axis.TMs are numbered. d, Crystal structure of the P1domain. e, NMR structure of the P4 domain aftertwenty superimpositions. The disordered regionsare shown in grey. f, Crystal structure of full-lengthSecDF, F form. g, Crystal structure of full-lengthSecDF, I form. The base subdomain of isolated P1was docked onto that of the F form, as shown in (f).DTTDTT+ +\u2013 \u2013 \u2013 \u2013 \u2013 \u2013 \u2013 \u2013+ + + + + +PMF + \u2013 \u2013 \u2013 \u2013+ + +\u2013 ATPYEGL5935S-OmpA (L59)DFAIMV SecD+ SecD1 SecD+ SecD1InputIntermediateInput+ ATPIntermediateIntermediateIntermediateMaturePrecursora 13 14 15 16 17 18 19 20 21 221 2383124173 4 5 6 7 8 9 10 11 12kDa+IntermediatePeriplasmCytoplasm +Intermediate +Intermediate + \u2013\u2013\u2013\u2013IntermediatePlasmid Vector DF D(D519N)F DF(R247M)MaturePrecursorb 131 2 3 4 5 6 7 8 9 10 11 12cIntermediateFigure 2 | SecDF-dependent translocation completion. a, Identification of aSecDF- and PMF-dependent translocation step. SecDF-deficient (secD1) IMVswere incubated with 35S-labelled pro-OmpA(L59) to generate translocationintermediates. Protein translocation was continued in the presence or absenceof ATP, PMF and dithiothreitol (DTT) before SDS-PAGE and phosphorimaging. Arrowheads indicate translocation intermediates. b, Schematicdepiction of the translocation intermediate of pro-OmpA(L59). A, SecA; DF,SecDF; YEG, SecYEG. c, Completion of pro-OmpA(L59) translocation usingIMVs from the secD1 (Cs) mutant expressing no additional protein (vector), E.coli SecDF or the SecDF derivatives SecD(D519N)F and SecDF(R247M).RESEARCH LETTER2 3 6 | N A T U R E | V O L 4 7 4 | 9 J U N E 2 0 1 1Macmillan Publishers Limited. All rights reserved\u00a92011The 4.0-A\u02da resolution structure of DrSecDF revealed the overallstructure of the I form in which the N-terminal regions exceptfor the P1-head (TM1-6 and P1-base) and the C-terminal regions(TM7-12 and P4) are assembled pseudo-symmetrically, similarto those of the previously reported F form (Tsukazaki et al.,2011), whereas the P1-head is situated just on the periplasmside of the P1-base (Figure 1; Figure S1). However, this 4.0- \u02dastructure did not further elucidate the function of SecDF. There-fore, to improve the stability of the flexible P1 domain in the crys-tals, we introduced two cysteine residues at positions 143 and268, referring to the 4.0-A\u02da structure, to fix the P1-head withhigher crystallographic B-factors onto the P1-base by disulfidebond formation (Figure S1A). The mutant SecDF crystalscompletely formed intramolecular disulfide bonds and diffractedX-rays to over 2.5-A\u02da resolution, enabling us to solve the struc-tures with space groups C2 and P212121 at 2.6- and 2.8-A\u02daresolution, respectively (Table S1; Figure S1). The asymmetricFigure 1. Crystal Structures of SecDF(A) SecDF in I form (MolB). The TM helices arenumbered. A part of PEG, O-(CCO)4, in the P1-head cavity is depicted in blue. Th tunnel in theTM region is shown as a black surface model.(B) Structural comparison of the I form (MolB,colored as in A) with the F form (PDB ID 3AQP,gray).(C) P1-head in MolB. The PEG molecule iscaptured by the cavity. The surface model iscolored according to hydrophobicity (from white[hydrophilic] to red [hydrophobic]).See also Figures S1\u2013S3 and Table S1.unit of the P212121 crystal contains twodistinct molecules designated MolA andMolB, whereas that of the C2 crystal con-tains one molecule, MolC. In Figure 1A,the crystal structure of MolB is shownas a representative I form of SecDF.As the structures in I form did not differsignificantly (Figure S1), we hereaftermainly discuss SecDF in I form using thehigher-resolution MolA-C structures.Each orientation of the P1-head in I formas well as in the previously determinedP1 domain fragment is structurallydistinct (Figure S1), presumably becauseof its intrinsic flexibility even in I form.Superimposit on of the I (MolB) and Fforms (PDB ID 3AQP) revealed anapproximately 100! rigid-body rotationof the P1-head domain (Figure 1B), witha swinging motion of the P1-head on theP1-base allowed by the two connectinghinge loops.Interaction Site of PeriplasmicRegionUnique features of the M lB crystalstructure include an ambiguous electrondensity at the P1-head cavity, potentially from a part of the poly-ethylene glycol (PEG) used as a precipitant, whereas there wasno electron density map (Fo-Fc) in the same region of the MolAand MolC structures (Figure 1C; Figure S3A). The interactionsite in the cavity provides amphipathic surfaces (Figure 1C).The P1-head structure implied that the cavity captures theemerging polypeptide from the SecYEG translocon during pro-tein translocation. We examined this hypothesis using site-directed ultraviolet-cross-linking of pBPA, an ultraviolet-reac-tive amino acid derivative (Mori and Ito, 2006). pBPA oleculeswere introduced at each indicated position in the cavity or theexterior periplasmic surface of the Escherichia coli SecDF com-plex (Ec ecDF) (Figures S3A and S3B). Following irradiationduring protein translocation, 35S-labeled polypeptide cross-linking products were only observed upon introduction ofpBPA at position 407, corresponding to Q253 in the DrSecDFcavity (Figures S3C\u2013S3E), suggesting that the P1-head cavity896 Cell Reports 19, 895\u2013901, May 2, 2017AB    21  Strikingly, the channel is only open when the P1 domain is in the extended \"I\" conformation, suggesting that the channel activity of SecDF may be coupled to movement of the P1 domain (Tsukazaki 2018, Furukawa et al. 2017). There is evidence that movement of the P1 domain is coupled to protein translocation. Mutations which abolish or limit conformational dynamics of the P1 domain decrease rates of protein translocation in vivo (Furukawa et al. 2017). Additionally, the P1 domain contains a hydrophobic cavity which directly contacts a translocating polypeptide as it emerges on the periplasmic side of the Sec translocon (Figure 1.8) (Tsukazaki 2018, Furukawa et al. 2017).   These observations lead to a possible model of how SecDF increases the rate of protein translocation through SecYEG.  First, a nascent polypeptide exiting on the periplasmic side of SecYEG encounters SecDF in the \"F\" conformation. The nascent preprotein binds to the hydrophobic cavity of the P1 domain. Second, the P1 domain changes conformation from the \"F\" state into the extended \"I\" state, pulling the translocating polypeptide further into the periplasm (Figure 1.9). This conformational change exposes the proton channel through SecDF, allowing protons to pass from the periplasm into the cytoplasm through SecDF. By an as-yet-unknown mechanism, proton flow through SecDF causes the P1 domain to release the polypeptide, and reset into the \"F\" conformation (Figure 1.9) (Furukawa et al. 2018, Furukawa et al. 2017, Tsukazaki 2018). This preliminary model is based on biochemical and structural evidence. Clearly, more detailed experiments - perhaps using single-molecule FRET - are needed to monitor the conformational changes of SecDF in real-time. This will allow researchers to assess whether the sequence of events postulated above is correct.      22   Figure 1.9 A working model of the PMF-driven translocation enhancement by SecDF (A) F form, capturing state. (B) I form, holding state. (C) I to F transition and substrate-releasing state. The two essential charged residues of EcSecDF are highlighted. SecDF is coloured as in Fig. 8. SecYEG, grey; SecA, green; preprotein, black line; proton movement, white arrow. Reproduced with permission from (Tsukazaki et al. 2011).  1.5.2 YidC  A second well-characterized SecYEG interactor is the membrane insertase YidC. YidC is conserved throughout evolution. A eukaryotic analogue of YidC - Oxa1 - resides in the mitochondrial membrane and is required for co-translational insertion of certain proteins within the mitochondrial inner membrane. Another YidC analogue - Alb3 - performs a similar function in the chloroplast thylakoid membrane (Hennon et al. 2015). In E. coli, YidC acts in concert with SecYEG during the biogenesis of many inner membrane proteins. YidC is also able to catalyze membrane integration of a subset of integral membrane proteins independently of SecYEG (Komar et al. 2016, Hennon et al. 2015). which presumably exist in different protonation states20. Likewise, pro-tonation of the key charged residues of SecDF could induce the twistingof TM4 and TM10. This would be transmitted to the conserved P1-TM4 linker region (Supplementary Fig. 7) and would trigger the con-formational transition of the P1 head subdomain.Althoug direct evidence f r the PMF-dependent conformationaltransition of SecDFwill await further structural and functional studies,we have shown that SecDF is a component of the Sec machinery thatutilizes the PMF to complete protein translocation after the ATP-dependent function of SecA.M THODS SUMMARYX-ray diffraction analysis of TtSecDF was described previously17. The initialphases were determined by the single-wavelength anomalous dispersion method.The initial model was m nually built, using the structures of the TM regions ofAcrB and the separately-determined P1 and P4 domain structures as references(Supplementary Methods). The model was substantially improved by using thezonal scaling25 and methionine-marking26 methods and was finally refined toRwork5 29.8% and Rfree5 31.9% at 3.3 A\u02da resolution.To monitor the later steps of translocation, 35S-labelled pro-OmpA(L59) andIMVs, prepared as described previously18,27, were incubated to form translocationintermediates which were then incubated under several conditions and treatedwith proteinase K. The translocation status of 35S-pro-OmpA was examined byphosphor imaging after SDS-polyacrylamide gel electrophoresis (SDS\u2013PAGE).For the complementation test, E. coli secD1 (Cs) mutant cells carrying plasmidsencoding HA-tagged EcSecD and EcSecF mutants were spotted onto LB agarplates (pH adjusted to ,7.5) and incubated at 20 uC. The efficiency of proteinexport in vivo was assessed by the 35S-methionine pulse-labelling procedures.To monitor the protein transport activity of Vibrio SecDF-1, E. coli secD1 cellsexpressing VaSecD-1 andVaSecF-1 were pulse-labelled in the presence or absenceof Na1. To detect single channel activity, electric currents were measured withinside-out membrane patches excised from TtSecDF-containing E. coli giantspheroplasts, using an Axopatch 200B amplifier (Axon CNS, Molecular Devices).Received 11 January 2010; accepted 9 March 2011.Published online 11 May 2011.1. van denBerg, B. et al.X-ray structure of a protein-conducting channel.Nature427,36\u201344 (2004).2. Zimmer, J., Nam, Y. & Rapoport, T. A. Structure of a complex of the ATPase SecAand the protein-translocation channel. Nature 455, 936\u2013943 (2008).3. Tsukazaki, T. et al. Conformational transition of Sec machinery inferred frombacterial SecYE structures. Nature 455, 988\u2013991 (2008).4. duPlessis, D. J., Nouwen,N.&Driessen, A. J. TheSec translocase.Biochim. Biophys.Acta 1808, 851\u2013865 (2011).5. Driessen, A. J. & Wickner, W. Proton transfer is rate-limiting for translocation ofprecursor proteins by the Escherichia coli translocase.Proc. Natl Acad. Sci. USA 88,2471\u20132475 (1991).6. Shiozuka, K., Tani, K., Mizushima, S. & Tokuda, H. The proton motive force lowersthe level of ATP required for the in vitro translocation of a secretory protein inEscherichia coli. J. Biol. Chem. 265, 18843\u201318847 (1990).7. Pogliano, J. A. & Beckwith, J. SecD and SecF facilitate protein export in Escherichiacoli. EMBO J. 13, 554\u2013561 (1994).8. Sagara, K., Matsuyama, S. & Mizushima, S. SecF stabilizes SecD and SecY,components of the protein translocation machinery of the Escherichia colicytoplasmic membrane. J. Bacteriol. 176, 4111\u20134116 (1994).9. Hand, N. J., Klein, R., Laskewitz, A. & Pohlschroder, M. Archaeal and bacterial SecDand SecF homologs exhibit striking structural and functional conservation. J.Bacteriol. 188, 1251\u20131259 (2006).10. Matsuyama, S., Fujita, Y. & Mizushima, S. SecD is involved in the release oftranslocated secretory proteins from the cytoplasmic membrane of Escherichiacoli. EMBO J. 12, 265\u2013270 (1993).11. Economou, A., Pogliano, J. A., Beckwith, J., Oliver, D. B. & Wickner, W. SecAmembrane cycling at SecYEG is driven by distinct ATP binding and hydrolysisevents and is regulated by SecD and SecF. Cell 83, 1171\u20131181 (1995).12. Arkowitz, R. A. & Wickner, W. SecD and SecF are required for the protonelectrochemical gradient stimulation of preprotein translocation. EMBO J. 13,954\u2013963 (1994).13. Duong, F. &Wickner,W. The SecDFyajCdomain of preprotein translocase controlspreprotein movement by regulating SecA membrane cycling. EMBO J. 16,4871\u20134879 (1997).14. Nouwen, N., Piwowarek, M., Berrelkamp, G. & Driessen, A. J. The large firstperiplasmic loopofSecDandSecFplays an important role inSecDF functioning. J.Bacteriol. 187, 5857\u20135860 (2005).15. Pogliano, K. J. & Beckwith, J. Genetic and molecular characterization of theEscherichia coli secD operon and its products. J. Bacteriol. 176, 804\u2013814 (1994).16. Tseng, T. T. et al. The RND permease superfamily: an ancient, ubiquitous anddiverse family that includes human disease and development proteins. J. Mol.Microbiol. Biotechnol. 1, 107\u2013125 (1999).17. Tsukazaki, T. et al. Purification, crystallization and preliminary X-ray diffraction ofSecDF, a translocon-associatedmembrane protein, from Thermus thermophilus.Acta Crystallogr. F 62, 376\u2013380 (2006).18. Uchida, K., Mori, H. & Mizushima, S. Stepwise movement of preproteins in theprocess of translocation across the cytoplasmic membrane of Escherichia coli. J.Biol. Chem. 270, 30862\u201330868 (1995).19. Schiebel,E.,Driessen,A.J.,Hartl,F.U.&Wickner,W.DmH1andATPfunctionatdifferentsteps of the catalytic cycle of preprotein translocase.Cell64,927\u2013939 (1991).20. Murakami,S., Nakashima,R., Yamashita, E.,Matsumoto, T. &Yamaguchi, A.Crystalstructures of a multidrug transporter reveal a functionally rotating mechanism.Nature 443, 173\u2013179 (2006).21. Seeger, M. A., von Ballmoos, C., Verrey, F. & Pos, K. M. Crucial role of Asp408 in theproton translocation pathway of multidrug transporter AcrB: evidence from site-directed mutagenesis and carbodiimide labeling. Biochemistry 48, 5801\u20135812(2009).22. Ha\u00a8se, C. C. &Barquera, B. Role of sodiumbioenergetics in Vibrio cholerae.Biochim.Biophys. Acta 1505, 169\u2013178 (2001).23. Sasaki, M., Takagi, M. & Okamura, Y. A voltage sensor-domain protein is a voltage-gated proton channel. Science 312, 589\u2013592 (2006).24. Hattori, M. et al.Mg21-dependent gatingof bacterialMgtE channel underliesMg21homeostasis. EMBO J. 28, 3602\u20133612 (2009).25. Vassylyev, D. G. et al. Structural basis for substrate loading in bacterial RNApolymerase. Nature 448, 163\u2013168 (2007).26. Inaba,K. et al.Crystal structureof theDsbB-DsbAcomplex reveals amechanismofdisulfide bond generation. Cell 127, 789\u2013801 (2006).27. Matsuo, E., Mori, H., Shimoike, T. & Ito, K. Syd, a SecY-interacting protein, excludesSecA from the SecYE complex with an altered SecY24 subunit. J. Biol. Chem. 273,18835\u201318840 (1998).Supplementary Information is linked to the online version of the paper atwww.nature.com\/nature.AcknowledgementsWe thank Y. Akiyama, R. Suno, Y. Morimoto, T. Minamino,K. Namba, K. Inaba, M. Hattori and H. Nishimasu for suggestions; T. Sakamoto andA. Kurabayashi for assistance with sample preparation; R. Yamasaki, M. Sano,K. Mochizuki, K. Yoshikaie, K. Imayoshi and T. Adachi for technical support;M. Hommaand S. Kojima for providing the Vibrio genomic DNA; the beamline staff members atBL41XU of SPring-8 (Hyogo, Japan) and at NW12 of KEK PF-AR (Tsukuba, Japan) fortechnical help during data collection and M. Ibba for comments on our manuscript.This work was supported by aGrant-in-Aid for Scientific Research (S) from theMinistryofEducation, Culture,Sports, ScienceandTechnology (MEXT) toO.N., byaCRESTgrantfrom JST to K.I., by a BIRD grant from JST to H.M. and R.I., by a grant for the NationalProject on Protein Structural and Functional Analyses to O.N., by NIH grants to D.G.V.and by grants from MEXT to T.Tsukazaki, H.M., R.I., S.F. and K.I.Author Contributions T.Tsukazaki performed the structural determination and thebiochemical experiments with SecDF. H.M. performed the functional analyses ofSecDF. Y.E. solved the crystal structure of the SecDF P1 domain and assisted with thefunctional analysis of SecDF. R.I., S.F., D.G.V. and O.N. assisted with the structuraldetermination. A.D.M. performed patch clamp and pH fluorescence experiments.T.Tanaka and T.K. solved the structure of the P4 domain by NMR. A.P. and D.G.V.assisted with the crystallization and data collection of SecDF. All authors discussed theresults andcommentedon themanuscript.O.N. andK.I. supervised thework andwroteand edited the manuscript.Author Information The coordinates and structure factors have been deposited in theProtein Data Bank under the accession codes 3AQP for the entire TtSecDF protein and3AQO for the P1 domain. The PDB and BMRB codes for the deposited P4 domain are2RRN and 11426 respectively. Reprints and permissions information is available atwww.nature.com\/reprints. The authors declare no competing financial interests.Readers are welcome to comment on the online version of this article atwww.nature.com\/nature. Correspondence and requests for materials should beaddressed to O.N. (nureki@ims.u-tokyo.ac.jp) or K.I. (kito@cc.kyoto-su.ac.jp).PeriplasmCytoplasmI formF formR247D519H+I \u2192 F formPreproteinba cYEG DFATM1\u20136 TM7\u201312P4P1 baseP1 headFigure 4 | A working model of the PMF-driven translocation enhancementby SecDF. a, F form, capturing state. b, I form, holding state. c, I to F transitionand substrate-releasin state. The two essential char ed residues of EcSecDF arehighlighted. SecDF is coloured as in Fig. 2. SecYEG, grey; SecA, green; p eprotein,black line; protonmovement,white arrow. See themain text discussion fordetails.RESEARCH LETTER2 3 8 | N A T U R E | V O L 4 7 4 | 9 J U N E 2 0 1 1Macmillan Publishers Limited. All rights reserved\u00a92011                  A B C    23   A number of crystal structures of the YidC membrane insertase have been presented recently (Kumazaki, Chiba, et al. 2014, Kumazaki, Kishimoto, et al. 2014, Xin et al. 2018). The structures reveal that YidC contains 6 transmembrane helices, with a large periplasmic P1 domain (Figure 1.10) (Kumazaki, Kishimoto, et al. 2014, Xin et al. 2018). Unlike with SecDF, the functional role of the YidC P1 domain is unclear. Indeed, YidC remains functional even if a large portion of the P1 domain is deleted (Jiang et al. 2003).    Figure 1.10 Crystal structure of YidC The architecture of YidC is revealed, with six transmembrane segments and a large periplasmic (P1) domain. The smaller cytosolic domain is termed C1. The hydrophilic groove between TMs 3 (coloured red) and 5 (coloured purple) is indicated. Reproduced with permission from (Kiefer and Kuhn 2018). 2 FEMS Microbiology Letters, 2018, Vol. 365, No. 12Figure 1. The membrane insertase YidC of E. coli. (A) The YidC protein (green)consists of the \u03b2-folded periplasmic sandwich, the membrane-spanning partwith the transmembrane segments and of the coiled coil region in the cyto-pl sm. The hydrophobic slide of TM3 (purple) and TM5 (re ) a e highlighted.The periplasmic P1 domain (\u03b2-fold) and the cytoplasmic C1 domain (coiled coil)are indicated. (B) Top view showing the hydrophilic groove that is closed by theperiplasmic loop P2 and the hydrophobic slide of TM3 (purple) and TM5 (red).Structure and function of YidC from Escherichia coliIn the inner membrane of E. coli, YidC is a six-spanningmembrane protein with a large periplasmic domain P1 in itsN-terminal half. Whereas many homologues of YidC lack thisdomain, particularly the variants found in Gram-positive bacte-ria and in the organelles, the family members of Gram-negativebacteria consistently contain the P1 domain. This periplasmicdomain resembles a \u03b2-folded sandwich that can bind glycols(Ravaud et al. 2008) and is involved in the interaction with SecF(Xie et al. 2006). Nevertheless in E. coli, most of this domaincan be deleted without leading to a defect (Jiang et al. 2003).The membrane-spanning 5 helix-bundle is the most conservedpart among YidC and the related members of this family (Kuhnet al. 2003). To test the function of the 5 helix-bundle of therelated members in E. coli, several hybrid proteins have beenconstructed where this part is replaced by the homologoussequences (Jiang et al. 2002; van Bloois et al. 2007; Xin et al.2018). In E. coli, the 5 helix-bundle of YidC forms a hydrophilicgroove in the membrane that is open from the cytoplasmic sideand closed to the periplasm (Fig. 1; Kumazaki et al. 2014b). It isassumed that the groove transiently harbors hydrophilic proteindomains of substrates that are translocating to the periplasm.On its own, YidC is limited to the translocation of only smallperiplasmic regions. Possibly, this is also a consequence of thelimited space within the hydrophilic groove.All family members have a characteristic cytoplasmiccoiled-coil domain as C1 loop (Fig. 1A). Several studies havedocumented that this domain is essential for YidC function.However, its actual role is unknown so far. Recent cross-linkingexperiments have shown that this domain contacts SRP, the SRPreceptor, SecY andweakly SecE and SecG (Petriman et al. 2018). Adynamic conformational change was proposed for the orienta-tion of this domain based on the structural differences obtainedby cryo-electron microscopy and X-ray (Wickles et al. 2014).The hydrophobic slide of TM3 and TM5Early after their synthesis, the prospective membrane-spanningsequence of the substrate protein interactswith YidC and specif-ically contacts the transmembrane segments 3 and 5 (TM3 andTM5). The two alpha helices TM3 and TM5 of YidC form a 6A\u02da wide transmembrane gap that might clamp an incomingprotein chain between them (Fig. 2). The residues of TM3 andFigure 2. Model of membrane insertion of Pf3 coat protein by the membraneinsertase YidC. Model showing the binding (A), insertion of the Pf3 N-terminalregion (blue) into the groove (B) and the release of the protein into the bilayer(C). The hydrophobic part of Pf3 coat protein (gold) slides between TM3 (red) andTM5 (purple).TM5 involved in this interaction have hydrophobic side chains,suggesting a hydrophobic sliding mechanism for the incomingprotein chain. Using radiolabeled pulse experiments as short as1 min, it was possible to observe the early contacts between thePf3 coat protein and YidC in vivo (Klenner et al. 2008; Klennerand Kuhn 2012). Similar contact sites of TM3 and TM5 werefound for other substrate proteins like FtsQ and MscL (Yu et al.2008; Neugebauer et al. 2012).In vitro, with purified YidC that was reconstituted into purelipid vesicles, membrane insertion experiments were successfuland a YidC-dependent catalytic insertion of the purified Pf3 sub-strate protein by YidC could be documented (Serek et al. 2004).Also, with fluorescently labeled proteins the membrane inser-tion was very efficient in such an in vitro system (Ernst et al.2011). Based on these results, a single molecule set-up was es-tablished, where YidC-mediatedmembrane insertion of fluores-cently labeled substrate proteins could be observed in a smallwater droplet (50 \u00b5L), as the proteoliposomes were diffusingthrough a confocal volume of about 10 fL (Fig. 3). The diffusiontime of the proteoliposomes through the confocal volume wasabout 50msec and binding events occurring between a substrateand a YidC-containing proteoliposome could be observed in thistime span. The single-molecule system was used to determinethe insertion kinetics of the Pf3 coat protein by the changingfluorescence resonance energy transfer (FRET) (Winterfeld et al.2013). From this experiment the binding to YidC, the membranetranslocation of the periplasmic substrate region and the releaseFigure 3. Setup of the FCS experiments for observing the membrane insertion ofsingle proteins. For fluorescence correlation spectroscopy, a water droplet con-taining YidC proteoliposomes is placed onto a cover slip to detect a confocalvolume (green). The diffusion time of the proteoliposomes through the confocalvolume is about 50 msec. When purified and fluorescently-labeled Pf3 coat pro-tein is added to the droplet containing the proteoliposomes, its interaction withYidC can be observed by FRET.Downloaded from https:\/\/academic.oup.com\/femsle\/article-abstract\/365\/12\/fny106\/4980910 by University of British Columbia Library user on 12 July 2019              24  Curiously, the structures also reveal a conserved hydrophilic cavity between several of the transmembrane helices. This groove is open to the cytoplasm and to the membrane but is closed to the periplasm (Figure 1.10) (Hennon et al. 2015, Kumazaki, Kishimoto, et al. 2014). Interestingly, crosslinking studies have revealed that multiple residues within this hydrophilic groove come into direct contact with nascent YidC substrates (Kumazaki, Chiba, et al. 2014, Hennon et al. 2015). It is possible that the hydrophilic groove interacts with the hydrophilic regions of YidC substrates as they undergo transport across the membrane. The mechanism by which the substrate is released from YidC - as well as how the hydrophobic transmembrane portion is inserted into the lipid bilayer - remains unknown (Hennon et al. 2015, Xin et al. 2018). Obtaining a high-resolution structural snapshot of YidC in complex with a translocating substrate would shed valuable light on this important outstanding question in the field of membrane protein biogenesis.  1.5.3 The holo-translocon - SecYEG-SecDF-YidC  Recent work has shown that YidC, SecYEG and SecDF associate transiently together to form a membrane-embedded translocation holo-complex termed the \"holo-translocon\", or HTL (Collinson, Corey, and Allen 2015, Komar et al. 2016, Schulze et al. 2014, Botte et al. 2016). By using multiple affinity tags, the HTL can be isolated in detergent. The resultant complex is amenable to biophysical and structural studies (Botte et al. 2016, Schulze et al. 2014). Biochemical assays using purified HTL reveal that - during post-translational translocation - the HTL exhibits greater stimulation by the PMF compared with SecYEG alone (Schulze et al. 2014,     25  Komar et al. 2016). The HTL also displays greater insertase activity than SecYEG for certain membrane protein substrates (Komar et al. 2016). How do the multiple subunits of the HTL work synergistically to drive protein translocation and membrane protein insertion? Indirect evidence of the subunit organization within the HTL complex has arisen from site-directed crosslinking experiments. Recent work has shown that the YidC insertase interacts with SecYEG mainly via the SecY lateral gate (Sachelaru et al. 2017, Petriman et al. 2018). The interaction interface between SecYEG and SecDF, and the interaction interface between YidC and SecDF remain largely uncharacterized. Clearly, the determination of a high-resolution structure of the HTL complex would shed valuable light on these important questions. This remains an active area of research. Encouragingly, a low-resolution EM structure of the HTL was recently presented, giving researchers a preliminary idea of the subunit organization and stoichiometry of this complex. Given recent advances in cryo-electron microscopy, a high-resolution structure of the HTL complex appears to be within reach.   1.5.4 YfgM and PpiD  While the Sec ancillary subunits SecDF and YidC have been relatively well characterized in the literature, comparatively little is known about the membrane-bound periplasmic chaperones YfgM and PpiD. Both YfgM and PpiD have soluble periplasmic domains - although both are anchored to the inner membrane by a single N-terminal transmembrane segment (Figure 1.11) (Maddalo et al. 2011, G\u00f6tzke et al. 2014). In marked contrast to SecDF and YidC, neither YfgM nor PpiD is essential for cell viability (G\u00f6tzke et al. 2014).      26   Figure 1.11 Experimentally determined topologies of YfgM and PpiD Both proteins possess a single N-terminal transmembrane segment and a large soluble periplasmic domain. Reproduced with permission from (Maddalo et al. 2011).   The interaction between PpiD and SecYEG is documented in the literature. Previous in vivo site-directed crosslinking studies have shown that PpiD interacts directly with both the SecY and SecG subunits of the SecYEG complex.  Other studies have shown that PpiD is likely involved in the trafficking and biogenesis of E. coli outer membrane proteins, possibly as a member of the soluble network of E. coli periplasmic chaperones (F\u00fcrst et al. 2018, Sachelaru et al. 2014).   The role of YfgM in protein translocation has also not been thoroughly characterized to date. YfgM was first identified as a potential interactor of the Sec translocon using the \"guilt by association\" approach. As recently as 2011, YfgM had no annotated function, but was found to 1855 dx.doi.org\/10.1021\/pr101105c |J. Proteome Res. 2011, 10, 1848\u20131859Journal of Proteome Research ARTICLEmembrane proteins (see UniProt database). For YggE, YgiMand YhcB we explored the membrane topology experimentallyusing C-terminal fusions to alkaline phosphatase (PhoA) and thegreen fluorescent protein (GFP). These fusion partners are onlyactive in the periplasm or cytoplasm respectively46,47 and cantherefore be used to determine protein localization and topology(see refs 2, 36, and 48). When we expressed the PhoA andGFP fusion proteins in E. coli it was evident that the C-terminusof YggE was in the periplasm (i.e., GFP negative\/PhoA positive),while the C-termini of YgiM and YhcB were in the cytoplasmFigure 4. YfgM forms a complex with PpiD in the inner membrane of E. coli. (A) Cropped sections of the BN-\/SDS-PAGE showing the vertical channelcontaining YfgM and PpiD. Soluble andmembrane proteins were used to calibrate protein complexes in the BN-PAGE. As these two independent sets ofproteins give significantly different calibration curves (see text for details), we have calculated themolecular mass for each complex using both curves andreported a molecular mass range. All molecular mass markers are in kDa. (B) Normalized activity for C-terminal PhoA and fluorescence for GFP fusionsto YfgM and PpiD. Diagonal lines indicate cut-offs for the assignment of Cin (i.e., GFP positive\/PhoA negative) and Cout (i.e., GFP negative\/PhoApositive). Normalization of data and cut off values were described previously.2 (C) Experimentally constrained topology predictions of YfgM and PpiD.(D) Purification of YfgM-His8 by immobilized metal affinity chromatography and size exclusion chromatography, and analysis by SDS-PAGE. Theprotein that copurified was identified by mass spectrometry as PpiD.    27  associate tightly with PpiD within the bacterial membrane (Maddalo et al. 2011). This lead to the hypothesis that YfgM, like PpiD, may be an ancillary subunit of the Sec translocon (G\u00f6tzke et al. 2015, Maddalo et al. 2011). A subsequent study using classical co-immunoprecipitation and 2D gel electrophoresis techniques identified YfgM as well as PpiD as bona fide interactors of the Sec translocon (G\u00f6tzke et al. 2014).   To further elucidate the biological role of YfgM, the same research group employed a mass-spectrometry based approach to identify periplasmic proteins whose translocation may be influenced by YfgM (G\u00f6tzke et al. 2015). The authors compared the whole cell proteome between wild-type E. coli and an E. coli strain in which YfgM was deleted. They found that several periplasmic proteins were depleted in the YfgM-- strain, which implies that YfgM may be an important factor that regulates their translocation into the periplasm. Alternatively, it could be that YfgM is required for these proteins to fold correctly after translocation is completed (G\u00f6tzke et al. 2015).  In the absence of structural data for either YfgM or PpiD, it is not known how these putative chaperones may interact with translocating polypeptides as they emerge on the periplasmic side of the Sec translocon. The precise interaction interface between PpiD, YfgM and SecYEG is also unclear. Preliminary site-directed in vivo crosslinking studies suggest that PpiD may interact with SecYEG via either the lateral gate or via SecG. However, it is unknown if other interaction sites are possible. It is similarly unknown how the presence of YfgM may influence this interaction.      28  A further key unanswered question in this area of research is: are YfgM and PpiD constituents of the \"holo-translocon\" complex? Interactions between PpiD, YfgM and SecDF have not been reported in the literature to date. However, such an interaction seems eminently plausible, given that both PpiD and SecDF appear to influence translocation of Sec-dependent substrates on the periplasmic side of the membrane.   Another key question in the field is: are there additional ancillary subunits of the Sec translocon which have not yet been detected by classical proteomic or biochemical means? To address this important question, I turned to proteomics to identify and validate potentially novel membrane-bound interactors of the Sec translocon.   1.6 Overview of methods for identifying transient interactors of membrane proteins  In-depth analysis of membrane protein interaction networks - or the \"membrane interactome\" - has lagged behind characterization of the soluble interactome, due primarily to the insoluble nature of the cell membrane as well as the transient nature of many protein-protein interactions within the lipid bilayer (Babu et al. 2018, Papanastasiou et al. 2013, Papanastasiou et al. 2016). High throughput methods such as Yeast 2-Hybrid (Y2H) and Protein Complementation Assays (PCA) have been used in large-scale interaction studies for soluble proteomes, but have not been widely applied towards studying the membrane interactome (Rattray and Foster 2019, Rajagopala et al. 2014).       29  1.6.1 Blue Native PAGE  Most methods for analyzing the membrane proteome involve first extracting membrane proteins from the lipid bilayer using mild detergent. One classical approach which has been employed with some success involves separation of detergent-solubilized native membranes by non-denaturing Blue Native PAGE. Potential complexes are revealed as spots or bands on the gel. MS analysis is then performed on the observed protein spots to identify individual components of each potential complex (Figure 1.12) (Stenberg et al. 2005, Maddalo et al. 2011, Scott et al. 2017). Although this method does involve the use of potentially disruptive detergents and acidic blue dyes, it can be performed under \"native\" conditions, without genetic manipulation.        30   Figure 1.12 Native PAGE analysis of the E. coli membrane proteome Detergent-solubilized E. coli membrane proteins were analyzed by Blue Native (BN) PAGE. Gel bands were excised, and the protein components identified by mass spectrometry. The approximate migratory positions of known molecular weight markers is indicated on the left-hand side of the gel. Reproduced with permission from (Stenberg et al. 2005).        Protein Complexes in the InnerMembraneBioenergetic Complexes\u2014The majority of the proteins solubilizedfrom the IMVs were from complexes involved in bioenergetic pro-cesses. This is not surprising, since E. coli possesses a large, modularrespiratory chain consisting of 15 primary dehydrogenases and 10 ter-minal reductases\/oxidases (reviewed in Ref. 33). We were able to iden-tify two terminal oxidases, three primary dehydrogenases, and theF1-F0-ATP synthase. All but one of these known complexes resolvedintact in the gels and are described in detail below.Succinate dehydrogenase is a known heterotetramer, SdhABCD,which trimerizes to form a functional complex with a predictedmolecular mass of 355 kDa (34). From the SDS gel, we could identifythe 66-kDa flavoprotein subunit (SdhA) and the 26-kDa iron-sulfurprotein (SdhB) (Fig. 3A). Two other unidentified proteins thatcorrespond in molecular mass to that predicted for SdhC (14 kDa)and SdhD (15 kDa) were also detected in the same channel. Basedon the size of the succinate dehydrogenase complex in the BN gel(Fig. 1A) and the fact that all four constituent proteins werepresent in the SDS gel, we conclude that the (SdhABCD)3 complex isintact.The cytochrome bo3 ubiquinol oxidase is also a known heterotet-rameric complex, CyoABCD, with a predicted molecular mass of 145kDa (35). In the SDS gel, we could identify CyoA (35 kDa) (Fig. 3B).Three other proteins in the same channel could not be identified, buttheirmolecularmasses correspond to that predicted forCyoB (apparentmolecular mass of 45 kDa\/predicted molecular mass 74 kDa), CyoC (22kDa) and CyoD (12 kDa). Again, based on the size of the complex in theBN gel (Fig. 1A) and the fact that all four constituent proteins werepresent in the SDS gel, we conclude that the cytochrome bo3 ubiquinoloxidase is intact.Glucose dehydrogenase (Gdh) is a monomeric protein with fivetransmembrane segments (36). Although not in an oligomeric complex,FIGURE 1. Coomassie-stained BN-PAGE separa-tion of membrane protein complexes frominner (A) and outer (B) membrane vesicles.Approximately 100 !g of protein was loaded ineach lane. Commercially available molecular massmarkers are indicated to the left and a densitomet-ric scan showing protein abundance to the right.NADH dehydrogenase is not intact and is markedwith a number symbol.Protein Complexes of the E. coli Cell EnvelopeOCTOBER 14, 2005\u2022VOLUME 280\u2022NUMBER 41 JOURNAL OF BIOLOGICAL CHEMISTRY 34411 at University of British Columbia on January 3, 2019http:\/\/www.jbc.org\/Downloaded from     31  1.6.2 Affinity Purification Mass Spectrometry (AP\/MS)  Another detergent-based method for characterizing the membrane proteome involves affinity purification (AP) of a tagged membrane protein target after solubilization of the membrane proteome with mild detergent. Co-purifying interactors of the target protein are then identified by mass spectrometry (MS) analysis (Figure 1.13). This combination of affinity purification (AP) followed by mass spectrometry (MS) analysis is commonly referred to in the literature as AP\/MS. In a recent landmark study, Babu et. al. employed a high-throughput AP\/MS-based approach to establish a comprehensive understanding of membrane protein interaction networks in the model organism Escherichia coli (E. coli) (Babu et al. 2018). In this study, the authors performing large-scale AP\/MS experiments on chromosomally tagged membrane proteins after solubilization of the cell envelope under non-denaturing conditions. (Babu et al. 2018). This study was a follow-up on a previous paper from the same group which used a similar approach to examine membrane protein interactions in Saccharomyces cerevisiae (Babu et al. 2012). Although this single-tagged bait AP\/MS approach has been employed with some success in the past, key drawbacks of the method include the use of detergents, which may dissociate transiently associated multimeric protein complexes, as well as the use of affinity tags which may disrupt the structure and\/or function of the target protein (Hu et al. 2009, Babu et al. 2018, Babu et al. 2012, Gingras et al. 2007, Gingras, Abe, and Raught 2019, Lee et al. 2018).       32   Figure 1.13 Schematic of the classical detergent-based AP\/MS workflow Detergent solubilized membrane extracts containing the tagged bait protein of interest are incubated with affinity resin. Extensive washing removes non-specific contaminant proteins. The tagged bait protein and its transient interactors are then eluted. After digestion, the co-purifying interactors are identified by LC-MS\/MS. Adapted with permission from (Gingras, Abe, and Raught 2019).    Making sense of proximity-dependent biotinylation results Gingras, Abe and Raught 45Figure 1celllysatedensitygentlecell lysisgentlecell lysiscell lysateexpressing baitliquidchromatographygradientultra-centrifugationliquid chromatography - mass spectrometry(LC-MS)reconstruct elutionprofilesscore against controlsdata visualizationorganelle enrichmentand fractionationprotein complexfractionationA280[protein]variousprotein multimers(a) Classical interaction and organellar proteomics (b) Affinity purification (AP)(c) Proximity-dependent biotinylation (PDB) (d) Mass spectrometry analysispeptide generationwash away non-interacting proteinspurify interactingproteinsbait epitopetaganti-epitopebeadstrypsin(protease)intensitytimeHPLC MS1intensitym\/zMS2intensitym\/zpurifyproximity-interactorsin vivo  proteinbiotinylationKKKYYYYYYYYKKKKKKKKYYAPEXbiotin3-24 hrbiotin-phenol H2O230 mins 1 minBioIDsurface-exposedlysine residuessurface-exposedtyrosine residuesbaitbaitperoxidasebiotinligase diminishingbiotinylationstrengthpositiveinteractorsPCA plotstreptavidin-conjugatedbeads dot plotHDAC6PPP2R1ANUDCHSPA4ECH1ACACBPPP6R3PPP6R2PPP6R1ANKRD28STK3STK4harshlysisYYYYKKKKKKCurrent Opinion in Chemical BiologyOrganellar and interaction proteomics approaches. (a) Schematic of classical organellar (top) and protein complex (bottom) fractionationapproaches. A cell is first lysed, and the prepared lysate separated based on biophysical properties of its constituents. For comprehensivecharacterization, the entire elution profile is analyzed by quantitative MS (see d). (b) Standard affinity purification (AP). A bait is often fused to anepitope tag (here, the common 3x-FLAG tag is shown) and expressed in a relevant setting. After cell lysis under mild conditions, the epitope tag iscaptured on an affinity support, resulting in the purification of the bait and its stable interaction partners. (c) In proximity-dependent biotinylationwww.sciencedirect.com Current Opinion in Chemical Biology 2019, 48 :44\u201354                       Detergent-solubilized  membrane containing  tagged bait proteinAffinity PurificationLC-MS\/MS to identify  co-purifying interactors        Detergent-based AP\/MS workflow    33  The AP\/MS-based experimental workflow is often combined with SILAC (Stable isotope labeling by amino acids in cell culture) labeling to enable better quantification of the results. In a SILAC labeling experiment, two proteomes are generated: a \"heavy\" experimental condition which is grown in the presence of a isotopically labeled amino acid (typically lysine or arginine); and a \"light\" reference control proteome which is grown in the presence of non-labeled amino acids (Gokhale et al. 2012, Zhang et al. 2012, Rees, Lilley, and Jackson 2015). In an AP\/MS experiment, the affinity-tagged protein of interest is expressed in one condition, while the other condition expresses either an empty vector control, or else the protein of interest with no affinity tag. The \"heavy\" and \"light\" proteomes are treated identically and subjected to affinity purification side-by-side (Figure 1.14). Theoretically, the \"heavy\" elution should contain the target protein, its transient interactors, as well as any background contaminants which interact non-specifically with the resin. The \"light\" elution, by contrast, should contain only background contaminant proteins (Figure 1.14). The elutions are pooled and digested before mass spectrometry analysis (Gokhale et al. 2012, Zhang et al. 2012). Peptides corresponding to each individual protein are identified, and a \"heavy\"\/\"light\" ratio is determined. Non-specific contaminant proteins should have ratios close to 1, while the target protein and its co-purifying interactors should have ratios greater than 1 (Gokhale et al. 2012, Zhang et al. 2012, Rees, Lilley, and Jackson 2015).  The incorporation of SILAC labeling into the classical AP\/MS workflow enables researchers to differentiate more easily between co-purifying transient interactors and non-specific contaminants (Rees, Lilley, and Jackson 2015, Gokhale et al. 2012).        34   Figure 1.14 Incorporation of SILAC labeling into the AP\/MS workflow The Heavy proteome contains the tagged bait protein and is grown in the presence of isotopically heavy Lysine or Arginine. The Light reference proteome expresses an empty vector control. The two proteomes are subjected to affinity purification side-by-side before being pooled, digested and analyzed by LC-MS\/MS to identify their protein contents. Adapted with permission from (Zhang et al. 2012).    and for subsequent proteomic analysis using the principledescribed above. The results presented here employ theSecYEG channel, the maltose transporter MalFGK2, and theinsertase YidC to demonstrate the utility of nanoscale lipidbilayers for membrane proteomic analysis.\u25a0 EXPERIMENTAL SECTIONMaterialsAmino acid isotopologues were purchased from CambridgeIsotope Laboratories, (Andover, MA) and all others fromSigma-Aldrich (St. Louis, MO). Dioleoyl-sn-glycero-3-phospho-(1\u2032-rac-glycerol) (DOPG) and Escherichia coli total lipid extractwere purchased from Avanti Polar Lipids (Alabaster, AL). Ni2+-NTA chelating Sepharose and Superdex 200 bead suspensionswere purchased from GE Life Sciences (Uppsala, Sweden).Sequencing grade trypsin was purchased from Roche (Laval,QC, Canada).Preparation of SILAC Soluble Prey FractionsE. coli strain JW2806 [\u0394(araD-araB)567 \u0394lacZ4787(::rrnB-3)lambda\u2212 \u0394lysA763::kan rph-1 \u0394(rhaD-rhaB)568 hsdR514]10containing the lysA knockout was grown in LB media overnightat 37 \u00b0C. Following two washes in M9 minimal media, theculture was diluted 1:100 (relative to the original culture) intoM9 minimal media supplemented with 2 mM magnesiumsulfate, 0.4% w\/v glucose or 0.2% w\/v maltose, 10 \u03bcM CaCl2,4 \u03bcg\/mL of vitamin B1, and 0.2 mg\/mL of all amino acids exceptthe labeled cultures, which contained 0.3 mg\/mL of 13C615N2-lysine and 0.174 mg\/mL of 13C615N4-arginine. Cultures weregrown for approximately 6 h at 37 \u00b0C, which is equivalent to sixgenerations. Cells were solubilized in TSG buffer (50 mM Tris-HCl pH 7.9, 100 mM NaCl, 10% w\/v glycerol) and lysed bythree passes in a French pressure cell. The debris was removedby low-speed centrifugation (3000 rcf, 10 min), and the solublefraction was isolated by ultracentrifugation (126 000 rcf, 45 min).The unlabeled and heavy-labeled protein fractions were storedin \u221270 \u00b0C at a protein concentration of \u223c20\u221225 mg\/mL(determined by a Bradford assay) prior to use in pull-downexperiments.Protein and Nanodisc PreparationThe His6-tagged SecYEG complex was overexpressed, solubi-lized, and purified as described11 through immobilized metalaffinity chromatography, followed by anion exchange chroma-tography. N-terminal His6-tagged YidC was overexpressed fromthe plasmid pBAD22YidC in E. coli BL21(DE3) cells, and themembrane was solubilized with n-dodecyl \u03b2-D-maltoside andpurified under similar conditions as SecYEG. Furtherpurification was achieved by gel filtration chromatographyusing a Tricorn Superdex 200 HR 10\/300 column. MalFGK2complex was overexpressed from the plasmid pBAD22-FGKbearing a C-terminal His6-tag on MalK in E. coli BL21(DE3)cells. Membranes were solubilized under similar conditions asSecYEG except at 5 mg\/mL in buffer containing 20% instead of10% glycerol and purified by immobilized metal affinitychromatography and gel filtration chromatography as above.Nanodiscs were reconstituted as described in Dalal andDuong.11 The SecEYG:MSP1:lipid ratio employed was 1:4:40.The reconstitution of YidC and MalFGK was performedwithout added lipids at a protein:MSP1 ratio of 1:4 and 1:5,respectively. The prepared discs were purified by gel filtrationchromatography.11Pull-Down ExperimentsThe indicated nanodiscs (10 \u03bcg each), or the scaffold proteinMSP1 as control, were immobilized onto Ni2+-NTA sepharosebeads (20 \u03bcL suspension) in 0.5 mL of TSG buffer for 5 min atroom temperature with gentle shaking. After washing with0.5 mL of TSG buffer to remove excess unbound protein, thebeads were incubated with the unlabeled or labeled solubleprotein fractions (1 mg each) in 0.5 mL of TSG buffer. Beadsfrom both samples were then combined together and washedthree times in 1 mL of TSG buffer containing 50 mM imidazole,followed by elution with 0.1 mL of TSG buffer containing 600 mMimidazole. The pull-down efficiency was assessed by 12% SDS-PAGE and silver staining prior to digestion and analysis of thesample by mass spectrometry.Figure 1. Strategy employed to identify the interacting partners of amembrane protein using nanodisc and SILAC. Nanodiscs (right)containing a membrane protein of interest, or the His-taggedmembrane scaffold protein alone as a control (left), were used asbaits. Crude protein extract from cells grown in normal \u201clight\u201d orstable isotope \u201cheavy\u201d analogue media were used as prey. After affinitypull-down using Ni-NTA sepharose beads, the coeluted proteins wereanalyzed by SDS-PAGE and their identity revealed by LC\u2212MS\/MS.For each peptide identified, the peak intensity of the heavy form iscompared to the light form to determine whether a protein is a specificinteractor or nonspecific contaminant.Journal of Proteome Research Technical Notedx.doi.org\/10.1021\/pr200846y | J. Proteome Res. 2012, 11, 1454\u221214591455                              Light proteome No expression of baitHeavy proteome Bait expressed                             Affinity purification                                                          Tagged bait proteinNon-specific contaminant                                 35  While the conventional detergent-based methods described above have been used with success to study membrane protein interaction networks - or the \"membrane interactome\" - in the past, the deleterious effects of detergents on membrane protein stability and function are well known. In light of this, there is clearly a need for development of detergent-free methods for studying the membrane interactome. The nanodisc membrane mimetic has been previously used with some success to reconstitute the membrane proteome in a detergent-free environment (Marty et al. 2013, Roy et al. 2015). A drawback of using the nanodisc method to stabilize the membrane proteome, however, is that the diameter of proteins and complexes captured in the disc is influenced by the length of the nanodisc scaffold protein, MSP (Marty et al. 2013). An additional caveat of the method i\".s that formation of a nanodisc membrane protein library requires addition of lipids, the identity of which influences the membrane protein content isolated in the nanodisc library (Roy et al. 2015, Marty et al. 2013).    1.6.3 The Peptidisc Library  To address the potential drawbacks of the nanodisc method, our laboratory has recently developed the peptidisc as a flexible, \"one-size-fits-all\" membrane mimetic for membrane protein reconstitution (Figure 1.15) (Carlson et al. 2018, Carlson et al. 2019, Saville, Troman, and Duong Van Hoa 2019, Angiulli et al. 2020). Peptidisc reconstitution is based on a 37 amino acid amphipathic ApoA1-mimetic peptide termed \"NSP\" (nanodisc scaffold peptide), which was originally developed to form lipid nanodiscs (Kariyazono et al. 2016). To increase the water-solubility of NSP, we reversed its amino-acid sequence, creating the peptide \"NSPr\" (Carlson et     36  al. 2018, Angiulli et al. 2020).  Membrane protein reconstitution into peptidisc occurs when multiple copies of the Nanodisc Scaffold Peptide (NSPr) self-assemble around detergent-solubilized membrane proteins. Upon removal of detergent, incorporation of both endogenous lipids and solubilized membrane proteins into soluble detergent-free particles occurs in a single step (Carlson et al. 2018, Saville, Troman, and Duong Van Hoa 2019).  Critically, the number of wrapping peptide scaffolds adapts to fit the size of the protein target. Furthermore, peptidisc formation does not require addition of exogenous lipids. Our previous work has shown that the peptidisc is able to efficiently reconstitute both inner and outer membrane proteins of E. coli (Carlson et al. 2018). We have recently described the application of the peptidisc to trap the global detergent-solubilized E. coli membrane proteome directly into water soluble, detergent-free peptidiscs (Figure 1.15) (Carlson et al. 2019). The resulting membrane protein particles are termed a \"peptidisc library\". Our work shows that these peptidisc library preparations are soluble and amenable to biochemical fractionation (Carlson et al. 2019). Using the E. coli SecYEG translocon as a membrane marker, I show that an over-expressed target protein can be effectively purified from the library.  In Chapter 3, I present work done by myself and by others which show that the peptidisc is well-suited to trapping transient interactors of membrane proteins. By combining the peptidisc method with SILAC labeling, my work identifies a number of previously unclear aspects of membrane protein interactions.      37   Figure 1.15 Schematic of the peptidisc library workflow Biological membranes are briefly solubilized with detergent. After an ultracentrifugation step to remove aggregates and insoluble material, the detergent solubilized membrane extract is reconstituted into a peptidisc library. Detergent removal is accomplished by successive dilution and concentration steps.    Native membranesDetergent solubilizationLibrary reconstitution    38  1.6.4 BioID  Alternative techniques which do not require membrane proteins to be extracted with detergents are emerging in the literature. One such technique is BioID (Gingras, Abe, and Raught 2019, Rattray and Foster 2019, Lambert et al. 2015). BioID relies on fusion of a mutant bacterial biotin ligase - often E. coli BirA - to a membrane protein of interest. Upon expression of the membrane protein-biotin ligase fusion in an appropriate cell line, the biotin ligase releases a \"cloud\" of activated biotin, which will react with the exposed amine groups of nearby proteins (Figure 1.16). Thus, proximal interacting proteins of the membrane protein target will become biotinylated. These proteins can then be purified using streptavidin-based resins before being digested and identified by mass spectrometry (Figure 1.16) (Rattray and Foster 2019, Gingras, Abe, and Raught 2019, Roux et al. 2012, Lambert et al. 2015). The BioID technique - especially combined with more conventional AP\/MS strategies - is a powerful method for identifying transient and novel interactors of membrane proteins and complexes (Gingras, Abe, and Raught 2019, Lambert et al. 2015). The BioID method does, however, suffer from the drawback that each protein of interest must be individually tagged. Addition of such a large tag - in this case, a biotin ligase enzyme - can potentially disrupt the folding of the protein of interest, or else disrupt complex formation with native interactors. While the BioID method shows considerable promise, it is not used in this thesis. The use of BioID to identify transient interactors of E. coli membrane proteins and complexes is the main research project of one of my laboratory colleagues, Ms. Zhiyu (Katherine) Zhao.       39     Figure 1.16 Schematic of the BioID experimental workflow A bait protein of interest (shown in blue) is fused to a mutant biotin ligase (shown in dark blue) and is expressed in cells. The mutant biotin ligase releases a cloud of activated biotin, which reacts with lysine residues of proximal and interacting proteins. After cell lysis, biotinylated proteins can be isolated using streptavidin coated beads, digested and identified by LC-MS\/MS. Adapted with permission from (Gingras, Abe, and Raught 2019).           Making sense of proximity-dependent biotinylation results Gingras, Abe and Raught 45Figure 1celllysatedensitygentlecell lysisgentlecell lysiscell lysateexpressing baitliquidchromatographygradientultra-centrifugationliquid chromatography - mass spectrometry(LC-MS)reconstruct elutionprofilesscore against controlsdata visualizationorganelle enrichmentand fractionationprotein complexfractionationA280[protein]variousprotein multimers(a) Classical interaction and organellar proteomics (b) Affinity purification (AP)(c) Proximity-dependent biotinylation (PDB) (d) Mass spectrometry analysispeptide generationwash away non-interacting proteinspurify interactingproteinsbait epitopetaganti-epitopebeadstrypsin(protease)intensitytimeHPLC MS1intensitym\/zMS2intensitym\/zpurifyproximity-interactorsin vivo  proteinbiotinylationKKKYYYYYYYYKKKKKKKKYYAPEXbiotin3-24 hrbiotin-phenol H2O230 mins 1 minBioIDsurface-exposedlysine residuessurface-exposedtyrosine residuesbaitbaitperoxidasebiotinligase diminishingbiotinylationstrengthpositiveinteractorsPCA plotstreptavidin-conjugatedbeads dot plotHDAC6PPP2R1ANUDCHSPA4ECH1ACACBPPP6R3PPP6R2PPP6R1ANKRD28STK3STK4harshlysisYYYYKKKKKKCurrent Opinion in Chemical BiologyOrganellar and interaction proteomics approaches. (a) Schematic of classical organellar (top) and protein complex (bottom) fractionationapproaches. A cell is first lysed, and the prepared lysate separated based on biophysical properties of its constituents. For comprehensivecharacterization, the entire elution profile is analyzed by quantitative MS (see d). (b) Standard affinity purification (AP). A bait is often fused to anepitope tag (here, the common 3x-FLAG tag is shown) and expressed in a relevant setting. After cell lysis under mild conditions, the epitope tag iscaptured on an affinity support, resulting in the purification of the bait and its stable interaction partners. (c) In proximity-dependent biotinylationwww.sciencedirect.com Current Opinion in Chemical Biology 2019, 48 :44\u201354Making sense of proximity-dependent biotinylation results Gingras, Abe and Raught 45Figure 1celllysatedensitygentlecell lysisgentlecell lysiscell lysateexpressing baitliquidchromatographygradientultra-centrifugationliquid chromatography - mass spectrometry(LC-MS)reconstruct elutionprofilesscore against controlsdata visualizationorganelle enrichmentand fractionationprotein complexfractionationA280[protein]variousprotein multimers(a) Classical interaction and organellar proteomics (b) Affinity purification (AP)(c) Proximity-dependent biotinylation (PDB) (d) Mass spectrometry analysispeptide generationwash away non-interacting proteinspurify interactingproteinsbait epitopetaganti-epitopebeadstrypsin(protease)intensitytimeHPLC MS1intensitym\/zMS2intensitym\/zpurifyproximity-interactorsin vivo  proteinbiotinylationKKKYYYYYYYYKKKKKKKKYYAPEXbiotin3-24 hrbiotin-phenol H2O230 mins 1 minBioIDsurface-exposedlysine residuessurface-exposedtyrosine residuesbaitbaitperoxidasebiotinligase diminishingbiotinylationstrengthpositiveinteractorsPCA plotstreptavidin-conjugatedbeads dot plotHDAC6PPP2R1ANUDCHSPA4ECH1ACACBPPP6R3PPP6R2PPP6R1ANKRD28STK3STK4harshlysisYYYYKKKKKKCurrent Opinion in Chemical BiologyO ganellar and i teraction proteomics approaches. (a) Schematic of classical organellar (top) and protein complex (bottom) fractionationapproac s. A cell is first lysed, and the prepared lysate separated based on biophysical properties of its constituents. For comprehensivecharacterization, the entire elution profile is analyzed by quantitative MS (see d). (b) Standard affinity purification (AP). A bait is often fused to anepitope tag (here, the common 3x-FLAG tag is shown) and expressed in a relevant setting. After cell lysis under mild conditions, the epitope tag iscaptured on an affinity support, resulting in th  purifi ation of the bait and its stable interaction partners. (c) In proximity-dependent biotinylationwww.sciencedirect.com Current Opinion in Chemical Biology 2019, 48 :44\u201354Making sense of proximity-dependent biotinylation results Gingras, Abe and Raught 45Figure 1celllysatedensitygentlecell lysisgentlecell lysiscell lysateexpressing baitliquidchromatographygradientultra-centrifugationliquid chromatography - mass spectrometry(LC-MS)reconstruct elutionprofilesscore against controlsdata visualizationorganelle enrichmentand fractionationprotein complexfractionationA280[protein]variousprotein multimers(a) Classical interaction and organellar proteomics (b) Affinity purification (AP)(c) Proximity-dependent biotinylation (PDB) (d) Mass spectrometry analysispeptide generationwash away non-interacti g proteinspurify interactingproteinsbait epitopetaganti-epitopebeadstrypsin(protease)intensitytimeHPLC MS1intensitym\/zMS2intensitym\/zpurifyproximity-interactorsin vivo  proteinbiotinylationKKKYYYYYYYYKKKKKKKKYYAPEXbiotin3-24 hrbiotin-phenol H2O230 mins 1 minBioIDsurface-exposedlysine residuessurface-exposedtyrosine residuesbaitbaitperoxidasebiotinligase diminishingbiotinylationstrengthpositiveinteractorsPCA plotstreptavidin-conjugatedbeads dot plotHDAC6PPP2R1ANUDCHSPA4ECH1ACACBPPP6R3PPP6R2PPP6R1ANKRD28STK3STK4harshlysisYYYYKKKKKKCurrent Opinion in Chemical BiologyOrganellar and interaction proteomics approaches. (a) Schematic of classical organellar (top) and protein complex (bottom) fractionationappro ches. A cell is first lys d, and the prepared lysate s parated based on biophysical properties of its constitu nts. For comprehensivecharact rizatio , the entire lutio  profile is analyzed by qu ntitative MS (see d). (b) Standard affinity purificati  (AP). A bait is often fus d to anepitope tag (here, the common 3x-FLAG tag is shown) and expressed in a r levant set ing. After cell lysis under mild conditions, the pitop  tag isca tured on an affinity support, resulting in the purification of th  bait a d its stable interaction partners. (c) In proximity-dependent biotinylationwww.sciencedirect.com Current Opinion in Chemical Biology 2019, 48 :44\u201354       BioID workflowin vivo biotinylation                      purification of biotinylated interactors    40  1.6.5 Native Mass Spectrometry (nMS)  Recently, a method was proposed by the Robinson laboratory for studying membrane protein complexes by native mass spectrometry (nMS) which also does not require membrane protein extraction by detergent (Chorev et al. 2018). Membrane protein complexes are rather ejected from native membranes using sonication. The ejected complexes - which contain mixtures of proteins and lipids - are then analyzed by nMS (Chorev et al. 2018). In their initial study, Robinson and colleagues applied this method successfully to both bacterial and mitochondrial membranes.  Encouragingly, this nMS-based method has revealed some striking novel aspects of membrane biology.   In their analysis of E. coli membranes, Robinson and colleagues detect the core of the Sec translocon - SecYEG (Chorev et al. 2018). Surprisingly, they observe a stable association between the Sec translocon and the F1F0 ATP synthase complex (Chorev et al. 2018). This surprising and novel interaction has implications for our understanding of the mechanism of protein translocation. It is well known that protein transport through the Sec translocon is enhanced by the PMF across the membrane (Collinson, Corey, and Allen 2015, Duong and Wickner 1997b, a, Tsukazaki et al. 2011). Given that the F1F0 ATP synthase has a critical role of generating and maintaining the PMF, its physical interaction with the Sec translocon may represent a possible mechanism for how the PMF can be harnessed efficiently to accelerate Sec-dependent protein translocation.       41  Robinson and colleagues also reveal some novel and unexplored aspects of the outer membrane-embedded BamABCDE complex (Chorev et al. 2018). The Bam complex is required for biogenesis of outer membrane proteins in gram-negative bacteria and consists of a major outer membrane-integrated subunit BamA and four additional membrane-anchored lipoprotein subunits, BamB, C, D and E (Roman-Hernandez, Peterson, and Bernstein 2014, Gu et al. 2016). Structural studies of the Bam complex by X-ray crystallography have shown that the Bam complex consists of one copy each of BamA, B, C, D and E. The nMS results, however, suggests that the complex may in fact contain two copies of BamE, rather than just one (Gu et al. 2016, Chorev et al. 2018). A further novel observation detected by nMS is evidence that the BamE subunit binds to cardiolipin, which may represent a mechanism of localization of the Bam complex to specific regions of the cell membrane (Chorev et al. 2018).  This nMS-based approach for studying membrane protein complexes is clearly a powerful way to study the membrane proteome. Although this is well beyond the scope of this thesis, I anticipate that this nMS-based approach - combined with the peptidisc method - will be especially effective at unravelling previously unclear aspects of membrane biology.           42  1.7 Overview of Objectives  The work presented in this thesis is largely focused on the Sec translocon - a conserved membrane-bound complex of fundamental importance in cell biology. Although the Sec translocon has been extensively characterized both biochemically and structurally, a number of outstanding questions remain largely unaddressed in the literature. The questions we sought to address in this thesis investigation are outlined below.  The first question concerns the dynamics of SecA-SecYEG interactions during protein translocation. This work is presented in Chapter 2. Here, we aim to address conflicting reports in the literature about the importance of SecA cycling on-and-off the membrane (also referred to as \"SecA processivity\") and SecYEG while a preprotein substrate is undergoing transport. Is \"SecA processivity\" important for translocation of all Sec-dependent preprotein substrates? If so, does the importance of SecA processivity vary between individual preprotein substrates, or is it required by all substrates? To address these important questions, we reconstituted the translocation reaction in an in vitro system and compared the translocation activity between wild-type SecA and a non-processive mutant - PrlD23. We found that the importance of SecA processivity during the translocation reaction varies depending on the length of the substrate undergoing transport: translocation of longer substrates depends on SecA processivity, while translocation of shorter substrates does not.       43  In Chapter 3, we adopt a proteomics-based strategy to characterize the membrane interactome of SecYEG. A large body of work in the literature has shown that the Sec translocon interacts transiently with a number of membrane-bound ancillary subunits. Several of these ancillary subunits have been shown to augment the activity of the translocon during various stages of the translocation reaction. The key question we address in this chapter is: are there additional membrane-bound interactors of SecYEG which may have evaded detection by existing biochemical and proteomic methods? We employ the newly developed peptidisc method - combined with SILAC labeling and quantitative proteomics - to search for additional interactors of the Sec translocon. In addition to currently known and validated interactors of the Sec translocon, we also identify a number of novel and unexpected interactors. In this study, we define \"interactors\" either as: i) Functional interactors which modulate the structure and\/or activity of the Sec translocon; or ii) Proximal interactors which are localized near the Sec translocon within the cell membrane, but do not influence the activity of the translocon. The implications of these findings on our understanding of the Sec protein transport pathway are discussed.   The work described in Chapter 3 demonstrates that our peptidisc-based method for fractionating the membrane proteome is demonstrably superior than detergent-based methods. Specifically, we show that many transient interactions which are prone to dissociation in detergent are preserved in peptidisc. Chapter 4 focuses on how the peptidisc method can be further optimized for future studies on the membrane proteome. We noticed in our recent work (described in detail in Chapter 3) that our peptidisc library preparations are often contaminated with soluble proteins - particularly large, soluble complexes which co-sediment with the membrane fraction during     44  centrifugation. We rationalized that the presence of these soluble contaminants may interfere with fractionation and other downstream analyses of the membrane proteome.   Here, we functionalize the peptidisc scaffold with a His tag to enable purification of the reconstituted membrane proteome away from soluble contaminants. We demonstrate that the functionalized scaffold enables affinity purification of the bona fide membrane proteome. Specifically, we demonstrate that many high-abundance soluble contaminants such as ribosomal subunits are strongly depleted after an affinity purification step. Simultaneously, membrane proteins are significantly enriched following a purification step.   We further apply this method to survey changes in the membrane proteome upon SecDFyajC depletion. We observe an increase in the amount of membrane localized SecA upon SecDFyajC depletion, as well as strong up-regulation of YibN, an inner membrane protein with no annotated function. These findings highlight the utility of functionalized peptidiscs for detailed analysis of the bacterial membrane proteome with altered gene expression. Looking forward, this workflow can be applied multi-organelle mammalian cell lines to effectively monitor changes in the membrane proteome between wild-type and disease states.       45  Chapter 2: Investigating the stability of the SecA-SecYEG complex during protein translocation across the bacterial membrane  2.1  Introduction In E. coli, many periplasmic and extracellular proteins are transported across the inner membrane through the heterotrimeric SecYEG protein-conducting channel. Proteins can be transported either co-translationally or post-translationally (Park and Rapoport 2012, Collinson, Corey, and Allen 2015, Crane and Randall 2017, Koch et al. 2016). During post-translational translocation, the polypeptide substrate is synthesized by ribosomes before being directed toward SecYEG. The energy for protein translocation is then provided by the cytosolic ATPase SecA, which binds to the SecYEG channel and drives protein export through a series of ATP-dependent conformational changes (Park and Rapoport 2012, Morita, Tokuda, and Nishiyama 2012, Chatzi et al. 2014, Robson et al. 2007, Collinson, Corey, and Allen 2015, Cranford-Smith and Huber 2018). The proton motive force (PMF) across the inner membrane also contributes to driving protein translocation, particularly at the later stages of the transport reaction (Duong and Wickner 1997b, Tsukazaki et al. 2011, Crane and Randall 2017).   A landmark previous study reported that ATP binding to SecA triggers a \u201cpower stroke\u201d, resulting in forward movement of a polypeptide segment into the mouth of the SecYEG channel (Bauer et al. 2014). Following ATP hydrolysis, when ADP is still bound to the enzyme, SecA adopts a conformation allowing the polypeptide to slide across the SecYEG channel until the next ATP     46  binding event (Bauer et al. 2014, Erlandson et al. 2008). In this model, SecA exists mainly in an ADP-bound state when translocation is taking place (Bauer et al. 2014, Ding, Mukerji, and Oliver 2003). This power-stroke based model was further substantiated by the same research group using  a FRET-based experimental approach (Catipovic et al. 2019). Other FRET-based studies using a different experimental setup, however, led to an alternative model, in which protein translocation does not require a \"power stroke\" (Allen et al. 2016). Instead, SecA rather operates as a \"Brownian ratchet\", allowing substrates to passively diffuse within the SecY channel while the enzyme is in an ADP-state. When a steric blockage is reached due to a stretch of bulky or aromatic residue within the substrate, ATP binding to SecA leads to temporarily widening of the SecY channel, thus allowing the blockage to pass (Allen et al. 2016, Fessl et al. 2018).   What prevents substrates from simply diffusing backwards - or \"backsliding\" - through the SecY channel after initiation of translocation? A recent molecular dynamics-based study from the same research group has shown that translocating substrates adopt stable secondary structures after emerging on the periplasmic side of the SecY channel (Corey et al. 2019). Critically, formation of substrate secondary structure does not take place on the cytosolic side of the SecY channel. Once the partially translocated polypeptide has adopted secondary structure after emerging on the periplasmic side of SecY, it is unable to pass through the SecY channel in a backwards direction (Corey et al. 2019). Thus, adoption of substrate secondary structure post-translocation ensures that substrate diffusion is biased in the forward direction (Corey et al. 2019, Collinson 2019).  Both the \"power stroke\" and \"Brownian ratchet\" models, in which transport largely relies on passive sliding of substrates through the SecY channel without specific sequence recognition, explains why the     47  translocon is able to handle such a wide variety of proteins, each with a different sequence (Bauer et al. 2014, Allen et al. 2016, Fessl et al. 2018, Cranford-Smith and Huber 2018).   Despite these recent insights, some aspects of SecA-SecYEG interactions during the translocation reaction remain unclear. Studies by Morita et. al (2011) and Mao et. al. (2013) suggest that SecA is a highly dynamic enzyme which must undergo successive cycles of binding and dissociation from the SecYEG complex in order for translocation to proceed efficiently (Mao et al. 2013, Morita, Tokuda, and Nishiyama 2012). In both of these studies, SecA is hypothesized to dissociate from SecYEG while a polypeptide substrate is undergoing transit across the membrane (Morita, Tokuda, and Nishiyama 2012, Mao et al. 2013). Importantly, dissociation of SecA does not dislodge the substrate from SecYEG. Instead, a \"fresh copy\" of cytosolic SecA binds to SecYEG and completes translocation of the substrate (Morita, Tokuda, and Nishiyama 2012, Mao et al. 2013). We hereafter refer to these cycles of SecA binding to and dissociating from SecYEG as \"SecA processivity\" (Bauer et al. 2014). Other studies, however, reported that SecA processivity is not essential for efficient protein translocation. Whitehouse et. al. (2012) and Gold et. al. (2013) both showed that immobilization of SecA onto SecYEG via cysteine cross-linking results in the formation of a SecA-SecYEG complex that is still competent for translocation of proOmpA (Whitehouse et al. 2012, Gold et al. 2013). Furthermore, a fusion construct of SecA-SecYEG also is able to catalyze translocation of proOmpA (Sugano et al. 2017). Given the two differing views reported in the literature, the importance of SecA processivity warrants further investigation.      48  Here, we use a SecA mutant PrlD23 (SecA Y134S) that is less processive than wild-type SecA (Gouridis et al. 2013). The mutant exhibits elevated ATPase activity and associates more tightly with bacterial membranes and with SecYEG than does wild-type SecA (Gouridis et al. 2013, Schmidt et al. 2000). We reconstituted purified SecYEG with SecA or PrlD23 into proteoliposomes and assessed the in vitro translocation efficiency of proOmpA and proPhoA. As a complementary approach, we assessed translocation efficiency using a covalently cross-linked SecA-SecYEG complex. Interestingly, we observed an effect of SecA processivity on proPhoA translocation, but not on proOmpA translocation. To determine which features of the two substrates lead to this difference, we generated a series of chimeric substrates by exchanging protein domains between proPhoA and proOmpA. Our data indicate that SecA processivity is mainly influenced by the length of the preprotein substrate.  2.2 Materials and Methods  2.2.1 Protein Production and Purification  Unless otherwise stated, all proteins were expressed in E. coli BL21(DE3) (New England Biolabs) for 3 hr at 37\u00b0C after induction with either 0.2% Arabinose (for pBad plasmids) or 0.5 mM IPTG (for pET23 and Trc99a plasmids) at an OD of 0.4\u20130.7 in LB medium supplemented with required antibiotic. Cells were harvested by low-speed centrifugation (10,000 x g, 6 min) and resuspended in TSG buffer (50 mM Tris-HCl: pH 8; 100 mM NaCl; 10% glycerol) with douncing. Resuspended cells were treated with 1 mM phenylmethylsulfonyl fluoride (PMSF) and lysed     49  using a microfluidizer (Microfluidics) at 10,000 psi. Unbroken cell debris and other aggregates were removed by an additional low-speed centrifugation. Cytosolic and crude membrane fractions containing the overexpressed protein of interest were subsequently isolated by ultracentrifugation (100,000 x g, 45 min) and crude membrane fraction resuspended in TSG buffer.  Expression and purification of His-tagged SecYEG was performed from the plasmid pBad22-His-EYG as previously described (Dalal et al. 2009, Dalal et al. 2012, Carlson et al. 2018, Alami et al. 2007). Briefly, crude membranes were solubilized for 1 hr at 4\u00b0C in TSG buffer + 1% n-dodecyl \u03b2  D-maltoside (DDM). Solubilized material was clarified by ultra-centrifugation and passed over a 5 mL Ni2+-NTA column. After extensive washing in TSG buffer + 0.02% DDM, SecYEG was eluted in TSG + 0.02% DDM + 300 mM Imidazole. The fractions containing the SecYEG complex were pooled and diluted five-fold in TG buffer (50 mM Tris-HCl: pH 8, 10% glycerol) +0.02% DDM before being applied to a 5 mL Fast Flow S cation exchange column. After sample loading, the column was washed with 20 mL TG buffer + 0.02% DDM. SecYEG was eluted from the column in \"high salt\" (50 mM Tris-HCl: pH 8; 500 mM NaCl; 10% glycerol) buffer + 0.02% DDM. Fractions were stored at -80 and used within a month.   Inner Membrane Vesicles (IMVs) containing over-expressed SecYEG were prepared from the E. coli strain KM9 (unc: Tn10, rna10, relA1, spoT1 and metB1) as previously described (Dalal and Duong 2009, Dalal et al. 2012). Briefly, the plasmid pBad22-His-EYG was transformed into chemically competent KM9 cells. Protein expression and crude membrane isolation was     50  performed as described above. To further enrich the inner membrane fraction, the crude membrane fraction was layered onto a 2-step 20%-50% sucrose gradient in SW41 tubes and re-centrifuged at 200 000 g for 2 hours. The inner membrane (IMV) fraction was recovered as a distinct brown band near the middle of the gradient and diluted 4-fold in TSG before being pelleted by ultracentrifugation (100 000 g, 15 minutes). IMVs were stored at -80 and used in translocation assays within a month.  Expression and purification of wild-type and mutant variants of SecA was as previously described using the plasmid pET23 SecAHis (Dalal et al. 2012, Duong 2003). Briefly, cells were grown as described above to induce expression of His-tagged SecA. The clarified cytosol containing over-expressed SecA was applied to a 5 mL Ni2+-NTA column. After extensive washing in TSG buffer, SecA was eluted in TSG buffer + 300 mM Imidazole. Fractions containing SecA were pooled and diluted five-fold in TG buffer (50 mM Tris-HCl: pH 8, 10% glycerol) before being applied to a 5 mL Fast Flow S cation exchange column. After sample loading, the column was washed with 20 mL TG buffer. SecA was eluted from the column in \"high salt\" (50 mM Tris-HCl: pH 8; 500 mM NaCl; 10% glycerol) buffer. Fractions containing SecA were concentrated to ~ 1mL in an Amicon concentrator (100 kDa cut-off) and injected onto a 25 mL Superdex 200 10\/300 column equilibrated in TSG buffer + 0.1 mM DTT using an AKTA FPLC system. Protein abundance was monitored by UV at 280 nm. Collected fractions were pooled and stored at -80 until use. Mutant variants of SecA employed in this study have been described previously and were purified as described above (Gouridis et al. 2013, Bauer et al. 2014, Catipovic et al. 2019, Whitehouse et al. 2012).     51    The coding region of Alkaline Phosphatase A (proPhoA) was cloned into the plasmid pBad33 and transformed into E. coli strain BL21. A 2L culture of pBad33-PhoA (C190S) was grown in LB media supplemented with 50 \u03bcg\/mL chloramphenicol. At OD600 0.4, 0.2% (w\/v) L-Arabinose was added to induce protein expression. After 3 hours of induction, cells were harvested and lysed using a Microfluidizer (3 passes at 8000 psi). proPhoA was extracted from the resulting inclusion bodies in Buffer A (50mM Tris HCl pH 7.9 and 6M Urea). Expression and purification of proOmpA bearing a unique cysteine (mutant S129C) from the plasmid Trc99a-proOmpA (S129C) was carried out as previously described (Duong and Wickner 1998, Dalal et al. 2009). Cloning of all PhoA\/OmpA recombinant translocation substrates used in this study was carried out using the Polymerase Incomplete Primer Extension (PIPE) method (Klock and Lesley 2009). All constructs were verified by DNA sequencing (Genewiz).  Protein expression and purification of all derivatives was carried out as described above for proPhoA. All recombinant translocation substrates employed in this study are described in detail in Table 2.1.  2.2.2 Fluorescent Labeling  Dye labeling of translocation substrates was performed as previously described (Dalal et al. 2012). Briefly, 100 \u03bcg of purified protein was incubated in the dark for two hours with a molar excess of the cysteine-specific Alexa Fluor 680 labeling reagent (Invitrogen Molecular Probes). Labeling was     52  quenched by addition of 1mM DTT. Excess dye was removed by gel filtration as described in Dalal et. al. (2012) (Dalal et al. 2012).   2.2.3 SecY-SecA cross-linking  The SecYEG-SecA cross-linked complex was prepared in accordance with previously published protocols, with minor modifications (Bauer et al. 2014, Whitehouse et al. 2012). The SecY cysteine mutation K268C was introduced into pEYG by site-directed mutagenesis. After transformation into BL21, membrane extracts bearing the over-expressed SecY complex was isolated from 6L of culture and resuspended in TSG buffer. Endogenous SecA was removed by incubating the membranes on ice for one hour in Buffer A. The membrane fraction was recovered by ultracentrifugation (55k rpm, 4\u00baC, Beckman Ti70 rotor) and washed gently with 10 mL TSG to remove residual Urea. The membrane pellet was then resuspended in TSG. Urea-treated membranes (50 mg) were incubated with purified \"N95\" 795C SecA (50 mg) on ice for 1 hour in TSG buffer supplemented with 5mM MgCl2. After the initial hour-long incubation, 0.1mM Copper Phenanthroline (CP3) was added to induce cross-link formation. After a further 1-hour incubation, membranes were pelleted by ultracentrifugation (55k rpm, 4\u00baC, Beckmann Ti70 rotor), and the membranes gently washed in 10 mL TSG buffer to remove residual CP3. The membrane pellet was then resuspended in TSG buffer supplemented with 1% DDM. The cross-linked complex was then purified by Nickel affinity chromatography as described above for SecYEG.      53  2.2.4 Proteoliposome reconstitutions  Total E. coli lipid extract (100mg; Avanti Polar Lipids) dissolved in chloroform was dried under a stream of nitrogen and dessicated overnight. The resulting lipid film was resuspended in Buffer C (50mM Tris HCl pH 7.9, 50mM NaCl) to a final concentration of 10mg\/mL and sonicated to homogeneity. Unilamellar vesicles were prepared by extrusion through a 100nm filter at 300 psi using a Lipex 10\/1.5mL Thermobarrel and Regular Barrel Extruder (Northern Lipids Inc.). The size of the resulting liposomes was verified by Dynamic Light Scattering (DLS) using a Zetasizer (Malvern Technologies).  A solution of Triton X-100 was titrated into a 0.5mg aliquot of the extruded liposomes to determine the concentration of detergent required to swell but not disrupt the vesicles. Liposome swelling was assessed by monitoring the turbidity of the solution at OD550 using a UV spectrophotometer. Optimal swelling was observed at a concentration of 0.1% Triton X-100. To form proteoliposomes, 0.5mg extruded liposomes was treated with 0.1% Triton X-100. The purified SecYEG complex (1nmol) was added to this detergent-lipid mixture and incubated for 10 minutes on a rocking platform at 4\u00baC. Liposomes were incubated with a 1\/10 volume of BioBeads overnight to facilitate detergent removal. Proteoliposomes were recovered by ultracentrifugation (55krpm, Beckman TLA55 rotor) and resuspended in Buffer C. Co-reconstitution of SecYEG with wild type or PrlD23 SecA was performed as previously described with minor modifications (Bariya and Randall 2018, Mao et al. 2013). Briefly, purified SecYEG (1nmol) was mixed with either wild-type or PrlD mutant SecA (1nmol each) for 30 minutes at 40C before being added to the lipids. The co-reconstituted preparations are termed SecYEG-SecA and SecYEG-PrlD throughout this chapter.      54  2.2.5 Protein translocation assays  Translocation assays were performed essentially as previously described (Dalal et al. 2009, Dalal et al. 2012). Briefly, SecYEG proteoliposomes (2\u03bcg) or SecYEG-containing IMVs (5\u03bcg) were mixed on ice with SecA (1.5\u03bcg), fluorescently labeled substrate (2ng) and ATP (2mM) in 50\u03bcL of TL buffer (50mM Tris HCl pH 7.9, 50mM NaCl, 5mM MgCl2). A negative control was performed in parallel containing all reaction components except ATP. Reactions were incubated at 30\u00baC in a water bath for the indicated time before being chilled on ice. All un-transported substrate was degraded by incubation with Proteinase K (5\u03bcg; Bioshop) on ice for 10 minutes. The reactions were precipitated with ice-cold TCA (17% final, 30 minutes) followed by centrifugation (13.2 krpm, 10 minutes, 4\u00baC). The pellets were washed with ice-cold acetone before being dissolved in SDS-PAGE sample buffer and analyzed on 15% SDS-PAGE. Transported proteins were visualized by fluorescence scanning using a LI-COR Odyssey (LI-COR Biosciences).   2.2.6 Analysis of Translocation Data  Densitometry analysis of the translocation assays was performed using the Li-COR Odyssey software. Densitometry for full-length translocation was determined by selecting the area of each lane labeled full-length (FL). The data was normalized to the densitometry obtained for fully translocated material with WT SecA at the final time-point in each assay. The fluorescent intensity of this signal was arbitrarily assigned a value of 100. Densitometry for the translocation intermediates was obtained by selecting the area of each lane marked \u201cintermediates\u201d. As for full-    55  length translocation, the fluorescent intensity within this area was plotted relative to the intensity obtained for fully translocated material with WT SecA at the final time-point in each assay. Each assay was repeated at least two times to ensure reproducibility. The resulting data was plotted using GraphPad Prism 6 software (GraphPad, San Diego CA). Error bars represent standard deviations (SD).   Plasmid name Protein encoded Position of cysteine residues used for labeling Remarks Reference pBad22-hisEYG SecYEG N\/A None (Young and Duong 2019, Carlson et al. 2018, Dalal et al. 2012) pBad22-hisEYG (SecY K268C) Mutant derivative of SecYEG N\/A Mutation K268C in SecY to facilitate crosslinking (Young and Duong 2019) pET23-SecAhis SecA N\/A None (Duong 2003) pET23-PrlD23his SecA PrlD N\/A contains a PrlD point mutation (Y134S) (Young and Duong 2019) pET23-SecA \"N95\" 795C SecA \"N95\" 795C N\/A SecA truncation missing the 95 C-terminal amino acid residues; cysteine mutation at 795C to facilitate covalent crosslinking to SecYEG (Young and Duong 2019, Whitehouse et al. 2012, Bauer et al. 2014)     56  pBad33-PhoA proPhoA 200C, 307C, 357C contains 3 native cysteine residues (Young and Duong 2019) pTrc99a-proOmpA proOmpA 129C contains an engineered cysteine residue at position 108 (Duong and Wickner 1997b, 1999, 1997a) pTrc99a-pOAss-PhoA Recombinant substrate pOAss-PhoA 200C, 307C, 357C proOmpA signal sequence (ss) fused to the PhoA mature sequence (Young and Duong 2019) pBad33-PhoA202-OmpA Recombinant substrate: PhoA202-OmpA  200C, 308C residues 1-202 of proPhoA fused with residues 22-346 of proOmpA (Young and Duong 2019) pTrc99a-pOA199-PhoA Recombinant substrate: proOmpA199-PhoA 129C, 304C, 354C residues 1-199 of proOmpA fused with residues 203-472 of proPhoA (Young and Duong 2019) pBad33-pOAFL-PhoACT Recombinant substrate: proOmpA fused with a C-terminal PhoA fragment 129C, 451C, 501C Full-length proOmpA fused with residues 203-472 of proPhoA (Young and Duong 2019) pET23-PhoA202 Recombinant substrate: PhoA202 200C Residues 1-202 of proPhoA (Young and Duong 2019, Dalal et al. 2012)  Table 2.1 List of constructs employed in Chapter 2 The table describes all plasmid constructs used in Chapter 2 of this thesis.         57  2.3 Results  2.3.1 Effect of a SecA mutation which stabilizes SecA-SecYEG interactions  We reconstituted purified E. coli SecYEG into extruded 100 nm E. coli liposomes as described in Materials and Methods. Translocation assays were performed in the presence of ATP, fluorescently labeled proPhoA, and either wild-type or PrlD SecA. The data is reported on Figure 2.1. With the PrlD mutant, a reproducible transport defect is observed - the amount of fully transported substrate measured in the presence of the PrlD mutant is only 60% of the amount observed with wild-type SecA. Additionally, there is greater accumulation of translocation intermediates - partially transported substrate fragments - in the presence of the PrlD mutant. The apparent molecular weight of these intermediates is ~35 kDa. Although some intermediates are formed in the presence of wild-type SecA, these are far less prominent and do not form a discrete band, in contrast to the intermediates seen with the PrlD mutant (Figure 2.1A, quantified in B). Similar results were obtained when we performed translocation assays using IMVs containing over-expressed SecYEG instead of proteoliposomes (Figure 2.1C, quantified in D). These data led us to hypothesize that the PrlD mutation stabilizes SecA-SecY interactions at the membrane surface, thereby reducing SecA processivity and as a consequence, increasing formation of translocation intermediates while reducing full translocation.        58      0 2 4 6 8 10020406080100120Time (minutes)+ SecA+ PrlDTranslocationActivityFull-Length Translocation0 2 4 6 8 10020406080100120Time (minutes)+ SecA+ PrlDTranslocationActivityTranslocation IntermediatesBWT SecA SecA Y134S (PrlD SecA) 10% PhoA inputSecA Added10 2 4 6 10 2 4 6 10 Time at 300 (min.)- + + ATP0 5 10 15050100150Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDensitometry Analysis of Full-Length Translocation0 5 10 15050100150Densitometry Analysis of TranslocationIntermediatesTime (minutes)WT SecA IntPrlD SecA IntTranslocationActivityFigure 1A.0 5 10 15020406080100120Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDensitometry Analysis of Full-Length Translocation0 5 10 15020406080100120Time (minutes)Densitometry Analysis of TranslocationIntermediatesWT SecA IntPrlD SecA IntTranslocationActivityB.0 5 10 15020406080100120Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDensitometry Analysis of Full-Length Translocation0 5 10 15020406080100120Time (minutes)Densitometry Analysis of TranslocationIntermediatesWT SecA IntPrlD SecA IntTranslocationActivityC.WT SecA SecA Y134S (PrlD SecA) 10% PhoA inputSecA Added10 2 4 6 10 2 4 6 10 Time at 300 (min.)- + + ATP0 5 10 15050100150Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDensitometry Analysis of Full-Length Translocation0 5 10 15050100150Densitometry Analysis of TranslocationIntermediatesTime (minutes)WT SecA IntPrlD SecA IntTranslocationActivityi  0 5 10 15020406080100120Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDe sitometry Analysis of Full-Length Translocation0 5 10 15020406080100120Time (minutes)Densitometry Analysis of TranslocationIntermediatesWT SecA IntPrlD SecA IntTranslocationActivityB.0 5 10 15020406080100120Time (minutes)WT SecA FLPrlD SecA FLTranslocationActivityDensitometry Analysis of Full-Length Translocation0 5 10 15020406080100120Time (minutes)Densitometry Analysis of TranslocationIntermediatesWT SecA IntPrlD SecA IntTranslocationActivityC.ATP:     -      +      +       +      +    +       +       +       +Time:  10     2      4       6     10     2       4        6      10                                                                                            1       2     3       4      5   6       7        8       9   10%-46-32-58-25ASecA PrlDIntermediatesPhoA     FLA0 2 4 6 8 10020406080100120Time (minutes)+ SecA+ PrlDTranslocationActivityFull-length Translocation0 2 4 6 8 10050100150200250300Time (minutes)+ SecA+ PrlDTranslocationActivityTranslocation IntermediatesDATP:     -         +         +        +        +    +          +        +         +Time:    10     2         4        6       10    2          4        6        10                                                                                                        1         2        3        4        5   6        7        8        9   10%C-58-46-32-25SecA PrlDIntermediatesPhoA     FLC    59  Figure 2.1 Translocation activity of PrlD23 with the substrate PhoA (A) Translocation assays were performed using SecYEG proteoliposomes under conditions as described in Materials and Methods. Fluorescently labeled PhoA (10% of the total input for one reaction) was loaded onto the gel as a standard. Positions of fully translocated substrate (labeled FL) and of the translocation intermediates (labeled intermediates) are indicated with arrows. (B) Densitometry analysis of (A). Data were normalized to 100 units based on the densitometry obtained for fully translocated PhoA with WT SecA after 10 minutes at 300C (lane 5). The data represents the mean \u00b1 SD from 3 independent experiments. (C) Translocation assays were performed as in (A) but using IMVs containing over-expressed SecYEG. (D) Densitometry analysis of (C).  2.3.2 PrlD is able to out-compete wild-type SecA at SecYEG  We next assessed the relative stabilities of SecYEG-SecA and SecYEG-PrlD complexes using a competition assay. We reconstituted pre-formed SecYEG-SecA and SecYEG-PrlD complexes into proteoliposomes as described in Materials and Methods. The reconstitution efficiency of both preparations was comparable, as assessed by SDS-PAGE (Figure 2.2A). We also verified that SecA and SecYEG remains in complex during the co-reconstitution process by measuring the translocation activity without additional SecA. As seen earlier (Figure 2.1), the amount of the 35 kDa translocation intermediates formed with PrlD SecA (Figure 2.2B, lanes 5-8) is appreciably higher than the amount formed with the wild-type enzyme (lanes 1-4). In both cases however, translocation does not require additional SecA, showing that SecA and SecYEG remain a functional complex during the reconstitution process, in agreement with a previously published study (Mao et al. 2013, Bariya and Randall 2018).       60  The co-reconstituted SecYEG-SecA complexes were subsequently incubated with a stoichiometric excess of wild-type or PrlD SecA (Figure 2.2C).  With wild-type SecA, the protein translocation profile is similar to that seen in Figure 2.2B, left panel.  In contrast, with PrlD, there is less proPhoA transported and more translocation intermediates (Figure 2.2C, lanes 5-7; quantified in Figure 2.2D). The reciprocal experiment, wherein stoichiometric excess of SecA is added to the pre-formed SecYEG-PrlD complex, was performed in parallel (Figure 2.2C, quantified in Figure 2.2D). In that case, the protein translocation profile is unchanged with both SecA and PrlD. As expected, addition of PrlD SecA to the pre-formed SecYEG-PrlD complex has no effect (Figure 2.2C, lanes 12-14).      61   Figure 2    A     B   C     D                                     1  2     3    4      5        6        7        8       10%            Figure 2YEG+A YEG+PrlD Proteoliposomes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes)A.C. 2\u00b5g of each type of proteoliposomes used in the assay above were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue staining.  -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.Figure 2YEG+A YEG+PrlD Proteoliposomes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes)A.C. 2\u00b5g of each ty e of proteolipo om s used in the assay abov  were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue staining.  -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.PhoAIntermediates    ATP          -          +          +         +            -            +          +         +    Time        15         2          6        15          15           2          6        15                       Proteo         SecYEG-SecA                SecYEG-PrlD               1    2      3 4    5      6  7     8       9   10      11  12      13     14                     PhoAIntermediates15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.                              Proteo           ecYEG-SecA                   SecYEG-PrlD          ATP    -       +       +      +      +       +       +   -       +       +        +        +       +        +      Time   15      2       6     15     2       6      15  15     2       6       15       2       6       15                                                + SecA        + PrlD       + SecA        + PrlD0 5 10 15020406080100120Time (minutes)+ SecATranslocation Activity + PrlDYEG-A   Full Translocation0 5 10 15020406080100120Time (minutes)+ SecA+ PrlDTranslocation ActivityYEG-PrlD Full Translocation0 5 10 15020406080100120Time (minutes)YEG-A Translocation Intermediates+ SecA+ PrlDTranslocation Activity0 5 10 15020406080100120Time (minutes)+ SecA+ PrlDTranslocation ActivityYEG-PrlD Translocation Intermediates                                    1  2     3    4      5        6        7        8       10%            FiYEG+A YEG+PrlD Proteoliposomes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes).C. 2\u00b5g of each type proteolip somes used in the assay above were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue staini g.  -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.YEG+A YEG+PrlD Proteoliposomes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes).C. 2\u00b5g of each ty e of prote li o om s used in the assay abov  were analyz d by 15% SDS-PA  ed by Co massie Brilliant Blue staining. -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.PhoAIntermediates    ATP          -          +          +         +            -            +          +         +    Time        15         2          6        15          15           2          6        15                       Proteo         SecYEG-SecA                SecYEG-PrlD                                      1  2     3    4      5        6        7        8       10%        Figure 2YEG+A YEG+PrlD Proteoliposo es- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes)A.C. 2\u00b5g of each type of proteoliposomes used in the assay above were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue taining.  -SecA -SecY -SecE\/G +     YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification o PhoA Transl catioY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 202406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl LsTranslcation ctivity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Transl cation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.Figure 2+  + rl roteoliposo es- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 ime at 300 (minut s)A.C. 2\u00b5g of each ty e of proteolipo om s used in the assay abov  were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue staining.  -SecA -SecY -SecE\/G E +A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 0 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.PhoAIntermediates    ATP          -          +          +         +            -            +          +         +    Time        15         2          6        15          15           2          6        15                                     1    2      3 4    5      6  7     8       9   10      11  12      13     14                     PhoAIntermediates15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-lip somes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.                            Proteo           ecYEG-SecA                   SecYEG-PrlD          ATP    -       +       +            +       +       +   -       +       +        +        +       +        +      Time   15      2       6     15     2       6      15  15     2       6       15       2       6       15                                                + SecA        + PrlD       + SecA        + PrlD             1    2      3 4    5      6  7     8       9   10      11  12      13     14                     PhoAIntermediates15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.                              Proteo           ecYEG-SecA                   SecYEG-PrlD          ATP    -       +       +      +      +       +       +   -       +       +        +        +       +        +      Time   15      2       6     15     2       6      15  15     2       6       15       2       6       15                                                + SecA     + PrlD       SecA      + PrlD             1    2      3 4    5      6  7     8       9   10      11  12      13     14                  PhoAIntermediates15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.15 2 6 15 2 6 15 5% PhoA15 2 6 15 2 6 15 5% PhoA- + - + ATPYEG + PrlD SecA YEG + WT SecA Proteo-liposomes+ WT SecA + PrlD SecA + WT SecA + PrlD SecA SecA addedFigure 3A.                      Proteo          ecYEG-SecA             SecYEG-PrlD          ATP    -       +       +      +      +       +       +   -       +       +        +        +       +        +      ime   15      2       6     15     2       6      15  15     2       6       15       2       6       15                                                + SecA        + PrlD       + SecA        + PrlDFLFL\t \t \t \t \t \t \t \t \t \t  \t \t \t \t \t \t \t \t      \t \t \t \t \t \t \t \t \t \t     Proteoliposomes:      SecYEG-SecA   SecYEG-PrlDSecASecY          SecE\/G      Proteoliposomes YEG+A YEG+PrlDFigure 2YEG+A YEG+PrlD Proteoliposomes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes)A.C. 2\u00b5g of each type of proteoliposomes used in the assay above were analyzed by 15% SDS-PAGE followed by Coomassie Brilliant Blue staining.  -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation Inter ediatesY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl PLsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation IntermediatesY+A PLsY+Prl PLsTranslcation ActivityD.SecASecY   SecE\/G   Proteoliposomes YEG+A YEG+PrlDFigure 2YEG+A YEG+PrlD Proteolipos mes- + 5% PhoA - + 5% PhoA ATP15 2 6 15 15 2 6 15 Time at 300 (minutes)A.C. 2\u00b5g of each type of prote liposomes u ed in the ass y bove were analyzed by 15% SD -PAGE foll wed by Coomassie Brilliant Blue staining.  -SecA -SecY -SecE\/G YEG+A    YEG+PrlD B.C.0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl LsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation I termediatesY+A PLsY+Prl LsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA TranslocationY+A PLsY+Prl LsTranslcation Activity0 5 10 15 20020406080100120Time (minutes)Quantification of PhoA Translocation I termediatesY+A PLsY+Prl LsTranslcation ActivityD.- SecA- Se- SecE\/G    62  Figure 2.2 Competition assay(s) between PrlD SecA and wild-type SecA (A) To verify the efficiency of our co-reconstitution procedure, 4\u00b5g co-reconstituted SecYEG-SecA and 4\u00b5g co-reconstituted SecYEG-PrlD were analyzed by 15% SDS-PAGE followed by Coomassie Blue staining. (B) Translocation assays were performed with 4 \u00b5g SecYEG co-reconstituted in liposomes with SecA (SecYEG-SecA) or with PrlD SecA (SecYEG-PrlD) without addition of extra SecA.  (C) Assays were performed as described in (B) in the presence of 1.5 \u00b5g additional SecA or PrlD SecA, as indicated. (D) Densitometry analysis of (C). The plotted data represents the mean \u00b1 SD for 2 independent experiments.   These data indicate that PrlD can replace SecA during the translocation process but not the opposite, implying that the functional PrlD-SecYEG complex is more stable than SecA-SecYEG complexes. This is in agreement with an earlier study which showed that PrlD SecA stably inserts into the membrane at SecYEG to a much greater extent than does wild-type SecA (Gouridis et al. 2013). Importantly, the heightened stability of the SecYEG-PrlD complex is not due to an affinity difference between wild-type and PrlD SecA, but rather to the altered conformation of the PrlD mutant (Gouridis et al. 2013).     2.3.3 Covalent cross-linking of SecY-SecA mimics the effect of the PrlD mutation  We measured translocation of proPhoA when SecA processivity is limited via a cysteine crosslink between SecA and SecY. The cysteine variants were cross-linked together and purified according to the protocol of Whitehouse et. al (2012) before being reconstituted into proteoliposomes (Figure 2.3A) (Whitehouse et al. 2012). As expected, the SecA-SecY crosslink is reducible with DTT (Figure 2.3A). Translocation assays were then performed with the reconstituted cross-linked complex for 10 minutes at 30\u00b0C. In that case, there is accumulation of translocation     63  intermediates, which are not detected in the presence of DTT (Figure 2.3B, compare Lane 2 to Lane 3). These accumulated intermediates can however complete translocation upon addition of DTT (compare Lane 4 to Lane 5). Together, these results show that cleavage of the SecY-SecA covalent bond restores SecA processivity and thereby promotes the forward movement of the jammed translocation intermediates. Quantifications of the fully and partially transported species is reported on Figure 2.3C.      64   Figure 2.3 Translocation activity of a covalently linked SecA-SecYEG complex (A) The crosslinked SecYEG-SecA complex reconstituted into proteoliposomes (4\u00b5g) was analyzed by 12% SDS-PAGE and Coomassie blue staining with DTT (1mM) as indicated. (B) Translocation assays were performed as described for Figure 2.2. All incubations were for 10 minutes at 300C. The translocation intermediates formed in lane 3 were re-incubated for an additional 10 minutes with or without DTT as indicated (lanes 4 and 5, respectively). (C) Densitometry analysis of (B). Data were normalized to 100 units based on the densitometry obtained for fully translocated PhoA in lane 2. The plotted data represents the mean \u00b1 SD for 3 independent experiments.   Figure 3      A         C                                           -SecA x SecY                            -SecA                                  -SecY             DTT       -        + S cY-SecA - SecA -SecY - + 1mM DTT1 2 3 4 50100200300400500LaneFull-length TranslocationTranslocationActivity1 2 3 4 50100200300400500LaneTranslocationActivityTranslocation IntermediatesFL    65  2.3.4 The importance of SecA-SecY processivity varies between preprotein substrates  The results above indicate that SecA processivity is important for proPhoA translocation. To determine whether SecA processivity is required for translocation of a different substrate, we employed proOmpA. In contrast to proPhoA, translocation of proOmpA seems to be uninfluenced by SecA processivity. ProOmpA translocation proceeds modestly more efficiently in conditions where SecA processivity is restricted - either by the PrlD23 mutant (Figure 2.4A) or by the SecA-SecYEG crosslinked complex (Figure 2.4C).      66   Figure 2.4 Translocation activity of PrlD23 SecA with the substrate proOmpA (A) Fluorescently labeled proOmpA was employed in translocation assays as in Figure 2.1. (B) Quantification of (A). (C) ProOmpA was employed in translocation assays with the SecYEG-SecA cross-linked complex. Proteoliposomes, ATP and labeled substrate were incubated at 300C for the indicated times in the presence or absence of 1mM DTT. 50\u00b5L aliquots were withdrawn at the indicated timepoints before being analyzed as described for Figure 2.1. (D) Quantification of (B). 100 units of translocation activity is defined as the level of fully translocated proOmpA observed after 15 minutes in the presence of ATP and 1mM DTT (lane 7). Figure 4 A    B   C   D0 5 10 15020406080100120140Time (minutes) proOmpA Translocation+ SecA + PrlDTranslocationActivity                1        2       3     4                 5          6         7                 20%      Figure 5 Cont\u2019dFigure 4:  A. The importance of SecA-SecY processivity varies between preprotein substrates. Strikingly, the importance of SecA-SecY processivity varies depending on the identity of the translocation substrate. Inhibition of SecA-SecY processivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY processivity is not inhibited.  B. Artificial abolition of SecY-SecA processivity via the SecY-SecA covalent cross-link described for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010015Quan ification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DTT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 20020406080100120Time (minutes)Quantification of proOmpA Translocation-DTT+DTTTranslocationActivityFigure 4:  A. The importance of SecA-SecY processivity varies between preprotein substrates. Strikingly, the importance of SecA-SecY processivity varies depending on the identity of the translocation substrate. Inhibition of SecA-SecY processivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY processivity is not inhibited.  B. Artificial abolition of SecY-SecA processivity via the SecY-SecA covalent cross-link described for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationrl  SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DTT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 20020406080100120Time (minutes)Quantification of proOmpA Translocation-DTT+DTTTranslocationActivi y.C.i re 5 t\u2019Figure 4: A. The importance of SecA-SecY proces ivity varies betwe n preprotein substrates. Strikingly, the importance of SecA-SecY proces ivity varies depending on the identity of the translocation substrate. Inhibition of SecA-SecY proces ivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY proces ivity is not inhibited.  B. Artificial abolition of SecY-SecA proces ivity via the SecY-SecA covalent cros -link descr bed for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Qua ification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantif cation of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 2002040608010120Time (minutes)Quantification of proOmpA Translocation-DT+DTTranslocationActivityFigure 4: A. The i portance of SecA-SecY proces ivity varies betwe n preprotein substrates. Strikingly, the importance of SecA-SecY proces ivity varies depending on the identity of the translocation substrate. Inhibition of SecA-SecY proces ivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY proces ivity is not inhibited.  B. Artificial abolition of SecY-SecA proces ivity via the SecY-SecA covalent cros -link described for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 2002040608010120Time (minutes)Quantification of proOmpA Translocation-DT+DTTranslocationActivity..pOA  ATP        -            +            +           +                 +             +           + Time       15           2            6          15                2             6          15                       _- DTT      + DTT0 5 10 15020406080100120Time (minutes)- DTT+DTTTranslocationActivity proOmpA Translocation                 1        2        3       4       5    6            7          8          9         10        11       12   pOA  ATP       -         +          +         +         +          +                -     +            +            +           +            +  Time     15        2         4         6        10        15               15    2            4            6          10  15                                        WT SecA         PrlD SecA               1        2       3     4                 5          6         7                 20%      Figure 5 Cont\u2019dFigure 4:  A. The importance of SecA-SecY processivity varies between preprotein substrates. Strikingly, the importance of SecA-SecY processivity varies dep nding on the identity of the translocation substrate. Inhibition of SecA-SecY processivity via the PrlD mutation leads to higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY processivity is not inhibited.  B. Artificial abolition of SecY-SecA processivity via the SecY-SecA covalent cross-link described for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD S cA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quan ification of proOmpA Tr nslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 5 10 2 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DTT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 20020406080100120Time (minutes)Quantification of proOmpA Translocation-DTT+DTTTranslocationActivityFigure 4:  A. The importance of SecA-SecY processivity varies between preprotein substrates. Strikingly, the importanc  of A-SecY processivity varies d pending on the identity of the translocation substrate. Inhibition of SecA-SecY processivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY processivity is not inhibited.  B. Artificial abolition of SecY-SecA processivity via the SecY-SecA covalent cross-linkdescribed for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP10 15 20050100150Quantification of proOmpA Translocationi  ( i t )WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 20050100150Quantification of proOmpA TranslocationTime ( inutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10  r1  TT- +6 10 5 10 15 20020406080100120Time (minutes)Quantification of proOmpA Translocation-DTT+DTTTranslocationActivity.C.i re 5 t\u2019Figure 4: A. The importance of SecA-SecY proces ivity varies betwe n preprotein substrates. Strikingly, the importance of SecA-SecY proces ivity varies dep nding on the identity of the translocation substrate. Inhibition of SecA-SecY proces ivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY proces ivity is not inhibited.  B. Artificial abolition of SecY-SecA proces ivity via the SecY-SecA covalent cros -link described for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD S cA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Qua ification of proOmpA Tr nsloca ionTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA added 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantif cation of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10% proOmpA1mM DT- + + ATP15 2 6 15 2 6 15 Time at 300 (minutes)0 5 10 15 2002040608010120Time (minutes)Quantification of proOmpA Translocation-DT+DTTranslocationActivityFigure 4: A. The i portance of SecA-SecY proces ivity varies betwe n preprotein substrates. Strikingly, the importanc  of SecA-SecY proces i ity varies depending n the id ntity of the translocation substrate. Inhibition of ecA-Se Y proces ivity via the PrlD mutation leads to a higher level of proOmpA translocation (the precursor of Outer Membrane Porin A) than is observed when SecA-SecY proces ivity is not inhibited.  B. Artificial abolition of SecY-SecA proces ivity via the SecY-SecA covalent cros -linkdescribed for Figure 4 yields a similar result:  WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantification of proOmpA TranslocationTi e ( i t )WT SecA translocationPrlD SecA translocationTranslocationActivity WT SecA 10% pOASecA Y134S (PrlD SecA)10% pOASecA ad ed 10 2 4 6 10 15 10 2 4 6 10 15 Time at 300 (min.)- + - + ATP0 5 10 15 2005010150Quantification of proOmpA TranslocationTime (minutes)WT SecA translocationPrlD SecA translocationTranslocationActivity 37 kDa- + 10  r1 T- + +15 2 6 15 2 6 150 5 10 15 2002040608010120i  ( i t )Quantification of proOmpA Translocation-DT+DTTranslocationActivity..pOA  ATP        -            +            +           +                 +             +           + Time       15           2            6          15                2             6          15                           - DTT      + DTT                1        2        3       4       5    6            7          8          9         10        11       12   pOA  ATP       -         +          +         +         +          +                -     +            +            +           +            +  Time     15        2          4         6        10        15              15    2            4            6          10  15                                       WT SecA         PrlD SecAFLFL    67    2.3.5  Influence of substrate leader peptide and mature domain on SecA  Why is an effect of SecA processivity apparent with the proPhoA substrate but not with proOmpA? We determined the influence of both the substrate leader peptide and the mature domain on SecA processivity. We attached the signal sequence of proOmpA onto the PhoA mature domain (pOAss-PhoA; Figure 2.5A). As seen with proPhoA, a significant amount of pOAss-PhoA translocation intermediates (size ~ 35 kDa) is formed with PrlD SecA. Thus, variations in the substrate signal sequence do not influence SecA processivity.      68   Figure 2.5 Effect of the substrate leader peptide on SecA processivity (A) Cartoon representations of proPhoA and the pOAss-PhoA fusion protein. The leader peptide of proOmpA (depicted in red) was fused to the mature sequence of PhoA (depicted in blue). The molecular weights of each protein are indicated on the right-hand side of the figure. (B) pOAss-PhoA was fluorescently labeled and employed in translocation assays with SecYEG IMVs as described in Figure 2.1C. (C) The amount of fully translocated products and translocation intermediates were quantified as described for Figure 2.1.          A         B    C  1 PhoA (23-471)MW: 49 kDa1 471MW: 49 kDaAproPhoA:pOAssPhoA:471Figure 6 A   B         1         2         3        4       5           6           7          8          9       10%       pOAss-PhoAIntermediates  ATP         -           +            +            +             +              +            +             +            +  Time       10          2            4            6            10             2            4             6            10                                WT SecA          PrlD SecA         1         2         3        4       5           6           7          8          9       10%       pOAss-PhoAIntermediates  ATP         -           +            +            +             +              +            +             +            +  Time       10          2            4            6            10             2            4             6            10                                WT SecA          PrlD SecA0 2 4 6 8 10050100150200250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivityFL- 58- 46- 32- 25Figure 6 A   B         1         2         3        4       5           6           7          8          9       10%       pOAss-PhoAIntermediates ATP - +                          +  +      +  Time 10 2                          4  6     10                            PrlD SecA         1         2         3        4       5           6           7          8          9       10%       pOAss-PhoAIntermediates  ATP         -           +            +            +             +              +            +             +            +  Time       10          2            4            6            10             2            4             6            10                                WT SecA          PrlD SecA0 2 4 6 8 1050150250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 1050150250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivityFL- 58- 46- 32- 25    69  We next exchanged portions of the mature domains of proPhoA and proOmpA. We generated two constructs, PhoA202-OmpA and pOA199-PhoA, which have similar molecular weights as wild-type proPhoA (Fig. 2.6A). In the presence of the PrlD mutant, both of these substrates form the same ~ 35 kDa translocation intermediate observed with PhoA (Figure 2.6, panels B and D; quantified in C and E). This result suggests that the substrate sequence does not influence SecA processivity.      70   Figure 2.6 Effect of substrate mature domain on SecA processivity (A) Cartoon representations of proPhoA, proOmpA, and the two fusion substrates generated in this study. The molecular weights of each protein are indicated on the right-hand side of the figure. (B)  PhoA202-OmpA - a fusion between the N-terminal 202 residues of proPhoA and the mature sequence of OmpA - was fluorescently labeled and employed in translocation assays as   A        proPhoA:proOmpA:PhoA202-OmpA:pOA199\u2014PhoA:1 4711 3461 2021 199OmpA (22-346)PhoA (203-471)526467MW: 49 kDaMW: 37 kDaMW: 56 kDaMW: 51 kDa  A       B   C  D  EPhoA:proOmpA:PhoA202-OmpA:pOA199\u2014PhoA:1 202 4711 3461 2021 199199OmpA mature domain: 22-346PhoA C-terminus: 203-471       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP       -           +          +          +          +              +          +           +         + Time      10          2          4          6         10             2          4           6         10                             WT SecA        PrlD SecA        1         2         3        4      5         6          7           8        9                               pOA199 phoAIntermediates  ATP       -          +            +             +           +              +             +           +          + Time      10         2            4             6          10             2             4           6          10                                     WT SecA             PrlD SecA       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP       -           +          +          +          +              +          +           +         + Time      10          2          4          6         10             2          4           6         10                             WT SecA        PrlD SecA0 2 4 6 8 10050100150200250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivityFLFL       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP       -           +          +          +          +              +          +           +         + Time      10          2          4          6         10             2          4           6         10                             WT SecA        PrlD SecAMW:  49kDaMW:  51kDaMW:  56kDaMW:  37kDa- 58- 46- 32- 25  A       B   C  D  EPhoA:proOmpA:PhoA202-OmpA:pOA199\u2014PhoA:1 202 4711 3461 2021 199199OmpA mature domain: 22-346PhoA C-terminus: 203-471       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP      -          +         +         +        +             +          +           +         + Time      10          2          4          6         10             2          4           6         10                             WT SecA        PrlD SecA        1         2         3        4      5         6          7           8        9                               pOA199 phoAIntermediates  ATP       -          +            +             +           +              +             +           +          + Time      10         2            4             6          10             2             4           6          10                                     WT SecA             PrlD SecA       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP   -       +   +          +   +       +        +   +        + Time    10        2   4        6    10      2       4   6         10                   WT SecA        PrlD e0 2 4 6 8 10050100150200250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocationActivity0 2 4 6 8 10050100150200250Time (minutes)Translocation Intermediates+ SecA+ PrlDTranslocationActivityFLFL       1        2       3    4       5        6      7          8        9              PhoA202OmpAIntermediates  ATP       -           +          +          +          +              +          +           +         + Time      10          2          4          6         10             2          4           6         10                             WT SecA        PrlD SecAMW:  49kDaMW:  51kDaMW:  56kDaMW:  37kDa- 58- 46- 32- 25    71  described in Figure 2.5. (C) Quantification of (B). (D) proOmpA199-PhoA - a fusion between the N-terminal 1-199 residues of proOmpA and residues 203-471 of PhoA - was fluorescently labeled and employed in translocation assays as described in Figure 2.5. (E) Quantification of (D).  2.3.6 Influence of substrate length on SecA processivity  Given the difference in size between proOmpA (346 residues, 37 kDa) and proPhoA (471 residues, 49 kDa), we examined if SecA processivity is influenced by the length of the substrate. We constructed an extended variant of proOmpA by fusing a C-terminal fragment of PhoA (PhoA residues 203-471) onto proOmpA. This generated the substrate pOAFLPhoACT (614 residues, 68 kDa; Figure 2.7A).  We have shown in the section above that the 35 kDa translocation intermediates observed with proPhoA is not dependent on this fragment of PhoA (compare wild-type PhoA with PhoA202-OmpA). Thus, any new translocation intermediates we observe with this fusion protein can be attributed to its size, as opposed to its sequence. In the presence of wild-type SecA, we observe time-dependent accumulation of both fully translocated material as well as a ~50 kDa translocation intermediate. We observe less fully translocated material in the presence of the PrlD mutant, along with more pronounced accumulation of the 50 kDa translocation intermediate (Figure 2.7B and C). Thus, it appears that substrate length does influence SecA processivity.       72   Figure 2.7 Effect of substrate length on SecA processivity (A) Cartoon representations of the recombinant substrate pOAFL PhoACT, using the same colour scheme as Figure 2.7A. A cartoon representation of the substrate PhoA is included here for reference. The molecular weights of each protein are indicated on the right-hand side of the figure. (B)  pOAFL PhoACT was fluorescently labeled and employed in translocation assays as described in Figure 2.6. (C) Quantification of (B).                      A     1 346 PhoA (203-471) 614MW: 68 kDa1 471MW: 49 kDaAproOmpAproPhoA:pOAFLPhoACT:BC0 2 4 6 8 10020406080100120140Time (minutes)Full-length Translocation+ SecA+ PrlDTranslocation Activity0 2 4 6 8 10020406080100120140Time (minutes)+ SecA+ PrlDTranslocation ActivityTranslocation IntermediatesIntermediatesATP:    -     +      +     +    +      +      +Time:  10     2      6      10   2      6    10                                                                     1      2     3     4  5      6      7  10%-80-58-46-32  WT SecA   PrlD SecApOAFLPhoACT   FL    73  To accumulate more evidence for the possible role of substrate length, we also tested the effect of SecA processivity on translocation of an N-terminally truncated variant of proPhoA termed PhoA202 (202 residues, 22 kDa - Figure 2.8A) (Dalal et al. 2012). We expected that reducing the substrate length would decrease its dependence on SecA processivity. Consistent with this hypothesis, we find that translocation of PhoA202 is not influenced by SecA processivity. Indeed, as with proOmpA, we find that translocation of PhoA202 is modestly increased with the non-processive PrlD mutant (Figure 2.8B, quantified in 2.8C). We do not observe accumulation of translocation intermediates with this substrate. This was expected, given that the fluorescent probe is located close to the C-terminus of the protein (Figure 2.8, Table 2.1).      74   Figure 2.8A proPhoA truncation mutant is no longer dependent on SecA processivity (A) Cartoon representations of the substrate PhoA202. A cartoon representation of the substrate proPhoA is included here for reference. The molecular weights of each protein are indicated on the right-hand side of the figure. (B)  PhoA202 was fluorescently labeled and employed in translocation assays as described in Figure 2.6. (C) Quantification of (B).        1 471MW: 49 kDaAproPhoA:PhoA202:1 202MW: 22 kDaBCWT SecA PrlD SecA050100150200250300350400Time (minutes)PhoA202 Translocation + WT SecA + PrlD SecATranslocation Activity  1        2        3ATP:     -           +          +Time:    10        10        10SecA:   WT      WT     PrlD   10%-25-22-17-11ABC    75  2.4 Discussion  We have examined the importance of successive cycles of SecA binding and dissociation from SecYEG - termed SecA processivity - using a SecA mutant (PrlD23) that associates more tightly with bacterial membranes than does wild-type SecA (Gouridis et al. 2013). Although previous work has shown that SecA undergoes cycles of binding and dissociation from SecYEG, this study was motivated by conflicting reports as to whether or not this processivity is an absolute requirement for substrate translocation (Mao et al. 2013, Morita, Tokuda, and Nishiyama 2012, Bauer et al. 2014).  By monitoring accumulation of translocation intermediates, we show that SecA processivity influences translocation of the substrate proPhoA, but not of proOmpA. To unravel the reason for this difference between the two substrates, we examined the effect of the substrate signal sequence and mature domain on SecA processivity. We generated a series of chimeric substrates by exchanging substrate signal peptides as well as portions of the substrate mature domains. The sizes of the chimeric substrates are comparable to that of proPhoA (~50 kDa), yet these chimeric substrates display the same dependence on SecA processivity as proPhoA, leading us to conclude that substrate sequence is likely not the main factor determining SecA processivity.   We next tested the possible effect of substrate length. It has been speculated earlier that preprotein substrates of differing lengths and differing amino acid compositions may behave differently during the translocation reaction (Tomkiewicz et al. 2006). However, this question has     76  not been fully addressed, particularly in the context of SecA processivity. We extended the size of proOmpA beyond that of proPhoA, generating the substrate termed pOAFLPhoACT (68 kDa). In that case, we observed some accumulation of translocation intermediates when SecA is able to act in a processive manner. Accumulation of these intermediates is increased in the presence of the PrlD mutant, suggesting that increasing the length of proOmpA increases its dependence on SecA processivity.   We also tested whether decreasing the length of proPhoA would decrease its dependence on SecA processivity. As expected, translocation with the shorter PhoA202 is not dependent on SecA processivity, further supporting our conclusion about the importance of substrate length. We note that earlier studies by Tomkiewicz et al. (2006) and Fessl et. al. (2018) examined the effect of substrate size using extended proOmpA deriviatives, but these studies did not assess accumulation of translocation intermediates (Tomkiewicz et al. 2006, Fessl et al. 2018). The authors employed either a fluorescent probe or an epitope tag located at the extreme C-terminus of their substrates, thus any intermediates that may have accumulated would not have been detectable (Tomkiewicz et al. 2006). In our study, we suspect that the 35 kDa PrlD-dependent translocation intermediates observed with proPhoA, as well as the ~ 50 kDa intermediates seen with pOAFLPhoACT, may represent rate-limiting steps in the translocation reaction. It could be, for instance, that ~ 35 kDa is the maximum length of proPhoA that can be translocated by SecA before the enzyme needs to dissociate from the membrane and SecYEG. In the case of the longer substrate - pOAFLPhoACT, however, we no longer observe this ~ 35 kDa intermediate. Instead, the size of the observed intermediates is shifted up to ~ 50 kDa. This could be because the translocase is able to \"sense\" the     77  length of the substrate, and that the size of the rate-limiting step varies according to preprotein length.   Altogether, our findings are in line with the processive model of protein translocation that has been previously proposed, wherein SecA cycles on and off of the membrane and SecYEG during the translocation reaction (Mao et al. 2013, Morita, Tokuda, and Nishiyama 2012). We further report that the importance of this on\/off cycling varies depending on the length of the translocating substrate. However, this is demonstrated using only two substrates. Therefore, a more comprehensive analysis, testing multiple substrates - on the scale of the recent studies by Bariya and Randall (2018), Chatzi et. al (2017) and Tsirigotaki et. al. (2018) - will be needed to determine whether SecA processivity is a universal phenomenon (Chatzi et al. 2017, Tsirigotaki et al. 2018, Bariya and Randall 2018).       78  Chapter 3: Utilizing the peptidisc membrane mimetic to identify novel transient interactors of the E. coli SecY complex  3.1 Introduction Membrane proteins are instrumental in vital cellular processes including energy production, nutrient import, cell-cell signaling, and protein translocation. While they occasionally act alone, membrane proteins often transiently associate with each other in macromolecular complexes while carrying out their functions (Stenberg et al. 2005, Babu et al. 2018, Maddalo et al. 2011, Gokhale et al. 2012, Carlson et al. 2019, Hu et al. 2019, Zhang et al. 2012).   Development of methods to study membrane protein interaction networks - or the \"membrane interactome\" - in conditions which mimic the native membrane environment is therefore critical for our understanding of these diverse cellular processes. Traditionally, in-depth analysis of the membrane interactome has been complicated by the necessity of extracting proteins from the membrane using of detergents.  Even the mildest detergents are known to disrupt native protein conformations and\/or dissociate transiently associated complexes, causing many potential interactions to be lost (Babu et al. 2018, Babu et al. 2012, Gokhale et al. 2012, Carlson et al. 2019, Maddalo et al. 2011, Chorev et al. 2018, Lee et al. 2018).   Various membrane mimetics have been developed to facilitate study of membrane proteins in the absence of detergents - most notably the nanodisc and SMALPs (D\u00f6rr et al. 2014, Roy et al. 2015, Marty et al. 2013, Carlson et al. 2019). While the nanodisc has been applied with some     79  success for characterizing both prokaryotic and eukaryotic membrane proteomes, the method is not without drawbacks. These include the need for adding exogenous lipids, as well as tailoring the length of the scaffold protein used in the reconstitution - changes in either of these parameters can bias reconstitution towards proteins and complexes of a certain diameter (Roy et al. 2015, Marty et al. 2013).   Thus, there is a clear need for development of an unbiased, more streamlined approach for reconstituting the membrane interactome.  To address this, our laboratory recently developed the peptidisc as a flexible, \"one-size-fits-all\" scaffold for membrane protein reconstitution (Carlson et al. 2018, Carlson et al. 2019). Upon removal of detergent, incorporation of both endogenous lipids and solubilized membrane proteins into water-soluble, detergent-free particles occurs in a single step (Figure 3.1).       80   Figure 3.1 Preservation of labile interactors of the Sec translocon in peptidisc Schematic depicting the Holo-Translocon (HTL) complex within the membrane. Addition of non-ionic detergent such as Dodecyl Maltoside (DDM) extracts the proteins from the membrane, but the complex is prone to dissociation in detergent into individual subcomplexes - SecYEG (red) SecDF (purple and green, respectively) and YidC (cyan). Reconstitution of the detergent extract into peptidisc, however, stabilizes the HTL assembly.   SolubilizationComplexDissociationPeptidisc Reconstitution+    81  We recently showed that a detergent-solubilized E. coli membrane proteome can be directly reconstituted into water soluble peptidiscs (Carlson et al. 2019). The resulting membrane protein particles are termed a \"peptidisc library\". By combining SILAC labeling with protein correlation profiling (PCP), we were able to rapidly characterize the peptidisc-reconstituted membrane interactome by quantitative proteomics (Carlson et al. 2019). We fractionated the SILAC-labeled peptidisc library by size-exclusion chromatography and used a bioinformatic-based approach to analyze co-fractionation profiles (Stacey et al. 2018, Carlson et al. 2019, Stacey et al. 2017). Briefly, the co-fractionation data was analyzed using the PrInCE (Predicting Interactomes via Co-Elution) program - a software package designed for analyzing SILAC labeled fractionation datasets (Stacey et al. 2017). This allowed us to identify potential protein complexes through a principle of \u201cguilt-by-association\" (Carlson et al. 2019, Stacey et al. 2018, Stacey et al. 2017, Rattray and Foster 2019). Analysis of the SILAC-labeled peptidisc library in this manner allowed us to generate over 4900 possible binary interactions, each of which is characterized by a degree of precision (Carlson et al. 2019). Our interaction list is hereafter called the peptidisc interactome (Carlson et al. 2019). The generation and computational validation of the peptidisc interactome dataset using PrInCE has been described in detail in the PhD thesis of Dr. Michael Carlson, a recent PhD graduate from the Duong laboratory. To avoid repetition, these computational methods will not be described further in this thesis.  Here, we focus on experimentally validating the potentially novel interactors of the SecYEG protein translocation channel - or Sec translocon - which were identified in the peptidisc interactome. We selected the Sec translocon as a validation target because of its central     82  importance for protein translocation and membrane protein biogenesis. Numerous studies from our laboratory and others have demonstrated that the Sec translocon is a diverse protein complex assembly platform, interacting with a number of membrane-bound ancillary subunits (Duong and Wickner 1997a, G\u00f6tzke et al. 2014, Schulze et al. 2014, Botte et al. 2016, Park and Rapoport 2012, Rapoport, Li, and Park 2017, Wu, Cabanos, and Rapoport 2019, Collinson, Corey, and Allen 2015, Jau\u00df et al. 2019). Many of these previously identified interactors - including the membrane-bound chaperones PpiD and YfgM as well as the holo-translocon subunits YidC, YajC, SecD and SecF - were identified as interactors of SecYEG in our proteomic study. Remarkably, the peptidisc interactome also shows significant correlation between certain subunits of Sec and Bam complexes, suggesting an astonishing network of protein associations across the entire bacterial cell envelope (Carlson et al. 2019).  We present experimental data which further supports the existence of a Sec-Bam super-complex. Additionally, our data identifies the unannotated E. coli protein YibN as a novel potential interactor of the Sec translocon. We further show that the peptidisc method is demonstrably superior to detergent-based methods for preserving transient interactions. Taken together, these results reinforce the utility of the peptidisc for revealing novel insights into the membrane proteome, which may otherwise be lost using detergent-based methods.        83  3.2 Materials and Methods  3.2.1 Reagents  Tryptone, yeast extract, Na2HPO4, KH2PO4, NaCl, imidazole, Tris-base, acrylamide 40%, bis-acrylamide 2% and TEMED were obtained from Bioshop Canada. Amino acid isotopologues were purchased from Cambridge Isotope Laboratories.  Isopropyl \u03b2-D-1-thiogalactopyranoside (IPTG), ampicillin, kanamycin, and arabinose were purchased from GoldBio. n-dodecyl-\u03b2-d-maltoside (DDM) was purchased from Anatrace. Ni2+-NTA chelating Sepharose was obtained from Qiagen.  Peptide NSPr (purity >80%) was obtained from Peptidisc Biotech Canada.  All other chemicals were obtained from Fisher Scientific Canada.  3.2.2 Plasmids  The plasmids pBad22-hisEYG, pBad33-DFyajC, pBad22-hisYidC, and pBad33-YfgMHis-PpiD have been previously described (Young and Duong 2019, Carlson et al. 2019, Zhang et al. 2012, Duong and Wickner 1997a, b). All plasmids generated in this study were cloned using the PIPE method (Klock and Lesley 2009). The genes for SecDF were amplified from pBad33-DFyajC and inserted into pBad22-hisEYG, generating the plasmid pBad22-hisEYGDF. To construct pBad22-HTL, the gene for YidC without its affinity tag was amplified from pBad22-hisYidC and inserted into pBad22-hisEYGDF. To construct pBad22-hisBamA, the gene encoding BamA was amplified from the E. coli genome and inserted into pBad22. A his-tag was subsequently added     84  onto the N-terminus of the BamA mature domain immediately after the signal peptide. All constructs were verified by DNA sequencing (Genewiz).  3.2.3 Expression of target proteins in SILAC labeling conditions  SILAC labeling of cells and preparation of SILAC-labeled membranes was performed as previously described (Carlson et al. 2019, Zhang et al. 2012). Briefly, plasmids were chemically transformed into E. coli JW2806 [\u0394(araD-araB)567 \u0394lacZ4787(::rrnB-3) lambda\u2212 \u0394lysA763::kan rph-1 \u0394(rhaD-rhaB)568 hsdR514]. Cells were grown overnight in M9 media supplemented with either 0.3 mg\/mL \"heavy\" Lys4 or 0.3 mg\/mL \"light\" Lysine, as appropriate. The next morning, the cultures were diluted 1\/100 into fresh M9 media supplemented with either Lys4 or light lysine. Protein expression was induced with 0.01% arabinose once the cells had reached OD~0.4-0.6. The cultures were then shifted to 25\u00b0C and grown for a further 16 hours.  Cells were harvested and resuspended in TSG (50 mM Tris-HCl: pH 8; 100 mM NaCl; 10% glycerol) buffer containing 1 mM PMSF. Cell lysis and membrane isolation was performed as described in Chapter 2 of this thesis. Membranes were solubilized with 0.5% DDM in 1 mL on ice for 15 minutes. Insoluble material was removed by ultracentrifugation (100 000 g, 15 minutes). A 200 \u00b5L aliquot was incubated with 50 \u00b5L Ni-NTA resin for 30 minutes. The resin was washed with 10 CV of TSG buffer + 0.02% DDM, then eluted in TSG buffer + 0.02% DDM + 300 mM Imidazole. Eluted proteins were analyzed by 15% SDS PAGE and visualized by Coomassie Blue staining. The remaining detergent-solubilized material was mixed with a 4:1 excess of NSPR peptide in a 15 mL 100 kDa cutoff Amicon concentrator and reconstituted into     85  peptidisc libraries as previously described (Carlson et al. 2019). The resultant library was incubated with Ni-NTA resin for 30 minutes at 4\u00b0C. The resin was washed with 10 column volumes (CV) of TSG buffer, then eluted in TSG buffer + 300 mM Imidazole. Eluted proteins were analyzed by 15% SDS PAGE and visualized by Coomassie Blue staining.  The \"heavy\" and \"light\" elution fractions were pooled, denatured with 6M Urea and digested with trypsin and LysC before STAGE tipping and analysis by mass spectrometry.   3.2.4 Digestion of protein samples  Digestion of protein samples was performed as previously described (Carlson et al. 2019). Where applicable, detergent was first removed from protein samples by acetone precipitation.  Protein samples were mixed with 80% ice cold acetone, then left overnight on ice to precipitate.  Precipitated proteins were pelleted by low speed centrifugation (10,000 x g, 10 min, 4\u00b0C).  After removal of the supernatant, pellets were air-dried at 42\u00b0C for 10 minutes before storage at -20\u00b0C until digestion.  For peptidisc samples, detergent was removed during peptidisc assembly, so no acetone precipitation was necessary.    Immediately before digestion, samples were first denatured in 6M urea.  Denatured proteins were incubated with 5 mM DTT for 1 h at 25\u00b0C to reduce any cysteines.  Free cysteines were alkylated by addition of 20 mM iodoacetamide for 1 h at 25\u00b0C in the dark, the reaction was then quenched by addition of 40 mM DTT.  Samples were pre-cleaved by addition of 0.1 \u00b5g Lys-C for 1.5 hours at 25\u00b0C, followed by dilution to 1 M urea in 50 mM ammonium bicarbonate, pH 8.3. Proteomics grade trypsin (1 \u00b5g; Promega) was added to each sample, and the reactions left to digest overnight at 25\u00b0C.  Digested samples were     86  acidified to <pH 2.5 by addition of 1% trifluoroacetic acid and the resulting peptide supernatant purified using self\u2010made Stop\u2010and\u2010go\u2010extraction tips (StageTips) composed of C18 Empore material (3M) packed in to 200 \u03bcl pipette tips. Prior to addition of the peptide solution, StageTips were conditioned with methanol and equilibrated with 0.5% acetic acid (Buffer A3). Peptide supernatants were loaded onto columns and washed with three bed volumes of Buffer A3. Peptide samples were eluted with 80% acetonitrile, 0.5% acetic acid (Buffer B3) into microfuge tubes, dried down using a vacuum concentrator, and stored at -20\u00b0C.  3.2.5 Liquid chromatography and mass spectrometry analysis  Liquid chromatography and mass spectrometry analysis was performed as previously described (Carlson et al. 2019). Purified peptides were analyzed using a quadrupole \u2013 time of flight mass spectrometer (Impact II; Bruker Daltonics) on-line coupled to an Easy nano LC 1000 HPLC (ThermoFisher Scientific) using a Captive spray nanospray ionization source (Bruker Daltonics) including a 2cm-long, 100 \u03bcm-inner diameter fused silica fritted trap column, 75 \u03bcm-inner diameter fused silica analytical column with an integrated spray tip (6-8 \u03bcm diameter opening, pulled on a P-2000 laser puller from Sutter Instruments).  The trap column is packed with 5 \u03bcm Aqua C-18 beads (Phenomenex, www.phenomenex.com) while the analytical column is packed with 1.9 \u03bcm-diameter Reprosil-Pur C-18-AQ beads (Dr. Maisch, www.Dr-Maisch.com).  Buffer A consisted of 0.1% aqueous formic acid in water, and buffer B consisted of 0.1% formic acid in acetonitrile.  Samples were resuspended in buffer A and loaded with the same buffer.  Standard 45 min gradients were run from 0% B to 35% B over 90 min, then to 100% B over 2 min, held at     87  100% B for 15 min.  Before each run the trap column was conditioned with 20 \u03bcL buffer A, the analytical \u2013 with 4 \u03bcL of the same buffer and the sample loading was set at 20 \u03bcL.  When one column system was used the sample loading volume was set at 8 \u03bcL + sample volume.  The LC thermostat temperature was set at 7\u00b0C.  The sample was loaded on the trap column at 850Bar and the analysis was performed at 0.25 \u03bcL\/min flow rate.  The Impact II was set to acquire in a data-dependent auto-MS\/MS mode with inactive focus fragmenting the 20 most abundant ions (one at the time at 18Hz) after each full-range scan from m\/z 200Th to m\/z 2000Th (at 5Hz rate).  The isolation window for MS\/MS was 2 to 3Th depending on parent ion mass to charge ratio and the collision energy ranged from 23 to 65eV depending on ion mass and charge (Beck et al. 2015).  Parent ions were then excluded from MS\/MS for the next 0.4min and reconsidered if their intensity increased more than 5 times.  Singly charged ions were excluded since in ESI mode peptides usually carry multiple charges.  Strict active exclusion was applied.  Mass accuracy: error of mass measurement is typically within 5 ppm and is not allowed to exceed 10 ppm.  The nano ESI source was operated at 1900V capillary voltage, 0.20Bar CaptiveSpray nanoBooster pressure, 3L\/min drying gas and 150\u00b0C drying temperature. Analysis of Mass Spectrometry Data was performed using MaxQuant 1.5.3.30 (Cox and Mann 2008; Cox et al. 2014; Tyanova et al. 2014). The search was performed against a database comprised of the protein sequences from the source organism (E.coli K12) plus common contaminants using the following parameters: peptide mass accuracy 40 parts per million; fragment mass accuracy 0.05Da; trypsin enzyme specificity, fixed modifications - carbamidomethyl, variable modifications - methionine oxidation, deamidated N, Q and N-acetyl peptides.  Proteins were quantified from 1 peptide identification.  Only those peptides exceeding     88  the individually calculated 99% confidence limit (as opposed to the average limit for the whole experiment) were considered as accurately identified.  3.2.6 Validation of the HTL-YfgM-PpiD (HMD) interaction  The plasmid pBad33-YfgM PpiD (containing no affinity tag) was transformed into BL21(DE3) competent cells containing the plasmid pBad22-HTL. Protein expression and membrane preparation was performed as described above. To further enrich the inner membrane fraction, the crude membrane fraction was layered onto a 2-step 20%-50% sucrose gradient in SW41 tubes and re-centrifuged at 200 000 g for 2 hours. The inner membrane (IMV) fraction was recovered as a distinct brown band near the middle of the gradient and diluted 4-fold in TSG before being pelleted by ultracentrifugation (100 000 g, 15 minutes). IMVs were solubilized with 0.5% DDM in 1 mL volume on ice for 15 minutes. Insoluble material was removed by ultracentrifugation (100 000 g, 15 minutes). A 200 \u00b5L aliquot was incubated with 50 \u00b5L Ni-NTA resin for 30 minutes. The resin was washed with 10 CV of TSG buffer + 0.02% DDM, then eluted in TSG buffer + 0.02% DDM + 300 mM Imidazole. Eluted proteins were analyzed by 15% SDS PAGE and visualized by Coomassie Blue staining. The remaining detergent-solubilized material was mixed with a 4:1 excess of peptidisc peptide in a 15 mL 100 kDa cutoff Amicon concentrator and reconstituted into peptidisc libraries as previously described (Carlson et al. 2019). The resultant library was incubated with Ni-NTA resin for 30 minutes at 4\u00b0C. The resin was washed with 10 CV of TSG buffer, then eluted in TSG buffer + 300 mM Imidazole. Eluted proteins were analyzed by 15% SDS PAGE and visualized by Coomassie Blue staining.      89   3.3 Results  3.3.1 Experimental validation of binary interactions by affinity purification mass spectrometry (AP\/MS)  We performed a series of in vitro experiments centered on the Sec translocon to experimentally validate the peptidisc interactome dataset.  The goal was to use AP\/MS to confirm and potentially discover novel pairwise associations predicted from our peptidisc interactome datalist.  Principally, we aimed to confirm interactions between the core SecYEG complex and the membrane-anchored periplasmic chaperones YfgM and PpiD.  These interactions are detected in our peptidisc interactome datalist at high (>75%) precision (Carlson et al. 2019). We were also interested by the astonishing apparent interaction between the Sec and Bam complexes, also given in our datalist at high (>75%) precision. These include interactions between the SecY complex and the BamA, BamC and BamD subunits of the Bam complex (Carlson et al. 2019).  To perform these validation AP\/MS experiments, the his-tagged SecYEG complex was expressed in SILAC labeling conditions (experimental workflow shown in Figure 3.2). The membrane fraction was briefly solubilized with detergent and reconstituted into peptidisc library. The SecY complex was subsequently isolated by Ni-NTA and the co-isolated proteins were identified by LC-MS\/MS.  To measure protein enrichment and to control for non-specific co-purifying background contaminants, the pulldown experiments were performed in parallel using a detergent extract or peptidisc library prepared from cells transformed with the     90  empty vector. The results for each AP\/MS experiment are plotted as enrichment matrices in Figure 3.3. For each quantified protein, the Log2-transformed peptide intensity is plotted against the corresponding Log2-transformed SILAC ratio.     Figure 3.2 The peptidisc AP\/MS workflow Schematic showing a typical peptidisc SILAC AP\/MS workflow. Two identical proteomes are grown: one \"light\" proteome (\"light\" lysine) containing a vector control, and one \"heavy\" (Lys4) proteome with an over-expressed his-tagged bait protein. After lysis, the membrane proteomes of both cells are isolated and reconstituted in peptidisc libraries. Ni-NTA affinity pulldowns are performed in parallel and the elutions are mixed 50:50 by volume before digestion, STAGE tipping and LC-MS\/MS analysis.   + +Peptidisc library Peptidisc library\u201cLight\u201d Proteome \u201cHeavy\u201d Proteome+ Empty vector + His-tagged protein Affinity pulldown  Affinity pulldownPool elutionsMS analysis    91   Figure 3.3 Identification of SecYEG interactors by AP\/MS. A) Enrichment matrix of each quantified protein identified in the SecYEG detergent AP\/MS pulldown.  The Log(2) peptide intensity for each quantified protein is plotted against the corresponding Log(2) SILAC ratio. Arbitrary enrichment cutoffs were set for both the x and y axes; these are indicated on the plot as dashed lines to aid the eye. Proteins of interest are highlighted in red. The number of unique peptides detected for each protein of interest is given in parentheses. Each black dot is a protein quantified in the pulldown experiments B) As in A, but for proteins quantified in the SecYEG peptidisc library pulldown.  \t \t-4 -2 0 2 4161820222426Log(2) SILAC ratioLog(2) Total IntensitySecYEG Library AP\/MSSecY (7)SecG (2)OmpA (7)YfgM (10)PpiD (15)YidC (7)BamB (8)YajC (4)200 quantified proteinsBamC (7)-4 -2 0 2 4 6 8 10161820222426Log(2) SILAC ratioRplL (4)RplA (5)FepA (3)RplI (5)YkgM (3)RplT (5)RplK (4)RpmC (2)SecY (11)SecYEG Detergent AP\/MS154 quantified proteins-4 -2 0 2 4161820222426Log(2) SILAC ratioLog(2) Total IntensitySecYEG Library AP\/MSSecY (7)SecG (2)OmpA (7)YfgM (10)PpiD (15)YidC (7)BamB (8)YajC (4)200 quantified proteinsBamC (7)-4 -2 0 2 4 6 8 10161820222426Log(2) SILAC ratioRplL (4)RplA (5)FepA (3)RplI (5)YkgM (3)RplT (5)RplK (4)RpmC (2)SecY (11)SecYEG Detergent AP\/MS154 quantified proteinsAB    92  Experiments with the detergent extract shows that SecY is highly enriched after affinity pulldown (Figure 3.3A). There is also enrichment of several ribosomal proteins, which is not surprising given the intrinsic affinity of ribosomes for the Sec complex (Rapoport et al. 2017; Park and Rapoport 2012). However, many known membrane-bound interactors of the SecY complex are not enriched, likely due to their dissociation during the prolonged incubation with detergent. Accordingly, in peptidisc, several ancillary subunits of the Sec translocon, including the periplasmic chaperones PpiD and YfgM, as well as the holo-translocon subunits YidC and YajC are detected. There is also strong enrichment of the porin OmpA in addition to several subunits of the outer membrane Bam complex.  The BamB and BamC subunits in particular are significantly enriched (Figure 3.3B). The dynamics of a Sec-Bam complex interactions await further experimentation, but as it is this series of evidence validates the ability of the peptidisc PCP-SILAC and AP\/MS workflows to capture novel protein assemblies that are difficult to isolate in detergent.  3.3.2 Identification of Bam complex interactors  We next applied the AP\/MS workflow towards the protein BamA - the major subunit of the outer membrane-embedded Bam complex. Our interactome dataset identified the Sec ancillary subunits YidC and YajC as potential Bam interactors, as well as the cell surface-exposed lipoprotein RcsF (Figure 3.4A).  To explore the validity of these predicted interactions, we expressed his-tagged BamA in SILAC labeling conditions and analyzed the peptidisc library or detergent extract using the AP\/MS workflow described above.       93  Experiments with the detergent extract shows that BamC and BamD are the only subunits enriched along with BamA (Figure 3.4C). This finding is in agreement with an earlier study which showed that BamB is prone to dissociating from the rest of the complex in detergent solution (Han et al. 2016). The only other interactor that is significantly enriched is RcsF.  In peptidisc, by contrast, all 4 other subunits of the Bam complex (subunits B, C, D and E) are captured along with BamA (Figure 3.4B). Additionally, there is significant enrichment of the Sec translocon ancillary subunits YidC and YajC, thereby providing additional evidence to support this potentially novel interaction. The lipoprotein RcsF and the porin OmpA are also significantly present (Figure 3.4B). We note that BamA-OmpA interaction was not reported in our interactome datalist, probably due to the unusually broad SEC elution profile of OmpA, leading to false negative identification.  However, a series of recent publications have shown that OmpA is a bona fide interactor of both RcsF and BamA in the cell context (Hart et al. 2019; Konovalova et al. 2014).       94   Figure 3.4 Identification of BamA interactors by AP\/MS. A) Pair-wise co-elution plots of select BamA interactors as predicted from the peptidisc PCP-SILAC workflow. Pairwise interaction correlation values are shown above each plot. B) Enrichment matrix of each quantified protein quantified in the BamA detergent AP\/MS pulldown. The data was plotted and labeled as in Figure 3. C) As in A, but for proteins quantified in the BamA peptidisc AP\/MS pulldown.           AB C0 2 4 6 8 10 12012345Elution volumeIsotopologue Ratio (H\/L)BamA RcsF Co-elutionBamARcsFCorrelation: 0.951 +\/- 0.0070 2 4 6 8 10 12012345Elution volumeIsotopologue Ratio (H\/L) BamAYajCCorrelation: 0.9875 +\/- 0.002BamA YajC Co-elution0 2 4 6 8 10 12012345BamA YidC Co-elutionElution volumeIsotopologue Ratio (H\/L) BamAYidCCorrelation: 0.965 +\/-  0.0350 2 4 6 8 10 12012345Elution volumeIsotopologue Ratio (H\/L)BamA RcsF Co-elutionBamARcsFCorrelation: 0.951 +\/- 0.0070 2 4 6 8 10 12012345Elution volumeIsotopologue Ratio (H\/L) BamAYajCCorrelation: 0.9875 +\/- 0.002BamA YajC Co-elution0 2 4 6 8 10 12012345BamA YidC Co-elutionElution volumeIsotopologue Ratio (H\/L) BamAYidCCorrelation: 0.965 +\/-  0.035-2 0 2 4 6 8161820222426Log(2) SILAC ratioBamA detergent AP\/MSLog(2) Peptide Intensity62 quantified proteinsBamA (37)BamD (9)RcsF (3)BamC (10)AcrB (20)CyoB (6)CyoA (9)-2 0 2 4 6161820222426Log(2) SILAC ratioBamA library AP\/MS176 quantified proteinsBamA (48)BamB (9)BamC (11)BamD (9)RcsF (3)OmpA (5)BamE (1)YidC (8)FtsH (13)RodZ (5)YajC (4)Log(2) Peptide Intensity-2 0 2 4 6 8161820222426Log(2) SILAC ratioBamA detergent AP\/MSLog(2) Peptide Intensity62 quantified proteinsBamA (37)BamD (9)RcsF (3)BamC (10)AcrB (20)CyoB (6)CyoA (9)-2 0 2 4 6161820222426Log(2) SILAC ratioBamA library AP\/MS176 quantified proteinsBamA (48)BamB (9)BamC (11)BamD (9)RcsF (3)OmpA (5)BamE (1)YidC (8)FtsH (13)RodZ (5)YajC (4)Log(2) Peptide Intensity    95  3.3.3 Identification of YidC interactors  We were very encouraged to detect significant enrichment of YidC in our BamA peptidisc AP\/MS experiment. To further confirm this novel interaction, we next performed SILAC AP\/MS experiments using YidC as the bait protein. In detergent, the only protein significantly enriched besides YidC itself is an unannotated protein, YibN (Figure 3.5A). This finding is in agreement with our peptidisc interactome datalist, which identified YibN as an interactor of YidC at high precision (Figure 3.5C). In peptidisc, YibN is once again significantly enriched along with YidC, further confirming this novel interaction (Figure 3.5B). We also observe significant enrichment of the Bam complex subunits BamA and BamE, as well as the holo-translocon subunit SecD (Figure 3.5B). These later observations further validate the YidC-Bam interaction described above.         96   Figure 3.5 Identification of YidC interactors by AP\/MS. A) Enrichment matrix of each quantified protein quantified in the YidC detergent AP\/MS pulldown. The data was plotted and labeled as in Figure 3.4. B) As in A, but for proteins quantified in the YidC peptidisc AP\/MS pulldown. C) Pairwise co-elution profiles of YidC and the unannotated protein YibN as predicted from the peptidisc PCP-SILAC workflow from our previously published proteomic study.       -2 0 2 4 6 8 101618202224262830Log(2) SILAC ratioYidC Detergent AP\/MSLog(2) Intensity51 quantified proteinsYidC (29)YibN (11)-2 0 2 4 616182022242628Log(2) SILAC ratioYibN Detergent AP\/MSLog(2) Intensity49 quantified proteinsYibN (11)YidC (29)-2 0 2 4 6 8 10 121618202224262830Log(2) SILAC ratioYidC Library AP\/MSLog(2) Intensity52 quantified proteinsYidC (29)YibN (11)BamA (15)BamE (1)SecD (12)-2 0 2 4 61618202224262830Log(2) SILAC ratioYibN Library AP\/MSLog(2) IntensityYibN (11)YidC (29)BamA (15)BamB (4)BamE (1)SecD (12)116 quantified proteinsA BC0 2 4 6 8 10 1201234Elution volumeHeavy\/Light ratioYidC\/YibN co-elution profileYibNYidC    97  3.3.4 Identification of SecDF interactors  A recent preprint from the Collinson research group (doi: https:\/\/doi.org\/10.1101\/589077) suggested that SecDF may also play a critical role in mediating Sec-Bam interactions. To further investigate this possibility, we performed AP\/MS on SecDF. In SILAC-labeling conditions, we were unable to over-produce his-tagged SecDF (data not shown). However, we were able to co-produce tagged SecDF together with non-tagged SecYEG. In addition to SecY and SecDF, we observe significant enrichment of YfgM, PpiD as well as the BamB, C and D subunits of the Bam complex (Figure 3.6). We also observe enrichment of the AtpF subunit of the ATP synthase complex, which was recently identified as a possible interacting partner of the Sec translocon using native mass spectrometry (Chorev et al. 2018).      98   Figure 3.6 Identification of SecDF interactors by AP\/MS.  Peptidisc AP\/MS using SecDF as bait. The Log(2) peptide intensity was plotted against the Log(2) SILAC ratio. Proteins of interest are highlighted in red. Numbers in parentheses indicate the number of unique peptides per quantified protein.     0 2 4 6 8 10222426283032343638Log(2) SILAC ratioSecDF Library AP\/MSLog(2) IntensitySecD (51)SecF (15)RodZ (10)SecY (8)PpiD (12)BamB (6)YibN (8)BamC (8)BamD (4)FtsY (8)YfgM (14)348 quantified proteinsAtpF (13)    99  We also see significant enrichment of the unannotated protein YibN which was detected earlier as an interactor of YidC.  This observation raises the possibility that YibN may be as-yet uncharacterized interactor of the Sec translocon. Exciting preliminary data from our laboratory has revealed that YibN is an integral inner membrane protein of E. coli. However, the biological significance of a YibN-Sec interaction awaits further experimentation, as the function of YibN is currently unknown.  Taken together, our AP\/MS results suggest that SecYEG, YidC, YfgM, PpiD and SecDF interact together, forming a transient and dynamic inner membrane \"platform\" that interacts with the Bam complex and brings it into association with SecYEG. Given that this putative complex consists of the entire HTL complex together with the additional subunits YfgM and PpiD, we are terming this new complex the HMD complex (HTL+YfgM+PpiD).  3.3.5 Experimental validation of the HMD complex  To experimentally validate the HMD complex, we first engineered an expression vector that enables over-production of the bacterial holo-translocon (HTL) SecYEG-SecDF-YidC under the control of an Arabinose promoter as described in Materials and Methods. The SecE subunit is modified with an N-terminal 6x His-tag to facilitate affinity purification. We co-expressed the HTL with non-tagged YfgM and PpiD. Membranes were isolated as described above and batch affinity purifications were performed in both detergent and peptidisc (see Materials and Methods for details). Aliquots from each purification step were analyzed by 15% SDS-PAGE (Figure     100  3.7). All subunits are well produced and are present in the membrane fraction, as expected (Figure 3.7 A and B, lane \"S\"). In detergent, the His-tag on the SecE subunit enables efficient purification of the core SecYEG complex. However, the ancillary subunits SecD, SecF, YidC, PpiD and YfgM are not co-purified, but rather appear in the \"Flowthrough\" and \"Wash\" lanes. This suggests that the \"HMD\" complex is dissociated in detergent, likely due to the extensive washing (Figure 3.7A, lanes \"F\" and \"W1\"). In peptidisc, by contrast, we observe co-purification of SecD, SecF, YidC, PpiD and YfgM along with the His-tagged SecYEG complex, indicating that the complex is preserved in peptidisc (Figure 3.7B). This finding suggests that HMD is a bona fide complex in the bacterial inner membrane.         101   Figure 3.7 Experimental validation of the HTL-YfgM-PpiD (HMD) complex The HTL complex (from Figure 3.1) was co-expressed with untagged variants of the membrane-bound chaperones YfgM and PpiD. Membranes were solubilized in DDM and affinity purifications were performed in both detergent (A) and peptidisc (B).                             Detergent                      S        F      W1    W2      E   YfgM -  SecF -  SecY -  D\/C  PpiD   SecE\/G -- 37- 25- 20- 15- 50- 75                     Peptidisc                      S       F     W1    W2     E   PpiD   D\/C  SecY -  SecF -  YfgM -  SecE\/G -- 37- 50- 75- 25- 20- 15A B    102  3.4 Discussion  We recently described the peptidisc method as an unbiased \"one-size-fits-all\" scaffold for streamlined membrane protein reconstitution (Carlson et al. 2018, Carlson et al. 2019). Our previous work has shown that a peptidisc-reconstituted E. coli membrane proteome is soluble and amenable to biochemical fractionation. By combining quantitative proteomics with protein correlation profiling, we were able to identify over 4900 pairwise interactions out of >700 000 possible random interactions (Carlson et al. 2019). In our published study, we validate several of these potentially novel interactions using SILAC AP\/MS.  Here, we demonstrate using AP\/MS that the peptidisc is superior to conventional detergent-based methods for preserving and stabilizing transient interactions between membrane proteins (Carlson et al. 2019). In our published study, our biochemical fractionation and AP\/MS data point to the existence of a super-complex between the Sec translocon and the outer membrane-embedded Bam complex (Carlson et al. 2019). In this thesis, I employ the peptidisc AP\/MS method to more thoroughly explore the membrane interactome of the Sec translocon and to further probe the protein-protein interaction networks which make up the Sec-Bam super-complex.    I performed systematic AP\/MS experiments on multiple subunits of the Sec translocon to better characterize the protein-protein interaction networks which underlie the Sec-Bam super-    103  complex. We identify interactions between the Bam complex and several inner membrane-bound subunits of the Sec translocon - including SecYEG itself, YidC, SecDF, YfgM and PpiD.  Interactions between SecYEG and its ancillary subunits SecDF and YidC are well characterized in the literature, and form a transient macromolecular assembly termed the Holo-translocon, or HTL (Schulze et al. 2014, Botte et al. 2016, Komar et al. 2016). Our AP\/MS results suggest that YfgM and PpiD may interact transiently with the HTL, forming a complex which we term \"HMD\". The HMD complex may in turn interact transiently with the outer membrane embedded Bam complex. Our experimental data validates the HMD complex, demonstrating that it is a bona fide resident of the E. coli inner membrane. The discovery of an expanded \"holo-translocon\" complex - with two additional membrane-bound subunits - is not surprising. SecYEG is known to be a highly dynamic and versatile protein complex assembly platform (Crane and Randall 2017, Wu, Cabanos, and Rapoport 2019, Sachelaru et al. 2014, Petriman et al. 2018). In a live cell, it is very likely that SecYEG is constantly exchanging between different ancillary subunits. The previously characterized HTL complex - and the newly identified HMD complex - are merely static snapshots of what is likely a very dynamic assembly.   Our proteomic data also revealed the existence of YibN as a largely uncharacterized Sec ancillary subunit. Is YibN a functional interactor of the translocon, or merely a proximal interactor? It is possible that YibN localizes near the Sec translocon within the cell membrane, leading to its capture and identification in our AP\/MS workflow. It is also possible, however, that YibN interacts with the Sec translocon and\/or its ancillary subunits to influence their activity. The possible functional role of this potential additional Sec ancillary subunit awaits further     104  experimentation and is the subject of ongoing investigation in our laboratory. As our data stands, however, the peptidisc method is clearly a powerful tool for revealing novel potential interactors of a target membrane protein.       105  Chapter 4: His-tagged Peptidiscs Enable Affinity Purification of the Membrane Proteome for Downstream Mass Spectrometry Analysis  4.1 Introduction  Membrane proteins play vital roles in many cellular functions, including energy generation, cell division, protein translocation and nutrient import. In many cases, membrane proteins transiently associate with each other into macromolecular complexes in order to carry out their functions (Stenberg et al. 2005, Babu et al. 2018, Maddalo et al. 2011, Gokhale et al. 2012, Carlson et al. 2019, Hu et al. 2019, Zhang et al. 2012). Establishing an in-depth understanding of the membrane proteome and its interaction network, the so-called \"membrane interactome\", is therefore critical for our understanding of these diverse cellular processes. Traditionally, proteomics-based methods employed to analyze membrane interactomes rely extensively on biochemical fractionation of detergent-solubilized membrane extracts (Stenberg et al. 2005, Maddalo et al. 2011, Babu et al. 2012, Babu et al. 2018, Hu et al. 2019, Nielsen et al. 2005, Vuckovic et al. 2013).  These high-resolution fractionation techniques were originally developed for analyzing soluble proteomes, but they can also be adapted for analysis of membrane proteins by the inclusion of detergents (Havugimana et al. 2012, Havugimana, Wong, and Emili 2007, Hu et al. 2009, Moutaoufik et al. 2019, Babu et al. 2018, Babu et al. 2012). A common difficulty in studies analyzing membrane interactomes in detergent however is that prolonged exposure to detergents during fractionation often leads to dissociation of transient interactions, causing all but the most stable interactions to be disrupted and thus undetected (Carlson et al. 2019, Babu et     106  al. 2018, Babu et al. 2012, Vuckovic et al. 2013, D\u00edaz-Mej\u00eda, Babu, and Emili 2009, Marty et al. 2013). Additionally, detergents are not compatible with most mass spectrometry methods and must be removed from samples before analysis. Methods for detergent removal - such as acetone precipitation - often lead to loss of material, and thus fewer protein IDs (Carlson et al. 2019).   We recently introduced the peptidisc as an alternative to detergent to maintain membrane proteins in solution (Carlson et al. 2018, Carlson et al. 2019). We further reported that the E. coli cell envelope can be analyzed directly by mass spectrometry once the solubilized membrane proteome is reconstituted in these water-soluble particles, termed a \"peptidisc library\" (Carlson et al. 2019).  We showed that the peptidisc library minimizes the deleterious effects of detergent on labile membrane complexes because fractionation occurs in detergent-free conditions (Carlson et al. 2019). Bioinformatic analysis of the co-fractionation data  coupled to affinity pull-down revealed a number of novel and unexpected membrane protein interactions, often undetected in previous detergent-based analyses (Carlson et al. 2019). Altogether, the peptidisc emerged as an advantageous tool to survey the membrane interactome.  Despite this advance, work is still needed to optimize membrane proteomics methods and in particular the peptidisc library workflow. One outstanding issue is how to effectively enrich the low-abundance membrane proteome while removing high abundance cytosolic components from membrane preparations (Papanastasiou et al. 2013, Papanastasiou et al. 2016, Carlson et al. 2019).  Ribosomes and other soluble proteins often co-sediment with the membrane fraction during ultracentrifugation, and these \u201ccontaminants\u201d remain strongly prevalent across multiple     107  fractions  during size-exclusion chromatography (Carlson et al. 2019). During mass spectrometry analysis, the abundance of contaminants masks detection of peptides from less abundant integral membrane proteins (Nielsen et al. 2005, Vuckovic et al. 2013, Roy et al. 2015). Previous studies have attempted to remove ribosomes and other soluble complexes by pre-treating membrane extracts with solutions of high ionic strength, or even denaturants such as urea (Papanastasiou et al. 2013, Papanastasiou et al. 2016, Wu et al. 2011). However, these wash conditions are rather stringent and potentially disrupt or dissociate labile membrane-bound complexes.   We present here a simple approach to selectively isolate the membrane proteome using a peptidisc functionalized with a hexa-histidine tag. After reconstitution of the E. coli K12 membrane proteome in peptidiscs, the bona fide membrane proteome is purified by Ni-NTA chromatography in detergent-free buffer (Figure 4.1). Peptidisc-encapsulated integral membrane proteins, as well as peripherally bound membrane-associated, proteins are captured on the resin whereas soluble contaminants are washed away (Figure 4.1).  We present our results showing that the His-tagged peptidisc effectively enables purification of the E. coli membrane proteome. We then apply this method to study the membrane proteome of the strain BL325, which contains the SecDFyajC operon under the control of an arabinose-inducible promoter (Duong and Wickner 1997a, b, Economou et al. 1995). SecDFyajC is a well-characterized membrane-integrated ancillary subunit of the SecYEG translocon which is essential for biogenesis of the bacterial membrane (Rapoport, Li, and Park 2017, Tsukazaki 2018, Tsukazaki et al. 2011). Previous studies have revealed that cells depleted in SecDF are less viable than wild-type E. coli and exhibit reduced rates of protein transport (Kato, Nishiyama, and Tokuda 2003, Duong and     108  Wickner 1997a, Pogliano and Beckwith 1994). In light of these earlier  in vivo observations, we rationalized that there may be significant changes in the membrane proteome as a result of SecDF depletion. We address this biological question by surveying the membrane proteome using the His-tagged peptidisc workflow. Raw data from this study are available via ProteomeXchange with identifier PXD017242.      109   Figure 4.1 Overview of the functionalized peptidisc workflow. A crude detergent extract containing both membrane proteins and soluble proteins is reconstituted into the his-tagged peptidisc scaffold peptide. The resulting membrane protein library - or \"peptidisc library\" - is then passed over Ni-NTA resin. Soluble proteins are eliminated during the wash steps, while the bona fide membrane proteome is eluted off the resin.                                           Figure 1. Overview of the functionalized eptidis  w rkflow.  A crude detergent xtract containing both membrane proteins and soluble proteins is reconstituted into the his-tagged peptidisc scaffold peptide. The resulting membran  protein library - or \"peptidisc library\" - is then passed over Ni-NTA resin. Soluble proteins are eliminated during the wash steps, while the bona fide membrane proteome is eluted off the resin.      110  4.2 Materials and Methods  4.2.1 Reagents and Plasmids  The E. coli strains W3110 and BL325 were from our laboratory collection. The plasmid pET28-HisMsbA was previously described (Carlson et al. 2019). A non-tagged variant of MsbA encoded on pET28-MsbA was generated by deletion of the hexa-histidine tag using the PIPE method (Klock and Lesley 2009). The peptidisc peptide and the His-tagged derivative (purity >95%) were obtained from Peptidisc Biotech Canada.  Detergent n-dodecyl-\u03b2-d-maltoside (DDM) was purchased from Anatrace. The resin Ni2+-NTA chelating Sepharose was obtained from Qiagen. Tryptone, yeast extract, NaCl, imidazole, Tris-base, acrylamide 40%, bis-acrylamide 2% and TEMED were obtained from Bioshop Canada. Isopropyl \u03b2-D-1-thiogalactopyranoside (IPTG), Arabinose and kanamycin were purchased from GoldBio. All other chemicals were obtained from Fisher Scientific Canada.     4.2.2 Preparation of native E. coli membranes   E. coli K12 strain W3110 was grown on a LB plate overnight at 37 \u00b0C. A single colony was picked and grown overnight in 10 mL LB media at 37\u00b0C with shaking. The next day, the overnight culture was diluted 1:100 into 1 L fresh media and grown at 37\u00b0C for an additional 6     111  hours. Cells were harvested and resuspended in TSG buffer. Cell lysis and membrane isolation was performed as described below for MsbA.  4.2.3 Expression and purification of MsbA Unless otherwise stated, all centrifugation and incubation steps in this study were performed at 4\u00b0C. His-tagged MsbA and its non-tagged derivative were expressed in E. coli BL21(DE3) (New England Biolabs) for 2 h at 37\u00baC after induction with 0.5 mM IPTG at an OD of 0.4 in LB medium supplemented with 25 \u00b5g\/mL kanamycin.  Cells were harvested by low speed centrifugation (6,000 x g, 6 min) and resuspended in TSG buffer (50 mM Tris-HCl, pH 7.8; 100 mM NaCl; 10% Glycerol).  Cells were treated with 1 mM phenylmethylsulfonyl fluoride (PMSF) and lysed during 3 passages through a Microfluidizer at 10,000 psi.  Unbroken cells and large aggregates were removed by an additional low speed centrifugation. The crude membrane fraction containing overexpressed MsbA was subsequently isolated by ultracentrifugation (100,000 x g, 30 minutes) in a Ti70 rotor.  The crude membrane fraction was resuspended in TSG buffer at ~ 5mg\/mL. To isolate the inner membrane (IMV) fraction, MsbA crude membrane was layered onto an 11 mL 20-50% step sucrose gradient and centrifuged (200,000 x g, 2 hours) in an SW41 rotor. The IMV fraction was recovered as a distinct brown band in the middle of the gradient. Sucrose was removed by diluting the IMV fraction 4-fold in TSG buffer and re-centrifuging the membranes (100,000 x g, 15 minutes). Finally, the IMV pellet was resuspended in TSG buffer, aliquoted and stored at -80\u00b0C to be used within a month. For affinity purification, MsbA crude membrane (5 mg\/mL) was solubilized with 1% DDM for 30 minutes at 4\u00b0C.  The solubilized material was clarified by ultracentrifugation (100,000 x g, 30 minutes). The     112  detergent-solubilized supernatant was poured over a gravity column containing 5 mL Ni-NTA affinity resin. After collecting unbound material, the resin was washed with 20 column volumes of TSG buffer supplemented with 0.02% DDM. Bound proteins were eluted with TSG buffer supplemented with 0.02% DDM plus 300 mM Imidazole. The eluate was stored at 4\u00b0C until use.  4.2.4 On-gradient reconstitution of MsbA   About 100 \u00b5g of detergent-purified MsbA was mixed with a 4:1 mass excess of either peptidisc peptide or His-tagged peptidisc peptide and layered onto a 2.2 mL 5-20% linear sucrose gradient buffered in TSG. Gradients were centrifuged at 200,000 x g for 15 hours using a TLS55 rotor in a Beckman tabletop ultracentrifuge. Gradients were manually fractionated in 200 \u00b5L aliquots and analyzed by native PAGE followed by Coomassie blue staining of the gel. The protocol and recipe for native gel electrophoresis has been described in detail previously (Carlson et al. 2018, Dalal et al. 2009, Dalal et al. 2012).   4.2.5 Depletion of SecDFyajC  SecDFyajC was depleted in the strain BL325 as previously described with some modifications (Duong and Wickner 1997a, Economou et al. 1995). The strain BL325 (BL21 tgt::kan\u2010araC+ \u2010PBAD::yajCsecDF) was revived on a LB agar plate supplemented with 0.02% arabinose and 25 \u00b5g\/mL kanamycin. No growth was observed in the absence of arabinose, as expected. Colonies     113  were picked and grown overnight in 10 mL LB media at 37\u00b0C with shaking. As a DF+ control, a second flask supplemented with 0.02% arabinose was prepared in parallel. The next day, the cultures were diluted 1:100 into 1 L fresh media supplemented with antibiotic (with or without arabinose, as needed) and grown at 37\u00b0C for 6 hours. DF+ and DF- cells were harvested and resuspended in TSG buffer. Cell lysis and membrane isolation was performed as described above. Effective depletion of SecDF was verified by Western blotting using a SecF-specific antibody as previously described (Economou et al. 1995).   4.2.6 Preparation of peptidisc libraries  Peptidisc libraries were prepared as described in Chapter 3. Briefly, membranes (1 mL volume, ~ 1 mg\/mL) were solubilized on ice for 15 minutes with 0.5% DDM. The insoluble material was pelleted by ultracentrifugation (100,000 x g, 15 minutes). A 4:1 mass excess of His-tagged peptidisc peptide was added to the detergent extracts. The solubilized membrane-peptide mixture was rapidly diluted to 10 mL in TSG buffer over a 100 kDa cut-off centrifugation filter (Amicon). The mixture was concentrated to ~ 1 mL (3,000 x g, 10 minutes) at 4\u00b0C before being diluted back to 10 mL and re-concentrated to ~ 1 mL. Purification of the His-tagged peptidisc library was performed as follows: 500 \u00b5L of His-tagged peptidisc library was incubated with 50 \u00b5L Ni-NTA beads for 30 minutes with shaking. Unbound material was collected in the flow through fraction. The resin was washed 3 times with 500 \u00b5L TSG buffer to remove non-specifically bound contaminants. The purified membrane proteome was eluted with 100 \u00b5L TSG     114  buffer supplemented with 300 mM Imidazole. Aliquots of all fractions from the purification were analyzed by 15% SDS PAGE followed by Coomassie blue staining of the gel.  4.2.7 Protein digestion and LC-MS\/MS analysis Protein digestion and LC-MS\/MS analysis were performed as previously described (Carlson et al. 2019). Samples were first denatured in 6M urea. Denatured proteins were incubated with 5 mM DTT for 1 h at 25\u00b0C to reduce any cysteines.  Free cysteines were alkylated by addition of 20 mM iodoacetamide for 1 h at 25\u00b0C in the dark, and the reaction was then quenched by addition of 40 mM DTT.  Samples were pre-cleaved by addition of 0.1 \u00b5g Lys-C or 1.5 hours at 25\u00b0C, followed by dilution to 1 M urea in 50 mM ammonium bicarbonate, pH 8.3. Proteomics grade trypsin (1 \u00b5g; Promega) was added to each sample, and the reactions left to digest overnight at 25\u00b0C.  Digested samples were acidified to < pH 2.5 by addition of 1% trifluoroacetic acid and the resulting peptide supernatant purified using self\u2010made Stop\u2010and\u2010go\u2010extraction tips (StageTips) composed of C18 Empore material (3M) packed in to 200 \u03bcl pipette tips. Prior to addition of the peptide solution, StageTips were conditioned with methanol and equilibrated with 0.5% acetic acid (Buffer A3). Peptide supernatants were loaded onto columns and washed with three bed volumes of Buffer A3. Peptide samples were eluted with 80% acetonitrile, 0.5% acetic acid into microfuge tubes, dried down using a vacuum concentrator, and stored at -20\u00b0C.      115  4.2.8 Liquid chromatography and mass spectrometry analysis Purified peptides were analyzed using a quadrupole \u2013 time of flight mass spectrometer (Impact II; Bruker Daltonics) on-line coupled to an Easy nano LC 1000 HPLC (ThermoFisher Scientific) using a Captive spray nanospray ionization source (Bruker Daltonics) including a 2cm-long, 100 \u03bcm-inner diameter fused silica fritted trap column, 75 \u03bcm-inner diameter fused silica analytical column with an integrated spray tip (6-8 \u03bcm diameter opening, pulled on a P-2000 laser puller from Sutter Instruments).  The trap column was packed with 5 \u03bcm Aqua C-18 beads (Phenomenex, www.phenomenex.com) while the analytical column was packed with 1.9 \u03bcm-diameter Reprosil-Pur C-18-AQ beads (Dr. Maisch, www.Dr-Maisch.com).  Buffer A consisted of 0.1% aqueous formic acid in water, and buffer B consisted of 0.1% formic acid in acetonitrile.  Samples were resuspended in buffer A and loaded with the same buffer.  Standard 45 min gradients were run from 0% B to 35% B over 90 min, then to 100% B over 2 min, held at 100% B for 15 min.  Before each run, the trap column was conditioned with 20 \u03bcL buffer A, and the analytical with 4 \u03bcL of the same buffer. The sample loading was set at 20 \u03bcL. When one column system was used the sample loading volume was set at 8 \u03bcL + sample volume.  The LC thermostat temperature was set at 7\u00b0C.  The Captive Spray Tip holder was modified similarly to an already described procedure (Beck et al. 2015) \u2013 the fused silica spray capillary was removed (together with the tubing which holds it) to reduce the dead volume, and the analytical column tip was fitted in the Bruker spray tip holder.  The sample was loaded on the trap column at 850Bar and the analysis was performed at 0.25 \u03bcL\/min flow rate.  The Impact II was set to acquire in a data-dependent auto-MS\/MS mode with inactive focus fragmenting the 20 most abundant ions (one at the time at 18Hz) after each full-range scan from m\/z 200Th to m\/z     116  2000Th (at 5Hz rate).  The isolation window for MS\/MS was 2 to 3Th depending on parent ion mass to charge ratio and the collision energy ranged from 23 to 65eV depending on ion mass and charge (Beck et al. 2015).  Parent ions were then excluded from MS\/MS for the next 0.4min and reconsidered if their intensity increased more than 5 times.  Singly charged ions were excluded since in ESI mode peptides usually carry multiple charges.  Strict active exclusion was applied.  Mass accuracy: error of mass measurement is typically within 5 ppm and is not allowed to exceed 10 ppm.  The nano ESI source was operated at 1900V capillary voltage, 0.20Bar CaptiveSpray nanoBooster pressure, 3L\/min drying gas and 150\u00b0C drying temperature.  4.2.9 Analysis of Mass Spectrometry Data   Analysis was performed using MaxQuant 1.5.3.30 (Cox and Mann 2008; Cox et al. 2014; Tyanova et al. 2014). The search was performed against a database comprised of the protein sequences from the source organism (E.coli K12) plus common contaminants using the following parameters: peptide mass accuracy 40 parts per million; fragment mass accuracy 0.05Da; trypsin enzyme specificity, fixed modifications - carbamidomethyl, variable modifications - methionine oxidation, deamidated N, Q and N-acetyl peptides. Proteins were quantified from 1 peptide identification.  Only those peptides exceeding the individually calculated 99% confidence limit (as opposed to the average limit for the whole experiment) were considered as accurately identified. Proteins were annotated as membrane-annotated if they contain the word \"membrane\" in their Gene Ontology (GO) term. Subcellular localization of     117  identified proteins (IM, OM, lipoprotein or peripherally bound protein) was determined based on their Subcellular Localization (CC) data available on UniProtKB.   4.3 Results  4.3.1 Protein reconstitution with His-tagged peptidisc peptides  The His-tagged derivative of the peptidisc scaffold was obtained from Peptidisc Biotech. To verify that addition of the His-tag does not adversely affect the property of the peptide, we performed reconstitution tests using the model membrane protein MsbA (Carlson et al. 2019). Dodecyl maltoside-purified MsbA was incubated with an excess of His-tagged or tag-less peptide scaffolds and reconstituted side-by-side using the \"on-gradient\" method as described in the Materials and Methods. Gradients were fractionated, and aliquots were analyzed by native-PAGE followed by Coomassie blue staining of the gels (Figure 4.2A).  Based on the very similar gel migration profiles of the two preparations, we conclude that addition of a His-tag onto the peptidisc scaffold does not interfere with the efficiency of membrane protein reconstitution.  We then created a peptidisc library using E. coli overproducing the wild type untagged version of MsbA. Inner membranes vesicles (IMVs) were isolated, solubilized with DDM and the membrane proteome was reconstituted with the His-tagged peptide.  Analysis by native-PAGE indicates that the library is water-soluble and that MsbA is trapped in peptidisc (Figure 4.2C, lane 1). To determine if MsbA can be isolated by affinity chromatography, an aliquot of the     118  starting library (labeled SL; Figure 4.2C) was incubated with Ni-NTA agarose beads. Fractions corresponding to flow-through (FT), wash (W) and the eluted purified library (PL) were analyzed by native-PAGE (Figure 4.2C) and 12% SDS-PAGE (Figure 4.2B). Qualitative assessment of the Coomassie-blue stained gels indicate that MsbA is effectively recovered after incubation of the library with Ni-NTA agarose beads. Together, these results provide initial evidence that the His-tag on the peptidisc scaffold enables isolation of the membrane proteome.       119   Figure 4.2 Reconstitution efficiency of MsbA with the functionalized peptidisc scaffold. A) Detergent-purified MsbA was mixed with the peptidisc peptides and layered onto 5-20% Tris-buffered sucrose gradients. Following ultra-centrifugation and gradient fractionation, proteins are analyzed by native-PAGE followed by Coomassie blue staining of the gel.  B) and C) E. coli purified inner membranes enriched with plasmid-encoded MsbA were reconstituted into peptidisc libraries using the His-tagged peptidisc peptide scaffold. The starting library (SL) was subjected to nickel-affinity chromatography and fractions corresponding to flow-through material (F), wash material (W), and the eluted purified library (PL) were analysed by gel electrophoresis as indicated, followed by Coomassie blue staining of the gels.   Figure 2. Reconstitution efficiency of MsbA with the functionalized peptidisc scaffold. A) Detergent-purified MsbA was mixed with the peptidisc peptides and layered onto 5-20% Tris-buffered sucrose gradients. Following ultra-centrifugation and gradient fractionation, proteins are analyzed by native-PAGE.  B) and C) E. coli purified inner membranes enriched with plasmid-encoded MsbA were reconstituted into peptidisc libraries using the His-tagged peptidisc peptide scaffold. The starting library (SL) was subjected to nickel-affinity chromatography and fractions corresponding to flow-through material (F), w sh m terial (W), and the elute  purified library (PL) were analysed by gel e ectrophoresis as indicated, followed by Coomassie blue staining ofth  gels.    1    2    3    4    5    6    7    8    9   10  11                  peptidisc                    gradient   1     2    3    4    5    6     7    8    9   10  11 gradient                  his-peptidisc                    - MsbA-peptidisc - MsbA-peptidisc B               MsbA                         his-tagged peptidisc L       F       W       EC       MsbA        his-tagged peptidisc L     F    E 12% SDS-PAGE Native PAGE MsbA - MsbA-peptidisc -ASL PL S PL    120  4.3.2  Proteomic analysis of the affinity purified peptidisc library  We turned to proteomics to quantify membrane protein enrichment and to verify effective depletion of non-membrane components, such as ribosomal and other high-abundance soluble proteins. The total cell envelope proteome from E. coli strain W3110 (containing both inner and outer membrane proteins) was briefly solubilized with DDM and immediately reconstituted into His-tagged peptidiscs.  The resulting library was incubated with Ni-NTA agarose beads and protein aliquots taken from the different purification steps were analyzed by SDS-PAGE to visually assess the quality of the purification (Figure 4.3A). In parallel, aliquots from the starting library (SL) and from the purified library (PL) were trypsin-digested, desalted with STAGE-tips and analyzed by LC-MS\/MS.  In the starting library (SL), 673 proteins are detected, of which 384 (57%) contain the word \u2018membrane\u2019 in their Gene Ontology (GO) term (Figure 4.3B).  Of these 384 proteins, 239 proteins (~2\/3) are predicted to be localized to the inner membrane. The remaining third corresponds to outer membrane (OM) proteins, peripherally bound and lipid-anchored proteins (23, 66 and 32, respectively). Of the 289 non-membrane annotated proteins (43% of the proteins detected in this sample), 42 correspond to ribosomal subunits (6% of the total protein detected). In the purified library (PL), a total of 505 proteins are detected, of which 363 (72%) are membrane-annotated and 142 are non-membrane annotated (28%) including 35 ribosomal proteins (7% of the total protein detected). Taken together, these data indicate a membrane protein enrichment of 20% and a non-membrane components depletion of 35% upon library     121  purification.  There is no apparent bias in the membrane enrichment since, in both the starting and purified libraries, 2\/3 of the proteins detected are associated to the inner membrane and 1\/3 corresponds to outer, peripheral, and lipid-anchored membrane proteins. Further comparison of the starting and purified libraries reveals considerable overlap, with 408 proteins being common to both and 306 (~75%) being membrane annotated (Figure 4.3C). Importantly, of the 264 proteins unique to the non-purified library, 70% are non-membrane annotated (169 cytosolic proteins and 12 ribosomal proteins). The affinity purification is therefore effective at removing non-membrane annotated proteins from the peptidisc library.  To obtain a more global view, we looked at the peptide intensity values. The peptide intensity value for each protein is derived from precursor ion intensities, which is indicative of protein abundance (Zhu, Smith, and Huang 2010).  We initially plotted the peptide intensity values for the top 100 most abundant proteins in the starting and purified library samples (Figure 4.4). While the overall intensity values for membrane proteins remains similar before and after purification, it is evident that there is a general decrease in the intensities of cytosolic proteins (Figure 4.4; compare A with B). We next compared the peptide intensity values for a select subset of the most abundant cytosolic and ribosomal proteins in the starting and purified libraries. For this subset, we observe between 5 to 10-fold decrease in intensity after library purification (Table 4.1). In contrast, the peptide intensities and number of unique peptides detected for a subset of well-characterized integral membrane proteins and peripherally bound membrane proteins - including multiple subunits of the Sec translocon - are increased up to 10-fold following purification (Table 4.2).     122    Figure 4.3 The his-tagged peptidisc scaffold enables purification of the E. coli membrane proteome. A) E. coli W3110 total membranes were reconstituted into his-tagged peptidisc libraries and purified over Nickel affinity resin as in Figure 2. Aliquots from each step of the purification were analyzed by 15% SDS-PAGE followed by Coomassie blue staining of the gel. The starting library (SL) and purified library (PL) samples (indicated by \"*\") were further analyzed by LC-MS\/MS. B) Graphical representation of the protein content of the two analyzed samples. \"Total IDs\" represent the total number of proteins identified in each sample; \"Mem. IDs\" represents the number of IDs with a \"membrane\" GO term. C) Venn diagram representation showing extent of overlap between the starting library (grey circle) and the purified library (red circle). Altogether, 769 proteins were detected across the two samples, with 408 proteins present in both. The number of proteins with a \"membrane\" GO term in each sample is also indicated.   Figure 3. The his-tagged peptidisc scaffold enables purification of the E. coli membrane proteome. A) E. coli W3110 total membran s were reconstituted into his-tagged peptidisc libraries and purified over Nickel affinity resin as in Figure 2. Aliquots from each step of the purification were analyzed by 15% SDS-PAGE followed by Coomassie blue staining of the gel. Th  starting library (SL) and purif ed library (PL) samples (indicated by \"*\") were further nalyzed by LC-M \/MS. B) Graphical representation of the protein cont nt of the two analyzed s mples. \"Total IDs\" represent the total number of proteins identified in ach sample; \"Mem. IDs\" represents the number of IDs wit  a \"membra e\" GO term. C) Venn diagram representation howing xtent of overlap between the tarting library (grey circle) and the purified library (red circle). Altogether, 769 proteins were detected across the two samples, with 408 proteins present in both. The umber of proteins with a \"membrane\" GO term in each sample is also indicat d.            123    Figure 4.4 Global changes in peptide intensity following library purification. (A) The peptide intensity values for the top 100 proteins in the starting library (SL) sample were plotted. Proteins with a \"Membrane\" GO term are in green; proteins with a non-membrane GO term are in red. (B) As in A, but for the purified library sample.                 A B    124    Protein  Starting Library   Purified Library # unique peptides  Intensity (x105) # unique peptides Intensity (x105) RplL 10 1737 8 272 GroL 30 1136 10 35.6 RpsA 28 784 1 1.7 AceF 27 575 6 9.5 RplA 11 493 7 57.3 RplM 11 400 6 123 RplC 14 359 8 50.0 SucB 15 245 7 23.3 RplV 8 239 4 35 HupB 4 223 1 7.5 MaeB 25 202 - - LacZ 8 192 - - RplQ 7 191 3 22.6  Table 4.1 Depletion of soluble proteins following library purification over Ni-NTA The table shows the number of unique peptides and the intensity values for a representative subset (full list in supplementary data table 1) of the most abundant soluble proteins present in both the starting and purified library samples.              125    Protein  Purified Library   Starting Library # unique peptides  Intensity (x105) # unique peptides Intensity (x105) CydA 26 1873 22 940.4 PtsG 20 1632 21 1618.5 SecD 38 1422 26 170 FrdA 39 1239 29 938.2 ManX 15 1100 8 304.4 SecF 12 797 5 46.5 PpiD 24 520.5 20 263 AcrB 27 447.7 20 382 AtpF 11 450.7 11 365 SecY 13 369.5 8 153 YfgM 13 245.3 8 177 YidC 14 236.7 10 200 OmpA 9 155.3 7 123 BamA 18 132.8 14 97.4  Table 4.2 Enrichment of membrane proteins following library purification over Ni-NTA The table shows the number of unique peptides and the intensity values for select membrane proteins (full list in supplementary data table 1) in both the starting and purified library samples.              126  4.3.3 Effect of SecDFyajC depletion on the E. coli membrane proteome  We applied the peptidisc workflow to characterize changes in the E. coli membrane proteome upon depletion of the SecDFyajC complex, a critical ancillary subunit of the Sec translocon (Tsukazaki et al. 2011, Tsukazaki 2018). Previous studies show that cells depleted for SecDFyajC are markedly less viable and exhibit a cold-sensitive phenotype (Pogliano and Beckwith 1994, Kato, Nishiyama, and Tokuda 2003, Economou et al. 1995). In light of this, we rationalized that SecDFyajC depletion may lead to a specific change in the global E. coli membrane proteome. To address this question, we employed the E. coli B strain BL325, in which the operon encoding for SecDFyajC is controlled by an arabinose-inducible promoter (Duong and Wickner 1997a, b). Cells were grown with or without arabinose and the cell membranes (labeled DF+ and DF-, respectively) were analyzed by Western blotting using a SecF-specific antibody. Consistent with previous reports, SecDFyajC is depleted below detectable levels in the DF- sample (Figure 4.5A) (Economou et al. 1995).  To assess changes in the membrane proteome, equal amounts of DF+ and DF- membrane samples were solubilized, reconstituted into His-tagged peptidiscs, purified by nickel-affinity chromatography and analyzed by mass spectrometry. We first compared the starting and purified libraries in the DF+ strain. A total of 486 proteins are detected across both samples, with 192 proteins present in both (Figure 4.5B). Of those, 126 (or 65%) are membrane annotated. As above with E. coli K12, there is a significant global decrease in the intensity of soluble proteins (Figure 4.6A). The peptide intensity value of multiple well-characterized integral membrane     127  proteins is increased after library purification while the intensity of soluble contaminants is strongly reduced (Table 4.3). We next compared the starting and purified libraries in the DF- strain. A total of 410 proteins are detected across both samples, with 213 proteins present in both (Figure 4.5C). Of the 213 proteins, 146 (or 69%) are membrane annotated. Once again, we observe a considerable increase in the peptide intensities of membrane proteins and strong depletion of soluble contaminants following the purification step (Figure 4.6B, Table 4.3). We note that SecD and SecF are still detected in the DF- condition, suggesting that SecDFyajC is still present to some levels in the cell membrane, even though it is depleted below the detection limit of a Western blot.  We next compared the DF+ and DF- purified libraries to identify membrane proteome changes caused by the depletion. A total of 424 proteins are detected across both samples, with 236 proteins identified in both (Figure 4.5D). Of the 236 proteins, 174 (74%) are membrane annotated.  We ranked these common membrane proteins by peptide intensity. In the DF- purified library, the two highest ranked proteins are the motor SecA ATPase and the unannotated inner membrane protein YibN (Table 4.3).  The peptide intensity of SecA in the DF- condition is increased 8-fold over the DF+ condition (Table 4.3). Similarly, the peptide intensity of YibN is increased 4-fold over the DF+ condition. The peptide intensities for several other Sec ancillary subunits - SecG, YidC, PpiD and YfgM - are comparable between the DF+ and DF- samples (Table 4.3). We note that the increase of membrane-localized SecA in the DF- sample is in line with previous data showing that defects in SecD and SecF result in increased levels of cellular SecA (Rollo and Oliver 1988, Riggs, Derman, and Beckwith 1988). The biological significance     128  for the increased localization of YibN remains to be explored, but as it is, these results validate the utility of the His-tagged peptidisc for identifying novel insights in membrane biology.       129    Figure 4.5 Proteomic analysis of a SecDFyajC depletion strain A) E. coli BL325 cells was grown in the presence or absence of arabinose and the membrane fractions were analyzed side-by-side by 15% SDS-PAGE followed by Western blotting using a SecF-specific antibody. B) and C) Venn diagrams showing overlap between the DF+ and DF- starting and purified libraries. The diagram is labeled as in Figure 3C. D) Venn diagram showing overlap between the DF+ and DF- purified libraries.                                         Figure 4. Proteomic analysis of a SecDFyajC depletion strain. A) E. coli BL325 cells was grown in the presence or absence of arabinose and the membrane fractions were analyzed side-by-side by 15% SDS-PAGE followed by Western blotting using a SecF-specific antibody. B) and C) Venn diagrams showing overlap between the DF+ and DF- starting and purified libraries. The diagram is labeled as in Figure 3C. D) Venn diagram showing overlap between the DF+ and DF- purified libraries.     130     Figure 4.6 Global changes in peptide intensities after library purification. (A) Data for the DF+ SL and DF+ PL samples were plotted as in Figure 4 to show global changes in protein intensities after library purification. (B) As in A, but for the DF- SL and DF- PL samples.          AB    131   Protein  DF-  DF+  DF- Starting Library  DF+  Purified Library  Purified Library Starting Library  # unique peptides Intensity (x 105) # unique peptides Intensity (x 105) # unique peptides Intensity (x 105)  # unique peptides Intensity (x 105) SecA 37 204.6 13 25.2 37 150.9 9 8.3 YibN 6 198.7 5 54.5 4 84.6 4 8.1 AtpF 12 132.8 12 185.9 10 70.6 10 33.9 OmpA 7 131.2 6 29.1 8 74.8 5 14.1 PpiD 18 103.8 20 151.5 14 29.5 12 14.6 SecD 20 81.4 21 111 18 27.1 15 15.8 YidC 8 75.9 8 60.5 10 32 7 12.8 AcrB 14 67.1 14 69.6 6 8.5 6 4.4 BamA 18 63.8 17 66.1 18 31.1 13 12.9 YfgM 8 38 12 76.9 5 8.5 3 5.2 SecF 4 22.4 6 33.3 3 6.6 2 2.3 SecY 2 9.4 1 3.5 2 6.9 - -  Table 4.3 Comparison of protein intensities between the DF-\/DF+ starting and purified libraries. The table compares the number of peptides detected and the intensity values for the membrane proteins listed above in the DF+ and DF- purified libraries. The values for the DF- and DF+ starting libraries are also presented.                     132  4.4 Discussion  Ultracentrifugation after cell lysis is the most common method for isolating the membrane proteome out of a raw cell lysate (Maddalo et al. 2011, Stenberg et al. 2005, Wu et al. 2011, Nielsen et al. 2005).  During ultracentrifugation however, large soluble complexes such as the ribosome and other high-abundance cytosolic proteins tend to co-sediment with the membrane fraction (Papanastasiou et al. 2013, Papanastasiou et al. 2016). These protein-rich ribosomal particles and soluble factors contaminate the membrane fractions and complicate mass spectrometry analyses since the signal intensities from these proteins masks the less abundant peptides derived from membrane proteins (Nielsen et al. 2005, Vuckovic et al. 2013, Roy et al. 2015).  To address this issue, we have explored a novel fractionation method to enrich the E. coli membrane proteome and to deplete large soluble complexes from crude membrane extracts. We have employed a peptidisc scaffold functionalized with a His-tag, which enables affinity purification of the membrane proteome after peptidisc library reconstitution. Our results show that the His-tagged peptidisc scaffold is fully functional for reconstitution of both purified membrane proteins as well as the global, unpurified membrane proteome. Our proteomic analysis further reveals that multiple membrane proteins and complexes are enriched after the library purification (Table 2), whereas multiple ribosomal and other soluble metabolic proteins are strongly depleted (Figure 4, Table 1). Our detailed data analysis (presented Figure 3 and Table 2) reveals a number of membrane proteins that are detected only in the purified library.     133  These proteins are often represented with low peptide numbers and low peptide intensities. This low degree of detection is likely due to their low abundance in the starting biological material and peptide masking in the complex protein-rich starting library. This latter observation underscores the importance of library purification for increasing coverage of the membrane proteome.  As a biologically relevant application of the method, we have surveyed changes in the membrane proteome upon SecDFyajC depletion, and we report here two novel observations. Firstly, we detect that SecA, the peripherally bound subunit of the SecYEG complex, is far more abundant in the library upon SecDFyajC depletion. This observation is in line with the pioneering work from the Oliver and Beckwith research groups, which showed that secretion defects caused by mutations in SecD and SecF lead to increased expression of SecA (Riggs, Derman, and Beckwith 1988, Rollo and Oliver 1988). Other studies have shown using protease protection assays that SecDFyajC depletion leads to decreased level of the SecYEG-inserted form of SecA (Economou et al. 1995, Duong and Wickner 1997b). It therefore possible that the up-regulation of SecA in SecDFyajC depletion conditions serves to compensate for a defective mode of interaction with the SecYEG translocon or with the membrane.  Secondly, the other protein that is neatly enriched in the library upon SecDFyajC depletion is the protein YibN. This putative inner membrane protein contains a single N-terminal transmembrane segment but no annotated function. Our recent interactome study indicates that YibN is a high-probability interactor of multiple subunits of the Sec translocon - most notably the core SecY and     134  SecG subunits, as well as SecD and the membrane insertase YidC (Carlson et al. 2019).  A previous study also reported strong up-regulation of YibN upon depletion of YidC (Wickstr\u00f6m et al. 2011), whereas a recent detergent-based AP\/MS approach identified YibN in the list of potential interactors of the SecYEG translocon (Jau\u00df et al. 2019). The exact association and biological role of YibN is still unknown but this growing body of evidence suggests a relationship of YibN with the Sec translocon, making this finding a promising avenue of future research.  In summary, the functionalized peptidisc scaffold is a valuable additional tool to advance research in membrane proteome biology and the findings presented in this pilot study already highlight its utility for detailed analysis of the bacterial membrane proteome with altered gene expression.  We note that the changes in levels of SecA was drastic enough to be detected before the library purification step, yet the purification step is advantageous because it greatly reduces the levels background noise level introduced by the high-abundance soluble contaminants (Figure 6). In other situations where changes in the membrane proteome may be more complex and the proteins less abundant, we anticipate that the His-peptidisc step will be critical to reveal with precision the subtle alterations in the membrane proteome. Looking forward, this workflow is amenable to multi-organelle mammalian cell lines to effectively monitor changes in membrane proteome and interactome in both wild-type and disease states.        135  Chapter 5: Conclusions and Future Directions  Sec-mediated protein translocation is a highly conserved process across all organisms and has been thoroughly characterized - particularly in bacteria - over many years. Previous studies in both prokaryotes and eukaryotes have revealed the identity of many of the key components of the Sec pathway through a combination of genetic, proteomic and biochemical evidence (Crane and Randall 2017, Park and Rapoport 2012, Rapoport, Li, and Park 2017).   Recent biochemical and structural studies on the bacterial Sec translocon have yielded valuable insights into the mechanism(s) of Sec-mediated protein translocation, as well as into interactions between the core Sec translocon and its known soluble binding partners (Bauer et al. 2014, Erlandson et al. 2008, Catipovic et al. 2019, Li et al. 2016, Ma et al. 2019, Allen et al. 2016, Fessl et al. 2018, Corey et al. 2019). The majority of these studies have focussed on interactions between the membrane embedded SecY complex and the cytosolic motor ATPase SecA, which binds onto the SecY complex and drives unfolded preprotein substrates across the membrane at the expense of ATP.  The mechanistic insights derived from these studies have given rise to two different mechanistic models for SecA-mediated protein translocation. Although these models differ in certain respects, they are not mutually exclusive. Both contend that preprotein translocation occurs largely through passive forward diffusion of the translocating preprotein substrate through the SecY channel, rather than on active ATP-dependent \"pushing\" by SecA (Bauer et al. 2014, Catipovic et al. 2019, Allen et al. 2016, Fessl et al. 2018). Both models     136  explain why the translocon is able to handle such a wide variety of proteins with no evident sequence similarity.   Despite these recent advances, some aspects of SecA-SecYEG interactions during the translocation reaction remain unclear. There is a body of evidence in the literature which suggests that SecA is a highly dynamic enzyme which must undergo successive cycles of binding and dissociation from the SecY complex in order for translocation to proceed efficiently (Mao et al. 2013, Morita, Tokuda, and Nishiyama 2012). Other studies, however, reported that SecA cycling - or \"SecA processivity\" - is not required for efficient protein translocation (Whitehouse et al. 2012, Sugano et al. 2017, Bauer et al. 2014). This mechanistic controversy is addressed in Chapter 2. Using in vitro protein translocation assays, we compared the translocation activity between wild-type SecA and a non-processive mutant - PrlD23. We reveal that the importance of SecA processivity during the translocation reaction varies depending on the length of the substrate undergoing transport: translocation of longer substrates depends on SecA processivity, while translocation of shorter substrates does not.   A further open question in the field of Sec-mediated protein translocation concerns the protein-protein interaction network, or the \"interactome\", of the Sec translocon. Previous studies from our lab and others have revealed that the Sec translocon interacts dynamically with multiple different membrane-bound ancillary subunits. The best characterized of these are the YidC membrane insertase and SecDF, which uses the energy of the proton motive force across the inner membrane to help drive protein translocation through the SecY complex (Duong and     137  Wickner 1997a, b, Kumazaki, Kishimoto, et al. 2014, Kumazaki, Chiba, et al. 2014, Tsukazaki et al. 2011, Tsukazaki 2018, Furukawa et al. 2017). These interactions have been studied extensively over many years, culminating with the isolation of a stable \"holo-complex\" consisting of one copy each of SecYEG, SecDF and YidC (Komar et al. 2016, Schulze et al. 2014, Botte et al. 2016, Collinson, Corey, and Allen 2015).  More recent studies, however, have suggested that the interactome of the Sec translocon may be more diverse than previously realized. A recent series of proteomics-based studies revealed that the membrane-bound periplasmic chaperones YfgM and PpiD also interact with the SecY complex (Maddalo et al. 2011, G\u00f6tzke et al. 2014, G\u00f6tzke et al. 2015, Jau\u00df et al. 2019). The ATP synthase complex has also recently been implicated as a potential interactor of the SecY complex (Chorev et al. 2018, Carlson et al. 2019). It has also recently been suggested that the Sec translocon may interact transiently with the BamABCDE complex in the bacterial outer membrane (Wang et al. 2016).  The key question addressed in Chapter 3 is: \"Does the Sec translocon have additional, as-yet-unidentified interacting partners?\" Previously characterized interacting partners and ancillary subunits of the Sec translocon were identified using conventional detergent-based methods (Duong and Wickner 1997a, G\u00f6tzke et al. 2014, Schulze et al. 2014, Jau\u00df et al. 2019). However, the destabilizing effects of detergents on multi-subunit membrane protein complexes is well-known: prolonged exposure to detergents during fractionation of the membrane proteome often leads to dissociation of transient interactions, causing all but the most stable interactions to be     138  disrupted (Babu et al. 2018, Babu et al. 2012, D\u00edaz-Mej\u00eda, Babu, and Emili 2009, Vuckovic et al. 2013). To circumvent the drawbacks of detergents, our laboratory recently developed the peptidisc as a streamlined, \"one-size-fits-all\" membrane mimetic to stabilize membrane proteins in a detergent-free environment. Our published work has demonstrated that the peptidisc is able to stably reconstitute both purified membrane proteins as well as the un-purified global membrane proteome (Carlson et al. 2018, Saville, Troman, and Duong Van Hoa 2019, Carlson et al. 2019).  The experimental work in Chapter 3 demonstrates that the peptidisc is superior compared to conventional detergent-based methods for preserving and stabilizing multi-subunit membrane protein complexes. We over-expressed the Sec translocon in SILAC labeling conditions and identified its co-purifying interacting partners by AP\/MS. In peptidisc, we identify a number of known interactors of the SecY complex, including YfgM, PpiD, and YidC. We also identify the outer membrane embedded Bam complex and the porin OmpA as interactors of the Sec translocon. An inner membrane protein with no annotated function, YibN, is also identified. By contrast, these interactions are largely undetected in a classical detergent-based workflow.  We were particularly intrigued by our observation that the Bam complex interacts with the Sec translocon, as this interaction would span across the bacterial periplasm. To further validate this observation, we performed systematic AP\/MS experiments on the Bam complex, and the Sec ancillary subunits YidC and SecDF.  We obtain compelling evidence that the Bam complex does     139  indeed interact with the Sec translocon, and that this interaction may be mediated at least in part by the large periplasmic domains of YidC and SecDF.   Our observation that the Sec translocon and the Bam complex interact directly is novel and exciting but requires additional experimental validation. Efforts are currently underway in our lab and others to isolate the Sec-Bam super-complex in sufficient yield and purity for structural and biochemical characterization. We note a recent preprint article from the Collinson group (https:\/\/doi.org\/10.1101\/589077) which seeks to isolate and stabilize the Sec-Bam super-complex in detergent solution using chemical crosslinkers. An alternative approach currently underway in our laboratory is to isolate the complex from native E. coli membranes, without resorting to over-expression.   Mohan Babu's research group (University of Regina) has constructed a library of E. coli strains in which every membrane protein ORF has been independently modified with a C-terminal 3 x FLAG tag (Babu et al. 2018). We will obtain strains containing the FLAG tagged Sec ancillary subunits employed in this study: SecY, SecD\/SecF, YfgM, PpiD, and YidC. We will perform small-scale FLAG purifications from these strains in peptidisc to evaluate which one(s) shows the most promising results for capturing the Sec-Bam super-complex. In parallel, we will also assess the ability of the FLAG-tagged Bam complex to co-purify with the Sec translocon in peptidisc. Our first major goal in this project is to determine the subunit composition and stoichiometry of the Sec-Bam super-complex. We will also seek to identify the main interaction interface(s) governing assembly of the Sec-Bam interaction. If we are able to obtain promising     140  results for these preliminary objectives, we will invest more time and effort to pursue a medium- to high-resolution structure of the Sec-Bam super-complex.  One central unresolved question in the field is: \"what are the implications of the existence of the Sec-Bam super-complex for the mechanism of outer membrane protein (OMP) insertion?\" The long-established model is that nascent OMPs are translocated into the periplasm via the Sec machinery, and are sequestered by chaperones until they are delivered to the Bam complex for insertion into the outer membrane (Hagan, Kim, and Kahne 2010, Roman-Hernandez, Peterson, and Bernstein 2014, Gu et al. 2016). The existence of a Sec-Bam super-complex suggests a second possible, more integrated mechanism: perhaps nascent OMPs can be transferred directly to the Bam complex after emerging on the periplasmic side of the Sec translocon? This important question clearly warrants additional exploration but is well beyond the scope of this thesis.   The focus of Chapter 4 is on improving the peptidisc library method for fractionation and characterization of the membrane proteome. In our work in Chapter 3, we noticed that our peptidisc library preparations were contaminated with numerous high-abundance soluble proteins, including ribosomal subunits and other large soluble complexes such as GroEL. These large soluble complexes often co-sediment with the membrane fraction during ultracentrifugation.  Critically for our studies, the abundance of these soluble contaminants during mass spectrometry analysis can potentially mask detection of peptides from less abundant integral membrane proteins (Nielsen et al. 2005, Vuckovic et al. 2013, Roy et al. 2015). Previously developed methods for removing soluble contaminants from membrane preparations     141  are relatively invasive, requiring solutions of high ionic strength, or even denaturants such as urea. (Papanastasiou et al. 2013, Papanastasiou et al. 2016, Wu et al. 2011). Thus, our goal in Chapter 4 was to develop a gentler method for purifying the membrane proteome away from high-abundance soluble contaminants.  With this goal in mind, we functionalized the peptidisc scaffold with a hexa-histidine tag. We demonstrate that the functionalized scaffold is fully functional for membrane protein reconstitution. Our proteomic data show that after formation of a peptidisc library, the bona fide membrane proteome can be purified by Ni-NTA chromatography in detergent-free buffer. Peptidisc-encapsulated integral membrane proteins, as well as peripherally bound membrane-associated, proteins are captured on the resin whereas soluble contaminants are washed away.   As a simple yet relevant case study, we applied this method to study changes in the membrane proteome caused by depletion of SecDFyajC, a well-characterized ancillary subunit of the Sec translocon which is essential for biogenesis of the bacterial membrane (Rapoport, Li, and Park 2017, Tsukazaki 2018, Tsukazaki et al. 2011). Previous studies have revealed that cells depleted in SecDFyajC display a strong growth defect (Kato, Nishiyama, and Tokuda 2003, Duong and Wickner 1997a, Pogliano and Beckwith 1994). Given these previous observations, we rationalized that depletion of SecDFyajC may cause a global change in the membrane proteome.   Application of our functionalized peptidisc workflow to this biological question reveals two important observations. First, we observe that SecA, the peripherally bound subunit of the     142  SecYEG complex, is far more abundant in the library upon SecDFyajC depletion. This observation is in line with previous in vivo work from the Oliver and Beckwith research groups, which showed that secretion defects caused by mutations in SecD and SecF lead to increased expression of SecA (Riggs, Derman, and Beckwith 1988, Rollo and Oliver 1988). Second, we observe strong enrichment of an unannotated inner membrane protein, YibN. YibN has been implicated as an interactor of the Sec translocon in multiple previous studies, including our recent peptidisc interactome study (Carlson et al. 2019, Jau\u00df et al. 2019). We further note a previous proteomic study which reported up-regulation of YibN upon depletion of the YidC membrane insertase, another well-known interactor of the Sec translocon (Wickstr\u00f6m et al. 2011).   What is the biological role of YibN, and what is the significance of its interaction(s) with the Sec translocon? A recent study reported up-regulation of YibN upon depletion of the YidC ancillary subunit of the Sec translocon (Wickstr\u00f6m et al. 2011). This earlier finding - together with our observation that SecDFyajC depletion also leads to upregulation of YibN - suggest that YibN may be involved in mediating protein translocation and\/or membrane protein insertion, possibly by modulating the activity of YidC and\/or SecDFyajC. Perhaps additional proteomic studies focussed on strains harbouring a YibN deletion may shed additional light on this important question. Since YibN is a non-essential gene, it may be necessary to combine YibN deletion with depletion of SecDFyajC and\/or YidC in order to see a strong effect (Baba et al. 2006).        143  As our data stands, we have demonstrated that the functionalized peptidisc scaffold is a valuable additional tool for researchers studying membrane proteomes. We have shown that the membrane proteome can be trapped in water-soluble particles in a single step. Soluble contaminants which co-sediment with the membrane fraction can be easily removed in an affinity purification step post-reconstitution. In light of the initial success of this method in our pilot study, the functionalized peptidisc workflow should next be applied toward characterization of the membrane proteome of mammalian cells. A more far-reaching goal is to catalogue changes in the membrane proteome between wild-type and disease states.      144  Bibliography  Ahdash, Z., E. Pyle, W. J. Allen, R. A. Corey, I. Collinson, and A. Politis. 2019. \"HDX-MS reveals nucleotide-dependent, anti-correlated opening and closure of SecA and SecY channels of the bacterial translocon.\"  Elife 8. doi: 10.7554\/eLife.47402. Alami, M., K. Dalal, B. Lelj-Garolla, S. G. Sligar, and F. Duong. 2007. \"Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA.\"  EMBO J 26 (8):1995-2004. doi: 10.1038\/sj.emboj.7601661. Allen, W. J., R. A. Corey, P. Oatley, R. B. Sessions, S. A. Baldwin, S. E. Radford, R. Tuma, and I. Collinson. 2016. \"Two-way communication between SecY and SecA suggests a Brownian ratchet mechanism for protein translocation.\"  Elife 5. doi: 10.7554\/eLife.15598. Angiulli, G., H. S. Dhupar, H. Suzuki, I. S. Wason, F. Duong Van Hoa, and T. Walz. 2020. \"New approach for membrane protein reconstitution into peptidiscs and basis for their adaptability to different proteins.\"  Elife 9. doi: 10.7554\/eLife.53530. Baba, T., T. Ara, M. Hasegawa, Y. Takai, Y. Okumura, M. Baba, K. A. Datsenko, M. Tomita, B. L. Wanner, and H. Mori. 2006. \"Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection.\"  Mol Syst Biol 2:2006.0008. doi: 10.1038\/msb4100050. Babu, M., C. Bundalovic-Torma, C. Calmettes, S. Phanse, Q. Zhang, Y. Jiang, Z. Minic, S. Kim, J. Mehla, A. Gagarinova, I. Rodionova, A. Kumar, H. Guo, O. Kagan, O. Pogoutse, H. Aoki, V. Deineko, J. H. Caufield, E. Holtzapple, Z. Zhang, A. Vastermark, Y. Pandya, C. C. Lai, M. El Bakkouri, Y. Hooda, M. Shah, D. Burnside, M. Hooshyar, J. Vlasblom, S. V. Rajagopala, A. Golshani, S. Wuchty, J. F Greenblatt, M. Saier, P. Uetz, T. F Moraes, J. Parkinson, and A. Emili. 2018. \"Global landscape of cell envelope protein complexes in Escherichia coli.\"  Nat Biotechnol 36 (1):103-112. doi: 10.1038\/nbt.4024. Babu, M., J. Vlasblom, S. Pu, X. Guo, C. Graham, B. D. Bean, H. E. Burston, F. J. Vizeacoumar, J. Snider, S. Phanse, V. Fong, Y. Y. Tam, M. Davey, O. Hnatshak, N. Bajaj, S. Chandran, T. Punna, C. Christopolous, V. Wong, A. Yu, G. Zhong, J. Li, I. Stagljar, E. Conibear, S. J. Wodak, A. Emili, and J. F. Greenblatt. 2012. \"Interaction landscape of membrane-protein complexes in Saccharomyces cerevisiae.\"  Nature 489 (7417):585-9. doi: 10.1038\/nature11354. Bariya, P., and L. L. Randall. 2018. \"Co-assembly of SecYEG and SecA fully restores the properties of the native translocon.\"  J Bacteriol. doi: 10.1128\/JB.00493-18. Bauer, B. W., and T. A. Rapoport. 2009. \"Mapping polypeptide interactions of the SecA ATPase during translocation.\"  Proc Natl Acad Sci U S A 106 (49):20800-5. doi: 10.1073\/pnas.0910550106. Bauer, B. W., T. Shemesh, Y. Chen, and T. A. Rapoport. 2014. \"A \"push and slide\" mechanism allows sequence-insensitive translocation of secretory proteins by the SecA ATPase.\"  Cell 157 (6):1416-29. doi: 10.1016\/j.cell.2014.03.063.     145  Botte, M., N. R. Zaccai, J. L. Nijeholt, R. Martin, K. Knoops, G. Papai, J. Zou, A. Deniaud, M. Karuppasamy, Q. Jiang, A. S. Roy, K. Schulten, P. Schultz, J. Rappsilber, G. Zaccai, I. Berger, I. Collinson, and C. Schaffitzel. 2016. \"A central cavity within the holo-translocon suggests a mechanism for membrane protein insertion.\"  Sci Rep 6:38399. doi: 10.1038\/srep38399. Carlson, M. L., R. G. Stacey, J. W. Young, I. S. Wason, Z. Zhao, D. G. Rattray, N. Scott, C. H. Kerr, M. Babu, L. J. Foster, and F. V. H. Duong. 2019. \"Profiling the E. coli membrane interactome captured in peptidisc libraries.\"  Elife 8. doi: 10.7554\/eLife.46615. Carlson, M. L., J. W. Young, Z. Zhao, L. Fabre, D. Jun, J. Li, H. S. Dhupar, I. Wason, A. T. Mills, J. T. Beatty, J. S. Klassen, I. Rouiller, and F. Duong. 2018. \"The Peptidisc, a simple method for stabilizing membrane proteins in detergent-free solution.\"  Elife 7. doi: 10.7554\/eLife.34085. Catipovic, M. A., B. W. Bauer, J. J. Loparo, and T. A. Rapoport. 2019. \"Protein translocation by the SecA ATPase occurs by a power-stroke mechanism.\"  EMBO J 38 (9). doi: 10.15252\/embj.2018101140. Chatzi, K. E., M. F. Sardis, A. Economou, and S. Karamanou. 2014. \"SecA-mediated targeting and translocation of secretory proteins.\"  Biochim Biophys Acta 1843 (8):1466-74. doi: 10.1016\/j.bbamcr.2014.02.014. Chatzi, K. E., M. F. Sardis, A. Tsirigotaki, M. Koukaki, N. \u0160o\u0161tari\u0107, A. Konijnenberg, F. Sobott, C. G. Kalodimos, S. Karamanou, and A. Economou. 2017. \"Preprotein mature domains contain translocase targeting signals that are essential for secretion.\"  J Cell Biol 216 (5):1357-1369. doi: 10.1083\/jcb.201609022. Chorev, D. S., L. A. Baker, D. Wu, V. Beilsten-Edmands, S. L. Rouse, T. Zeev-Ben-Mordehai, C. Jiko, F. Samsudin, C. Gerle, S. Khalid, A. G. Stewart, S. J. Matthews, K. Gr\u00fcnewald, and C. V. Robinson. 2018. \"Protein assemblies ejected directly from native membranes yield complexes for mass spectrometry.\"  Science 362 (6416):829-834. doi: 10.1126\/science.aau0976. Collinson, I. 2019. \"The Dynamic ATP-Driven Mechanism of Bacterial Protein Translocation and the Critical Role of Phospholipids.\"  Front Microbiol 10:1217. doi: 10.3389\/fmicb.2019.01217. Collinson, I., R. A. Corey, and W. J. Allen. 2015. \"Channel crossing: how are proteins shipped across the bacterial plasma membrane?\"  Philos Trans R Soc Lond B Biol Sci 370 (1679). doi: 10.1098\/rstb.2015.0025. Corey, R. A., Z. Ahdash, A. Shah, E. Pyle, W. J. Allen, T. Fessl, J. E. Lovett, A. Politis, and I. Collinson. 2019. \"ATP-induced asymmetric pre-protein folding as a driver of protein translocation through the Sec machinery.\"  Elife 8. doi: 10.7554\/eLife.41803. Crane, J. M., and L. L. Randall. 2017. \"The Sec System: Protein Export in.\"  EcoSal Plus 7 (2). doi: 10.1128\/ecosalplus.ESP-0002-2017. Cranford-Smith, T., and D. Huber. 2018. \"The way is the goal: how SecA transports proteins across the cytoplasmic membrane in bacteria.\"  FEMS Microbiol Lett 365 (11). doi: 10.1093\/femsle\/fny093. Dalal, K., H. Bao, and F. Duong. 2010. \"Modulation of the SecY channel permeability by pore mutations and trivalent cations.\"  Channels (Austin) 4 (2):83-6. doi: 10.4161\/chan.4.2.10533.     146  Dalal, K., C. S. Chan, S. G. Sligar, and F. Duong. 2012. \"Two copies of the SecY channel and acidic lipids are necessary to activate the SecA translocation ATPase.\"  Proc Natl Acad Sci U S A 109 (11):4104-9. doi: 10.1073\/pnas.1117783109. Dalal, K., and F. Duong. 2009. \"The SecY complex forms a channel capable of ionic discrimination.\"  EMBO Rep 10 (7):762-8. doi: 10.1038\/embor.2009.87. Dalal, K., N. Nguyen, M. Alami, J. Tan, T. F. Moraes, W. C. Lee, R. Maurus, S. S. Sligar, G. D. Brayer, and F. Duong. 2009. \"Structure, binding, and activity of Syd, a SecY-interacting protein.\"  J Biol Chem 284 (12):7897-902. doi: 10.1074\/jbc.M808305200. Ding, H., I. Mukerji, and D. Oliver. 2003. \"Nucleotide and phospholipid-dependent control of PPXD and C-domain association for SecA ATPase.\"  Biochemistry 42 (46):13468-75. doi: 10.1021\/bi035099b. Duong, F. 2003. \"Binding, activation and dissociation of the dimeric SecA ATPase at the dimeric SecYEG translocase.\"  EMBO J 22 (17):4375-84. doi: 10.1093\/emboj\/cdg418. Duong, F., and W. Wickner. 1997a. \"Distinct catalytic roles of the SecYE, SecG and SecDFyajC subunits of preprotein translocase holoenzyme.\"  EMBO J 16 (10):2756-68. doi: 10.1093\/emboj\/16.10.2756. Duong, F., and W. Wickner. 1997b. \"The SecDFyajC domain of preprotein translocase controls preprotein movement by regulating SecA membrane cycling.\"  EMBO J 16 (16):4871-9. doi: 10.1093\/emboj\/16.16.4871. Duong, F., and W. Wickner. 1998. \"Sec-dependent membrane protein biogenesis: SecYEG, preprotein hydrophobicity and translocation kinetics control the stop-transfer function.\"  EMBO J 17 (3):696-705. doi: 10.1093\/emboj\/17.3.696. Duong, F., and W. Wickner. 1999. \"The PrlA and PrlG phenotypes are caused by a loosened association among the translocase SecYEG subunits.\"  EMBO J 18 (12):3263-70. doi: 10.1093\/emboj\/18.12.3263. D\u00edaz-Mej\u00eda, J. J., M. Babu, and A. Emili. 2009. \"Computational and experimental approaches to chart the Escherichia coli cell-envelope-associated proteome and interactome.\"  FEMS Microbiol Rev 33 (1):66-97. doi: 10.1111\/j.1574-6976.2008.00141.x. D\u00f6rr, J. M., M. C. Koorengevel, M. Sch\u00e4fer, A. V. Prokofyev, S. Scheidelaar, E. A. van der Cruijsen, T. R. Dafforn, M. Baldus, and J. A. Killian. 2014. \"Detergent-free isolation, characterization, and functional reconstitution of a tetrameric K+ channel: the power of native nanodiscs.\"  Proc Natl Acad Sci U S A 111 (52):18607-12. doi: 10.1073\/pnas.1416205112. Economou, A., J. A. Pogliano, J. Beckwith, D. B. Oliver, and W. Wickner. 1995. \"SecA membrane cycling at SecYEG is driven by distinct ATP binding and hydrolysis events and is regulated by SecD and SecF.\"  Cell 83 (7):1171-81. doi: 10.1016\/0092-8674(95)90143-4. Erlandson, K. J., E. Or, A. R. Osborne, and T. A. Rapoport. 2008. \"Analysis of polypeptide movement in the SecY channel during SecA-mediated protein translocation.\"  J Biol Chem 283 (23):15709-15. doi: 10.1074\/jbc.M710356200. Fessl, T., D. Watkins, P. Oatley, W. J. Allen, R. A. Corey, J. Horne, S. A. Baldwin, S. E. Radford, I. Collinson, and R. Tuma. 2018. \"Dynamic action of the Sec machinery during initiation, protein translocation and termination.\"  Elife 7. doi: 10.7554\/eLife.35112.     147  Furukawa, A., S. Nakayama, K. Yoshikaie, Y. Tanaka, and T. Tsukazaki. 2018. \"Remote Coupled Drastic \u03b2-Barrel to \u03b2-Sheet Transition of the Protein Translocation Motor.\"  Structure 26 (3):485-489.e2. doi: 10.1016\/j.str.2018.01.002. Furukawa, A., K. Yoshikaie, T. Mori, H. Mori, Y. V. Morimoto, Y. Sugano, S. Iwaki, T. Minamino, Y. Sugita, Y. Tanaka, and T. Tsukazaki. 2017. \"Tunnel Formation Inferred from the I-Form Structures of the Proton-Driven Protein Secretion Motor SecDF.\"  Cell Rep 19 (5):895-901. doi: 10.1016\/j.celrep.2017.04.030. F\u00fcrst, M., Y. Zhou, J. Merfort, and M. M\u00fcller. 2018. \"Involvement of PpiD in Sec-dependent protein translocation.\"  Biochim Biophys Acta 1865 (2):273-280. doi: 10.1016\/j.bbamcr.2017.10.012. Gingras, A. C., K. T. Abe, and B. Raught. 2019. \"Getting to know the neighborhood: using proximity-dependent biotinylation to characterize protein complexes and map organelles.\"  Curr Opin Chem Biol 48:44-54. doi: 10.1016\/j.cbpa.2018.10.017. Gingras, A. C., M. Gstaiger, B. Raught, and R. Aebersold. 2007. \"Analysis of protein complexes using mass spectrometry.\"  Nat Rev Mol Cell Biol 8 (8):645-54. doi: 10.1038\/nrm2208. Gokhale, A., P. Perez-Cornejo, C. Duran, H. C. Hartzell, and V. Faundez. 2012. \"A comprehensive strategy to identify stoichiometric membrane protein interactomes.\"  Cell Logist 2 (4):189-196. doi: 10.4161\/cl.22717. Gold, V. A., S. Whitehouse, A. Robson, and I. Collinson. 2013. \"The dynamic action of SecA during the initiation of protein translocation.\"  Biochem J 449 (3):695-705. doi: 10.1042\/BJ20121314. Gouridis, G., S. Karamanou, M. F. Sardis, M. A. Sch\u00e4rer, G. Capitani, and A. Economou. 2013. \"Quaternary dynamics of the SecA motor drive translocase catalysis.\"  Mol Cell 52 (5):655-66. doi: 10.1016\/j.molcel.2013.10.036. Gu, Y., H. Li, H. Dong, Y. Zeng, Z. Zhang, N. G. Paterson, P. J. Stansfeld, Z. Wang, Y. Zhang, W. Wang, and C. Dong. 2016. \"Structural basis of outer membrane protein insertion by the BAM complex.\"  Nature 531 (7592):64-9. doi: 10.1038\/nature17199. G\u00f6tzke, H., C. Muheim, A. F. Altelaar, A. J. Heck, G. Maddalo, and D. O. Daley. 2015. \"Identification of putative substrates for the periplasmic chaperone YfgM in Escherichia coli using quantitative proteomics.\"  Mol Cell Proteomics 14 (1):216-26. doi: 10.1074\/mcp.M114.043216. G\u00f6tzke, H., I. Palombo, C. Muheim, E. Perrody, P. Genevaux, R. Kudva, M. M\u00fcller, and D. O. Daley. 2014. \"YfgM is an ancillary subunit of the SecYEG translocon in Escherichia coli.\"  J Biol Chem 289 (27):19089-97. doi: 10.1074\/jbc.M113.541672. Hagan, C. L., S. Kim, and D. Kahne. 2010. \"Reconstitution of outer membrane protein assembly from purified components.\"  Science 328 (5980):890-2. doi: 10.1126\/science.1188919. Havugimana, P. C., G. T. Hart, T. Nepusz, H. Yang, A. L. Turinsky, Z. Li, P. I. Wang, D. R. Boutz, V. Fong, S. Phanse, M. Babu, S. A. Craig, P. Hu, C. Wan, J. Vlasblom, V. U. Dar, A. Bezginov, G. W. Clark, G. C. Wu, S. J. Wodak, E. R. Tillier, A. Paccanaro, E. M. Marcotte, and A. Emili. 2012. \"A census of human soluble protein complexes.\"  Cell 150 (5):1068-81. doi: 10.1016\/j.cell.2012.08.011. Havugimana, P. C., P. Wong, and A. Emili. 2007. \"Improved proteomic discovery by sample pre-fractionation using dual-column ion-exchange high performance liquid     148  chromatography.\"  J Chromatogr B Analyt Technol Biomed Life Sci 847 (1):54-61. doi: 10.1016\/j.jchromb.2006.10.075. Hennon, S. W., R. Soman, L. Zhu, and R. E. Dalbey. 2015. \"YidC\/Alb3\/Oxa1 Family of Insertases.\"  J Biol Chem 290 (24):14866-74. doi: 10.1074\/jbc.R115.638171. Hu, L. Z., F. Goebels, J. H. Tan, E. Wolf, U. Kuzmanov, C. Wan, S. Phanse, C. Xu, M. Schertzberg, A. G. Fraser, G. D. Bader, and A. Emili. 2019. \"EPIC: software toolkit for elution profile-based inference of protein complexes.\"  Nat Methods. doi: 10.1038\/s41592-019-0461-4. Hu, P., S. C. Janga, M. Babu, J. J. D\u00edaz-Mej\u00eda, G. Butland, W. Yang, O. Pogoutse, X. Guo, S. Phanse, P. Wong, S. Chandran, C. Christopoulos, A. Nazarians-Armavil, N. K. Nasseri, G. Musso, M. Ali, N. Nazemof, V. Eroukova, A. Golshani, A. Paccanaro, J. F. Greenblatt, G. Moreno-Hagelsieb, and A. Emili. 2009. \"Global functional atlas of Escherichia coli encompassing previously uncharacterized proteins.\"  PLoS Biol 7 (4):e96. doi: 10.1371\/journal.pbio.1000096. Jau\u00df, B., N. A. Petriman, F. Drepper, L. Franz, I. Sachelaru, T. Welte, R. Steinberg, B. Warscheid, and H. G. Koch. 2019. \"Non-competitive binding of PpiD and YidC to the SecYEG translocon expands the global view on the SecYEG interactome in E. coli.\"  J Biol Chem. doi: 10.1074\/jbc.RA119.010686. Jiang, F., M. Chen, L. Yi, J. W. de Gier, A. Kuhn, and R. E. Dalbey. 2003. \"Defining the regions of Escherichia coli YidC that contribute to activity.\"  J Biol Chem 278 (49):48965-72. doi: 10.1074\/jbc.M307362200. Kariyazono, H., R. Nadai, R. Miyajima, Y. Takechi-Haraya, T. Baba, A. Shigenaga, K. Okuhira, A. Otaka, and H. Saito. 2016. \"Formation of stable nanodiscs by bihelical apolipoprotein A-I mimetic peptide.\"  J Pept Sci 22 (2):116-22. doi: 10.1002\/psc.2847. Kato, Y., K. Nishiyama, and H. Tokuda. 2003. \"Depletion of SecDF-YajC causes a decrease in the level of SecG: implication for their functional interaction.\"  FEBS Lett 550 (1-3):114-8. doi: 10.1016\/s0014-5793(03)00847-0. Kiefer, D., and A. Kuhn. 2018. \"YidC-mediated membrane insertion.\"  FEMS Microbiol Lett 365 (12). doi: 10.1093\/femsle\/fny106. Klock, H. E., and S. A. Lesley. 2009. \"The Polymerase Incomplete Primer Extension (PIPE) method applied to high-throughput cloning and site-directed mutagenesis.\"  Methods Mol Biol 498:91-103. doi: 10.1007\/978-1-59745-196-3_6. Koch, S., J. G. de Wit, I. Vos, J. P. Birkner, P. Gordiichuk, A. Herrmann, A. M. van Oijen, and A. J. Driessen. 2016. \"Lipids Activate SecA for High Affinity Binding to the SecYEG Complex.\"  J Biol Chem 291 (43):22534-22543. doi: 10.1074\/jbc.M116.743831. Komar, J., S. Alvira, R. J. Schulze, R. Martin, J. A. Lycklama A Nijeholt, S. C. Lee, T. R. Dafforn, G. Deckers-Hebestreit, I. Berger, C. Schaffitzel, and I. Collinson. 2016. \"Membrane protein insertion and assembly by the bacterial holo-translocon SecYEG-SecDF-YajC-YidC.\"  Biochem J 473 (19):3341-54. doi: 10.1042\/BCJ20160545. Komarudin, A. G., and A. J. M. Driessen. 2019. \"SecA-Mediated Protein Translocation through the SecYEG Channel.\"  Microbiol Spectr 7 (4). doi: 10.1128\/microbiolspec.PSIB-0028-2019. Kumazaki, K., S. Chiba, M. Takemoto, A. Furukawa, K. Nishiyama, Y. Sugano, T. Mori, N. Dohmae, K. Hirata, Y. Nakada-Nakura, A. D. Maturana, Y. Tanaka, H. Mori, Y. Sugita,     149  F. Arisaka, K. Ito, R. Ishitani, T. Tsukazaki, and O. Nureki. 2014. \"Structural basis of Sec-independent membrane protein insertion by YidC.\"  Nature 509 (7501):516-20. doi: 10.1038\/nature13167. Kumazaki, K., T. Kishimoto, A. Furukawa, H. Mori, Y. Tanaka, N. Dohmae, R. Ishitani, T. Tsukazaki, and O. Nureki. 2014. \"Crystal structure of Escherichia coli YidC, a membrane protein chaperone and insertase.\"  Sci Rep 4:7299. doi: 10.1038\/srep07299. Lambert, J. P., M. Tucholska, C. Go, J. D. Knight, and A. C. Gingras. 2015. \"Proximity biotinylation and affinity purification are complementary approaches for the interactome mapping of chromatin-associated protein complexes.\"  J Proteomics 118:81-94. doi: 10.1016\/j.jprot.2014.09.011. Lee, Y. C., J. A. B\u00e5\u00e5th, R. M. Bastle, S. Bhattacharjee, M. J. Cantoria, M. Dornan, E. Gamero-Estevez, L. Ford, L. Halova, J. Kernan, C. K\u00fcrten, S. Li, J. Martinez, N. Sachan, M. Sarr, X. Shan, N. Subramanian, K. Rivera, D. Pappin, and S. H. Lin. 2018. \"Impact of Detergents on Membrane Protein Complex Isolation.\"  J Proteome Res 17 (1):348-358. doi: 10.1021\/acs.jproteome.7b00599. Li, L., E. Park, J. Ling, J. Ingram, H. Ploegh, and T. A. Rapoport. 2016. \"Crystal structure of a substrate-engaged SecY protein-translocation channel.\"  Nature 531 (7594):395-399. doi: 10.1038\/nature17163. Ma, C., X. Wu, D. Sun, E. Park, M. A. Catipovic, T. A. Rapoport, N. Gao, and L. Li. 2019. \"Structure of the substrate-engaged SecA-SecY protein translocation machine.\"  Nat Commun 10 (1):2872. doi: 10.1038\/s41467-019-10918-2. Maddalo, G., F. Stenberg-Bruzell, H. G\u00f6tzke, S. Toddo, P. Bj\u00f6rkholm, H. Eriksson, P. Chovanec, P. Genevaux, J. Lehti\u00f6, L. L. Ilag, and D. O. Daley. 2011. \"Systematic analysis of native membrane protein complexes in Escherichia coli.\"  J Proteome Res 10 (4):1848-59. doi: 10.1021\/pr101105c. Maillard, A. P., S. Lalani, F. Silva, D. Belin, and F. Duong. 2007. \"Deregulation of the SecYEG translocation channel upon removal of the plug domain.\"  J Biol Chem 282 (2):1281-7. doi: 10.1074\/jbc.M610060200. Mao, C., C. E. Cheadle, S. J. Hardy, A. A. Lilly, Y. Suo, R. R. Sanganna Gari, G. M. King, and L. L. Randall. 2013. \"Stoichiometry of SecYEG in the active translocase of Escherichia coli varies with precursor species.\"  Proc Natl Acad Sci U S A 110 (29):11815-20. doi: 10.1073\/pnas.1303289110. Marty, M. T., K. C. Wilcox, W. L. Klein, and S. G. Sligar. 2013. \"Nanodisc-solubilized membrane protein library reflects the membrane proteome.\"  Anal Bioanal Chem 405 (12):4009-16. doi: 10.1007\/s00216-013-6790-8. Morita, K., H. Tokuda, and K. Nishiyama. 2012. \"Multiple SecA molecules drive protein translocation across a single translocon with SecG inversion.\"  J Biol Chem 287 (1):455-64. doi: 10.1074\/jbc.M111.301754. Moutaoufik, M. T., R. Malty, S. Amin, Q. Zhang, S. Phanse, A. Gagarinova, M. Zilocchi, L. Hoell, Z. Minic, M. Gagarinova, H. Aoki, J. Stockwell, M. Jessulat, F. Goebels, K. Broderick, N. E. Scott, J. Vlasblom, G. Musso, B. Prasad, E. Lamantea, B. Garavaglia, A. Rajput, K. Murayama, Y. Okazaki, L. J. Foster, G. D. Bader, F. S. Cayabyab, and M. Babu. 2019. \"Rewiring of the Human Mitochondrial Interactome during Neuronal     150  Reprogramming Reveals Regulators of the Respirasome and Neurogenesis.\"  iScience 19:1114-1132. doi: 10.1016\/j.isci.2019.08.057. Nielsen, P. A., J. V. Olsen, A. V. Podtelejnikov, J. R. Andersen, M. Mann, and J. R. Wisniewski. 2005. \"Proteomic mapping of brain plasma membrane proteins.\"  Mol Cell Proteomics 4 (4):402-8. doi: 10.1074\/mcp.T500002-MCP200. Papanastasiou, M., G. Orfanoudaki, M. Koukaki, N. Kountourakis, M. F. Sardis, M. Aivaliotis, S. Karamanou, and A. Economou. 2013. \"The Escherichia coli peripheral inner membrane proteome.\"  Mol Cell Proteomics 12 (3):599-610. doi: 10.1074\/mcp.M112.024711. Papanastasiou, M., G. Orfanoudaki, N. Kountourakis, M. Koukaki, M. F. Sardis, M. Aivaliotis, K. C. Tsolis, S. Karamanou, and A. Economou. 2016. \"Rapid label-free quantitative analysis of the E. coli BL21(DE3) inner membrane proteome.\"  Proteomics 16 (1):85-97. doi: 10.1002\/pmic.201500304. Park, E., and T. A. Rapoport. 2011. \"Preserving the membrane barrier for small molecules during bacterial protein translocation.\"  Nature 473 (7346):239-42. doi: 10.1038\/nature10014. Park, E., and T. A. Rapoport. 2012. \"Mechanisms of Sec61\/SecY-mediated protein translocation across membranes.\"  Annu Rev Biophys 41:21-40. doi: 10.1146\/annurev-biophys-050511-102312. Petriman, N. A., B. Jau\u00df, A. Hufnagel, L. Franz, I. Sachelaru, F. Drepper, B. Warscheid, and H. G. Koch. 2018. \"The interaction network of the YidC insertase with the SecYEG translocon, SRP and the SRP receptor FtsY.\"  Sci Rep 8 (1):578. doi: 10.1038\/s41598-017-19019-w. Pogliano, J. A., and J. Beckwith. 1994. \"SecD and SecF facilitate protein export in Escherichia coli.\"  EMBO J 13 (3):554-61. Pohlschr\u00f6der, M., E. Hartmann, N. J. Hand, K. Dilks, and A. Haddad. 2005. \"Diversity and evolution of protein translocation.\"  Annu Rev Microbiol 59:91-111. doi: 10.1146\/annurev.micro.59.030804.121353. Rajagopala, S. V., P. Sikorski, A. Kumar, R. Mosca, J. Vlasblom, R. Arnold, J. Franca-Koh, S. B. Pakala, S. Phanse, A. Ceol, R. H\u00e4user, G. Siszler, S. Wuchty, A. Emili, M. Babu, P. Aloy, R. Pieper, and P. Uetz. 2014. \"The binary protein-protein interaction landscape of Escherichia coli.\"  Nat Biotechnol 32 (3):285-290. doi: 10.1038\/nbt.2831. Rapoport, T. A., L. Li, and E. Park. 2017. \"Structural and Mechanistic Insights into Protein Translocation.\"  Annu Rev Cell Dev Biol 33:369-390. doi: 10.1146\/annurev-cellbio-100616-060439. Rattray, D. G., and L. J. Foster. 2019. \"Dynamics of protein complex components.\"  Curr Opin Chem Biol 48:81-85. doi: 10.1016\/j.cbpa.2018.11.003. Rees, J. S., K. S. Lilley, and A. P. Jackson. 2015. \"SILAC-iPAC: a quantitative method for distinguishing genuine from non-specific components of protein complexes by parallel affinity capture.\"  J Proteomics 115:143-56. doi: 10.1016\/j.jprot.2014.12.006. Riggs, P. D., A. I. Derman, and J. Beckwith. 1988. \"A mutation affecting the regulation of a secA-lacZ fusion defines a new sec gene.\"  Genetics 118 (4):571-9. Robson, A., A. E. Booth, V. A. Gold, A. R. Clarke, and I. Collinson. 2007. \"A large conformational change couples the ATP binding site of SecA to the SecY protein channel.\"  J Mol Biol 374 (4):965-76. doi: 10.1016\/j.jmb.2007.09.086.     151  Rollo, E. E., and D. B. Oliver. 1988. \"Regulation of the Escherichia coli secA gene by protein secretion defects: analysis of secA, secB, secD, and secY mutants.\"  J Bacteriol 170 (7):3281-2. doi: 10.1128\/jb.170.7.3281-3282.1988. Roman-Hernandez, G., J. H. Peterson, and H. D. Bernstein. 2014. \"Reconstitution of bacterial autotransporter assembly using purified components.\"  Elife 3:e04234. doi: 10.7554\/eLife.04234. Roux, K. J., D. I. Kim, M. Raida, and B. Burke. 2012. \"A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells.\"  J Cell Biol 196 (6):801-10. doi: 10.1083\/jcb.201112098. Roy, J., H. Pondenis, T. M. Fan, and A. Das. 2015. \"Direct Capture of Functional Proteins from Mammalian Plasma Membranes into Nanodiscs.\"  Biochemistry 54 (41):6299-302. doi: 10.1021\/acs.biochem.5b00954. Sachelaru, I., N. A. Petriman, R. Kudva, and H. G. Koch. 2014. \"Dynamic interaction of the sec translocon with the chaperone PpiD.\"  J Biol Chem 289 (31):21706-15. doi: 10.1074\/jbc.M114.577916. Sachelaru, I., L. Winter, D. G. Knyazev, M. Zimmermann, A. Vogt, R. Kuttner, N. Ollinger, C. Siligan, P. Pohl, and H. G. Koch. 2017. \"YidC and SecYEG form a heterotetrameric protein translocation channel.\"  Sci Rep 7 (1):101. doi: 10.1038\/s41598-017-00109-8. Sardis, M. F., A. Tsirigotaki, K. E. Chatzi, A. G. Portaliou, G. Gouridis, S. Karamanou, and A. Economou. 2017. \"Preprotein Conformational Dynamics Drive Bivalent Translocase Docking and Secretion.\"  Structure 25 (7):1056-1067.e6. doi: 10.1016\/j.str.2017.05.012. Sato, K., H. Mori, M. Yoshida, M. Tagaya, and S. Mizushima. 1997. \"Short hydrophobic segments in the mature domain of ProOmpA determine its stepwise movement during translocation across the cytoplasmic membrane of Escherichia coli.\"  J Biol Chem 272 (9):5880-6. doi: 10.1074\/jbc.272.9.5880. Saville, J. W., L. Troman, and F. Duong Van Hoa. 2019. \"PeptiQuick, a One-Step Incorporation of Membrane Proteins into Biotinylated Peptidiscs for Streamlined Protein Binding Assays.\"  J Vis Exp (153). doi: 10.3791\/60661. Schiebel, E., A. J. Driessen, F. U. Hartl, and W. Wickner. 1991. \"Delta mu H+ and ATP function at different steps of the catalytic cycle of preprotein translocase.\"  Cell 64 (5):927-39. doi: 10.1016\/0092-8674(91)90317-r. Schmidt, M., H. Ding, V. Ramamurthy, I. Mukerji, and D. Oliver. 2000. \"Nucleotide binding activity of SecA homodimer is conformationally regulated by temperature and altered by prlD and azi mutations.\"  J Biol Chem 275 (20):15440-8. doi: 10.1074\/jbc.M000605200. Schulze, R. J., J. Komar, M. Botte, W. J. Allen, S. Whitehouse, V. A. Gold, J. A. Lycklama A Nijeholt, K. Huard, I. Berger, C. Schaffitzel, and I. Collinson. 2014. \"Membrane protein insertion and proton-motive-force-dependent secretion through the bacterial holo-translocon SecYEG-SecDF-YajC-YidC.\"  Proc Natl Acad Sci U S A 111 (13):4844-9. doi: 10.1073\/pnas.1315901111. Scott, N. E., L. D. Rogers, A. Prudova, N. F. Brown, N. Fortelny, C. M. Overall, and L. J. Foster. 2017. \"Interactome disassembly during apoptosis occurs independent of caspase cleavage.\"  Mol Syst Biol 13 (1):906. doi: 10.15252\/msb.20167067.     152  Stacey, R. G., M. A. Skinnider, J. H. L. Chik, and L. J. Foster. 2018. \"Context-specific interactions in literature-curated protein interaction databases.\"  BMC Genomics 19 (1):758. doi: 10.1186\/s12864-018-5139-2. Stacey, R. G., M. A. Skinnider, N. E. Scott, and L. J. Foster. 2017. \"A rapid and accurate approach for prediction of interactomes from co-elution data (PrInCE).\"  BMC Bioinformatics 18 (1):457. doi: 10.1186\/s12859-017-1865-8. Stenberg, F., P. Chovanec, S. L. Maslen, C. V. Robinson, L. L. Ilag, G. von Heijne, and D. O. Daley. 2005. \"Protein complexes of the Escherichia coli cell envelope.\"  J Biol Chem 280 (41):34409-19. doi: 10.1074\/jbc.M506479200. Sugano, Y., A. Furukawa, O. Nureki, Y. Tanaka, and T. Tsukazaki. 2017. \"SecY-SecA fusion protein retains the ability to mediate protein transport.\"  PLoS One 12 (8):e0183434. doi: 10.1371\/journal.pone.0183434. Tam, P. C., A. P. Maillard, K. K. Chan, and F. Duong. 2005. \"Investigating the SecY plug movement at the SecYEG translocation channel.\"  EMBO J 24 (19):3380-8. doi: 10.1038\/sj.emboj.7600804. Tanaka, Y., Y. Sugano, M. Takemoto, T. Mori, A. Furukawa, T. Kusakizako, K. Kumazaki, A. Kashima, R. Ishitani, Y. Sugita, O. Nureki, and T. Tsukazaki. 2015. \"Crystal Structures of SecYEG in Lipidic Cubic Phase Elucidate a Precise Resting and a Peptide-Bound State.\"  Cell Rep 13 (8):1561-8. doi: 10.1016\/j.celrep.2015.10.025. Tomkiewicz, D., N. Nouwen, R. van Leeuwen, S. Tans, and A. J. Driessen. 2006. \"SecA supports a constant rate of preprotein translocation.\"  J Biol Chem 281 (23):15709-13. doi: 10.1074\/jbc.M600205200. Tsirigotaki, A., K. E. Chatzi, M. Koukaki, J. De Geyter, A. G. Portaliou, G. Orfanoudaki, M. F. Sardis, M. B. Trelle, T. J. D. J\u00f8rgensen, S. Karamanou, and A. Economou. 2018. \"Long-Lived Folding Intermediates Predominate the Targeting-Competent Secretome.\"  Structure 26 (5):695-707.e5. doi: 10.1016\/j.str.2018.03.006. Tsirigotaki, A., J. De Geyter, N. \u0160o\u0161taric, A. Economou, and S. Karamanou. 2017. \"Protein export through the bacterial Sec pathway.\"  Nat Rev Microbiol 15 (1):21-36. doi: 10.1038\/nrmicro.2016.161. Tsukazaki, T. 2018. \"Structure-based working model of SecDF, a proton-driven bacterial protein translocation factor.\"  FEMS Microbiol Lett 365 (12). doi: 10.1093\/femsle\/fny112. Tsukazaki, T., H. Mori, Y. Echizen, R. Ishitani, S. Fukai, T. Tanaka, A. Perederina, D. G. Vassylyev, T. Kohno, A. D. Maturana, K. Ito, and O. Nureki. 2011. \"Structure and function of a membrane component SecDF that enhances protein export.\"  Nature 474 (7350):235-8. doi: 10.1038\/nature09980. Uchida, K., H. Mori, and S. Mizushima. 1995. \"Stepwise movement of preproteins in the process of translocation across the cytoplasmic membrane of Escherichia coli.\"  J Biol Chem 270 (52):30862-8. doi: 10.1074\/jbc.270.52.30862. Vuckovic, D., L. F. Dagley, A. W. Purcell, and A. Emili. 2013. \"Membrane proteomics by high performance liquid chromatography-tandem mass spectrometry: Analytical approaches and challenges.\"  Proteomics 13 (3-4):404-23. doi: 10.1002\/pmic.201200340. Wang, Y., R. Wang, F. Jin, Y. Liu, J. Yu, X. Fu, and Z. Chang. 2016. \"A Supercomplex Spanning the Inner and Outer Membranes Mediates the Biogenesis of \u03b2-Barrel Outer     153  Membrane Proteins in Bacteria.\"  J Biol Chem 291 (32):16720-9. doi: 10.1074\/jbc.M115.710715. Whitehouse, S., V. A. Gold, A. Robson, W. J. Allen, R. B. Sessions, and I. Collinson. 2012. \"Mobility of the SecA 2-helix-finger is not essential for polypeptide translocation via the SecYEG complex.\"  J Cell Biol 199 (6):919-29. doi: 10.1083\/jcb.201205191. Wickstr\u00f6m, D., S. Wagner, P. Simonsson, O. Pop, L. Baars, A. J. Ytterberg, K. J. van Wijk, J. Luirink, and J. W. de Gier. 2011. \"Characterization of the consequences of YidC depletion on the inner membrane proteome of E. coli using 2D blue native\/SDS-PAGE.\"  J Mol Biol 409 (2):124-35. doi: 10.1016\/j.jmb.2011.03.068. Wu, F., D. Sun, N. Wang, Y. Gong, and L. Li. 2011. \"Comparison of surfactant-assisted shotgun methods using acid-labile surfactants and sodium dodecyl sulfate for membrane proteome analysis.\"  Anal Chim Acta 698 (1-2):36-43. doi: 10.1016\/j.aca.2011.04.039. Wu, X., C. Cabanos, and T. A. Rapoport. 2019. \"Structure of the post-translational protein translocation machinery of the ER membrane.\"  Nature 566 (7742):136-139. doi: 10.1038\/s41586-018-0856-x. Xin, Y., Y. Zhao, J. Zheng, H. Zhou, X. C. Zhang, C. Tian, and Y. Huang. 2018. \"Structure of YidC from Thermotoga maritima and its implications for YidC-mediated membrane protein insertion.\"  FASEB J 32 (5):2411-2421. doi: 10.1096\/fj.201700893RR. Young, J., and F. Duong. 2019. \"Investigating the stability of the SecA-SecYEG complex during protein translocation across the bacterial membrane.\"  J Biol Chem 294 (10):3577-3587. doi: 10.1074\/jbc.RA118.006447. Zhang, X. X., C. S. Chan, H. Bao, Y. Fang, L. J. Foster, and F. Duong. 2012. \"Nanodiscs and SILAC-based mass spectrometry to identify a membrane protein interactome.\"  J Proteome Res 11 (2):1454-9. doi: 10.1021\/pr200846y. Zhu, W., J. W. Smith, and C. M. Huang. 2010. \"Mass spectrometry-based label-free quantitative proteomics.\"  J Biomed Biotechnol 2010:840518. doi: 10.1155\/2010\/840518. Zimmer, J., Y. Nam, and T. A. Rapoport. 2008. \"Structure of a complex of the ATPase SecA and the protein-translocation channel.\"  Nature 455 (7215):936-43. doi: 10.1038\/nature07335.  ","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/hasType":[{"value":"Thesis\/Dissertation","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#dateIssued":[{"value":"2020-05","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt":[{"value":"10.14288\/1.0389991","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/language":[{"value":"eng","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline":[{"value":"Biochemistry and Molecular Biology","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/provider":[{"value":"Vancouver : University of British Columbia Library","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/publisher":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/rights":[{"value":"Attribution-NonCommercial-NoDerivatives 4.0 International","type":"literal","lang":"*"}],"https:\/\/open.library.ubc.ca\/terms#rightsURI":[{"value":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/","type":"literal","lang":"*"}],"https:\/\/open.library.ubc.ca\/terms#scholarLevel":[{"value":"Graduate","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/title":[{"value":"Exploring the interactome of the bacterial Sec translocon","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/type":[{"value":"Text","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#identifierURI":[{"value":"http:\/\/hdl.handle.net\/2429\/74194","type":"literal","lang":"en"}]}}