UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of interactions for BtuB, Colicin E3, and HslT Mills, Allan 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_november_mills_allan.pdf [ 12.52MB ]
Metadata
JSON: 24-1.0319168.json
JSON-LD: 24-1.0319168-ld.json
RDF/XML (Pretty): 24-1.0319168-rdf.xml
RDF/JSON: 24-1.0319168-rdf.json
Turtle: 24-1.0319168-turtle.txt
N-Triples: 24-1.0319168-rdf-ntriples.txt
Original Record: 24-1.0319168-source.json
Full Text
24-1.0319168-fulltext.txt
Citation
24-1.0319168.ris

Full Text

Characterization of interactions for BtuB, Colicin E3, and HslTbyAllan MillsB.Sc., Vancouver Island University, 2010M.Sc., The University of British Columbia, 2013A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OF Doctor of PhilosophyinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)October 2016©Allan Mills, 2016Abstract The outer membrane of Gram-negative bacteria acts as a physical barrier against the dangers of the extracellular environment. The outer membrane contains a number of porins and transporters to facilitate the import of nutrients while simultaneously protecting cells from extracellular assault. How these proteins transport nutrients and how they can be subverted are still areas of investigation. In the first study the mechanisms of transport through the vitamin B12 transporter BtuB is investigated. BtuB was found to interact in a 1:1 molar ratio with the inner membrane protein TonB, which is required for transport of vitamin B12 (cobalamin). Binding of TonB, in turn, alters the binding dynamics of the ligand with BtuB and slows the dissociation of ligand. In the second study transport of the antimicrobial protein colicin E3 across the outer membrane was investigated. Colicin E3 is a ribosomal nuclease that exists in a complex with an inhibitor, immunity protein Im3. Denaturation of colicin E3 was found to facilitate the interaction of the colicin with its outer membrane binding partners by dissociation of Im3. Release of Im3 from colicin E3 allows the nuclease domain of colicin E3 to interact with lipopolysaccharide as part of the transport process. Finally, OmpC and HslT from the Gram-negative Salmonella enterica serovar Typhimurium are hypothesized to interact to protect persistent infectious cells from the oxidizing assault of the immune system. No direct interaction between OmpC and HslT was detectable, possible explanations for this lack of interaction are discussed. These results are discussed in the context of how both ligands and antimicrobial compounds are transported across the outer membrane.iiPrefaceA version of chapter 2 has been published.•Mills, A., Le, H. T., & Duong, F. (2016). TonB-dependent ligand trapping in the BtuB transporter. Biochimica et Biophysica Acta (BBA)-Biomembranes,1858(12), 3105-3112.In the submitted manuscript I performed all experiments and prepared all figures. I jointly wrote the manuscript with my supervisor. Figure 2-1 through Figure 2-6 were all based on figures from this article.A version of chapter 3 is being prepared for submission.• Mills, A., and F. Duong. (2016). Colicin E3 binds lipopolysaccharide after release of the immunity protein. In progress. In the article I performed all experiments and prepared all figures. The article was co-written with my supervisor. Chapter 4 was prepared as part of a collaboration with the Finlay lab.• van der Heijden, J., Reynolds, L. A., Scholz, R., Mills, A. T., Imami, K., Foster, L. J., Franck Duong, F., and B. Finlay. (2016) Salmonella rapidly regulates membrane permeability to survive oxidative stress and form persisters. mBio, 7(4), e01238-16.iiiTable of ContentsAbstract .........................................................................................................................................iiPreface ..........................................................................................................................................iiiTable of Contents..........................................................................................................................iv...............................................................................................................................List of Tables viii................................................................................................................................List of Figures ix...........................................................................................................................List of Equations xi............................................................................................List of Symbols and Abbreviations xii.....................................................................................................................Acknowledgements xiv.....................................................................................................................................Dedication xv................................................................................................................Chapter 1: Introduction 1...............................................................................................1.1 TonB dependent transporters 3.....................................................................................................................................1.2 BtuB 5.....................................................................................................................................1.3 TonB 9............................................................................................1.3.1 TonB Transport Models 11..........................................................................................................................1.4 Colicin E3 12...................................................................................................1.4.1 Colicin E3 structure 13.............................................................................................1.4.2 Transport of Colicin E3 17...........................................................1.6 Outer membrane permeability and oxidative stress 20...........................................................................................................................1.7 Nanodiscs 21............................................................................................1.8 Scintillation Proximity Assay 22..........................................................................................................................1.9 Objectives 23.................................Chapter 2: TonB alters the binding kinetics of vitamin B12 with BtuB 25........................................................................................................................2.1 Introduction 25iv.......................................................................................................2.2 Materials and Methods 27...................................................................................................................2.2.1 Materials 27..................................................................................................2.2.2 Purification of BtuB 27..................................................................................................2.2.3 Purification of TonB 28..................................................................................................2.2.4 Purification of BtuF 29.........................................................................................2.2.5 Purification of Colicin E3 30.............................................................................................2.2.6 Nanodisc reconstitution 30.......................................................................................2.2.7 Multi-angle light scattering 31.............................................................................................2.2.8 BtuF competition assay 31..........................................................................2.2.9 Scintillation proximity assay (SPA) 32......................................................2.2.10 Trypsinization of TonB-dependent transporters 33.........................................................................................................2.2.11 Other methods 34................................................................................................................................2.3 Results 34.....................................................................2.3.1 Reconstitution of BtuB into Nanodiscs 34..................2.3.2 Ligand promotes recruitment of TonB to Nd-BtuB in a 1:1 molar ratio 36......................................................2.3.3 TonB alters the binding of ligand with Nd-BtuB 39................................................2.3.4 TonB does not increase the binding affinity of BtuB 44..............................................2.3.5 TonB slows the dissociation of ligand from Nd-BtuB 46.....................................................2.3.6 Deletion of TonB box abolishes binding of TonB 48.................................2.3.7 Mutations of valine 90 abolishes retention of ligand by TonB 49.......................................................2.3.8 Structural changes in BtuB upon ligand binding 52..........................................................................................................................2.4 Discussion 56.....................................................................2.4.1 Reconstitution of BtuB into Nanodiscs 56..................................................................2.4.2 Interaction of TonB with Nanodisc BtuB 57v.....................................2.4.3 TonB-dependent retention of ligand within Nanodisc BtuB 58..............................................................2.4.4 Conformational changes in Nanodisc BtuB 61.........................Chapter 3: Transport of bacteriocin colicin E3 across the outer membrane 64........................................................................................................................3.1 Introduction 64.......................................................................................................3.2 Materials and Methods 65................................................................................................3.2.1 Purification of OmpF 65...................................................................................................3.2.2 Purification of TolB 66..........................................................3.2.3 Expression of colicin E3A33C and E3D381C 66.............................................3.2.4 Expression and nanodisc reconstitution of BtuBT11C 67................................................................................................3.2.5 Fluorescence labeling 67............................................................................3.2.6 Purification of lipopolysaccharide 67.............................................................................................3.2.7 Purification of C96Im3 68.............................................................3.2.8 Purification of Im3-free colicin E3 and C96 69....................................................................................3.2.9 Purification of colicin E3-TR 69..............................................................................3.2.10 Isothermal titration calorimetry  70................................................................................................................................3.3 Results 70....................................3.3.1 Binding of colicin E3 to BtuB and TolB without detergents 70..............................................................................3.3.2 Binding of colicin E3 and OmpF 76.........................3.3.3 Interaction of colicin E3 nuclease domain with lipopolysaccharide 84..........................................................................................................................3.4 Discussion 903.4.1 Denaturation promotes the interaction of colicin E3 with outer membrane proteins............................................................................................................................................ 903.4.2 Colicin E3 interacts with lipopolysaccharide after dissociation of the immunity ................................................................................................................................protein 92.........Chapter 4: Interactions of OmpC and HslT from Salmonella enterica Typhimurium 96vi........................................................................................................................4.1 Introduction 96.......................................................................................................4.2 Materials and Methods 97.................................................................................................4.2.1 HslT Expression test 97.................................................................................4.2.2 HslT purification and refolding 97....................................................................................................4.2.3 OmpC Purification 98..................................................................................4.2.4 Pulldown of OmpC with HslT 99............................................................................4.2.5 Crosslinking of OmpC and HslT 100..............................................................................................................................4.3 Results 100.............................................................................4.3.1 HslT Purification and Refolding 100..................................................................................................4.3.2 OmpC Purification 104..............................................................4.3.3 HslT-OmpC interaction test on BN-PAGE 105.............................................................................................4.3.4 HslT-OmpC Pulldown 107.........................................................................4.3.5 HslT-OmpC chemical crosslinking 109..................................................................4.3.6 HslT-OmpC interaction on SDS-PAGE 110........................................................................................................................4.4 Discussion 112............................................................................Chapter 5: Summary and future directions 115...................................................................................................................................References 120viiList of TablesTable 2-1. Kinetic constants estimated from Scintillation Proximity Assay of Nd-BtuB.............48Table 2-2. Dissociation Rate Constants Estimated from Scintillation Proximity Assay...............52Table 3-1. Thermodynamic binding parameters of colicin-LPS interactions................................89viiiList of FiguresFigure 1-1: Outer membrane porins................................................................................................3Figure 1-2: Structure of BtuB..........................................................................................................5Figure 1-3: Model of Vitamin B12 Transport...................................................................................8Figure 1-4: Dimeric crystal structures of TonB.............................................................................10Figure 1-5: Competing TonB dependent transport models............................................................12Figure 1-6: Structure of colicin E3................................................................................................16Figure 1-7: Model of colicin E3 transport.....................................................................................20Figure 2-1: Reconstitution of BtuB into nanodiscs.......................................................................36Figure 2-2: Nd-BtuB interacts with TonB as a monomer..............................................................39Figure 2-3: TonB alters ligand binding dynamics.........................................................................43Figure 2-4: TonB does not increase receptor binding affinity.......................................................46Figure 2-5: TonB slows the dissociation of ligand from BtuB......................................................47Figure 2-6: Dissociation of B12 from BtuB∆6-12 and BtuBV90R.......................................................51Figure 2-7: Conformational changes in BtuB upon ligand binding...............................................54Figure 2-8: Conformational changes and plug movement in TonB-dependent transporters.........55Figure 3-1: Labeling or colicin E3 and BtuB.................................................................................72Figure 3-2: Colicin E3 binds BtuB reconstituted into nanodiscs...................................................74Figure 3-3: Colicin E3 spontaneously crosslinks with TolB..........................................................76Figure 3-4: Colicin E3 binds OmpF under denaturing conditions.................................................77Figure 3-5: SDS promotes the binding of colicin E3 to OmpF......................................................78Figure 3-6: SDS critical micelle concentration is required to promote binding............................79ixFigure 3-7: SDS promotes interaction only with colicin E3.........................................................80Figure 3-8: Colicin E3 binds OmpF in a 1:1 ratio........................................................................81Figure 3-9: OmpF, BtuB, and TolB must be structured to interact with colicin E3......................82Figure 3-10: Colicin E3 binds the OmpF periphery......................................................................83Figure 3-11: Colicin E3 interacts with lipopolysaccharide...........................................................85Figure 3-12: Isolation of colicin E3 from the immunity protein...................................................86Figure 3-13: Binding isotherm of colicin E3 to LPS.....................................................................87Figure 3-14: Binding isotherm of nuclease domain C96 to LPS...................................................88Figure 3-15: Binding isotherm of colicin E3-TR to LPS...............................................................90Figure 4-1: Total expression of HslT from E. coli BL21.............................................................101Figure 4-2: HslT localization in E. coli cells...............................................................................102Figure 4-3: HslT purification.......................................................................................................103Figure 4-4: HslT Refolding.........................................................................................................104Figure 4-5: OmpC purification from Salmonella enterica..........................................................105Figure 4-6: HslT-OmpC binding on non-denaturing gel.............................................................106Figure 4-7: Excision of protein complexes from BN-PAGE.......................................................107Figure 4-8: HslT-OmpC pulldown repeat....................................................................................108Figure 4-9: Chemical crosslinking of OmpC and HslT...............................................................110Figure 4-10: HslT interaction with OmpC on SDS-PAGE..........................................................111Figure 4-11: HslT shift with OmpC in SDS................................................................................112xList of EquationsEquation 2-1...................................................................................................................................33Equation 2-2...................................................................................................................................33xiList of Symbols and AbbreviationsAAA  ATPase Associated with a variety of cellular ActivitiesATP  Adenosine triphosphateDDM  n-dodecyl-β-D-maltopyranosideH2O2  Hydrogen PeroxideIPTG  Isopropyl-β-D-thiogalactopyranosidekDa  kilodalton = 1000 gram mole-1KDO  2-Keto-3-DeoxyoctonateLB  Luria-Bertani brothLDAO  Lauryldimethylamine oxideLPS  LipopolysaccharideMSP  Membrane Scaffold ProteinNd-BtuB Nanodisc BtuBNd-FhuA Nanodisc FhuAOBS  OmpF Binding SiteOctyl-POE n-Octyl-PolyoxyethyleneOMP  Outer Membrane ProteinPAGE  Polyacrylamide Gel Electrophoresis PCR  Polymerase Chain ReactionPDB  Protein Data BankPMSF  Phenymethylsulfonyl fluoridePVT  Polyvinyl ToluenexiiROS  Reactive Oxygen SpeciesrRNase Ribosomal RNaseSDS   Sodium Dodecyl Sulfate SPA  Scintillation Proximity AssaySPR  Surface Plasmon ResonanceTBE  TolB Binding ElementxiiiAcknowledgements I would foremost give my thanks to my thesis advisor, Dr. Franck Duong. He is a very gifted scientist and it is a great privilege to work with him on this exciting project. I am grateful not only for his suggestions on my project but also for his training in manuscript and presentation preparation.  I am very fortunate for having the opportunity to work alongside a number of talented post-docs, graduate students, and lab technicians including Le Hai-Tuong, Badreddine Douzi, Michael Carlson, John Young, Huan Bao, Kush Dalal, Han-Sol Wan, Jean-François Montariol, Harpreet Sandhu, and Harvir Singh.  My committee members Dr. Joerg Gsponer, Dr. Calvin Yip, and Dr. Michel Roberge each deserve many thanks for their support and guidance in this project over the years.  This project was generously funded by the Natural Science and Engineering Program of Canada and the University of British Columbia. xivDedicationFor my familyxvChapter 1: Introduction Gram-negative bacteria have two separate membranes, an inner and outer membrane, separated by a small space termed the periplasm containing a thin layer of petidoglycan and specialized proteins (Nikaido, 2009). The outer membrane largely acts to protect the cell from the environmental conditions the cell may encounter and it is structurally unlike other lipid bilayers by having an asymmetric composition of the two leaflets. The inner leaflet, facing the periplasmic space between the inner and outer membranes, is composed of phospholipids. The outer leaflet, facing the extracellular environment, is composed primarily of lipopolysaccharide (LPS) (Nikaido, 2009).  LPS, also known as endotoxin, is composed of three major components: lipid A, core oligosaccharide, and O-antigen (Schletter et al. 1995). Lipid A forms the lipid interface with the outer membrane and has immunogenic activity (Galanos et al. 1985). The core oligosaccharide is composed of 2-keto-3-deoxyoctonate (KDO) sugars and may contain a combination of phosphates, pyrophosphates, and 2-aminoethylphosphate/pyrophosphate. This high density of negative charges attracts divalent cations, such as Ca2+, from the extracellular environment to maintain integrity of the outer membrane (Rietschel et al. 1994). O-antigens are composed of variable lengths of repeating glycosyl residues. Bacteria produce LPS of varying O-antigen lengths such that LPS preparations from a single colony are heterogeneous (Rietschel et al. 1994; Chart et al. 2000). Long lengths of O-antigens also inhibit the binding of both bacteriophages 1and antimicrobial peptides to protect cells, presumably by steric hindrance (Van der Ley et al. 1986).  The outer membrane of Gram-negative bacteria is also more impermeable than other lipid bilayers. Hydrophobic molecules, normally able to cross lipid bilayers by dissolving into the hydrophobic interior, are greatly slowed crossing the outer membrane. This is because the LPS of the outer leaflet have reduced fluidity compared to phospholipids due to multiple saturated fatty acid chains connected to individual head groups. Gram-positive bacteria are susceptible to hydrophobic antibiotics such as penicillin G, rifamycin, and erythromycin whereas these antibiotics have little effect on Gram-negative bacteria (reviewed in Nikaido, 2009).  The limited permeability of the outer membrane poses the problem of how to transport nutrients and other molecules that the bacterial cell requires. The outer membrane hosts a number of transmembrane proteins that serve as pores and transporters for nutrients. The family of porins include monomeric proteins such as OmpA and trimeric proteins such as LamB, OmpF and OmpC (Nikaido, 2003). Transporters of the outer membrane, with the exception of Wza (Dong et al. 2006), are β-barrel proteins with a central lumen connecting the extracellular environment to the periplasm. Channel proteins in the outer membrane permit small molecules such as water, salt, sugars and other molecules below approximately 600 daltons to transit the lumens of these porins by passive diffusion (Figure 1-1). Iron siderophores and vitamin B12 (also referred to as cobalamin) are larger than 600 daltons (>700 Da to approximately 1 kDa) (Nikaido, 2003). These nutrients are also present at low concentrations rendering passive diffusion an inefficient 2method of absorbing these nutrients (Raymond et al. 2003). To solve these problems Gram-negative bacteria have evolved the active transport system of TonB-dependent transporters.Figure 1-1: Examples of outer membrane porins. Crystal structures of the trimeric porin OmpF (Efremov and Sazanov, 2012) (PDB: 3POX) depicted facing into the extracellular surface (A). Monomeric OmpA (Cierpicki et al. 2006) (PDB: 2GE4) depicted in the plane of the membrane with extracellular loops pointing upwards and periplasmic loops pointing downwards (B). 1.1 TonB dependent transporters TonB-dependent transporters are active transporters for iron siderophores and vitamin B12. Structurally, these proteins are β-barrel proteins with a channel lumen connecting the extracellular space and the periplasm. However, TonB-dependent transporters are 22-stranded β-barrels, unlike the smaller 16-stranded β-barrels of porins such as OmpF (Nikaido, 2003), and have an N-terminal globular plug domain occluding passage through the lumen (Ferguson and Deisenhofer, 2004). The larger barrels of these proteins would permit passage of large solutes and render the cell susceptible to toxic environmental compounds. Deletion of large internal 3sequences of one TonB-dependent transporter, FepA, allowed large antibiotics to penetrate the outer membrane (Rutz et al. 1992). The β-barrels of TonB-dependent transporters have long extracellular loops and short periplasmic turns between the β-strands (Figure 1-2A). To prevent diffusion of solutes through the barrel lumen an approximately 150 residue “cork” or “plug” domain occludes the channel of the barrel after folding in from the periplasmic side (Figure 1-2B). Residues of both the plug domain and the extracellular loops contribute to the formation of a high-affinity ligand binding site. Binding of ligands occurs in a two-step induced fit process in the case of FepA (Payne et al. 1997), ShuA (Cobessi et al. 2010), and FyuA (Noinaj et al. 2010) in which the transporter first binds the ligand in a low affinity binding event before undergoing conformational changes to sequester the ligand with a high affinity. In addition to induced fit, other conformational changes occur in TonB-dependent transporters after binding of ligand. An N-terminal motif termed the “TonB box” extending from the plug domain rests within the barrel of the transporter. Upon binding of ligand this TonB box emerges from the barrel into the periplasm (reviewed in Noinaj et al. 2010). This signals the ligand loaded state of the transporter to the periplasm allowing them to recruit inner membrane proteins. Most ligands transported by TonB-dependent transporters are iron siderophores. However, one non-iron siderophore ligand is vitamin B12 (cobalamin). B12 is a water soluble cobalt-containing enzyme cofactor that plays a critical role in such enzymes as methionine synthase (Drennan et al. 1994), ethanolamine-ammonia lyase, propanediol dehydratase, and glycerol dehydratase. However, unlike other bacterial species, B12 cannot be synthesized de novo by Escherichia coli (Lawrence and Roth, 1996). E. coli transports this nutrient to solve potential 4shortages through use of the outer membrane transporter BtuB and the TonB-dependent transport system. A BFigure 1-2: Structure of BtuB. A) Crystal structure of Ca2+-BtuB shown in the outer membrane. The β-barrel (blue) is shown with the extracellular loops (top) and periplasmic surface (bottom) and with residues 381-511 removed. The plug domain (residues 6-132) is depicted in orange. B) Top-down view from the extracellular side of BtuB. The plug domain (orange) is shown obstructing the lumen of the β-barrel. (PDB: 1NQG).1.2 BtuB Unlike other TonB-dependent transporters, BtuB transports vitamin B12 rather than iron siderophores (Di Masi et al. 1973), sugar (Neugebauer et al. 2005), or zinc (Calmettes et al. 2015). Vitamin B12 is approximately 1,355.37 Da in mass, larger than the 600 Da mass permitted through outer membrane porins (Nikaido, 2003). Therefore only TonB-dependent transporters are large enough to accommodate vitamin B12. BtuB is structurally similar to other TonB-dependent transporters with a 150-residue N-terminal plug domain occluding the channel of a 22-stranded C-terminal β-barrel (Chimento et al. 2003b). The vitamin B12 binding site on the 5extracellular side of the protein is composed of residues of both the extracellular loops of the β-barrel and the plug domain. The extracellular loops undergo conformational change after binding calcium (Chimento et al. 2003b). The binding of two calcium ions and the subsequent ordering of the extracellular loops facilitates the high affinity binding of vitamin B12 to BtuB (Bradbeer et al. 1986; Chimento et al. 2003). BtuB undergoes a number of conformational changes upon binding of ligands. Binding of B12 causes movement of an apical loop (residues 85 to 95) on the extracellular side of the BtuB plug domain toward the bound B12 by approximately 6Å. After binding of vitamin B12 BtuB undergoes a conformational change in the TonB box. This region of the protein is structured and sequestered within the periplasmic side of the barrel (Chimento et al. 2003b). After binding of B12, however, it becomes unstructured (Chimento et al. 2003b) and emerges from the barrel into the periplasmic space in which it is able to recruit TonB (Freed et al. 2010). Proline substitution mutations introduced into the TonB box prevent import of bound ligands by disrupting the region and inhibiting the binding of TonB (Gudmundsdottir et al. 1989). Binding of TonB to ligand-loaded BtuB also leads to additional movement of the apical loop by an additional 2Å toward the bound ligand (Shultis et al. 2006). The role of this apical loop movement is currently unknown.  BtuB and other TonB-dependent transporters act as gated channels and require dislocation of the plug domain to permit import of bound ligands. This is achieved in concert with the inner membrane protein complex of TonB-ExbB-ExbD (Braun, 1995). TonB is an inner membrane protein with a single N-terminal transmembrane anchor that spans the periplasm and 6forms a complex with inner membrane proteins ExbB and ExbD, which are homologues of the flagellar motor proteins MotA and MotB (Cascales et al. 2001). ExbB-ExbD harvests the proton motive force (PMF) across the inner membrane and uses it to power the import of ligands bound on the extracellular side of the transporters. Ligands imported into the periplasm are then bound by soluble binding proteins and presented to transporters in the inner membrane for transport into the cytoplasm (Figure 1-3). The structure of TonB and its interactions with both the inner and outer membrane are still areas of study.7Outer MembraneInner MembraneTonB ExbBExbDPeriplasmH+N-terminusTonB boxBtuCDBtuFBtuB Figure 1-3: Model of Vitamin B12 Transport. BtuB recruits the TonB-ExbB-ExbD complex after becoming ligand-loaded (Shultis et al. 2006 PDB: 2GSK). TonB (green) spans the periplasm (Domingo Köhler et al. 2010) to transduce energy from the proton gradient harvested by ExbB-ExbD across the inner membrane to BtuB. ExbB (purple) has three transmembrane domains (Kampfenkel and Braun, 1993) while ExbD (red) has a single transmembrane domain (Kampfenkel and Braun, 1992) (Garcia-Herrero et al. 2007 PDB: 2PFU). After entry into the periplasm vitamin B12 is bound by BtuF (violet) (Karpowich et al. 2003 PDB: 1N4A) and presented to the inner membrane complex BtuCD (cyan) (Korkhov et al. 2014 PDB: 4R9U) for transport into the cytoplasm. 81.3 TonB  TonB is a 239-residue protein that is composed of several distinct motifs. TonB has an inner membrane spanning anchor of residues 1 to 32. This anchor is presumably an α-helix (Karlsson et al. 1993). TonB is known to interface with ExbB and ExbD through a particular His20 residue of this transmembrane anchor. Mutations of this residue result in abrogation of transport (Larsen et al. 2007). TonB possesses a poly-proline tract between residues 66 to 100 wherein every third residue is a proline giving the protein a rigid and elongated structure (Evans et al. 1986). This rigidity of TonB allows it to remain within the inner membrane while interacting with the periplasmic surface of transporters (Domingo Köhler et al. 2010). The C-terminal region of TonB is a globular region that interfaces the TonB box of transporters. Crystal structures of the C-terminus of TonB were not obtained until Chang et al. 2001 truncated TonB from residues 155 to 239. This showed that the C-terminal domain exists in a dimeric state (Chang et al. 2001) (Figure 1-4A). A larger construct showed a different structure (Ködding et al. 2005) (Figure 1-4B). Dimers have been reported in vivo (Sauter et al. 2003) and dimerization may occur during interactions with some transporters (Khursigara et al. 2004; Khursigara et al. 2005) however Postle et al. 2010 demonstrated that these dimers of TonB do not exist in vivo. Moreover, TonB dimerization is not necessary for the interaction of TonB with transporters (Koedding et al. 2004) and TonB may interact solely as a monomer with TonB-dependent transporters (Freed et al. 2013; Mills et al. 2014).9Figure 1-4: Dimeric crystal structures of TonB. A) Tightly intertwined dimeric TonB (residues 164 to 239) (Chang et al. 2001) (PDB: 1IHR). B) Alternative dimeric crystal structure of TonB (residues 148 to 239) (Ködding et al. 2005) (PDB: 1U07). TonB interacts with transporters through the “TonB box” on the transporter. These motifs are near the N-terminus of the transporter and consist of a largely conserved sequence. A peptide with a sequence similar to the conserved TonB box consensus (Glu-Thr-Val-Ile-Val) was found to inhibit the growth of cells grown on low-iron media and the killing of cells by colicins (Tuckman and Osbourne, 1992). TonB binds the TonB box in a β-strand exchange (Pawelek et al. 2006; Shultis et al. 2006; Brillet et al. 2007). The accessibility of the TonB box and the mechanism in which it becomes bound by TonB varies between different TonB-dependent transporters (Kim et al. 2007). FhuA and FecA both possess a region called the switch helix near the TonB box, this helix has been shown to unwind upon binding of the iron siderophore ligand (Ferguson et al. 1998; Ferguson et al. 2002). The TonB box of FhuA is intrinsically unstructured regardless of the ligand-loaded state of the transporter (Ferguson et al. 1998). The TonB box of FhuA is sequestered inside the periplasmic end of the barrel and the unwinding of the switch helix may extend the TonB box into the periplasm wherein it can recruit TonB (Braun, 2009). BtuB has a different mechanism for signaling its ligand-loaded status. BtuB does not have a switch helix and the BtuB TonB box is structured in the absence of ligand. Upon binding of 10ligand, however, this region becomes unstructured after binding of vitamin B12 (Chimento et al. 2003b). This region becomes structured again after recruiting TonB and engaging in β-strand exchange (Shultis et al. 2006). 1.3.1 TonB Transport Models There are two major competing models for how TonB is able to drive the import of ligands into the periplasm after binding the TonB box.  The pulling model posits that TonB undergoes conformational changes in response to energization by ExbB-ExbD that retract TonB toward the inner membrane distorting the bound plug domain. In silico simulations showed the plug of the vitamin B12 TonB-dependent transporter BtuB becomes unfolded by a pulling force exerted by TonB (Gumbart et al. 2007). This model is partially supported by the different conformational changes that TonB undergoes during the transport process (Larsen et al. 1999). Atomic force microscopy of the BtuB-TonB complex supports a pulling model of transport with detection of the partial unfolding of the plug domain (Hickman et al. 2014) (Figure 1-5A) The (Rotational Surveillance and Energy Transfer) ROSET model proposes that energy is transferred from the inner to the outer membrane through rotational motion of TonB (Klebba, 2016). Fluorescence measurements of TonB-GFP fusion protein in vivo showed rotational motion that was halted by dissipation of the PMF (Jordan et al. 2013). Rotational motion of TonB was initially suggested by the homology of ExbB-ExbD to MotA-MotB of the flagellar motor complex (Cascales et al. 2001). The lesser fluidity of the outer membrane compared to the inner 11membrane would allow force to be transferred into the plug domain without inducing an overall rotation in the transporter (Naumann et al. 1987; Rassam et al. 2015). The ExbB-ExbD complex, unlike MotA-MotB complex, does not have a stator to prevent rotation within the membrane (De Mot and Vanderleyden, 1994). Likely, the mass of the ExbB-ExbD complex outweighs TonB causing TonB to rotate rapidly within the inner membrane while retarding the counter-rotation of ExbB-ExbD (Klebba, 2016) (Figure 1-5B).Figure 1-5: Competing TonB-dependent Transport models. A) The “Pulling” model proposes that TonB undergoes conformational changes during transport to alter the plug domain and permit entry of bound ligands. B) The ROSET model suggests that rotary motion of TonB occurs to alter the plug domain and drive import of bound ligands. 1.4 Colicin E3 Aside from binding and transporting vitamin B12, BtuB also binds other ligands such as colicins. Colicins are antimicrobial toxins produced by bacteria under SOS response (Jacob et al. 1952). Colicins have a variety of killing methods including pore formation (Baty et al. 1987), 12inhibition of cell wall synthesis (Harkness and Braun, 1989), DNase (Lau and Condie, 1989), transfer RNase (Yajima et al. 2006), and ribosomal RNase (Senior et al. 1970). Colicin expressing cells also co-express an immunity protein. This immunity protein inhibits colicin activity and prevents it from killing the producing cells. Immunity protein is also expressed by neighboring cells from the same colony to survive the toxin assault (Kleanthous and Walker, 2001).  Colicin E3 is one such colicin that binds BtuB and uses a toxic ribosomal RNase to kill Escherichia coli cells (reviewed by Cascales et al. 2007).1.4.1 Colicin E3 structure All colicins follow a similar domain organization with an N-terminal transport domain (T-domain), a central receptor binding domain (R-domain), and a C-terminal active domain that kills cells. The T-domain interacts with specific molecular machineries of the bacteria to facilitate transport into the cell, the R-domain binds to specific surface receptors with high affinity to both adhere and concentrate the colicin on the exterior of targeted bacteria, and the active domain acts to kill the cells (Braun et al. 1994; Cascales et al. 2007).  Colicin E3 is a Y-shaped protein with the cognate immunity protein (Im3) bound between the N- and C-terminal globular domains (Soelaiman et al. 2001). The T-domain is composed of residues 1 to 312 with the first 83 residues of the protein being unstructured (Soelaiman et al. 2001; Yamashita et al. 2008). This unstructured domain contains three individual sequences of two OmpF binding sites termed OBS1 (residues 2-18) and OBS2 (residues 53-64) flanking a 13single TolB binding element (TBE) (residues 32-47). OBS1 binds OmpF with a 2µM affinity while OBS2 binds with a 24µM affinity (Housden et al. 2010). The TBE binds the periplasmic protein TolB with an approximate 125nM affinity (Bonsor et al. 2009). The remainder of the N-terminal T-domain is a structured globular domain that acts as a binding site for Im3 (Soelaiman et al. 2001).  The R-domain binds BtuB with high affinity (Cavard, 1994; Kurisu et al. 2003). The R-domain is composed of residues 313-450 and is a coiled coil stalk that extends ~100Å away from the rest of the protein. The tip of the coiled coil forms a hairpin studded with hydrophobic residues (Soelaiman et al. 2001). The hydrophobic tip of the R-domain interacts with the vitamin B12 binding site of BtuB and orients the protein approximately 45º perpendicular to the membrane surface. Upon binding BtuB the N- and C- terminal ends of the R-domain become disordered (Kurisu et al. 2003). Upon binding BtuB the immunity protein dissociates from the N-terminal T-domain but remains attached to the C-terminal nuclease domain (Walker et al. 2003; Zakharov et al. 2008) (Figure 1-6A).  The C-terminal active domain of colicin E3 is a ribosomal RNase (Senior et al. 1970; 1971). The rRNase specifically cleaves the 16S ribosomal subunit between nucleotides A1493 and G1494 within the ribosomal A site (Bowman et al. 1971; Senior and Holland, 1971). Cleavage of the ribosomes effectively prevents protein synthesis and ultimately kills the cell. Interestingly, the active site of the rRNase (residues D510, H513, E515, and E517 from Ng, et al. 2010) is not blocked by Im3. Prevention of ribosome cleavage is achieved by both steric 14hindrance of the rRNase domain from the large ribosome target and through charge repulsion of the negatively charged Im3 (calculated pI=3.96) against the negatively charged ribosome (Soelaiman et al. 2001) (Figure 1-6B). Thermodynamic investigation of the binding affinities between colicin E3 and Im3 found that the binding affinity was in the near femtomolar range (Kd = 10-14 M) but that the C-terminal rRNase domain alone binds in a weaker picomolar range (Kd = 10-12 M) (Walker et al. 2003).15 Figure 1-6: Structure of colicin E3. A) Co-crystal structure of colicin E3 (blue) and the immunity protein, Im3 (green). The N-terminal 83 residues are disordered and are not shown. Globular N-terminal T-domain and C-terminal nuclease domain sandwich the Im3 in a high affinity binding interaction. The central BtuB binding R-domain is a coiled-coil extending below the protein (Soelaiman et al. 2001) (PDB: 1JCH). B) Succession of the T-domain (black, residues 1-313), the R-domain (dark grey, residues 313-447), and the nuclease domain (light grey, residues 447-551) with the corresponding isoelectric points of the domains.161.4.2 Transport of Colicin E3 Colicin E3 must cross both the inner and outer membrane barriers to access the 16S ribosomal subunit in the cytoplasm (Bowman et al. 1971). How colicin E3 is transported across the outer membrane remains unknown. However, much has been elucidated for the initial events in the transport process. The initial event in transport of E3 is binding of colicin E3 to BtuB on the exterior of the cell (Di Masi et al. 1973; Taylor et al. 1998). Colicin E3 has a higher affinity for BtuB than vitamin B12 ensuring that colicin E3 is effectively adhered to the cell surface (Cavard, 1994). After binding, colicin E3 then undergoes a conformational change that dissociates Im3 from the N-terminal binding site but it remains attached to the C-terminal active domain with a femtomolar affinity (Walker et al. 2003; Zakharov et al. 2008). This frees the unstructured T-domain to effectively scan the outer membrane surface for the trimeric membrane protein OmpF. The T-domain then penetrates through the lumen of one monomer of the OmpF trimer by first inserting the OBS1 into one channel of OmpF with a 2µM affinity. Dissociation of OBS1 from the OmpF lumen eventually permits entry of the T-domain into the periplasm such that the TolB binding site, TBE, is exposed to the periplasm (Housden et al. 2010). TolB is part of the Tol system, a homologue of the Ton system, without a clearly defined function (Kleanthous, 2010). TolB is a soluble β-propeller periplasmic protein that associates with outer membrane proteins, including OmpF (Rigal et al. 1997), but largely remains with the outer membrane protein Pal (Clavel et al. 1998). TolB binds Pal with an affinity of KD ≅ 47 nM (Loftus et al. 2006). Colicins such as E3 and E9 bind TolB with a KD ≅ 125 nM (Bonsor et al. 2009). Calcium increases the 17affinity of colicin sufficiently to compete with Pal for TolB binding (Loftus et al. 2006). After dissociating from Pal the N-terminal region of TolB becomes unstructured and allows it to bind to TolA. This disordering occurs upon recruitment by colicins and allows them to interact with the rest of the Tol system (Bonsor et al. 2009). It is at this point that energy from the inner membrane transmitted to the colicin leads to irreversible cell damage (Jetten et al. 1975). This shift occurs in response to energy provided by the Tol complex and results in the release of bound immunity protein from the C-terminus (Vankemmelbeke et al. 2009). Immunity protein release in colicin E9, homologous to colicin E3, results from a force activated ‘trip-switch’ wherein remodeling of the colicin allows fast dissociation of Im3 (Farrance et al. 2013).  After release of Im3 colicin E3 is then able to be transported across the outer membrane. However, the majority of the colicin actually remains on the cell surface and only the C-terminal domain enters into the cytoplasm (Bénédetti et al. 1992; Chauleau et al. 2011). The C-terminal nuclease domain of colicin E3 occludes the channel of OmpF, implying that E3 may enter the periplasm through an OmpF channel after the release of the immunity protein (Zakharov et al. 2006). However, the nuclease domain of the related DNase of colicin E9 is a pore-forming domain that may be able to penetrate the outer membrane (Mosbahi et al. 2002). Colicin E3 nuclease domain also associates with and penetrates membranes, as do other colicin nuclease domains (Vankemmelbeke et al. 2012). The electropositivity and hydrophobic character of colicin nuclease domains makes them similar to antimicrobial peptides, which have the ability to pass over microbial membranes through association with the negatively charged head groups of lipids (reviewed by Zasloff, 2002). Interestingly, there is proteolytic cleavage of the colicin 18during transport that separates the nuclease domain from the rest of the colicin (Liao et al. 2001; Shi et al. 2005; Duché et al. 2007; Walker et al. 2007). Transport across the inner membrane appears to occur through the AAA+ ATPase protease FtsH (Walker et al. 2007) and only involves the nuclease domain of E3 (Chauleau et al. 2011) (Figure 1-7). How the nuclease domains crosses the outer and inner membrane intact is still not fully understood. 19Figure 1-7: Model of colicin E3 transport. Colicin E3Im3 complex binds BtuB (blue) and the unstructured N-terminal threads through the channels of OmpF (orange). This unstructured domain interacts with TolB (purple), which recruits the TolA-TolR-TolQ complex and drives the dissociation of the immunity protein (green). The free colicin nuclease domain is then believed to enter through an OmpF channel (dotted line) to interact with the inner membrane anionic lipids and the AAA+ FtsH protease to enter the cytoplasm (PDB: 1JCH).1.6 Outer membrane permeability and oxidative stress The outer membrane offers protection against stresses from the extracellular environment. Porins in the outer membrane permit access of antibiotics and other antimicrobial compounds to the periplasm (Delcour, 2009). An alteration in the permeability of the outer membrane is implicated in the development of antibiotic resistance (Masi, 2013). Porins of the 20outer membrane can be modulated to rapidly close by changes in transmembrane voltage or pH (Müller and Engel, 1999) or by exposure to polyamines (Iyer and Delacour, 1997). Decreased membrane permeability may also be implicated in the formation of “persisters”, non-replicating pathogenic cells that survive antibiotic treatment to reestablish infections (Kester and Fortune, 2014). Persisters also resist immune assault (Helaine et al. 2014), which includes reactive oxygen species (ROS). OmpC, a homologue of OmpF, is a trimeric outer membrane porin that acts as a passage through which ROS can access the cell interior. The development of a redox-sensitive GFP-based assay allowed investigation of ROS-induced stress in Salmonella cells (van der Heijden et al. 2015). This assay revealed that Salmonella cells undergo a rapid transition to decrease membrane permeability against ROS. Deletion of the outer membrane protein OmpC resulted in resistance to oxidizing agents. Intriguingly, deletion of the heat shock protein HslT abrogated resistance to oxidation (van der Heijden et al. 2016b). HslT is a homologue of the E. coli heat shock protein IbpA, which responds to heat shock as well as oxidation (Kitagawa et al. 2002). It is possible that HslT and OmpC directly interact through protein-protein interactions. 1.7 Nanodiscs Investigation of membrane proteins in vitro has classically relied on the use of detergents, liposomes, and planar lipid bilayers. Each of these approaches have limitations. Detergents can affect the oligomeric state of some membrane proteins (le Maire et al. 2000) and have been shown to alter protein-protein binding interactions of TonB-dependent transporters (Choul-Li et al. 2008; Mills et al. 2014). TonB dimerization is also affected by the type of detergent. TonB dimerizes in the presence of LDAO (Moeck and Letellier, 2001) and Tween 20 (Khursigara et al. 212004). Detergents also tend to co-concentrate with purified membrane proteins, which can elicit protein denaturation (Bayburt and Sligar, 2010). Liposomes and planar lipid bilayers have the disadvantage of only permitting one surface of the protein to be probed through interaction studies and can be difficult to prepare. Nanodiscs are a solution to these problems, allowing investigation of purified reconstituted membrane proteins in an aqueous state in the absence of detergents. Nanodiscs are composed of apolipoprotein A-1 engineered to form a membrane scaffold protein (MSP) that encircles a small disc of lipid bilayer that can encompass a membrane protein (Bayburt et al. 2002, Bayburt and Sligar, 2003). Purified membrane proteins in detergents are combined with the MSP and detergents are removed by detergent adsorbing hydrophobic materials (Bayburt and Sligar, 2010). Aside from the advantages of detergent removal, the nanodisc MSP can also be tagged to attach membrane proteins to surfaces for investigation using surface plasmon resonance (SPR) (Goluch et al. 2008) or scintillation proximity assay (SPA) (Nasr and Singh, 2013). Nanodiscs have been used to investigate bacteriorhodopsin (Bayburt et al. 2006), cytochrome P450 (Denisov and Sligar, 2011), G-protein coupled receptors (Leitz et al. 2006), the FGK2 maltose transporter (Bao and Duong, 2012), and the SecYEG translocase (Alami et al. 2007). Recently, I have also employed nanodiscs to investigate the binding interactions of the TonB-dependent transporter FhuA (Mills et al. 2014). 1.8 Scintillation Proximity Assay  Scintillation proximity assay (SPA) is a technique to measure the binding of radio-labeled ligands to proteins. SPA consists of homogeneous beads that can adhere proteins to the bead surface by 6His-tags, protein A/G, or biotin-streptavidin (PerkinElmer). These beads contain a scintillant core the becomes excited to emit light by the proximity of radioactive ligands binding 22the protein receptor on the bead surface. By titrating radio-labeled ligands the binding affinity of a protein can be determined while dissociation kinetics can be assessed by addition of excess  unlabeled ligand (Khawaja et al. 2008). This technique has been applied to soluble proteins (Sun et al. 2005), membrane receptors (Grandy et al. 1995), and transporters (Harder and Fotiadis, 2012). Nanodisc reconstituted membrane transporters have also been investigated using SPA. The amino-acid transporter LeuT was reconstituted into nanodiscs and adhered to the surface of SPA beads. The binding of LeuT for radio-labeled [3H]Leu and [3H]Ala was compared in detergent and in nanodiscs and LeuT was more active in nanodiscs (Nasr and Singh, 2013). This shows that the bilayer environment of nanodisc preparations are ideal for the SPA. 1.9 Objectives Despite the efforts made in investigating outer membrane transport there remains a number of mysteries surrounding transport. Namely, what events occur that result in the transport of these two BtuB ligands, vitamin B12 and colicin E3, and how is the permeability of the outer membrane changed to protect the cell from environmental stressors? 1) How is vitamin B12 transported through BtuB and across the outer membrane?2) How are the porins of outer membrane subverted by colicins to enter the cell?3) How does the cell protect itself by changing the permeability of the outer membrane? To investigate the first problem BtuB was reconstituted into nanodiscs and subjected to native gel electrophoresis and multi-angle light scattering to determine the stoichiometry of 23interaction between nanodisc (Nd)-BtuB and TonB. The binding dynamics of vitamin B12 with BtuB were investigated using a combination of native gel electrophoresis and autoradiography. The binding affinity and dissociation rate of vitamin B12 from BtuB were analyzed using the scintillation proximity assay (SPA). The mechanism of ligand retention within BtuB was investigating by introducing mutations into BtuB with deletion of the TonB box and mutation of a specific residue at the apex of an apical loop, valine 90, to an arginine.  To investigate how the nuclease domain of colicin E3 enters the periplasm interaction studies between colicin E3, BtuB, TolB, and OmpF were conducted both in the presence and absence of detergents. Binding of colicin E3 to its interaction partners was promoted by sodium dodecyl sulphate (SDS). SDS was found to both elicit the release of the colicin immunity protein Im3 and to chemically mimic LPS. Gel electrophoresis and isothermal titration calorimetry (ITC) studies show that the nuclease domain of colicin E3 interacts directly with LPS upon release of Im3. OmpC was purified intact as a trimer from the membranes of Salmonella enterica Typhimurium and HslT was purified and refolded from E. coli. A variety of techniques including nondenaturing gels, protein pulldown, and chemical crosslinking all failed to demonstrate a direct interaction between these two proteins. The implications and potential explanations of these results are discussed. 24Chapter 2: TonB alters the binding kinetics of vitamin B12 with BtuB2.1 Introduction Transport through TonB-dependent transporters largely relies on the interaction between ligand-loaded transporters with the TonB-ExbB-ExbD complex (Ferguson et al. 1998). Ligand binding induces a number of conformational changes in these transporters. The TonB box of BtuB becomes unstructured and exposed to the periplasmic space (Chimento et al. 2003; Freed et al. 2010). A number of structural alterations occurs on the extracellular face, an apical loop of residues 85 to 95 moves 6Å towards the bound substrate. In other TonB-dependent transporters, such as FepA, the extracellular loops of the β-barrel go through coordinated movements to “close” the transporter over the bound substrate in a form of induced fit (Payne et al. 1997; Smallwood et al. 2014). These structural changes have been reported in a number of other TonB-dependent transporters: ShuA (Cobessi et al. 2010), FecA (Yue et al. 2003), FhuA (Faraldo-Gomez et al. 2003), and FyuA (Noinaj et al. 2010). TonB binding also induces conformational changes in transporters. In FhuA the binding of TonB alters the conformation of extracellular loops 3, 4 and 5 (James et al. 2008) and the apical loop of residues 85 to 95 undergoes an additional 2Å movement toward bound substrate after binding of TonB (Shultis et al. 2006). What effect this binding of TonB has on the kinetics of ligand binding in these transporters is not fully known. Interestingly, the binding of TonB alters the binding kinetics of the pyoverdin transporter FpvA in Pseudomonas aeruginosa depending on the iron-loaded status of pyoverdin (Clément et al. 2004). This demonstrates that TonB can alter the binding kinetics of transporters 25depending on certain conditions. However, the effect of TonB on binding kinetics of BtuB has not been fully studied.  The effect of TonB on the binding kinetics of vitamin B12 (also referred to as cobalamin) ligand with BtuB and the mechanisms therein were investigated by reconstitution of BtuB into nanodiscs. Nanodiscs permit the investigation of membrane proteins in an aqueous state without the need for detergents. Results presented below show that i) BtuB can be reconstituted into nanodiscs, ii) Nd-BtuB interacts with a monomer of TonB in a ligand-dependent manner, iii) binding of TonB drastically slows the dissociation of vitamin B12 from BtuB, iv) trapping of B12 within BtuB by TonB requires both the TonB box and an apical loop (residue 85 to 95) of the plug domain, and v) binding of B12 rather than colicin E3 alters the conformation of BtuB and the plug domain can be denatured to elicit the dissociation of B12.262.2 Materials and Methods2.2.1 Materials Dodecyl-β-D-maltopyranoside (DDM) was purchased from Anatrace. Lauryldimethylamine-n-oxide (LDAO) was obtained from Sigma-Aldrich (St. Louis, MO). Triton-X-100 was acquired from Bioshop Canada (Burlington, ON). Amberlite XAD 2 beads were purchased from Supelco, Inc (Bellafonte, PA). 2.2.2 Purification of BtuB The gene btuB was amplified from Escherichia coli K-12 genome by PCR and cloned into the vector pBAD22 using polymerase incomplete primer extension (PIPE) (Klock and Lesley, 2009). Mutations were introduced by site-directed mutagenesis and verified by sequencing. BtuB (wild-type and mutants BtuB∆6-12 and BtuBV90R) was expressed in BL21 cells. Cells were grown for three hours to OD600 ≅ 0.3 in M9 minimal media at 37℃ supplemented with 80µg/ml ampicillin then induced with 0.2% arabinose. Cells were grown three additional hours to OD600 ≅ 1.0 and then harvested by centrifugation at 5,000 rpm for 10 minutes, 4℃ (JLA-10.500 rotor, J-E centrifuge, Beckman-Coulter). Cell pellets were resuspended in buffer A (50mM Tris-HCl, pH=7.9, 50mM NaCl, 10% v/v glycerol) supplemented with 1mM PMSF and lysed at 10,000 psi by three passages through a microfluidizer (Microfluidics Corp.). Lysate was cleared by centrifugation at 6,000 rpm, 10 minutes, 4℃ (Beckman-Coulter JA-25.50 rotor) and supernatant was centrifuged 45,000 rpm, 40 minutes, 4℃ (Beckman Type 60 Ti rotor) to collect membranes. Membrane pellets were resuspended in buffer A and diluted to 3.0mg/ml, then 27solubilized with 1% Triton-X-100 for one hour at room temperature (~21℃) with gentle agitation. Solubilized membrane was centrifuged at 45,000 rpm, 40 minutes, 4℃ to collect the outer membrane pellet. Pellets were resuspended in buffer A to 3.0mg/ml and solubilized with 1% LDAO overnight at 4℃ with gentle agitation. Solubilized membrane was centrifuged a final time at 45,000 rpm, 40 minutes, 4℃ and supernatant was applied to strong anion exchange on a 5mL HiTrap Q FF column (GE Healthcare) equilibrated in buffer B (50mM Tris-HCl, pH=7.9, 50mM NaCl, 10% v/v glycerol, 0.1% LDAO). BtuB protein was eluted using a gradient of 1M NaCl in buffer B over 10 column volumes. Fractions were analyzed by SDS-PAGE and Coomassie brilliant blue staining to determine purity of BtuB. The yield of BtuB was approximately 2mg per litre of culture. 2.2.3 Purification of TonB BL21 cells transformed with TonB-pET28 and expressing a truncated periplasmic domain of TonB (residues 32-239 hereafter referred to as TonB) was already available in the lab and was described previously in Mills, 2013 and Mills et al. 2014. Cells were grown for three hours at 37℃ in Luria-Bertani (LB) broth supplemented with 25µg/ml kanamycin to OD600 ≅ 0.6 then induced with 1mM IPTG. Cells were grown three more hours to OD600 ≅ 1.0 and harvested by centrifugation at 5,000 rpm for 10 minutes, 4℃, and the cell pellet was resuspended in buffer A. Cells were lysed by three passages through a microfluidizer at 10,000 psi in buffer A supplemented with 1mM PMSF. Cell lysate was centrifuged at 45,000 rpm, 40 minutes, 4℃, to pellet insoluble materials. Cleared supernatant was passed over an immobilized metal affinity chromatography column packed with cobalt agarose beads (high density) (Gold Biotechnology) 28equilibrated in buffer A. TonB was eluted with a step gradient of buffer C (50mM Tris-HCl, pH=7.9, 300mM NaCl, 10% v/v glycerol, 600mM imidazole). A yield of approximately 10mg of TonB per litre of culture.  For scintillation proximity assay experiments the 6His-tag on TonB was cleaved off by thrombin (Sigma-Aldrich, St. Louis, MO). Thrombin was added to TonB fractions directly from IMAC purification and dialyzed overnight in buffer A at 4℃. Thrombin was removed from TonB by ion exchange by strong cation exchange on 1mL SP FF column (GE Healthcare) equilibrated in buffer A. Protein was eluted by a 1M NaCl gradient of 5 column volumes in buffer A.2.2.4 Purification of BtuF BL21 transformed with BtuF-pBAD33 was previously produced by Spencer MacDonald and already available in the lab. Cells were grown in LB broth supplemented with 25µg/ml chloramphenicol for three hours to OD600 ≅ 0.6 then induced with 0.2% arabinose. Cells were grown an additional three hours to OD600 ≅ 1.0 before harvesting by centrifugation at 6,000 rpm, 5 minutes, 4℃. Cell pellets were resuspended in buffer A and lysed in a microfluidizer at 10,000 psi in buffer A supplemented with 1mM PMSF. Cell lysate was cleared by centrifugation at 45,000 rpm, 40 minutes, 4℃ and the supernatant was passed over a column packed with cobalt agarose beads equilibrated in buffer A. BtuF was eluted with a step gradient of buffer C. BtuF yielded approximately 1.5mg per litre of culture. 292.2.5 Purification of Colicin E3 The open reading frame of colicin E3 and the immunity protein (Im3) was cloned from plasmid ColE3-CA38 (generously provided by Dr. Benjamin Kerr of the University of Washington) into pET28 between the NcoI and the XhoI site with the 6His-tag attached to the C-terminus of the immunity protein. The resulting plasmid, ColE3-pET28, was transformed into BL21 and grown in LB supplemented with 25µg/ml kanamycin. Cells were grown to OD600 ≅ 0.6 and then induced with 1mM IPTG. Cells were then grown to OD600 ≅ 1.0 then harvested by centrifugation at 6,000 rpm, 5 minutes, 4℃. Cell pellets were resuspended in buffer A and lysed in a microfluidizer at 10,000 psi after being supplemented with 1mM PMSF. Cell lysate was cleared by centrifugation at 45,000 rpm, 40 minutes, 4℃. Supernatant was passed over a column containing cobalt agarose resin equilibrated in buffer A. The complex of colicin E3-Im3 was eluted with a step gradient of buffer C. Colicin E3-Im3 heterodimer was produced at approximately 50mg per litre of culture.  2.2.6 Nanodisc reconstitution Purified MSP-L156 (Mills, 2013; Mills et al. 2014) was combined with purified BtuB protein at a molar ratio of 1nmol of BtuB with 4nmol of MSP-L156 in a 100µl volume that was diluted with TSGD buffer (50mM Tris-HCl, pH=7.9, 100mM NaCl, 10% v/v glycerol, 0.1% v/v DDM). One-third volume of Amberlite beads was added to initiate the reconstitution of BtuB into nanodiscs and was incubated at 4℃ for 16 hours with gentle rocking to remove all detergents. 30 Nanodisc preparations were concentrated using a 30 kDa MWCO centrifugal filter (Millipore, USA) and injected onto a gel filtration Superdex 200 HR 10/30 column equilibrated in buffer A to remove aggregates and unincorporated proteins. Fractions eluted from the column were analyzed by 15% SDS-PAGE or 4-12% non-denaturing gels. Fractions were flash frozen in liquid nitrogen and stored at -80℃.2.2.7 Multi-angle light scattering Samples were injected onto a Superdex 200 HR 10/30 column equilibrated in 50mM Tris-HCl, pH=7.9, 100mM NaCl, 2% v/v glycerol and connected in-line to a MiniDAWN TREOS multi-angle light scattering apparatus and a T-rEX differential refractive index apparatus (Wyatt Technologies). Flow rate was maintained at 0.400 ml/min and all data was collected real-time and analyzed using ASTRA V software (Wyatt Technologies). Molecular masses were calculated using a Debye fit method. 2.2.8 BtuF competition assay Purified BtuF at approximately 13µM was incubated with approximately 2.2nM [57Co]-cobalamin (50mM sodium borate buffer 1.75µCi/ml; MP Biomedicals, Cat. No. 06B-430000) for 10 minutes at room temperature. The [57Co]-cobalamin-BtuF complex was desalted through a 1mL Sephadex G25 spin column (GE Healthcare) equilibrated in buffer A to remove excess [57Co]-cobalamin. Purified Nd-BtuB (BtuBWT and BtuB∆6-12) (2µM) samples mixed with an increasing concentration of TonB (~0-10µM) and 8µl of these mixtures were incubated with 6µl aliquots of the [57Co]-cobalamin-BtuF complex for 10 minutes at room temperature. Samples 31were then migrated on clear native (CN)-PAGE and the gel was fixed in 50% methanol, 7% glacial acetic acid for 3 minutes prior to drying on filter paper using a Model 583 gel dryer (BioRad) for 50 minutes. The dried gel was visualized by overnight exposure to a phosphoimaging screen and imaged by a Typhoon 8600 phosphorimager (GE Healthcare). Control experiments were conducted using the same procedure without Nd-BtuB. 2.2.9 Scintillation proximity assay (SPA) Nd-BtuB (wild-type and mutants) were pre-incubated with Cu2+ polyvinyltoluene (PVT) scintillation proximity assay (SPA) beads. BtuB does not have a polyhistidine tag, but the scaffold protein MSP-L156 has an N-terminal hexahistidine tag. This allows the nanodisc to be immobilized to the surface of the bead.  For equilibrium binding affinity 1.3µM Nd-BtuB were immobilized onto 20mg of PVT SPA beads in buffer A supplemented with 1µM CaCl2. Nd-FhuA was substituted for Nd-BtuB as a nonspecific binding control according to prior SPA studies of Auld et al. 2004. Proteins were allowed to bind beads over 30 minutes at 4℃. 90µl aliquots were incubated with either 40µl of either buffer A or TonB with the 6His-tag removed by thrombin (~8.6µM final concentration). Samples were incubated for one hour at room temperature. 10µl aliquots of these mixtures, with and without TonB, were incubated with 30µl volumes of different [57Co]-cobalamin dilutions. The protein-bead suspension was allowed to incubate with the radio-ligand for 20 minutes with shaking at room temperature. The samples were centrifuged 3,000 rpm, 5 minutes to pellet the beads. Samples were then measured in a 1450 MicroBeta TriLux counter (PerkinElmer). Specific 32binding was determined after subtracting nonspecific from total binding. Data were analyzed using a one-site specific nonlinear regression model with the following formula:           Eq. 2-1Data was analyzed using GraphPad Prism 6.0 (GraphPad, San Diego, CA).  Ligand dissociation measurements were performed by incubating Nd-BtuB (0.4µM) with 20mg of Cu2+ PVT SPA beads in buffer A supplemented with 1µM CaCl2 for 1 hour at room temperature. [57Co]-cobalamin (~0.2nM) was added in the presence or absence of a molar excess of TonB. After incubation for one hour at room temperature (two hours in the case of Nd-BtuBV90R) the scintillation was measured on the Microbeta TriLux apparatus. 100nM unlabeled vitamin B12 was then added to initiate exchange of radio-ligand for unlabeled ligand and measurements of the dissociation rate were taken periodically over a 24 hour period. Data was analyzed using a nonlinear regression one-phase dissociation model given by the following formula:           Eq. 2-2Data were analyzed using Prism GraphPad 6.0. 2.2.10 Trypsinization of TonB-dependent transporters Purified Nd-BtuB and Nd-FhuA were incubated in the presence or absence of ligands vitamin B12 (2µM), ferricrocin (5µM), colicin E3 (12µM), or colicin M (7µM) and was digested with 17µM tosylamide phenylethyl chloromethyl ketone-treated trypsin for 30 minutes (or indicated times) at 37℃. Trypsinolysis was stopped by adding 5µl of 5X Laemmli loading buffer 33and heating samples at 98℃ for five minutes. Samples were analyzed by migration on 10% SDS-PAGE and Coomassie brilliant blue staining.   Urea was added to samples where indicated and nanodiscs analyzed on BN-PAGE without trypsinolysis. Trypsinized samples in urea were analyzed by 10% SDS-PAGE. 2.2.11 Other methods Protein concentrations were measured using the Bradford assay (Bradford, 1976). Denaturing and nondenaturing gels and electrophoresis conditions were described previously (Dalal and Duong, 2010). Nondenaturing gels were all gradient gels of 4-12% and 10%, 12%, and 15% SDS-PAGE gels were used where indicated. 2.3 Results2.3.1 Reconstitution of BtuB into Nanodiscs Purified BtuB protein was reconstituted into nanodiscs using the 156-residue truncated membrane scaffold protein (MSP-L156) at a BtuB:L156 ratio of 1:4. LDAO detergent was removed and nanodisc was formed by incubation of the BtuB-L156 mixture with detergent adsorbing Amberlite beads overnight at 4℃. The reconstituted BtuB was separated from unincorporated L156 and aggregates using gel filtration chromatography on a Superdex 200 HR 10/30 column. Two peaks were observed eluting from the column at approximately 14 mL and 16 mL (Figure 2-1A). Fractions from gel filtration were then migrated on a 15% SDS-PAGE gel and BtuB was found exclusively in the initial peak at ~14 mL alongside L156 while the second 34peak corresponded to L156 alone (Figure 2-1B). The first peak of BtuB with L156 was assumed to be nanodisc reconstituted BtuB (Nd-BtuB). To confirm formation of Nd-BtuB fractions from the first peak isolated from size exclusion was analyzed using non-denaturing electrophoresis BN-PAGE. BtuB purified in LDAO detergent migrates as both a monomer and a dimer, presumably because of a detergent-dye interaction effect. L156 alone migrates near the dye front on BN-PAGE, distinct from the BtuB monomer. Nd-BtuB eluted from size exclusion chromatography migrated as a monodisperse species different from the BtuB monomer and dimer and distinct from the L156 (Figure 2-1C). 35Figure 2-1: Reconstitution of BtuB into nanodiscs A) Gel filtration profile of nanodisc-reconstituted BtuB (Nd-BtuB, ~700µg) applied to a Superdex 200 HR 10/300 column. B) Elution fractions from size exclusion (fractions 12 to 18) analyzed by 15% SDS-PAGE and visualized by Coomassie blue staining. C) Blue Native (BN)-PAGE of isolated BtuB in detergent (lane 1), MSP-L156 (lane 2), and nanodisc reconstituted BtuB (lane 3).2.3.2 Ligand promotes recruitment of TonB to Nd-BtuB in a 1:1 molar ratio BtuB has been reconstituted into nanodiscs. To evaluate the functionality of Nd-BtuB the nanodisc was interacted with its molecular partner TonB.  Binding between Nd-BtuB and an equimolar amount of purified TonB was assayed using clear native (CN)-PAGE electrophoresis. Nd-BtuB migrates as a single monodisperse band on CN-PAGE (Figure 2-2A, lane 1). Incubation of Nd-BtuB with vitamin B12 prior to loading on the 36gel yielded no change in migration on CN-PAGE, likely because of the small size of the ligand (~1kDa) (Figure 2-2A, lane 2). Incubation of Nd-BtuB with TonB resulted in only a small percentage of the Nd-BtuB forming a complex with TonB; however, addition of vitamin B12 resulted in almost complete formation of a complex with TonB (Figure 2-2A, compare lanes 3 to 4).  Previous work with TonB-dependent transporters has implicated TonB functioning as a dimer (Chang et al. 2001; Khursigara et al. 2004; Khursigara et al. 2005). However, recent work has demonstrated that TonB may function as a monomer upon interaction with ligand-loaded BtuB even if the TonB is pre-dimerized (Freed et al. 2013). To assess the TonB stoichiometry of interaction of BtuB in nanodiscs two approaches were used: 1) interaction between disulphide crosslinked TonB dimers on native electrophoresis and 2) multi-angle light scattering.  TonB with a cysteine substitution in the linker between the hexahistidine tag and the N-terminus of truncated TonB was found to purify from IMAC as both a monomer and dimer (Khursigara et al. 2004; Mills et al. 2014). Dimeric TonB was further isolated by passage through size exclusion on Superdex 200 HR 10/30 to give dimeric TonB2. Monomeric wild-type TonB and dimeric disulphide crosslinked TonB was incubated with the same concentration of Nd-BtuB in the presence of vitamin B12 and in the presence or absence of 5mM DTT.  The migration of the Nd-BtuB-TonB complex was unaffected by the presence or absence of DTT and migrated to the same level within the gel (Figure 2-2B, lanes 2 and 3). However, in 37the the absence of DTT the dimeric Nd-BtuB-TonB2 migration was retarded as a larger molecular weight complex (Figure 2-2B, lane 4). Addition of DTT to Nd-BtuB-TonB2 causes the complex to migrate to the same position as monomeric Nd-BtuB-TonB (Figure 2-2B, lane 5).  This change in migration of Nd-BtuB-TonB2 from Nd-BtuB-TonB indicates that TonB interacts with BtuB as a monomer. The artificially dimerized TonB2 likely interacts with BtuB through only a single TonB monomer. Addition of DTT breaks the disulphide crosslink and allows the second TonB to dissociate unincorporated from the complex.  To ascertain the molecular masses of both proteins and molecular complexes multi-angle light scattering was employed. Nd-BtuB and TonB were individually analyzed and a Nd-BtuB-TonB-B12 complex was also determined. The molecular mass of Nd-BtuB was found to be 120kDa ± 4% and the molecular mass of monomeric TonB was 24kDa ± 6%. The molecular mass of TonB was in good agreement with previously published molecular masses of this truncated TonB (Khursigara et al. 2004). The molecular mass of the Nd-BtuB-TonB-B12 complex was measured as 146kDa ± 5% (Figure 2-2C). This mass corresponds to a single ~120kDa Nd-BtuB with a TonB monomer of ~24kDa. 38ATonB2---+--+-+-+--++Nd-BtuBNd-BtuB-TonBNd-BtuB-TonB21 2 3 4 5BTonB - - + +B12 - + - +Nd-BtuBNd-BtuB-TonB1 2 3 4TonBDTT8 10 12 14 16 18 8 10 12 14 16 18 8 10 12 14 16 18Molecular Mass (g/mol)106105104103106105104103106105104103Nd-BtuB TonB Nd-BtuB-TonB-B12CFigure 2-2: Nd-BtuB interacts with TonB as a monomer A) Nd-BtuB (2µM) incubated with TonB (4µM) in the presence or absence of vitamin B12 (2µM) assayed using CN-PAGE and visualized with Coomassie blue. B) Nd-BtuB (2µM) incubated with wild-type and disulphide crosslinked TonB (TonB2) samples were treated with 1mM DTT where indicated and assayed by CN-PAGE and visualized with Coomassie blue. Vitamin B12 was included in all lanes. C) Size exclusion multi-angle light scattering (SEC-MALS) analysis of Nd-BtuB, TonB, and Nd-BtuB-TonB-B12. Molecular masses were determined to be: Nd-BtuB ~120kDa ± 4%, TonB ~24kDa ± 6%, Nd-BtuB-TonB ~146kDa ± 7%.2.3.3 TonB alters the binding of ligand with Nd-BtuB BtuB undergoes a number of conformational changes upon binding of vitamin B12 (Chimento et al. 2003b) and undergoes additional conformational changes upon binding of TonB (Shultis et al. 2006; Sikora and Cafiso, 2016). Conformational changes induced by TonB have also been implicated in other TonB-dependent transporters (James et al. 2008). However, these 39structural changes remain largely uncharacterized and what effect they have on ligand binding is unknown. To investigate possible changes in binding dynamics a combination of native-gels, competition assays, and autoradiography were employed. Vitamin B12 binding dynamics with Nd-BtuB were investigated using competition between radio-labeled [57Co]-cobalamin and unlabeled B12 on CN-PAGE and autoradiography. Nd-BtuB was pre-incubated with [57Co]-cobalamin in the presence or absence of TonB. These mixtures were then aliquoted and incubated with an increasing concentration of unlabeled B12. Samples were then separated on CN-PAGE and the gel visualized using autoradiography. In the absence of TonB the intensity of Nd-BtuB bands decreases as the concentration of unlabeled B12 increases indicating exchange of labeled B12 within the BtuB binding pocket for unlabeled B12 in solution. Band intensity decreased from 100% in 0µM unlabeled B12 to 55% with 1.7µM unlabeled B12 according to densitometry. However, under the same conditions with TonB in complex with Nd-BtuB the band intensity only decreased from 100% to 78% in 0µM to 1.7µM B12 (Figure 2-3A). This suggests that the binding of TonB alters either the binding affinity or the binding kinetics of BtuB for B12.  After the transport of B12 across the outer membrane it is bound by a soluble periplasmic protein BtuF. BtuF binds B12 with a high affinity of approximately 15nM (Cadieux et al. 2002) and presents it to the inner membrane protein complex BtuCD for transport into the cytoplasm (Locher and Borths, 2004). To further investigate the binding dynamics of B12 to BtuB an assay was devised in which BtuF was competed against Nd-BtuB in solution.  40 A complex of [57Co]-cobalamin-BtuF was isolated by elution from a 1mL G25 spin column. This isolated [57Co]-cobalamin-BtuF complex was then mixed with Nd-BtuB and aliquots portioned into tubes containing an increasing concentration of TonB. After five minutes of incubation the samples were loaded onto CN-PAGE and the gel was visualized by autoradiography. In the absence of TonB the radio-labeled B12 is shared between the two proteins with a predominance of ligand with Nd-BtuB (Figure 2-3B, lane 4). This predominance of ligand with Nd-BtuB is likely because of the higher binding affinity of BtuB (~0.3nM, Bradbeer et al. 1986) for B12 than BtuF (15nM, Cadieux et al. 2002). However, as the concentration of TonB increases there is, in correspondence with the formation of complex between Nd-BtuB and TonB, a decrease in the intensity of the BtuF band (Figure 2-3B, lanes 5 to 12).  BtuF forms a complex with TonB as part of the B12 transport chain (James et al. 2009). This discovery did not report any changes in ligand-binding interactions between BtuF alone or in complex with TonB. In order to control against the possibility that TonB is altering the B12 binding to BtuF a gradient of TonB was added to an isolated complex of [57Co]-cobalamin-BtuF and run on CN-PAGE and visualized by autoradiography. No decrease in the BtuF band intensity  occurred upon addition of increasing concentrations of TonB (Figure 2-3C). This implies that a TonB-BtuF interaction does not significantly change the ligand binding interactions between BtuF and B12. 41 It is unknown if other ligands of BtuB are also retained by the binding of TonB. To confirm that binding of TonB elicits this retention of vitamin B12 alone and not for other BtuB ligands a competition assay between B12 and colicin E3 was performed. Nd-BtuB was incubated with 200µM B12 and an increasing concentration of TonB for five minutes at room temperature. Colicin E3 was then added and loaded onto CN-PAGE. Colicin E3 has higher binding affinity for BtuB than B12 (Cavard, 1994), even in the presence of 200µM B12 a complex between Nd-BtuB and colicin E3 still forms (Figure 2-3D, lane 5). TonB prevents formation of a complex between BtuB and colicin E3 in the presence of B12 (Figure 2-3D, lanes 6 to 12). The mechanism preventing dissociation of B12 from BtuB is specific for the B12 ligand because only B12 elicits binding of TonB whereas colicin E3 does not (Cadieux et al. 2003).42AB12 (µM)Nd-BtuBNd-BtuB-TonB-TonB - - - - - + + + + + +100 94 83 80 60 55 100100 100 95 91 78Intensity (%)BNd-BtuBTonBBtuF-Nd-BtuB-TonB1 2 3 4 5 6 7 8 9 10 11 12TonB -1 2 3 4 5 6 7 8 9 10C DColE3Nd-BtuB-E3Nd-BtuB-TonBNd-BtuBTonB- +- +-B12- +- --+ + + + + + +- - 1.70.80.40.20.11.70.80.40.20.12 31 54 6 7 8 9 10 11 12Figure 2-3: TonB alters ligand binding dynamics. A) 2.5µM Nd-BtuB was incubated with ~0.4nM [57Co]-cobalamin in the presence and absence of TonB for 10 minutes. Vitamin B12 (0-1.7µM) was then added and visualized on CN-PAGE by autoradiography. [57Co]-cobalamin density was quantified using ImageJ with 0µM unlabeled B12 defined as 100%. B) 10µM [57Co]-cobalamin-BtuF (~8.1 x 103 Disintegration Per Minute (dpm)/µM ) incubated with 2µM Nd-BtuB and titrated with TonB (0-10µM). Samples were visualized on CN-PAGE by autoradiography. Nd-BtuB, BtuF, and Nd-BtuB-TonB were loaded as controls. C) Control experiment of 10µM [57Co]-cobalamin-BtuF titrated with TonB (0-10µM) and visualized by CN-PAGE autoradiography. D) 3µM Nd-BtuB with 200µM B12 was incubated with an increasing concentration of TonB (0-8µM) prior to addition of 0.5µM colicin E3 then visualized by CN-PAGE and Coomassie blue staining. Nd-BtuB, colicin E3, Nd-BtuB-TonB, Nd-BtuB-colicin E3 are loaded as references.432.3.4 TonB does not increase the binding affinity of BtuB TonB clearly affects the binding dynamics of ligand binding to BtuB. The effect may be caused by increased binding affinity or by alteration of binding kinetics between BtuB and B12. To distinguish these possibilities the Scintillation Proximity Assay (SPA) used by Nasr and Singh was employed. SPA consists of Cu2+ immobilized metal affinity beads with a central core composed of a polyvinyl toluene (PVT) scintillant. Proteins and protein complexes with a histidine tag can be immobilized on the surface of these beads. Radio-labeled ligands added to these beads are only detected once bound to their corresponding protein attached on the bead surface. Radio-labeled ligands held proximal to the bead surface allows radiation to penetrate to the scintillant which emits detectable blue light in response (Figure 2-4A).  Nd-BtuB wild-type and mutants were incubated with the beads in the presence or absence  of TonB with [57Co]-cobalamin. Controls consisted of beads alone, TonB alone, and Nd-FhuA alone in which no [57Co]-cobalamin binding was detectable with a background level of approximately 100 Disintegrations Per Minute (DPM). Nd-BtuB wild-type and mutants with and without TonB elicited approximately 1000 DPM (Figure 2-4B).  Nd-BtuB attached to SPA beads was incubated with an increasing concentration of [57Co]-cobalamin in the presence or absence of TonB. The binding affinity of Nd-BtuB for B12 was approximately 0.3nM, with variability due to the imprecise concentration of [57Co]-cobalamin. This binding affinity is in good agreement with previously reported BtuB binding affinity (Bradbeer et al. 1986). However, in the presence of TonB the binding of [57Co]-44cobalamin was determined to be approximately 2.7nM (Figure 2-4C). This approximately nine-fold lower binding affinity of BtuB for B12 in the presence of TonB is surprising but unpublished data regarding conformational changes in extracellular loops of BtuB generating a more open binding pocket and suggesting a lower binding affinity (Sikora and Cafiso, 2016). Interestingly, the Bmax of binding in the presence of TonB was calculated to be approximately twice the Bmax of Nd-BtuB alone. Analysis of the binding data by Scatchard plot revealed a possible second binding site for B12 on Nd-BtuB (Figure 2-4D). The possibility that TonB itself provides a second B12 binding site was discounted by controls of TonB alone (Figure 2-4B). The first binding site has a ~0.2nM affinity while the second site is calculated to have a ~5.3nM binding affinity (Table 2-1). Unfortunately, there is no evidence of a second B12 binding site in the crystal structures of BtuB (Chimento et al. 2003; Shultis et al. 2006). Nevertheless, another explanation is required to explain the apparent retention of ligand within Nd-BtuB in the presence of TonB. 45Beads030060090012001500DPMTonBNd-FhuANd-BtuBNd-BtuB + TonBNd-BtuB∆6-12Nd-BtuB∆6-12 + TonBNd-BtuBV90RNd-BtuBV90R + TonBBA[57Co]-cobalaminSPA BeadBlue LightRadiationNd-BtuB-B12TonB[57Co]-cobalamin (nM)0 200 400 6000500100015002000Bound (fmol/mg)Bound/Free (fmol/mg)Nd-BtuBNd-BtuB + TonBD0 1 2 3 4050010001500[57Co]-cobalamin (nM)Specific Binding (dpm)Nd-BtuBNd-BtuB + TonBC57Co]-cobalamin (n )Figure 2-4: TonB does not increase receptor binding affinity. A) Schematic of the scintillation proximity assay. B) 20mg of SPA beads were incubated with 0.4µM Nd-BtuB (wild-type and mutants) and ~0.2nM [57Co]-cobalamin for one hour. Data represent three independent experiments and standard deviation. C) Comparison of [57Co]-cobalamin binding Nd-BtuB (●) versus Nd-BtuB with TonB (○). D) Scatchard plot of [57Co]-cobalamin binding Nd-BtuB (●) versus Nd-BtuB with TonB (○).2.3.5 TonB slows the dissociation of ligand from Nd-BtuB Dissociation rate constant (koff) is a measurement of the rate at which ligand dissociates from a receptor. A 100nM excess of unlabeled B12 was added to a complex of [57Co]-cobalamin-Nd-BtuB attached to the SPA beads to measure the dissociation rate. In the absence of TonB the dissociation rate was 8.6 ± 0.7 x 10-3 min-1 whereas in the presence of TonB the dissociation rate was 1.5 ± 0.5 x 10-3 min-1 (Figure 2-5A; Table 2-1). This corresponds to a complex half-life of 46approximately 80 minutes in the absence of TonB and approximately 460 minutes in the presence of TonB (Table 2-1). The dissociation data was analyzed on a semi-logarithmic plot to investigate the possibility of a second ligand binding site. Dissociation of ligand did not follow a biphasic dissociation rate and did not support the possibility of a second binding site (Figure 2-5B).Figure 2-5: TonB slows the dissociation of ligand from BtuB. A) Dissociation of ~0.2nM [57Co]-cobalamin from 0.4µM Nd-BtuB (●) and from 0.4µM Nd-BtuB and 2.8µM TonB (○). B) Semi-logarithmic analysis of dissociation rate data from Nd-BtuB (●) and from Nd-BtuB-TonB (○).47Table 2-1. Kinetic constants estimated from Scintillation Proximity Assay of Nd-BtuBKd (nM) koff (min-1) kon (nM -1 min-1) t1/2 (min)Nd-BtuB 0.3 ±0.02 8.6 ±0.7 x 10-3 2.8 x 10-2 80 ±7.1Nd-BtuB+TonB 2.7 ±0.7 1.5 ±0.5 x 10-3 5.5 x 10-4 4.6 ±2.3 x 102Equilibrium binding and kinetics ascertained in vitro from scintillation proximity assay at 21ºC in buffer A. Standard error derived from three independent experiments. Statistical significance was determined using ANOVA (P < 0.05). Half-lives are determined from the relationship t1/2 = ln(2)/koff. 2.3.6 Deletion of TonB box abolishes binding of TonB TonB elicits the retention of ligand within the B12 binding site of BtuB. The mechanism of ligand retention was investigated by deletion of the TonB box from BtuB. Proline substitution mutations in the TonB box of BtuB prevents transport of B12 implying that TonB cannot be recruited to this ligand-loaded mutant BtuB (Cadieux et al. 2000). To prevent TonB binding to BtuB a deletion of residues 6 to 12 (6-DTLVVTA-12) representing the entire TonB box was constructed resulting in BtuB∆6-12 (this mutant was generated and provided by Dr. Badreddine Douzi). BtuB∆6-12 was reconstituted into nanodiscs and the competition assay between Nd-BtuB∆6-12 and an isolated complex of [57Co]-cobalamin-BtuF was performed. TonB was titrated into the mixture of Nd-BtuB∆6-12 and [57Co]-cobalamin-BtuF and samples were subjected to CN-PAGE and autoradiography. Addition of TonB to Nd-BtuB∆6-12 did not result in formation of a complex confirming that the deletion of the TonB box prevents high affinity interaction between TonB and BtuB (Figure 2-6A, lane 10). In contrast to wild-type Nd-BtuB the intensity of BtuF 48bands did not decrease as TonB concentration increased (Figure 2-6A, lanes 4 to 10). Without the TonB box TonB is unable to form a complex with BtuB and, concurrently, is unable to retain B12 within the BtuB ligand binding site.  To further confirm this result the dissociation rate was determined using the SPA. Nd-BtuB∆6-12 was attached to beads in the presence of [57Co]-cobalamin with or without TonB. 100nM of unlabeled B12 was added to the solution and the decrease in scintillation was measured over time. In contrast to wild-type BtuB, Nd-BtuB∆6-12 dissociates bound ligand at a rate of 5.3 ± 0.7 x 10-3 min-1 with a complex half-life of approximately 2.2 hours (Figure 2-6B, Table 2-2). This is nearly a full hour longer than wild-type BtuB. In the presence of TonB the ligand dissociates at a rate of 3.4 ± 0.5 x 10-3 min-1 giving a complex half life of about 3.4 hours. While this dissociation rate appears somewhat slower in the presence of TonB the dissociation rate standard error shows that the dissociation rates are not significantly different (Figure 2-6B, Table 2-2). 2.3.7 Mutations of valine 90 abolishes retention of ligand by TonB  The mechanism through which TonB elicits retention of vitamin B12 within BtuB is largely uncharacterized. The TonB box is essential for TonB-dependent ligand retention but what conformational changes occur in BtuB that cause B12 retention are unknown. BtuB undergoes structural changes in the presence of calcium and when ligand-loaded with vitamin B12 (Chimento et al. 2003b). BtuB also undergoes conformational changes upon binding of TonB. An apical loop on the extracellular side of plug domain (residues 85 to 95) forms part of the B12 49binding site goes through a structural movement 6Å closer to bound B12 upon ligand binding and moves 2Å closer to the substrate after TonB binding (Shultis et al. 2006). This structural change may act to retain B12 ligand within the binding site of BtuB upon TonB binding. To investigate this the apical loop of residues 85 to 95 was disrupted by the substitution of the apex residue valine 90 to arginine. Arginine is more hydrophilic than the hydrophobic valine residue (Monera et al. 1995) and replacement of solvent exposed hydrophobic residues with arginine has been used to increase protein stability (Strub et al. 2004). In BtuB it can be anticipated that replacement of valine 90 with arginine will disrupt the conformation of this apical loop in the B12 binding site of BtuB and may alter the kinetics of B12 binding BtuB.  Substitution of valine 90 with arginine yielded BtuBV90R. BtuBV90R was reconstituted into nanodiscs and it was found that Nd-BtuBV90R binds [57Co]-cobalamin at the same concentration as Nd-BtuBWT both on CN-PAGE (Figure 2-6D) and on SPA (Figure 2-6B). Therefore, Nd-BtuBV90R binds [57Co]-cobalamin to the same capacity as wild-type BtuB. SPA was used to measure the dissociation rate of B12 from Nd-BtuBV90R. In the absence of TonB the dissociation rate from Nd-BtuBV90R is 3.1 ± 0.3 x 10-1 min-1 and in complex with TonB the dissociation rate is 1.1 ± 0.1 x 10-1 min-1. This corresponds to a complex half-life of 2.2 minutes and 6.3 minutes, respectively (Figure 2-6C, Table 2-2). 5050 100 150020406080100Time (min)B/Bo (%)Nd-BtuB∆TonB BoxNd-BtuB∆TonB Box + TonBBATonB - --Nd-BtuB∆6-12BtuFNd-BtuB∆6-12 + TonBC% B/Bo10 20 30020406080100Time (min)Nd-BtuBV90RNd-BtuBV90R + TonBDNd-BtuBV90RNd-BtuBWTNd-BtuBWT/V90RNd-BtuB∆6-12Figure 2-6: Dissociation of B12 from BtuB∆6-12 and BtuBV90R. A) 10µM [57Co]-cobalamin-BtuF (~8.1 x 103 dpm/µM ) incubated with 2µM Nd-BtuB∆6-12 and titrated with TonB (0-10µM). Samples were visualized on CN-PAGE by autoradiography. Nd-BtuB and BtuF were loaded as controls. B) Dissociation of ~0.2nM [57Co]-cobalamin from 0.4µM Nd-BtuB∆6-12 (●) and from 0.4µM Nd-BtuB∆6-12 and 2.8µM TonB (○). C) Dissociation of ~0.2nM [57Co]-cobalamin from 0.4µM Nd-BtuBV90R (●) and from 0.4µM Nd-BtuBV90R and 2.8µM TonB (○). D) Comparison of [57Co]-cobalamin binding to Nd-BtuBWT (2µM) or Nd-BtuBV90R (2µM) visualized on CN-PAGE by autoradiography.51Table 2-2. Dissociation Rate Constants Estimated from Scintillation Proximity Assay.Transporter koff (min-1) t1/2 (min)BtuB∆6-12 5.3 ±0.7 x 10-3 1.3 ±0.2 x 102BtuB∆6-12 +TonB 3.4 ±0.5 x 10-3 2.0 ±0.4 x 102BtuBV90R 3.1 ±0.3 x 10-1 2.2 ±0.3BtuBV90R +TonB 1.1 ±0.1 x 10-1 6.3 ±0.6Rate constants ascertained in vitro from scintillation proximity assay. Rates were determined at 21ºC in buffer A. Standard error derived from three independent experiments. Statistical significance was determined using ANOVA (BtuB∆6-12 P > 0.05, BtuBV90R P < 0.05). Half-lives are determined from the relationship t1/2 = ln(2)/koff.2.3.8 Structural changes in BtuB upon ligand binding  The steps in TonB-dependent transport that follow TonB binding are largely unknown. Plug domain unfolding or dislocation are clear necessities for the transport of bound ligands. A new technique was required to investigate conformational changes in the plug domain that result in dissociation of ligand.  All TonB-dependent transporters undergo conformational changes upon binding of their cognate ligands (Moeck et al. 1996; Ferguson et al. 1998; Locher et al. 1998; Chimento et al. 2003b). A trypsinization assay was previously devised for the TonB-dependent iron-siderophore transporter FhuA. The pattern of fragments of FhuA digested by trypsin was altered depending on the ligand-loaded state of FhuA. Conformational changes in FhuA induced by ligand binding changes the accessibility of some parts of the FhuA structure for trypsin (Moeck et al. 1996). This trypsinization accessibility assay has not been used to determine the ligand loaded status of any other TonB-dependent transporter. This would also permit the investigation of not only the 52structural changes induced by ligand binding, but also the movement of the plug domain to permit import of bound ligand. BtuB was reconstituted into nanodiscs and was incubated with trypsin for varying times in the presence and absence of vitamin B12 at 37℃. Digestion was quenched with 5X Laemmli loading buffer and samples were visualized on SDS-PAGE. BtuB was completely digested in the absence of B12 after only five minutes (Figure 2-7, lane 3). However, in the presence of vitamin B12 BtuB was largely undigested by trypsin even after one hour (Figure 2-7, lane 12). To control for the possibility that vitamin B12 inhibits trypsin digestion Nd-BtuB and Nd-FhuA (Mills et al. 2014) were both incubated with vitamin B12, the FhuA ligand ferricrocin, or colicin M and trypsinolyzed at 37℃. BtuB was undigested in the presence of B12 but was fully digested in all other conditions while FhuA was undigested in the presence of ferricrocin and colicin M in agreement with previous results (Mills, 2013) (Figure 2-8A). 53Figure 2-7: Conformational changes in BtuB upon ligand binding. 2.5µM Nd-BtuB was incubated with trypsin at 37℃ for the indicated times in the presence or absence of 2µM vitamin B12. Samples were run on 10% SDS-PAGE and visualized by Coomassie blue staining. Interestingly, FhuA undergoes the same conformational changes when in complex with colicin M or ferricrocin. BtuB was also tested for conformational changes when in complex with colicin E3 using the trypsinolysis assay. BtuB was nearly completely digested in the presence of colicin E3. A faint band is still visible at the same position on SDS-PAGE as BtuB. However, colicin E3 also migrates on SDS-PAGE at the same position as BtuB therefore this band may represent undigested colicin E3 (Figure 2-8B).  Opening of the FhuA channel using denaturation by urea allows passage of ligands through the FhuA barrel (Udho et al. 2012). The opening of the BtuB channel for passage of bound ligand was investigated using urea treatment of Nd-BtuB in concert with trypsinolysis. Nd-BtuB was incubated with a gradient of urea and loaded onto BN-PAGE gel (Figure 2-8C). Nd-BtuB remained intact up to 4M urea indicating that both the nanodisc and the β-barrel are 54resistant to high concentrations of urea. Trypsinolysis of these samples revealed that bound B12 had been lost at 4M urea allowing most of the protein to become digested by trypsin. Urea (M) 0 0.5 1.0 2.0 4.0BN-PAGETrypsinBtuBFhuAB12 - + - - - + - -Fc - - + - - - + -Colicin M - - - + - - - +BtuB/Colicin E3B12 - - + -Colicin E3Trypsin- - - +- + + +BtuB FhuAA BCColicin MFigure 2-8: Conformational changes and plug movement in TonB-dependent transporters. A) 2.5µM Nd-BtuB and 2.5µM Nd-FhuA were incubated at 37℃ with trypsin and ligands 2µM vitamin B12, 2µM ferricrocin, and ~5µM colicin M before visualization on 12% SDS-PAGE with Coomassie blue staining B) ~2.5µM Nd-BtuB was incubated at 37℃ with trypsin and the ligands vitamin B12 or colicin E3 and visualized on 12% SDS-PAGE with Coomassie blue staining C) ~2.5µM Nd-BtuB was exposed to an increasing concentration of urea (0-4M) with vitamin B12 and visualized on BN-PAGE or the samples were trypsinized at 37℃ and visualized by 12% SDS-PAGE with Coomassie blue.552.4 Discussion2.4.1 Reconstitution of BtuB into Nanodiscs Escherichia coli rely on TonB-dependent transporters to bind iron-siderophores and vitamin B12 and then transport these ligands across the outer membrane. While many molecular details have been elucidated concerning how transport of bound ligands occurs many details remain unknown. Binding of TonB to ligand-loaded transporters elicits uncharacterized conformational changes in those transporters (Pawelek et al. 2006; Shultis et al. 2006; James et al. 2008). Previous research has employed a lipid mimetic such as detergents (Chimento et al. 2003b; Fanucci et al. 2003) or liposomes (Fanucci et al. 2002) for the investigation of BtuB. Detergents alter the conformation of BtuB (Fanucci et al. 2003a) and detergents alter the binding of TonB to the TonB dependent transporter FptA (Choul-Li et al. 2008) and liposomes do not permit access to both sides of the membrane simultaneously, complicating biochemical studies. Reconstitution of BtuB into nanodiscs negates the problems of these alternative approaches. Nd-BtuB is monodisperse (Figure 2-1C) and able to interact with ligands and protein partners (Figure 2-2A, B, and Figure 2-3D). BtuB was reconstituted without addition of lipids. BtuB co-purifies with a number of lipids and lipopolysaccharides encircling the β-barrel (Ferguson et al. 1998; Chimento et al. 2003a; 2003b) and these would co-reconstitute within the nanodisc. Addition of lipids would not recapitulate the composition of the outer membrane because of the high concentration of lipopolysaccharide in the outer leaflet. Nor would addition of lipopolysaccharide recapitulate the environment of the outer membrane because 56lipopolysaccharide would not reconstitute exclusively to the outer leaflet. Therefore, no additional lipids were reconstituted with BtuB.2.4.2 Interaction of TonB with Nanodisc BtuB In CHAPS detergent BtuB has high affinity binding to TonB (62nM ± 5 nM) in the absence of calcium and vitamin B12. Addition of calcium and vitamin B12 increased binding affinity to approximately 26nM ± 2 nM (Freed et al. 2013). However, Nd-BtuB does not bind an equimolar concentration of TonB in the absence of ligand, but addition of 2µM B12 leads to complete complex formation of Nd-BtuB-TonB-B12. Thus, BtuB only recruits TonB after becoming ligand-loaded. The transport of siderophore and cobalamin compete in live cells (Kadner and Heller, 1995) implying that TonB dissociates from transporters after completion of the TonB-dependent transport cycle and is then recruited to other ligand-loaded transporters.  TonB has also been implicated in functioning as a dimer during the TonB-dependent transport process (Chang et al. 2001; Sauter et al. 2003; Khursigara et al. 2004; 2005). More recent research has cast doubt on whether TonB functions as a dimer during the transport process (Postle et al. 2010; Freed et al. 2013; Mills et al. 2014). The observed change in migration of Nd-BtuB-TonB2 from Nd-BtuB-TonB indicates that wild-type TonB interacts with BtuB as a monomer. The artificially dimerized TonB2 likely interacts with BtuB through only a single TonB monomer. Addition of DTT breaks the disulphide crosslink and allows the second TonB to dissociate unincorporated from the complex (Figure 2-2B). Ergo, TonB interacts as a monomer 57with BtuB during the initial ligand-induced recruitment of TonB that ultimately elicits retention of ligand within BtuB.  The results from static light scattering also suggest that TonB interacts with BtuB as a monomer. The molecular mass of the Nd-BtuB-TonB-B12 complex is approximately 146kDa while Nd-BtuB and TonB alone are approximately 120kDa and 24kDa, respectively (Figure 2-2C). The mass of the complex corresponds to a monomer of TonB interacting with Nd-BtuB, reinforcing the conclusion that TonB interacts as a monomer with the transporter.2.4.3 TonB-dependent retention of ligand within Nanodisc BtuB Competition assays between labeled and unlabeled B12 show increased retention of the ligand within BtuB when in complex with TonB (Figure 2-3A). The competition between BtuB and BtuF for B12 also shows that in the absence of TonB the B12 is shared between BtuB and BtuF. However, when TonB is present it forms a complex with the ligand-loaded BtuB and causes retention of B12 with BtuB (Figure 2-3B). This confirms that TonB induces a conformational change in BtuB causing the ligand to be retained. Indeed, this effect allows BtuB to retain B12 strongly enough to resist binding of colicin E3 (Figure 2-3D). Whether TonB increases the overall affinity of BtuB for B12 or alters the binding kinetics of BtuB for B12 was unknown.  The affinity of Nd-BtuB for B12 was found to be decreased when in complex with TonB (Figure 2-4C). This may have been due to the opening of a second lower-affinity B12 binding site  58(Figure 2-4D); however, this was unsupported by crystal structure data (Shultis et al. 2006). It also does not explain the apparent retention of ligand within BtuB. The dissociation of ligand from BtuB was found to be greatly slowed by the presence of TonB (Figure 2-5A). This suggests that B12 takes much longer to dissociate from BtuB when complexed with TonB. This result shows why unlabeled B12 does not displace [57Co]-cobalamin on native gels or on SPA when in complex with TonB. It also shows why [57Co]-cobalamin is not shared back to BtuF during the competition assay and why colicin E3 cannot replace B12 on Nd-BtuB-B12-TonB complexes. BtuB clearly exchanges high affinity B12-binding for a slow dissociation of bound B12. Semilogarithmic plots of the dissociation data do not show the contribution of a second B12 binding site (Figure 2-5B). Ergo, the mechanism of how TonB retains ligand in BtuB is unknown. The TonB box of BtuB was deleted to investigate the retention of B12 within BtuB. The overall dissociation rate of B12 was slower from the BtuB∆6-12 versus BtuBWT (Table 2-1 versus Table 2-2). Addition of TonB, however, did not significantly affect this dissociation (Figure 2-6A and B; Table 2-2). Therefore, if TonB cannot bind BtuB no conformational change was induced in BtuB to retain ligand.  An apical loop of BtuB (residues 85 to 95) within the B12 binding site moves in response to TonB binding (Shultis et al. 2006). As a possible mechanism for ligand retention within BtuB the apex residue of this loop, valine 90, was mutated to an arginine to disrupt the conformation of 59this part of the binding site. Overall dissociation from this mutant BtuBV90R was significantly faster than wild-type. But this was largely unaffected by the addition of TonB, much like with BtuB∆6-12 (Figure 2-6C; Table 2-2). The dramatic increase in the dissociation rate of B12 from Nd-BtuBV90R shows that the movement of this apical loop is required to retain the B12 substrate. Disruption of the apical loop does not lead to prevention of B12 binding, but it does prevent the retention of ligand. Complex formation between Nd-BtuBV90R and TonB also has no substantial effect on the retention of ligand implicating this apical loop as the mechanism through which TonB acts to retain B12 in BtuB. Many TonB-dependent transporters retain bound ligands using an induced fit of the receptor for the ligand (Payne et al. 1997; Faraldo-Gomez et al. 2003; Yue et al. 2003; Cobessi et al. 2010). TonB has not been implicated in the induced fit of these transporters. Retention of ligand by binding of TonB to transporters has previously been proposed for the TonB-dependent transporter FhuA (James et al. 2009) and TonB has been implicated in altering the binding kinetics of FpvA depending on the iron-loaded status of the pyoverdin ligand (Clément et al. 2004). The decreased binding affinity between BtuB and B12 upon binding of TonB suggests that BtuB sacrifices high affinity binding in exchange for a slower dissociation rate. There are a number of reasons TonB may induce retention of ligand within BtuB. One reason may be to engage the TonB-dependent transport process. TonB exists in a complex with ExbB and ExbD in the inner membrane in a ratio of 1:7:2 ratio of TonB:ExbB:ExbD (Higgs et al. 2002) although more recent purification of the complex shows a 1:4:1 ratio (Sverzhinsky et al. 2015). This discrepancy could suggest that these proteins do not always exist in a functional complex or that 60the ratio of proteins changes during the transport cycle. This latter explanation is corroborated by the discovery that a homodimer of ExbD is cycled out of the complex during the energization process of TonB (Gresock et al. 2015). In this model TonB prevents bound ligand in the transporter from dissociating back into the extracellular space while the ExbB-ExbD complex is readied for driving transport.  Opening of the BtuB channel by unfolding of the plug domain would disrupt the B12 binding site (Figure 2-8; Udho et al. 2012). Disruption of the high affinity B12 binding site is necessary to allow dissociation of the ligand into the periplasm but no directionality of this dissociation has been proposed. TonB may be increasing the retention of ligand through movement of the apical loop (residues 85 to 95) (Shultis et al.  2006) in order to prevent B12 from dissociating back into the extracellular medium. Molecular dynamics simulations show that  the 85-95 apical loop is not disrupted by the partial unfolding of the plug domain (Gumbart et al. 2007). Therefore, TonB binding BtuB may cause movement of the apical loop to hold B12 within the binding site and may continue to hold B12 within BtuB while the plug domain is partially unfolded. This would allow B12 to dissociate into the periplasm and prevent dissociation back into the extracellular medium. 2.4.4 Conformational changes in Nanodisc BtuB TonB-dependent transporters undergo conformational changes upon binding of their ligands (Payne et al. 1997; Locher et al. 1998; Chimento et al. 2003b; Yue et al. 2003; Cobessi et al. 2010). Binding of cognate ligands to TonB-dependent transporters leads to conformational 61changes that make these transporters inaccessible to trypsin (Figure 2-7). Future work must address whether this is a common feature amongst TonB-dependent transporters. This seems likely given the highly homologous structures of TonB-dependent transporters (Noinaj et al. 2010). Intriguingly, colicins are also able to elicit conformational changes in TonB-dependent transporters (Cadieux et al. 2003; Mills, 2013). However, this appears to be specific for group B colicins that must pass through their cognate receptor and engage the TonB transport system to access the periplasm. Group A colicins, which engage the Tol system and only use BtuB as an anchor to attach to the cell surface, do not elicit conformational changes (reviewed in Cascales et al. 2008). Binding of the colicin E3 R-domain to BtuB, unlike binding of B12, will not lead to disorder of the TonB box or recruitment of TonB (Cadieux et al. 2003; Fanucci et al. 2003b; Freed et al. 2013).  This trypsin accessibility approach can also be used to investigate transport through the channel of BtuB, as well as other TonB-dependent transporters (Udho et al. 2012). Incubation of Nd-BtuB with 4M urea results in denaturation of the plug domain, but not the nanodisc or the barrel, because the protein becomes trypsin accessible after release of bound B12 substrate but does not denature on non-denaturing gel (Figure 2-8C). This indicates that denaturants, such as urea, can be used to reproduce the displacement of the plug and open the lumen of the β-barrel for passage of bound ligands by TonB.  Future research will uncover how the plug of BtuB is displaced to allow B12 to diffuse into the periplasm either partially unfolded (Gumbart et al. 2007) or as a rigid domain (Ma et al. 622007). Nanodiscs are a platform for those investigations because of they have no requirement for detergents, their included reconstitution of the lipid bilayer, and their resistance to denaturating conditions needed to displace the plug domain. 63Chapter 3: Transport of bacteriocin colicin E3 across the outer membrane3.1 Introduction Colicin E3 is a ribosomal RNase produced by some strains of E. coli to eliminate competing strains. In order to access the ribosomal target molecule colicin E3 must be transported across two separate membranes. Colicin E3 initially binds the vitamin B12 transporter BtuB to anchor itself onto the cell surface with high affinity (Di Masi et al. 1973; Cavard, 1994). The colicin then weaves the N-terminal 83 unstructured residues into the trimeric OmpF porin (Law et al. 2003; Yamashita et al. 2008; Housden et al. 2013) and interacts with the Tol system to elicit the dissociation of the bound immunity protein (Vankemmelbeke et al. 2009). The next steps of transport of the nuclease domain have not been fully elucidated. Upon release of the immunity protein the nuclease domain of colicin E3 blocks passage through the lumen of OmpF (Zakharov et al. 2006). The nuclease domain of the DNase colicin E9 interacts with the inner membrane protein FtsH, a AAA+ ATPase protease. This protease may be usurped by colicin to traverse the inner membrane (Walker et al. 2007). Processing of colicin E3 by FtsH shows only the nuclease domain enters the cytoplasm (Chauleau et al. 2011). The exact molecular details of how the colicin nuclease domain passes through the outer membrane is unknown.  To investigate the transport of colicin E3 across the outer membrane the binding of colicin to BtuB, OmpF, and TolB was investigated using a combination of both nanodiscs and detergents. SDS was found to promote the interaction of colicin E3 with OmpF. Colicin E3 likely interacts with OmpF through the periphery of the protein rather than weaving of N-terminal unstructured domain through the OmpF channel. When urea was used to denature colicin E3 it 64was found to bind lipopolysaccharide (LPS). Colicin E3 free of immunity protein (Im3) was found to bind LPS and the presence of Im3 inhibits this binding interaction. A truncation consisting of the 96 residues of the C-terminus of colicin E3, the nuclease domain, bound to LPS. In the presence of Im3 the nuclease domain did not interact with LPS. This shows that colicin E3 binds LPS as part of the transport across the outer membrane after the release of the immunity protein. 3.2 Materials and Methods3.2.1 Purification of OmpF OmpF was expressed from the Escherichia coli strain JW2203 (∆ompC) generously donated by Dr. Gerd Prehna (Prehna et al. 2012). Cells were grown overnight at 37℃ in LB broth supplemented with 25µg/ml kanamycin to OD600 ≅ 1.5. Cells were harvested by centrifugation at 5,000 rpm, 5 minutes, 4℃ then resuspended in buffer A (50mM Tris-HCl, pH=7.9, 50mM NaCl, 10% v/v glycerol) supplemented with 1mM PMSF and then lysed in a Microfluidizer at 10,000 psi. Lysate was cleared of insolubles with a centrifugation at 6,000 rpm, 5 minutes, 4℃ and membrane was pelleted by centrifugation at 45,000 rpm, 40 minutes, 4℃. Membrane pellets were resuspended in buffer A to a concentration of 3.0mg/ml and solubilized with 1% Triton-X-100 for one hour at room temperature with gentle rocking. Solubilized material was centrifuged at 45,000 rpm, 40 minutes, 4℃ to pellet the outer membrane fraction. Pellets containing the outer membrane were resuspended in buffer A to 3.0mg/ml and solubilized with 1% LDAO overnight at 4℃ with gentle rocking. Solubilized material was centrifuged a 65final time at 45,000 rpm, 40 minutes, 4℃ and the supernatant was applied to a 5ml HiTrap Q FF column equilibrated in buffer B. OmpF bound to the column was then eluted with a gradient of 1M NaCl in buffer B over 10 column volumes. The purified OmpF yielded approximately 3.5mg of protein per litre of culture. Fractions were analyzed on SDS-PAGE with Coomassie brilliant blue staining with and without boiling at 98℃ in 5x Laemmli buffer. 3.2.2 Purification of TolB The TolB gene was amplified from the genome of Escherichia coli K-12 by PCR and cloned into pET28 between the NcoI and XhoI restriction sites with a C-terminal 6His-tag. Site directed mutagenesis was used to introduce a cysteine substitution at the position of proline 201 to generate TolBP201C. BL21 transformed with the TolB-pET28 plasmid was grown in LB broth supplemented with 25µg/ml kanamycin to OD600 ≅ 0.5 then induced with 1mM IPTG. Cells were grown three more hours to OD600 ≅ 1.0 then harvested by centrifugation at 5,000 rpm, 10 minutes, 4℃. Cell pellets were resuspended in buffer A with 1mM PMSF and lysed at 10,000 psi with a Microfluidizer. Cell lysate was cleared by centrifugation at 45,000 rpm, 40 minutes, 4℃ and supernatant was passed over a Co2+ agarose column equilibrated in buffer A. TolB protein was eluted using a step gradient of buffer C. TolBP201C was purified using the same protocol. Approximately 12mg of TolB per litre of culture was produced from each purification.3.2.3 Expression of colicin E3A33C and E3D381C Site directed mutagenesis was used to introduce cysteine substitution mutations into the positions of alanine 33 (A33C) and aspartic acid 381 (D381C). Colicin E3 mutations were 66purified using the same protocol as wild-type. The colicin E3-Im3 heterodimer mutants were both produced at approximately 50mg of protein per litre of culture.3.2.4 Expression and nanodisc reconstitution of BtuBT11C Site directed mutagenesis was used to introduce cysteine substitution mutation into position threonine 11 of the TonB box of BtuB. BtuBT11C was purified using the same protocol as BtuB wild-type. Reconstitution of was BtuBT11C performed using the same protocol as wild-type BtuB.3.2.5 Fluorescence labeling BtuBT11C reconstituted into nanodiscs (Nd-BtuBT11C) and colicin E3 were labeled by addition of a five-fold molar excess of AlexaFluor 680 C2-maleimide (Molecular Probes-Invitrogen, Karlsruhe, Germany) (Ɛ679nm=175,000cm-1M-1) in 100µl volume. Proteins and dye were allowed to incubate for  one hour in the dark at room temperature. Labeled proteins were separated from excess dye by Superdex 200 gel filtration chromatography. Gels containing labeled proteins were visualized using LI-COR Odyssey IR Imaging (LI-COR Biosciences).3.2.6 Purification of lipopolysaccharide Lipopolysaccharide was isolated from E. coli strain DH5α following the protocol of Kramer et al. 2002. Cells were grown in 2L of LB broth to OD600 ≅ 0.6 then harvested by centrifugation. Cells were resuspended in buffer A and proteinase K was added to cells at a final concentration of 0.1mg/ml and incubated at 65℃ for one hour. DNase and RNase was added to 670.2mg/ml final concentration alongside 1µl/ml 20% MgSO4 and 4µl/ml chloroform and incubated overnight at 37℃. The cell mixture was mixed with an equal volume of 90% phenol and incubated at 65℃ for 15 minutes with shaking. Suspension was cooled on ice and and then centrifuged at 3,000 rpm for 15 minutes, 4℃. The upper layer of supernatant was pipetted into a new container and 300µl of ddH2O was added to extract residual phenol. 10 volumes of 95% ethanol and a final concentration of 0.5M sodium acetate were added to the suspension to initiate precipitate of LPS. LPS was allowed to precipitate at -20℃ overnight and the suspension was centrifuged at 3000 rpm, 10 minutes to pellet the LPS. Supernatant was removed and the LPS pellet was resuspended in ddH2O. LPS was dialyzed against 20mM Tris-HCl, pH=7.9, 50mM NaCl buffer (3500 Da MWCO). LPS purity and concentration was assessed by migration on SDS-PAGE and visualized with silverstaining.  3.2.7 Purification of C96Im3 The region of colicin E3-Im3 ORFs encoding the C-terminal 96 residues ribosomal nuclease domain of colicin E3 and the entirety of Im3 were cloned into pET28 between the NdeI and XhoI sites with an 6His-tag on the N-terminus of the C96 truncation. BL21 cells transformed with C96Im3-pET28 were grown to OD600 ≅ 0.5 and induced with 1mM IPTG. Cells were grown an additional three hours to OD600 ≅ 1.0 and harvested. Cells were resuspended in buffer A supplemented with 1mM PMSF and lysed at 10,000 psi in a microfluidizer. Cell lysate was cleared by centrifugation at 45,000 rpm, 40 minutes, 4℃. Supernatant was passed over a Co2+ agarose column equilibrated in buffer A. C96Im3 was eluted with a step gradient of buffer C. 683.2.8 Purification of Im3-free colicin E3 and C96 Colicin E3Im3 purified from IMAC was denatured with an 8M final concentration of urea (added as powder directly to purified E3Im3) to separate E3 and Im3 in solution. The denatured proteins were then separated on a Superdex 200 HR 10/30 equilibrated in buffer A supplemented with 6M urea. Fractions eluted were analyzed using 18% SDS-PAGE and fractions containing isolated colicin E3 were dialyzed against 20mM Tris-HCl, pH=7.9, 50mM NaCl.  To obtain Im3-free C96 the purification protocol of C96Im3 was followed as normal. Prior to elution with buffer C a step gradient of 6M guanidium-HCl was added to denature the C96Im3 complex. Im3 was eluted fully and the column was restored to buffer A. C96 was then eluted with a step gradient of buffer C. 3.2.9 Purification of colicin E3-TR  The Transport (T-) domain and Receptor binding (R-) domain of colicin E3 (residues 1 - 455) was cloned into pET28 between the NcoI and XhoI sites using PIPE with a 6His-tag at the C-terminus. This removed the C-terminal nuclease domain of colicin E3 and the immunity protein Im3. BL21 was transformed with the E3TR-pET28 plasmid and grown to OD600 ≅ 0.5 and then induced with 1mM IPTG. Cells were grown three additional hours to OD600 ≅ 1.0 and then harvested, lysed, and cleared by centrifugation at 45,000 rpm, 45 minutes, 4℃. Cleared lysate was passed over a Co2+ agarose column in buffer A. E3-TR protein was eluted using a step gradient of buffer C. 693.2.10 Isothermal titration calorimetry Isothermal titration calorimetry (ITC) experiments were performed on a MicroCal ITC-200 instrument (GE Healthcare). LPS and all proteins were dialyzed (MWCO 3500 Da and 12-14,000 Da) into ITC buffer composed of 20mM Tris-HCl, pH=7.9, 50mM NaCl. All binding experiments were initiated by a single 0.2µl injection of titrant. Binding experiments for the full length colicin E3 consisted of 20 injections (2µl volume) of ~500µM LPS injected into the cell containing either 20µM colicin E3 or 18µM colicin E3Im3. Binding experiments of the truncated C96 fragment consisted of 20 injections (2µl volume) of ~500µM LPS injected into the cell containing either 50µM C96 alone or 50µM C96Im3. All experiments were thermostated at 25℃. Heat of dilution was ascertained by injection of LPS ligand into a cell containing buffer alone. Binding isotherms were analyzed using a single-site binding model and errors were derived from chi-squared degrees of freedom on Origin 7 software (MicroCal, Inc.). 3.3 Results3.3.1 Binding of colicin E3 to BtuB and TolB without detergents Colicin E3 binds BtuB with an affinity high enough to outcompete the ligand B12 (Cavard, 1994; Kurisu et al. 2003). The functionality of purified colicin E3, both alone and in complex with the immunity protein (Im3) must be ascertained by interaction with purified BtuB. Detection of complex between colicin E3 and BtuB can be difficult because E3 and BtuB cannot be distinguished on SDS-PAGE (Housden et al. 2013). To investigate the binding of colicin E3 to BtuB reconstituted into nanodiscs the proteins were labeled using maleimide dye (Figure 703-1A). Neither BtuB nor colicin E3 has endogenous cysteines, a cysteine was introduced by site-directed mutagenesis into a threonine at position 11 of BtuB (Figure 3-1B). This residue is within the TonB box and is on the opposite face of the protein from the vitamin B12 and colicin E3 binding site. This ensures that the ligand binding site is unoccupied. Colicin E3 was labeled even without a cysteine residue, likely through amine-linkage (Sharpless et al. 1966). There is a cysteine within the immunity protein of the E3Im3 complex but it is not solvent accessible and cannot conjugate with the maleimide dye (Shoham and Djebli, 1992). 71E3Im3Maleimide dye Amine dyeBtuBT11CA B0 2 4 6 8 10Absorbance (mAU)Volume (mL)0 2 4 6 8 10Absorbance (mAU)Volume (mL)0 2 4 6 8 10Volume (mL)Fractions1 9 Fractions1 5 Fractions1 7Figure 3-1: Labeling or colicin E3 and BtuB. A) Size exclusion profiles and corresponding fractions of maleimide and amine dye labeled colicin E3Im3 complexes analyzed on 12% SDS-PAGE. B) Size exclusion profile and fractions of maleimide dye labeled BtuBT11C analyzed on 12% SDS-PAGE. Gels were visualized with a LI-COR Odyssey IR Imaging Scanner (LI-COR Biosciences). Maleimide dye conjugated Nd-BtuBT11C migrated to the same position on CN-PAGE gel as Nd-BtuBWT (Figure 3-2A, lane 1). Unlabeled colicin E3 and E3Im3 complex was not visible (Figure 3-2A, lane 2 and 3). Addition of colicin E3 either alone or in complex with Im3 resulted in formation of a complex between E3 and BtuB that was retarded as it migrated into the gel (Figure 3-2A, lane 4 and 5).  Vitamin B12 and colicin E3 compete for the same binding site on BtuB. To assess this competition vitamin B12 was added to samples with a final 2µM concentration. Very little 72inhibition of complex formation between Nd-BtuB and E3 could be observed (Figure 3-2A, lane 6 and 7). This is in agreement with prior work that shows that colicin E3 binds BtuB with higher affinity than the ligand vitamin B12 (Cavard, 1994).  Interestingly, colicin E3Im3 incubated with maleimide dye became labeled despite the absence of accessible cysteines in the native protein (Figure 3-2A). The ability of this labeled protein to bind unlabeled Nd-BtuB was assessed using CN-PAGE. Unlabeled Nd-BtuB was not visible on CN-PAGE (Figure 3-2B, lane 1). Labeled colicin E3Im3 migrated to the same position as unlabeled colicin E3Im3 on CN-PAGE (Figure 3-2B, lane 2). Addition of Nd-BtuB to E3Im3 resulted in a slight shift of colicin deeper into the gel (Figure 3-2B, lane 3). This is likely because colicin E3 has a pI of 9.04. Other proteins with high isoelectric points (such as TonB and Colicin M) do not migrate on CN-PAGE (Mills et al. 2014). The immunity protein in complex with colicin E3 likely acts to neutralize this positive charge and carry colicin E3 into the gel because the isoelectric point of Im3 is ~3.96. The charge of Nd-BtuB (pI~5.10) causes colicin E3 to migrate further into the gel even when in complex. Addition of vitamin B12 inhibits formation of a complex between colicin E3 and Nd-BtuB by occupying the B12 binding site. This, also, does not entirely obstruct the binding of E3 to Nd-BtuB because of the higher binding affinity of E3 for BtuB. 73Figure 3-2: Colicin E3 binds BtuB reconstituted into nanodiscs. A) Maleimide labeled Nd-BtuBT11C incubated with unlabeled colicin E3 or colicin E3Im3 complex in the presence or absence of vitamin B12 where indicated. Samples were assayed by CN-PAGE and visualized using Odyssey scanner. B) Maleimide labeled colicin E3 incubated with unlabeled Nd-BtuB in the presence or absence of ligand vitamin B12. Samples were visualized on CN-PAGE with an Odyssey scanner. TolB binding to colicin E3 was also assessed using nondenaturing gels. On CN-PAGE gels purified TolB is a band near the top of the gel (Figure 3-3A, lane 2). TolB and colicin E3Im3 were incubated together and a complex was observed forming between the two proteins (Figure 3-3A, lane 3). The entirety of colicin E3 was complexed with TolB given the high affinity (KD ~125nM) of colicin E3 for TolB (Bonsor et al. 2009). Previous work discussed the formation of spontaneous crosslinks between colicin E9 and TolB after substitution of residue alanine 33 in 74the colicin and proline 201 of TolB with cysteines (Housden et al. 2013). The same cysteine substitutions were introduced into TolB and colicin E3 to observe whether the same effect occurs with colicin E3. Purification of colicin E3A33CIm3 results in formation of a monomer and dimer (colicin E32) (Figure 3-3B, lane 1) while TolBP201C remains a monomer (Figure 3-3B, lane 2) on SDS-PAGE. Addition of 5mM DTT results in a monomer of colicin E3 (Figure 3-3B, lane 3). Incubation of colicin E3A33C with TolBP201C resulted in spontaneous formation of a complex, addition of oxidizer copper phenanthroline (CP3) to promote disulphide crosslinking only increased formation of crosslink slightly (Figure 3-3B, lane 5 compared to lane 6). Addition of 5mM DTT prevented the spontaneous crosslinking between these two proteins (Figure 3-3B, lane 7).  75E3Im3E3Im3-TolBTolBTolBP201CColicin E3A33CE3Im32E3A33C-TolBP201CColicin E3A33CTolBP201C+--++--+++++++A BColicin E3Im3TolB+--+++CP3DTT-----+-+--+--+Figure 3-3: Colicin E3 spontaneously crosslinks with TolB. A) ~3µM Colicin E3Im3 was incubated with ~5µM TolB and analyzed on CN-PAGE with Coomassie blue staining. B) ~3µM Colicin E3A33C and ~5µM TolBP201C were incubated together where indicated in the presence and absence of reducing agent DTT or oxidizing agent copper phenanthroline (CP3). Samples were run on 12% SDS-PAGE and visualized by Coomassie blue staining. 3.3.2 Binding of colicin E3 and OmpF  Interaction of colicin E3 with OmpF is a largely transient interaction. OBS1 of colicin E9 binds with 2µM affinity to OmpF while OBS2 binds with 24µM affinity (Housden et al. 2010). Interestingly, on SDS-PAGE there exists an interaction between maleimide-labeled colicin E3 and OmpF and BtuB, but not FhuA or colicin M (Figure 3-4A). OmpF trimer resists denaturation by SDS and exists as a trimer upon SDS-PAGE. However, boiling the trimeric OmpF in the presence of SDS will denature the protein into individual monomers (Rocque et al. 1987). OmpF and BtuB boiled at 98℃ for five minutes do not form a complex with labeled colicin E3 (Figure 763-4A). Coomassie stained SDS-PAGE shows complex formation between colicin E3 and OmpF, which was similar to results published for colicin N (Dover et al. 2000) (Figure 3-4B). Figure 3-4: Colicin E3 binds OmpF under denaturing conditions. A) Maleimide-labeled colicin E3 was incubated with ~2µM OmpF, ~2µM BtuB, ~5µM FhuA, or ~5µM colicin M. Samples were either untreated or boiled at 98℃ in SDS for 5 minutes. Samples were visualized on 10% SDS-PAGE with an Odyssey scanner. B) Binding of colicin E3 to OmpF visualized on 10% SDS-PAGE by Coomassie blue staining. Addition of SDS promoted complex formation between colicin E3 and OmpF. This effect was tested on non-denaturing gels by addition of 2% SDS to each sample. In the absence of SDS colicin E3Im3, OmpF, and TolB are smears on BN-PAGE (Figure 3-5, lanes 1 to 3). No formation of complex was detected between colicin E3 and OmpF under these conditions (Figure 3-5, lane 4), but there was an interaction between colicin E3 and TolB (Figure 3-5, lane 5). Addition of 1% SDS caused each protein to migrate as distinct bands (Figure 3-5, lanes 7 to 9) and allowed an interaction between colicin E3 and OmpF (Figure 3-5, lane 10). Complexes were also observed between colicin E3 and TolB (Figure 3-5, lane 11) and between OmpF and TolB 77(Figure 3-5, lane 12) in agreement with the results of Rigal et al. 1997. This shows that SDS promotes the interaction of colicin E3 with OmpF. -SDS -SDS +SDS +SDSColicin E3Im3OmpFTolB+ - - + + - + + -+ - -- + - + - + + - +- + -- - + - + + - + +- - +Colicin E3OmpFColicin E3-OmpFColicin E3-TolBOmpF-TolBTolBColicin E3Im3Colicin E3Im3-TolBFigure 3-5: SDS promotes the binding of colicin E3 to OmpF. Comparison of ~6µM colicin E3 binding ~3µM OmpF and ~4µM TolB in the presence and absence of 2% SDS where indicated and visualized on blue native (BN)-PAGE. A mixture of colicin E3Im3 and OmpF were incubated with a gradient of SDS and analyzed by BN-PAGE (Figure 3-6). A specific concentration of SDS (~0.2%) was required to form a complex between colicin E3 and OmpF. 0.2% corresponds to the critical micelle concentration (CMC) of SDS and it is at this point that denaturation of proteins occurs (Andersen et al. 2009). 78Figure 3-6: SDS critical micelle concentration is required to promote binding. ~6µM colicin E3 and ~3µM OmpF were incubated together and titrated with an increasing concentration of SDS (0-10%). Samples were visualized on BN-PAGE. To determine whether this effect of SDS is artifactual, colicin M, a FhuA binding protein, was incubated with OmpF. Colicin M does not interact with OmpF (reviewed in Cascales et al. 2007) and when incubated with SDS and OmpF no complex between the two proteins was formed (Figure 3-7, lanes 5 compared to lane 6). On the same gel colicin E3 incubated with OmpF in the presence of SDS continues to form a protein-protein complex (Figure 3-7, lanes 1 to 4). 79Colicin E3Colicin E3+OmpFColicin MColicin E3Im3OmpF+-TolB --Colicin M-+----+-++-----+-+-+1 2 3 4 5 6Figure 3-7: SDS promotes interaction only with colicin E3. Comparison of ~6µM colicin E3 and ~6µM colicin M binding to OmpF in 2% SDS analyzed on BN-PAGE.  The stoichiometry of the complex on BN-PAGE was found to be 1:1 ratio of OmpF trimer:colicin E3. Colicin E3 with a cysteine substituted for the aspartic acid residue at position 381 to yield colicin E3D381C. This cysteine was positioned within the tip of the R-domain at the BtuB binding site. This position is not expected to obstruct interactions with OmpF, unlike the E3A33C mutation. E3A33C is within the N-terminal unstructured OmpF interacting domain and dimers could interfere with OmpF interactions. E3D381C existed as a population of monomers and dimers after purification. Wild-type colicin E3 formed only a single complex with OmpF but the dimer of colicin E3 also formed a complex with OmpF. This reflects a 1:2 OmpF trimer:colicin 80E3 ratio. Addition of DTT to wild-type colicin E3-OmpF did not alter the migration of the complex. However, addition of DTT to the colicin E3 dimer-OmpF yielded a complex of the same size as wild-type (Figure 3-8A). This indicates that colicin E3 forms a 1:1 complex with the OmpF trimer. To confirm this result a gradient of colicin E3 with OmpF in the presence of 2% SDS on BN-PAGE showed that an excess of colicin E3 over OmpF did not produce any additional species of complex between E3 and OmpF (Figure 3-8B). Figure 3-8: Colicin E3 binds OmpF in a 1:1 ratio. A) ~6µM colicin E3WT or ~5µM colicin E3D381C were incubated with OmpF in 2% SDS in the presence or absence of 1mM DTT. Samples were visualized on BN-PAGE. B) Gradient of colicin E3 (0-8µM) with ~3µM OmpF in 2% SDS visualized on BN-PAGE. Colicin E3 does not interact with denatured outer membrane transporters or denatured TolB. OmpF, BtuB, and also TolB were boiled in the presence of SDS prior to addition of colicin E3. There is no formation of complex if the E3 binding partners after denaturation (Figure 3-9A, lanes 2 to 4). However, denaturation of the colicin E3Im3 complex by boiling at 98℃ in 2% 81SDS, then cooled and incubated with OmpF, BtuB, and TolB prior to analysis on BN-PAGE still permits the formation of complex between E3 and its binding partners (Figure 3-9A, lanes 5 to 7). Colicin E3 does not have to be structured to interact with its receptor binding partners.However, those receptors must be folded to interact with colicin E3. Indeed, in SDS a ternary complex composed of colicin E3-OmpF-TolB can be formed on BN-PAGE (Figure 3-9B). The order of addition appeared to require TolB to be added last after the addition of OmpF (Figure 3-9B, compare lane 7 to lane 8) suggesting that the N-terminal unstructured domain of E3 weaved within the OmpF channel.Figure 3-9: OmpF, BtuB, and TolB must be structured to interact with colicin E3. A) ~3µM OmpF, ~2µM BtuB, and ~4µM TolB were denatured by boiling in 2% SDS at 98℃ and cooled before addition of colicin E3. Alternatively, colicin E3 was also denatured in 2% SDS at 98℃ and cooled before addition of OmpF, BtuB, and TolB. Samples were analyzed on BN-PAGE. B) Ternary complex of colicin E3-OmpF-TolB formed in the presence of 2% SDS. Order of addition is indicated in the text. Samples were analyzed using BN-PAGE.  It was unknown what domain of colicin E3 binds to OmpF. The unstructured N-terminal domain, which contains the OmpF binding sites (Housden et al. 2010), is the presumed binding 82domain. To test this a technique was devised to isolate natively structured complexes from BN-PAGE. Colicin E3A33C mutant in 5mM DTT migrates as a single band as does TolBP201C (Figure 3-10A lanes 1 to 6). These proteins also spontaneously form crosslinks on BN-PAGE even in the presence of 8mM DTT (Figure 3-10A, lane 7 to 9). This E3A33C-TolBP201C crosslink was combined with OmpF in the presence of SDS and analyzed on BN-PAGE. The ternary complex of E3-OmpF-TolB is still formed (Figure 3-10B, lanes 3 and 4). This result suggests that the recruitment of colicin E3 to OmpF cannot be through the N-terminal unstructured T-domain because this was previously blocked by the crosslink with TolB. This suggests that another domain of the colicin is able to form a complex with outer membrane proteins in the presence of SDS. Figure 3-10: Colicin E3 binds the OmpF periphery. A) Preparative BN-PAGE gel of ~5µM colicin E3A33C, ~5µM TolBP201C, and ~3µM OmpF in 2% SDS and 8mM DTT. Gel was not destained in order to preserve protein conformation. Bands were excised and incubated with indicated partner proteins before loading onto the second BN-PAGE gel. B) Analytical BN-PAGE of isolated colicin E3A33C-TolBP201C crosslink interaction with OmpF in the presence of 2% SDS. Colicin E3A33C monomer, colicin E3-OmpF, and E3-OmpF-TolB are loaded as controls.833.3.3 Interaction of colicin E3 nuclease domain with lipopolysaccharide The promotion of colicin E3 binding to outer membrane partners by SDS suggests that SDS is mimicking some other binding component in the outer membrane. The most similar material in the outer membrane to fit these characteristics is lipopolysaccharide (LPS). LPS is an anionic surfactant similar to SDS (Rietschel et al. 1994), LPS replaces SDS in colicin-porin interaction studies (Dover et al. 2000). The interaction of colicin E3 with LPS was assessed by native PAGE. Colicin E3Im3 was incubated with an increasing concentration of LPS over a 30 minute period at 37℃ in the presence or absence of 3M urea. In the absence of urea only a moderate interaction between the colicin and LPS can be observed even at the highest concentration of LPS (Figure 3-11A, lanes 3 to 7). In the presence of 3M urea no change in migration was observed of colicin E3Im3 in low concentration of LPS (Figure 3-11A, lane 8), but an increasing concentration of LPS showed a retardation in migration between E3 and LPS (Figure 3-11A, lanes 9 to 12). Catalase (Amersham, Arlington Heights, IL) was used as a control for LPS interaction. Both colicin E3 and catalase migrate into clear native PAGE when untreated and when treated with 3M urea (Figure 3-11B, lanes 1 and 2 compared to lanes 3 and 4). LPS interacted slightly with the colicin E3Im3 complex, but did not interact with catalase (Figure 3-11B, lane 5 versus lane 6). Addition of both 3M urea and LPS resulted in the interaction of LPS with colicin E3 but not with catalase (Figure 3-11B, compare lane 7 to lane 8). 84Figure 3-11: Colicin E3 interacts with lipopolysaccharide. A) Colicin E3Im3 complex incubated with an increasing concentration of LPS (30 minutes 37℃) in the presence or absence of 3M urea. Samples were visualized on CN-PAGE with Coomassie blue staining. B) Comparison of colicin E3Im3 complex to catalase binding LPS. Samples were incubated with the indicated cofactors for 30 minutes at 37℃ then visualized on CN-PAGE with Coomassie blue staining. This interaction of colicin E3 with LPS in the presence of denaturant suggests that the dissociation of Im3 from E3 allows the binding of colicin E3 with LPS. To test this an Im3-free colicin E3 was obtained by denaturing the E3Im3 complex directly from IMAC with 6M urea. Denatured proteins were then subjected to size exclusion to separate colicin E3 (~60kDa) from 85Im3 (~10kDa) (Figure 3-12A). Colicin E3 and Im3 separated by this method were analyzed by BN-PAGE. The E3Im3 complex migrates as a smear (Figure 3-12B, lane 1) while E3 separated by this method migrates as a band at a slightly lower position in the gel (Figure 3-12B, lane 2). Isolated Im3 migrates near the dye front (Figure 3-12B, lane 3). The migration of colicin E3 and Im3 is comparable in position when the complex is denatured in 4M urea (Figure 3-12B, lane 4) or in SDS (Figure 3-12B, lane 5). AColicin E3Im3Colicin E3-Im3Colicin E3Im31 2 3 4 5-UreaSDS -----+--+BColicin E3Im3Fractions8 140 5 10 15 20Absorbance (mAU)Volume (mL)Figure 3-12: Isolation of colicin E3 from the immunity protein. A) Size exclusion chromatography profile of urea denature colicin E3 and Im3 on a Superdex 200 HR 10/300 column. Fractions were analyzed on 18% SDS-PAGE with Coomassie blue staining. Colicin E3Im3 complex is loaded in the first lane as a control. B) Colicin E3 in complex with Im3 (lane 1) or alone (lane 2) was compared to Im3 alone (lane 3), E3Im3 complex denatured in urea (lane 4), and E3Im3 complex denatured in 2% SDS (lane 6) visualized using BN-PAGE.86 Using the approach of Garidel et al. 2008, Henriksen and Andresen 2011, and Johnson et al. 2014 LPS interactions with colicin E3 were investigated using isothermal titration calorimetry  (ITC).  ~500µM LPS was injected into a cell containing either 20µM colicin E3 alone or 18µM colicin E3Im3 complex. In the absence of Im3 a binding affinity of ~2.4µM (±0.8µM) and a stoichiometry of approximately 1.07 (±0.06) is determined, although the heterogeneity of LPS makes it unlikely to determine an absolutely accurate stoichiometry (Figure 3-13A, Table 3-1). In the presence of Im3 only a modest binding between colicin E3 and LPS was detectable, suggesting that only a small population of the E3Im3 complex is able to interact with LPS (Figure 3-13B). This confirms that release of the Im3 from E3 allows the colicin to interact with LPS. Heat of dilution were ascertained by injection of LPS directly into buffer (Figure 3-13C).A B CFigure 3-13: Binding isotherm of colicin E3 to LPS. A) ~500µM LPS injected into 20µM colicin E3 alone or B) into 18µM colicin E3Im3 complex. C) Control of ~500µM LPS injected into buffer. Raw traces and integrated heats are presented. Im3 binds colicin E3 between the T-domain and C-domain (Soelaiman et al. 2001). To identify the domain of colicin E3 that binds LPS after release of Im3 a truncation of colicin E3 to 87the C-terminal 96 amino-acid residues representing the nuclease domain was produced. This truncation was termed C96. ITC binding of ~500µM LPS into 50µM C96 alone resulted in an affinity of approximately ~1.3µM (±0.4µM), however the stoichiometry was approximately 0.153 (±0.08) sites (Figure 3-14A, Table 3-1). This early saturation indicates that either a proportion of C96 is unavailable for binding LPS or that the LPS at ~500µM is more accessible to C96 than to the full colicin E3 protein. C96 embeds within phospholipid membranes (Vankemmelbeke et al. 2012, 2012b) and may, therefore, penetrate LPS aggregates. In the presence of Im3, however, there was no detectable binding between the C96Im3 complex to LPS beyond the dilution effects of LPS into buffer (Figure 3-14B compared to Figure 3-14C). A CBFigure 3-14: Binding isotherm of nuclease domain C96 to LPS. A) ~500µM LPS injected into 50µM C96 alone or B) into 50µM C96Im3 complex. C) Control of ~500µM LPS injected into buffer.88Table 3-1: Thermodynamic binding parameters of colicin-LPS interactionsInteractions Kd, µM N ∆H, (kcal mol-1) ∆S, (cal mol-1 deg -1)Colicin E3-LPS 2.4 ± 0.8 1.07 ± 0.063 8.3 ± 0.6 53.5C96-LPS 1.3 ± 0.4 0.153 ± 0.011 20.9 ± 2.1 97.2Errors for measurements are estimated errors from 𝜒2 minimized fit of the experimental data to a single-site binding model on Origin 7.0 Interaction experiments were also performed on a truncation of colicin E3 consisting of the T- and R-domains to determine whether these domains also interact with LPS. Binding between E3-TR truncation and LPS was not observed to occur. Titration of 500µM LPS into 20µM E3-TR was indistinguishable from controls of LPS titrated into buffer (Figure 3-15A compared to B). This indicates that colicin E3 does not interact with LPS through the T- or R-domain and that only the nuclease domain binds LPS. Altogether these results suggest that colicin E3 interacts with LPS through the C-terminal nuclease domain after the release of the immunity protein. 89A BFigure 3-15: Binding isotherm of colicin E3-TR to LPS. A) ~500µM LPS injected into 20µM colicin E3-TR. B) Control of ~500µM LPS injected into buffer control.3.4 Discussion 3.4.1 Denaturation promotes the interaction of colicin E3 with outer membrane proteins Colicin E3 forms complexes, including spontaneous disulphide crosslinks, with BtuB and TolB. Colicin E3 also forms complex with the trimeric protein OmpF in the presence of SDS. This SDS-promoted complex formation appears not to involve the N-terminal domain of colicin E3, unlike previous results from structure-function studies on the related colicin E9 (Housden et al. 2010; 2013). Indeed, interactions may be due to two combined effects of SDS in solution: i) SDS carries a negative charge, and ii) SDS has denaturing properties. Colicin E3 overall has a 90positive charge contributed by the C-terminal nuclease domain (Vankemmelbeke et al. 2012) and this domain also forms a complex with the immunity protein, Im3 (Soelaiman et al. 2001). Denaturing colicin E3 would result in release of the immunity protein and permit the interaction of the nuclease domain with the OmpF trimer (Zakharov et al. 2006).  These results are reminiscent of previous results obtained from studies of colicin N pore-forming domain interacting with trimeric proteins of the outer membrane (Dover et al. 2000). In these studies the binding of the colicin N pore-forming domain to OmpF, OmpC, and PhoE was performed in the presence of SDS. Indeed, later results of the colicin N-OmpF 2D crystals were prepared in buffer containing 0.1% w/v SDS and discovered that this colicin binds to the periphery of the OmpF trimer (Baboolal et al. 2008). Studies conducted with 1% (w/v) SDS have shown the pore-forming domain of colicin N fitting within the clefts in the OmpF trimer to interact at the protein-lipid interface (Clifton et al. 2012). SDS was replaced with LPS in experiments and was also found to increase the interaction between colicin N and trimeric outer membrane proteins (Dover et al. 2000).  Clearly, these effects of SDS are in part because of a molecular similarity as an anionic surfactant to LPS in the outer membrane (Dover et al. 2000). Unlike colicin N, however, colicin E3 exists in complex with the immunity protein Im3 (Soelaiman et al. 2001). The denaturation of colicin E3 by SDS allows the release of Im3, which also permits the interaction of colicin E3 with OmpF in these conditions. Therefore, the nuclease domain of colicin E3 interacts with OmpF after release of Im3.91 3.4.2 Colicin E3 interacts with lipopolysaccharide after dissociation of the immunity protein The colicin E3 nuclease domain was found to interact with LPS after the removal of the immunity protein Im3. When in complex with the immunity protein the colicin does not interact with LPS (Figure 3-13B, 3-14B). This effect may be due to electrostatics, the colicin has a isoelectic point of pI≈9.04 and gives it an overall positive charge at neutral pH. This net positive charge is primarily contributed by the C-terminal 96 residues of the nuclease domain (pI≈9.79; Vankemmelbeke et al. 2012) while the remainder of the colicin (including T- and R-domains) has a pI≈6.28. LPS has a net negative charge at neutral pH due to the presence of carboxylate group on the 2-keto-3-deoxyoctonate (KDO) sugars of the core oligosaccharide (Brandenburg et al. 2005). Therefore, the nuclease domain interacts with the core oligosaccharide of LPS through electrostatics. The immunity protein has a pI≈3.96 while the outer membrane surface has a pI≈3.85 (Sherbet and Lakshmi, 1973). The nuclease domain of colicin E3 clearly exchanges Im3 for LPS after interaction with the Tol system.  Interactions between colicin E3 with LPS is an endothermic interaction. This was previously observed with both colicin N binding LPS (Johnson et al. 2014) and with the antimicrobial peptide polymyxin B interaction with LPS (Brandenburg et al. 2005). A hydrophobic interaction indicates possible hydrophobic interactions between colicin E3 and LPS. 92LPS at <30℃ enters a gel phase (Brandenburg and Seydel, 1990; Brandenburg et al. 2005). ITC experiments of LPS-E3 interactions conducted were thermostated to 25℃ leaving the LPS in a gel state. Endothermic binding is, therefore, due not only to electrostatic interactions but may also involve hydrophobic interactions. The cytotoxic domains of nuclease colicins were found to insert into membranes with colicin E3 inserting the deepest into anionic lipids (Vankemmelbeke et al. 2012). The endothermic interaction between the nuclease domain and anionic lipids is the result of unfolding of the nuclease domain (Mosbahi et al. 2006), therefore binding of LPS is most likely the result of the nuclease domain unfolding.  The interaction of colicin E3 with LPS may represent a part of the transport process of the nuclease through the outer membrane. Much research has been conducted to elucidate the mechanisms through which the group A colicins, including colicin E3, cross the outer membrane. The current model of transport consists of i) the high affinity binding of colicin to BtuB to adhere to the cell surface (Cavard, 1994; Housden et al. 2005), ii) the extension of the N-terminal unstructured domain through the trimeric protein OmpF to deliver a binding epitope to the periplasm to recruit TolB (Bouveret et al. 1997; Housden et al. 2010; Housden et al. 2013), and iii) TolB then interacts with TolA of the Tol complex to drive the release of the bound immunity protein (Lazzaroni et al. 2002; Vankemmelbeke et al. 2009). It is then theorized that the colicin E3 nuclease domain then travels through the OmpF channel (Zakharov et al. 2006) and is then proteolytically cleaved to pass through FtsH into the cytoplasm (Chauleau et al. 2011). Unfortunately, the details of how the nuclease actually passes through the OmpF trimer is unknown. 93 The nuclease of colicin E3 interacts with LPS and this interaction may serve several roles. Binding of colicin E3 to BtuB orientates the colicin at 45º to the plane of the outer membrane (Kurisu et al. 2003) and interactions of the colicin with OmpF positions both the N- and C-terminus above the OmpF trimer (Kurisu et al. 2003; Housden et al. 2013). Release of Im3 frees the nuclease domain and may allow it to interact with the OmpF by first binding to the LPS at the periphery of the OmpF channel where the core oligosaccharide is exposed (Johnson et al. 2014; Patel et al. 2016). This would provide the first step in a ‘brownian ratchet’ of low-to-high affinity interactions to drive the nuclease through the OmpF pore (Journet et al. 2001). The colicin E3 nuclease binds anionic lipids with a high affinity (~8nM) (Mosbahi et al. 2006), and the substantially weaker binding (~3µM) of the nuclease to LPS would provide a first step for a brownian ratchet. Indeed, unfolding of the nuclease domain occurs during its interaction with anionic lipids may be important in crossing the inner membrane (Mosbahi et al. 2006; Vankemmelbeke et al. 2012). Unfolding may occur earlier in the transport process upon interaction with LPS and would be required for the nuclease to pass through the narrow OmpF channel (Zakharov et al. 2006; Yamashita et al. 2008). An intriguing alternative interpretation was put forward by Jakes, 2014, in which colicin N, when bound to the periphery of OmpF (Baboolal et al. 2008) may be able to penetrate through the outer membrane in the interface between the outer membrane and the circumference of an OmpF trimer. The electrostatic and hydrophobic interaction of the E3 nuclease, involving membrane insertion (Vankemmelbeke et al. 2012), with LPS in the outer membrane suggests that 94the nuclease domain interacts with the core oligosaccharide, then inserts into the membrane. This may allow the nuclease domain to traverse the membrane by penetrating at the OmpF-membrane interface.95Chapter 4: Interactions of OmpC and HslT from Salmonella enterica Typhimurium4.1 Introduction Gram-negative pathogens, such as Salmonella enterica Typhimurium, must survive the assault of reactive oxygen species (ROS) generated by the immune response (Imlay, 2013). The outer membrane is permeable for ROS through the porins (Masi, 2013) and regulation of porin expression is critical to cell survival against the immune response (Lavigne et al. 2012). Surviving against antibiotic treatment and the immune system results from decreased outer membrane permeability, which is a major characteristic of persistors. Persistors are non-reproducing cells, able to resist antibiotics, that can reinitiate infection after completion of treatment (Kaiser et al. 2014). Salmonella Typhimurium in an HpxF background deleted for catalases and peroxidases has a reduced capacity for detoxification of oxidative stressors (Hébrard et al. 2009) and a redox sensitive GFP-based assay developed to measure oxidative stress in cells (van der Heijden et al. 2015; 2016a) allowed investigation of outer membrane permeability. This suggests that the response to oxidative stress occurs post-translationally with porin-protein interactions. A common trimeric Gram-negative porin, OmpC, is one potential route exploited by damaging oxidative species. OmpC interacts with the periplasmic heat shock protein HslT, termed IbpA in Escherichia coli, in an unknown capacity (Butland et al. 2005). It is possible that this interaction is as a “plug” or “cork” for HslT to block passage through the channel of OmpC after encountering oxidative stress.96 To test the interaction of HslT and OmpC from Salmonella enterica Typhimurium these proteins were purified and a variety of biochemical assays performed to determine what conditions may promote binding.4.2 Materials and Methods4.2.1 HslT Expression test HslT expressing E. coli BL21 cells provided by Dr. van der Heijden and were grown to OD600 ≅ 0.5 in 50mL of LB broth supplemented with 25µg/ml kanamycin and induced with 1mM IPTG. 1mL samples of culture were removed at 0h (before induction), 1h, 2h, and 3h after induction. Cells were pelleted then solubilized in 150µl of 2% SDS. Lysate was pelleted and 20µl of supernatant were analyzed on 15% SDS-PAGE. A band at approximately 15kDa was observed expressing at 1h to 3h. Western blot was performed on 15% SDS-PAGE gels using a Mouse anti-His monoclonal antibody (ABM) observed His-tagged HslT in the whole cell lysate, associated with the membrane, and dissociated into supernatant after treatment of the membrane with 6M urea. 4.2.2 HslT purification and refolding 2L of HslT expressing BL21 were grown to OD600 ≅ 0.5 and induced with 1mM IPTG. Cells were grown for three more hours to OD600 ≅ 1.2 then harvested by centrifugation at 6,000 rpm, 10 minutes, 4℃. Cells were resuspended in a buffer D (50mM Tris-HCl, pH = 7.9, 100mM 97NaCl) and lysed in a microfluidizer at 10,000 psi after supplementing the cell solution with 1mM PMSF. The lysate was cleared by centrifugation at 45,000 rpm, 45 minutes, 4℃ to collect membrane. The membrane pellet was resuspended in denaturing buffer (buffer D with 6M Urea)  to a final membrane concentration of 3mg/ml concentration. Membrane was incubated for 30 minutes at 37ºC in the 6M urea buffer to denature the HslT and dissociate it from the membrane. The solution was then centrifuged at 45,000 rpm, 45 minutes, 4℃ to remove membrane. Supernatant was then passed over a Co2+ agarose column equilibrated in denaturing buffer. Denatured protein attached to the column was refolded by slowly diluting the 6M urea to 0M urea with a 10CV gradient with Buffer D. HslT was then eluted with 50mM Tris-HCl, pH = 7.9, 300mM NaCl, 10% glycerol, 600mM imidazole. Eluted HslT was then dialyzed against buffer D to remove the imidazole. The yield of HslT was approximately 2mg per litre of culture. 4.2.3 OmpC Purification 2L of OmpC expressing Salmonella enterica (∆ompF ∆ompD) were grown by Dr. Lisa Reynolds to high density over two days at 37℃. The cells were harvested at 6,000 x g and lysed in a french press at 10,000 psi in 50mM Tris-HCl, pH = 7.9, 50 mM NaCl, 10% glycerol (Buffer A) supplemented with 1mM PMSF. Cell lysate was centrifuged at 6,000 rpm, 10 minutes, 4℃ to pellet unbroken cells and insoluble materials. The supernatant was then centrifuged at 45,000 rpm, 45 minutes, 4℃ to pellet membrane. Membrane pellet was diluted in Buffer B to a protein concentration of ~3mg/ml and solubilized in 1% Triton X-100 at room temperature to solubilize the inner membrane. Solubilized membrane was then centrifuged at 45,000 rpm, 45 minutes, 4℃ to pellet the outer membrane. Outer membrane pellet was then resuspended in buffer A to a 98protein concentration of ~3mg/ml and solubilized overnight in 1% LDAO at 4ºC with gentle rocking. LDAO solubilized membrane was then centrifuged at 45,000 rpm, 45 minutes, 4℃ and the supernatant was passed over a HiTrap Q FF anion exchange column (GE Healthcare) equilibrated in buffer A supplemented with 0.1% LDAO. The OmpC was eluted using a salt gradient up to 1M NaCl in buffer A. Eluted OmpC was compared to the structurally similar Escherichia coli OmpF. OmpC is stable and remains trimeric even in SDS-PAGE gels unless boiled at 95ºC (Arockiasamy and Krishnaswamy, 2000). Samples of each fraction of purified OmpC were either left untreated or boiled at 95ºC for 5 minutes and then loaded on 12% SDS-PAGE gel. The yield of OmpC protein was 3mg per litre of culture. 4.2.4 Pulldown of OmpC with HslT HslT, which is his-tagged, can be pulled down by immobilized metal affinity chromatography beads.  HslT, OmpC, and HslT and OmpC mixed together were either left untreated or were pre-treated with 500µM H2O2, heating to 50ºC, or both. The proteins were then incubated with Co2+ IMAC beads (GoldBio) for 15 minutes at room temperature to allow binding of protein to the beads. Beads were pelleted and the supernatant/flowthrough was removed. Protein was eluted from beads using 50mM Tris-HCl, pH = 7.9, 300mM NaCl, 10% glycerol, 600mM imidazole. Beads were pelleted and supernatant removed. Both flowthrough and elution were run on 15% SDS-PAGE. In a repeat experiment beads were washed with buffer A supplemented with 0.1% LDAO and 20mM Imidazole prior to elution. 994.2.5 Crosslinking of OmpC and HslT HslT was dialyzed and OmpC was desalted into phosphate buffered saline to remove Tris for crosslinking experiments. HslT alone, OmpC alone, and both proteins incubated together were pre-treated by 500µM H2O2, heating at 50ºC, or both treatments simultaneously. Protein samples were then incubated with 1% formaldehyde or 1% glutaraldehyde for 10 minutes at room temperature before loading onto both 12% and 15% SDS-PAGE gels.4.3 Results4.3.1 HslT Purification and Refolding Salmonella Typhimurium HslT expression in E. coli strain BL21 was analyzed over a three-hour timeframe after induction using IPTG. HslT expression was minimal at zero-hour prior to induction, but a clear band becomes evident one hour after induction (Figure 4-1). A control of colicin E3-Im3 was loaded in lane 1 to compare the ~60kDa colicin E3 protein to the expression of HslT (Figure 4-1, lane 1). This confirms that this Salmonella protein can be expressed in Escherichia coli. 100Figure 4-1: Total expression of HslT from E. coli BL21. Samples of cells induced with IPTG grown at 37℃ to the indicated time, harvested, and lysed in SDS. Total protein was assessed by migration on 15% SDS-PAGE and Coomassie blue staining.  HslT/IbpA is a membrane associated protein that may be purified from the membrane by denaturation with urea (Kitagawa et al. 2002; Kuczynska-Wisnik et al. 2002). A test purification of crude membrane was performed on each step of the purification process and analyzed on SDS-PAGE by Coomassie blue staining and western blotting. His-tagged HslT was found in the whole cell, the crude membrane fraction, and in both the urea-treated membrane pellet and supernatant fractions (Figure 4-2A and B). Ergo, Salmonella HslT can be liberated from the membrane by urea denaturation. 101Figure 4-2: HslT localization in E. coli cells. BL21 cells expressing HslT were lysed and separated by centrifugation into soluble and crude membrane fractions. Crude membranes were urea-treated and centrifuged to separate membrane from membrane-associated protein. Samples from each step of the procedure were analyzed on 15% SDS-PAGE by Coomassie blue staining (A) and by western blotting with an anti-his antibody (ABM) (B). Molecular weight markers are indicated. Membrane was treated with urea, centrifuged to pellet the membrane, and supernatant containing denatured HslT was bound to a Co2+ IMAC column. HslT was refolded by slowly diluting the urea with a gradient of urea-free buffer and then eluted with imidazole. Analysis on SDS-PAGE of the elution fractions as well as the start material and flowthrough showed that not all the HslT was bound to the beads (Figure 4-3). This may be because some of the HslT remains in aggregates that sequester the His-tag from binding the Co2+ beads (Kitagawa et al. 2002). 102Figure 4-3: HslT purification. 15% SDS-PAGE gel of HslT fractions from Co2+ agarose IMAC visualized with Coomassie blue staining. Start material and flowthrough of IMAC purification were loaded, respectively. Fractions 1 through 10 were loaded in the indicated lanes. HslT refolding was analyzed by loading onto a non-denaturing blue native (BN)-PAGE gel either untreated, heated at 50ºC for 10 minutes, treated with 2% SDS, 1% DDM, or 1% LDAO. This gel showed that untreated HslT migrated near the top of the gel, presumably as a megadalton complex or aggregate described by Kitagawa et al. 2002 for IbpA (Figure 4-4A, lane 1). When treated with SDS or heated to 50ºC for 10 minutes the HslT migrated lower into the gel as a smaller complex or possible monomer (Figure 4-4A, lane 2 and 3). Treatment with DDM or LDAO detergents did not yield the lower molecular weight band but did disrupt the dye front near the bottom of the gel (Figure 4-4A, lane 4 and 5). Bands were excised from the BN-PAGE gel and were loaded onto a 15% SDS-PAGE gel. Compared to an HslT control, these lower bands represent monomers of HslT generated by denaturation by SDS and by heat treatment at 50ºC. Treatment with DDM does not produce monomers of HslT. 103 Salmonella HslT can be purified from E. coli and refolded during the purification process. Figure 4-4: HslT Refolding. A) Hslt refolding was assessed by migration on 4-12% BN-PAGE. Samples were treated with 2% SDS, 1% DDM, 1% LDAO, or heat treated to 50ºC before loading on BN-PAGE. B) Indicated bands from BN-PAGE were excised and migrated on 15% SDS-PAGE and visualized with Coomassie blue staining. HslT from IMAC was run as a control in lane 1.4.3.2 OmpC Purification Previous purification of OmpC from Salmonella species has used SDS to liberate the trimer protein from the outer membrane (Arockiasamy and Krishnaswamy, 2000). Another technique using a combination of Triton-X-100 and LDAO was developed to solubilize the trimer in less denaturing conditions. Salmonella Typimurium ∆ompF ∆ompD was grown to high density and lysed in presence of 1mM PMSF. Salmonella membranes were solubilized with Triton-X-100 to remove the inner membrane and solubilized with LDAO to solubilize the outer membrane. 104 OmpC was purified from strong anionic exchange column and the fractions were loaded onto SDS-PAGE. Each fraction was either boiled or left untreated and compared to a purified OmpF sample. OmpC, like OmpF, migrates as a trimeric complex near the top of the gel and boiling of the trimer in SDS at 95℃ resulted in OmpC monomers (Figure 4-5) (Arockiasamy and Krishnaswamy, 2000). Figure 4-5: OmpC purification from Salmonella enterica. OmpC was solubilized from outer membranes and purified using Q sepharose ion exchange. Fractions were visualized on 12% SDS-PAGE with Coomassie blue staining. Fractions were untreated or boiled at 98ºC for five minutes prior to loading on SDS-PAGE. Purified OmpF was loaded in the first lane as a control.4.3.3 HslT-OmpC interaction test on BN-PAGE  Samples of HslT and OmpC were treated both individually and together with either 500µM H2O2, heating to 50ºC, or both and then loaded onto BN-PAGE gel (Figure 4-6). Results show that HslT, again, migrates into the non-denaturing gel only after heating to 50ºC while 105OmpC results in a large smear that extends most of the length of the gel regardless of treatment. No shift in bands indicative of protein-protein interactions were observed under any conditions. Bands from this BN-PAGE gel were excised and run on 15% SDS-PAGE compared to matched controls of HslT and OmpC (Figure 4-7). OmpC excised from BN-PAGE and run on SDS-PAGE was denatured into monomers. The excision of bands near the top of the gel showed HslT and OmpC together (Figure 4-8, lane 9); however, the spread of OmpC throughout the gel would also explain both proteins migrating together. Therefore, no interaction between HslT and OmpC was detectable on blue native PAGE. 50ºCH2O2HslT OmpC HslTOmpC- - + + - - + + - - + +- + - + - + - + - + - +Figure 4-6: HslT-OmpC binding on non-denaturing gel. ~3µM HslT and ~5µM OmpC, individually or together, were treated with 500µM H2O2, heated to 50℃, or both. Samples were then migrated on a 4-12% BN-PAGE.106Figure 4-7: Excision of protein complexes from BN-PAGE. Bands from BN-PAGE (Figure 4-6) were excised and analyzed on 12% SDS-PAGE and visualized with Coomassie blue stain. HslT (lane 1), OmpC (lane 2), and boiled OmpC (lane 3) were run as controls.4.3.4 HslT-OmpC Pulldown A co-sedimentation assay was designed in which His-tagged HslT is bound to IMAC beads in the presence or absence of OmpC. Bead-protein complexes were pelleted and supernatant was removed. Protein was separated from beads by buffer containing high concentration of imidazole. A wash step with low concentration of imidazole to remove any nonspecifically bound protein from the beads. Flowthrough, wash, and elution were analyzed by SDS-PAGE (Figure 4-8). OmpC alone was incubated with IMAC beads under the indicated conditions as a control.  107 A detectable amount of HslT on SDS-PAGE does not bind to the beads efficiently and comes out with the flowthrough. However, after heating at 50ºC less HslT is detected in the flowthrough. This may be explained by HslT disassembling from a megadalton complex to become a monomer with a more exposed his-tag allowing more efficient capture on IMAC beads. OmpC, lacking a his-tag also comes out primarily in the flowthrough fractions whether in the presence or absence of HslT and regardless of pre-treatment with heat or oxidation. OmpC primarily came with the flowthrough, the remainder came with the wash buffer, and none could be observed in the elution. In this experiment the HslT could be observed in flowthrough, wash, and elution fractions. The absence of OmpC in elution fractions with HslT does not support the formation of a complex between these proteins (Figure 4-8).Figure 4-8: HslT-OmpC pulldown repeat. HslT and OmpC, individually and together, were incubated with Co2+ agarose beads (GoldBio) after treatment with 500µM H2O2. Beads were then washed and protein eluted with 600mM imidazole. Flowthrough, wash, and elution fractions are loaded on 15% SDS-PAGE and visualized with Coomassie blue staining.1084.3.5 HslT-OmpC chemical crosslinking Chemical crosslinking is a technique used to stabilize protein complexes (Fraenkel-Conrat and Olcott, 1948; Migneault et al. 2004). Crosslinking is also useful for the stabilization of transient protein complexes (Melcher, 2004). If the HslT and OmpC protein interaction is transient then chemical crosslinking may capture a complex of them both.  HslT and OmpC individually and together were pre-treated with 500µM H2O2 and/or heat (50℃), then incubated with formaldehyde or glutaraldehyde crosslinkers.  Samples were analyzed on 12% and 15% SDS-PAGE gels and do not show the formation of a complex between OmpC and HslT. No shift in migration was observed by either protein under any condition (Figure 4-9). In the event of formation of a complex between trimers of OmpC and HslT the stacking portion of the gel was not removed and remains at the top of the gel. The results from this experiment also do not show an interaction between OmpC and HslT. 109Figure 4-9: Chemical crosslinking of OmpC and HslT. OmpC and HslT incubated with 500µM H2O2 or heated at 50℃ were treated with either 1% formaldehyde or 1% glutaraldehyde to promote protein crosslinking. Complex formation between proteins was assessed by migration on either 12% or 15% SDS-PAGE visualized with Coomassie blue staining. OmpC and HslT were loaded in the first and second lanes as controls.4.3.6 HslT-OmpC interaction on SDS-PAGE The only potential interaction between HslT and OmpC is on SDS-PAGE in which the migration of HslT consistently shows retarded migration when combined with OmpC (Figure 4-8 and 4-9). Regardless of the conditions (+500µM H2O2, +1mM DTT) there is a slight shift of the HslT in response to incubation with OmpC (Figure 4-10A). To confirm that HslT is retarded in 110migration into the gel in response to OmpC a western blot was performed the shifted band was determined to be His-tagged HslT (Figure 4-10B). Figure 4-10: HslT interaction with OmpC on SDS-PAGE. HslT and OmpC, either individually or together, were treated with 500µM H2O2 or 1mM DTT. Protein samples were then analyzed by 15% SDS-PAGE. Gels were visualized with Coomassie blue staining (A) or by western blotting with an anti-His antibody (ABM) (B). To test for other factors that may affect HslT migration on SDS-PAGE, HslT was incubated with buffer containing 0.1% LDAO, OmpC, or BtuB. Analysis on SDS-PAGE shows retardation of migration of HslT in the presence of OmpC, but not in the presence of LDAO buffer or with BtuB (Figure 4-11). The shift in migration due to the addition of OmpC is uncharacterized and unknown. 111Figure 4-11: HslT shift with OmpC in SDS. HslT samples were incubated with buffer containing 0.1% LDAO, ~5µM OmpC, or ~5µM BtuB then analyzed using 15% SDS-PAGE and visualized with Coomassie blue. HslT was loaded in the first lane as a control. 4.4 Discussion HslT interactions with OmpC were predicted from multiple avenues. The HslT homologue from Escherichia coli, IbpA, was found to interact with OmpC (Butland et al. 2005). The rapid response of cells to oxidative stress suggests post-translational modifications are made to OmpC. Salmonella deletion strains of hslT, ompC, and hslTompC were tested for membrane permeability. Only the hslT deletion strain had increased permeability under oxidative stress and had a concurrent decrease in cell survivability. A molecular docking simulation proposed steric blocking of the OmpC pore by binding the periplasmic loop4 of the porin (van der Heijden et al. 2016b).112 No direct interaction between HslT and OmpC was detectable. This suggests a number of alternative interpretations: i) HslT affects unknown binding partners within the periplasm that, in turn, bind OmpC. The HslT homologue, IbpA, in concert with other proteins protects protein structure from both heat shock and from oxidizing conditions (Kitagawa et al. 2000; 2002; Matuszewska et al. 2008). An unidentified protein that acts to block OmpC may be activated in response to HslT binding after exposure to oxidizing conditions. Knockouts of HslT would prevent the activation of this protein and leave the OmpC channel open during encounters with reactive oxygen species. ii) HslT may function rapidly after exposure to oxidative stress to inhibit damage caused by oxidative stress, in an HpxF background. Oxidative stress does not dissociate the megadalton complex, unlike heating to 50℃ (Kitagawa et al. 2002), results that were reproduced here (Figure 4-4A and 4-6). The megadalton complex would, therefore, be able to bind proteins and enzymes affected by oxidative stress and prevent their denaturation. Unfortunately, this interpretation disagrees with the data that the diffusion of H2O2 across the membrane undergoes a “switching point” that would coincides a blocking of the outer membrane. HslT may affect the reaction to oxidative species, but not the diffusion rate. iii) HslT may interface with the membrane directly and affect membrane permeability. HslT, like the E. coli homologue IbpA, is membrane associated and may serve a function in preventing lipid bilayers from entering a non-bilayer state (Nakamoto and Vigh, 2007). Deletion of ompC affects membrane permeability directly by deletion of the porin. Knockouts of ibpA/ibpB in E. coli resulted in increased membrane permeability under heat stress conditions (Nakamoto and Vigh, 2007), therefore, deletion of hslT may result in increased membrane permeability. This does not explain the survival of ompChslT double knockouts after exposure to oxidizing conditions. 113Knockout of hslT may increase membrane permeability, while deletion of ompC conversely reduces permeability of the membrane causing this knockout strain to respond similarly to wild-type after the “switching point”. Stress response in pathogenic persisters against oxidative assault is a crucial area of research. Further investigations must be undertaken to ascertain how persisters survive stress and to develop therapeutics to eliminate them.114Chapter 5: Summary and future directions The outer membrane of Gram-negative bacteria forms the first layer of defense for bacteria against the extracellular environment. It also poses a difficulty for the cell in transporting the essential nutrients required for life. Antimicrobial compounds such as bacteriocins must subvert transport through the outer membrane to access their specific cellular targets. The immune system must also circumvent the outer membrane of infecting pathogens in order to kill infecting microbes. The broad scope of this thesis can be subdivided into three categories: i) investigating active transport through the TonB-dependent pathway using the outer membrane vitamin B12 transporter BtuB as a model, ii) to describe the molecular details of how a BtuB-binding bacteriocin, colicin E3, is transported across the outer membrane, and iii) whether the outer membrane becomes impermeable to chemical assault by interaction of stress proteins with the porins of the outer membrane.  Many details of the TonB-dependent transport process remain to be elucidated. Previous studies have shown that the binding of TonB to transporters alters the conformation of their extracellular and apical loops (Shultis et al. 2006; James et al. 2009). Previous studies have also employed detergents to investigate the interactions of TonB-dependent transporters with TonB (Khursigara et al. 2004; Khursigara et al. 2005; Freed et al. 2013) which may affect these binding interactions (Choul-Li et al. 2008; Mills et al. 2014). BtuB was reconstituted into nanodiscs to omit the problems associated with detergents, yielding a monodisperse population of BtuB within the disc. TonB interaction with BtuB is regulated by the vitamin B12 ligand as 115demonstrated by native gel. TonB binds BtuB in a ligand-dependent manner and in a 1:1 molar ratio as determined through both native gel and light scattering. This is in agreement with prior results with other TonB-dependent transporters (Koedding et al. 2004; Freed et al. 2013). A combination of non-denaturing gel autoradiography and scintillation proximity assay revealed that this binding of TonB was found to cause conformational changes in BtuB that retain the bound ligand. The retaining of ligand within BtuB likely allows the TonB-ExbB-ExbD complex to assemble and become engaged in transport. Alternatively, retention of ligand may prevent release of bound ligand back into the extracellular space when the plug domain is displaced during transport. Further studies are required to determine precisely what role the retention of ligand serves. Deletion of the TonB binding motif (the TonB box) and disruption of an apical loop of residues 85 to 95 both abrogated this retention effect of TonB. This shows that TonB must bind through the TonB box of BtuB to change the conformation of the apical loop to retain bound ligand.  The movement of the plug domain is necessary for transport (Gumbart et al. 2007; Ma et al. 2007). To investigate plug domain movement the approach of Udho et al. 2012 was employed. Denaturation of BtuB using urea caused the dissociation of bound ligand but did not destabilize the disc, indicating that the plug domain is dissociated but the β-barrel remains intact. The nanodisc is, therefore, an excellent platform for the future investigation of plug domain movement for transport of bound vitamin B12. 116 In chapter 3 the binding of BtuB-dependent colicin E3 to OmpF relies partially on the denaturation of the colicin. Colicin E3 interacts with BtuB in the absence of detergents, but denaturation with the anionic detergent SDS appeared to be required for the interaction of colicin E3 to OmpF. Indeed, denaturation of colicin E3 in SDS allowed the assembly of a complex of colicin E3-OmpF-TolB. However, colicin E3 disulphide crosslinked with TolB continued to bind OmpF, even though this would negate the passage of the N-terminal domain through the channel (Housden et al. 2010, 2013). This suggests that colicin E3 binds OmpF either through a separate domain from the T-domain or binds at the periphery of the OmpF trimer.  The interaction of colicin E3 with its partners in the presence of SDS suggests that colicin E3 interacts directly with LPS from the outer membrane because of the similarity of SDS to LPS as an anionic amphipathic molecule. Colicin binds LPS with low affinity (~3µM) and binds through the C-terminal nuclease domain after the dissociation of the high-affinity immunity protein. Release of the immunity protein by the Tol complex (Vankemmelbeke et al. 2009) allows the nuclease domain to interact with LPS. This binding of E3 nuclease to LPS likely elicits denaturation (Mosbahi et al. 2006), which would allow the nuclease to pass into the channel of OmpF. Interestingly, an alternative proposed for colicin N suggests that interaction with LPS may be due to penetration of the colicin through the interface between OmpF and the outer membrane (Jakes, 2014). Future investigation of the structure of the Im3-free colicin E3 with OmpF must determine whether the nuclease domain binds OmpF through the channel or at the periphery of the trimer. 117 In addition to the investigation of BtuB-dependent transport of B12 and colicin E3 the permeability of the outer membrane was also investigated. OmpC was anticipated to close in response to oxidative stress by interaction with the heat shock protein HslT (van der Heijden et al. 2016b). There was no evidence of a direct interaction between HslT and OmpC from native gels, pulldown assay, and chemical crosslinking. The only evidence of interaction appeared on SDS-PAGE with a slight shift in migration of HslT when in the presence of OmpC. Incubation of HslT with buffers containing LDAO or with BtuB did not result in a shift in migration. This slight shift may be due to post-translational modifications of HslT, further work on the interaction between HslT and OmpC must investigate possible conformational changes and other post-translational modifications.  The absence of a direct binding between HslT and OmpC suggests that other interactions occur during exposure to oxidizing stress. The HslT E. coli homologue IbpA responds to superoxide exposure as well as heat stress (Kitagawa et al. 2000). HslT may respond to the oxidative stress such that rendering the outer membrane impermeable is unnecessary. Alternatively, IbpA (and IbpB) is a heat shock protein that interacts with many proteins to prevent their inactivation. HslT may interact with an as yet unidentified protein partner during oxidative stress. This protein partner may then respond as a “plug” to block passage through the channel of OmpC. Additional work in this area will identify any possible partner that acts as an OmpC plug and it will also ascertain whether HslT acts to allow cells to rapidly respond to superoxide stress. 118 In summary, I have identified previously unknown steps in the transport of two ligands of the outer membrane transporter BtuB, vitamin B12 and colicin E3. Within a number of years the movement of the BtuB plug domain will be explained experimentally. This will distinguish whether the plug domain is partially unfolded (Gumbart et al. 2007) or moved out of the β-barrel  (Ma et al. 2007). Atomic force microscopy to induce the unfolding of the plug domain in a complex of BtuB-TonB is a likely candidate technique to elucidate plug domain movement (Chen et al. 2015). Reconstitution of TonB-dependent transporters into nanodiscs will greatly expedite research into plug movement. Also, unraveling additional details of how bacteriocins are transported through bacterial membranes will assist in the development of possible new therapeutics. Bacteriocins are currently being investigated as new antimicrobial agents (Cavera et al. 2015). Understanding how these bacteriocins transit the membranes of bacteria will allow us to create entirely novel polypeptides that subvert bacterial defenses (Lukacik et al. 2012). Altogether, this work opens up avenues and prospects for future investigations.119ReferencesAlami, M., Dalal, K., Lelj-Garolla, B., Sligar, S. G., & Duong, F. (2007). Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA. The EMBO journal, 26(8), 1995-2004.Andersen, K. K., Oliveira, C. L., Larsen, K. L., Poulsen, F. M., Callisen, T. H., Westh, P., ... & Otzen, D. (2009). The role of decorated SDS micelles in sub-CMC protein denaturation and association. Journal of molecular biology,391(1), 207-226.Arockiasamy, A., & Krishnaswamy, S. (2000). Purification of integral outer-membrane protein OmpC, a surface antigen from Salmonella typhi for structure–function studies: a method applicable to enterobacterial major outer-membrane protein. Analytical biochemistry, 283(1), 64-70.Auld, D. S., Farmen, M. W., Kahl, S. D., Kriauciunas, A., McKnight, K. L., Montrose, C., & Weidner, J. R. (2004). Receptor binding assays for HTS and drug discovery.Baboolal, T. G., Conroy, M. J., Gill, K., Ridley, H., Visudtiphole, V., Bullough, P. A., & Lakey, J. H. (2008). Colicin N binds to the periphery of its receptor and translocator, outer membrane protein F. Structure, 16(3), 371-379.Bao, H., & Duong, F. (2012). Discovery of an auto-regulation mechanism for the maltose ABC transporter MalFGK 2. PLoS One, 7(4), e34836.Baty, D., Knibiehler, M., Verheij, H., Pattus, F., Shire, D., Bernadac, A., & Lazdunski, C. (1987). Site-directed mutagenesis of the COOH-terminal region of colicin A: effect on secretion and voltage-dependent channel activity. Proceedings of the National Academy of Sciences, 84(5), 1152-1156.Bayburt, T. H., Grinkova, Y. V., & Sligar, S. G. (2002). Self-assembly of discoidal phospholipid bilayer nanoparticles with membrane scaffold proteins.Nano Letters, 2(8), 853-856.Bayburt, T. H., & Sligar, S. G. (2003). Self-assembly of single integral membrane proteins into soluble nanoscale phospholipid bilayers. Protein science, 12(11), 2476-2481.120Bayburt, T. H., Grinkova, Y. V., & Sligar, S. G. (2006). Assembly of single bacteriorhodopsin trimers in bilayer nanodiscs. Archives of biochemistry and biophysics, 450(2), 215-222.Bayburt, T. H., & Sligar, S. G. (2010). Membrane protein assembly into Nanodiscs. FEBS letters, 584(9), 1721-1727.Bénédetti, H., R. Lloubès, C. Lazdunski, and L. Letellier. (1992). Colicin A unfolds during its translocation in Escherichia coli cells and spans the whole cell envelope when its pore has formed. EMBO J. 11,441–447. Bonsor, D. A., Hecht, O., Vankemmelbeke, M., Sharma, A., Krachler, A. M., Housden, N. G., ... & Kleanthous, C. (2009). Allosteric β-propeller signalling in TolB and its manipulation by translocating colicins. The EMBO journal,28(18), 2846-2857.Bouveret, E., Rigal, A., Lazdunski, C., & Bénédetti, H. (1997). The N-terminal domain of colicin E3 interacts with TolB which is involved in the colicin translocation step. Molecular microbiology, 23(5), 909-920.Bowman, C. M., Dahlberg, J. E., Ikemura, T., Konisky, J., & Nomura, M. (1971). Specific inactivation of 16S ribosomal RNA induced by colicin E3 in vivo. Proceedings of the National Academy of Sciences, 68(5), 964-968.Bradbeer, C., Reynolds, P. R., Bauler, G. M., & Fernandez, M. T. (1986). A requirement for calcium in the transport of cobalamin across the outer membrane of Escherichia coli. Journal of Biological Chemistry, 261(6), 2520-2523.Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding.Analytical biochemistry, 72(1-2), 248-254.Brandenburg, K., & Seydel, U. (1990). Investigation into the fluidity of lipopolysaccharide and free lipid A membrane systems by Fourier-transform infrared spectroscopy and differential scanning calorimetry. European Journal of Biochemistry, 191(1), 229-236.Brandenburg, K., David, A., Howe, J., Koch, M. H., Andrä, J., & Garidel, P. (2005). Temperature dependence of the binding of endotoxins to the polycationic peptides polymyxin B and its nonapeptide. Biophysical journal,88(3), 1845-1858.Braun, V., Pilsl, H., & Groß, P. (1994). Colicins: structures, modes of action, transfer through membranes, and evolution. Archives of Microbiology, 161(3), 199-206.121Braun, V. (1995). Energy-coupled transport and signal transduction through the gram-negative outer membrane via TonB-ExbB-ExbD-dependent receptor proteins. FEMS microbiology reviews, 16(4), 295-307.Braun, V. (2009). FhuA (TonA), the career of a protein. Journal of bacteriology, 191(11), 3431-3436.Brillet, K., Journet, L., Célia, H., Paulus, L., Stahl, A., Pattus, F., & Cobessi, D. (2007). A β strand lock exchange for signal transduction in TonB-dependent transducers on the basis of a common structural motif. Structure,15(11), 1383-1391.Butland, G., Peregrín-Alvarez, J. M., Li, J., Yang, W., Yang, X., Canadien, V., ... & Davey, M. (2005). Interaction network containing conserved and essential protein complexes in Escherichia coli. Nature, 433(7025), 531-537.Cadieux, N., Bradbeer, C., & Kadner, R. J. (2000). Sequence changes in the ton box region of BtuB affect its transport activities and interaction with TonB protein. Journal of bacteriology, 182(21), 5954-5961.Cadieux, N., Bradbeer, C., Reeger-Schneider, E., Köster, W., Mohanty, A. K., Wiener, M. C., & Kadner, R. J. (2002). Identification of the periplasmic cobalamin-binding protein BtuF of Escherichia coli. Journal of bacteriology,184(3), 706-717.Cadieux, N., Phan, P. G., Cafiso, D. S., & Kadner, R. J. (2003). Differential substrate-induced signaling through the TonB-dependent transporter BtuB.Proceedings of the National Academy of Sciences, 100(19), 10688-10693.Calmettes, C., Ing, C., Buckwalter, C. M., El Bakkouri, M., Lai, C. C. L., Pogoutse, A., ... & Moraes, T. F. (2015). The molecular mechanism of Zinc acquisition by the neisserial outer-membrane transporter ZnuD. Nature communications, 6.Cascales, E., Lloubes, R., & Sturgis, J. N. (2001). The TolQ–TolR proteins energize TolA and share homologies with the flagellar motor proteins MotA–MotB. Molecular microbiology, 42(3), 795-807.Cascales, E., Buchanan, S. K., Duché, D., Kleanthous, C., Lloubes, R., Postle, K., ... & Cavard, D. (2007). Colicin biology. Microbiology and Molecular Biology Reviews, 71(1), 158-229.122Cavard, D. (1994). Rescue by vitamin B12 of Escherichia coli cells treated with colicins A and E allows measurement of the kinetics of colicin binding on BtuB. FEMS microbiology letters, 116(1), 37-42.Cavera, V. L., Arthur, T. D., Kashtanov, D., & Chikindas, M. L. (2015). Bacteriocins and their position in the next wave of conventional antibiotics.International journal of antimicrobial agents, 46(5), 494-501.Chang, C., Mooser, A., Plückthun, A., & Wlodawer, A. (2001). Crystal structure of the dimeric C-terminal domain of TonB reveals a novel fold.Journal of Biological Chemistry, 276(29), 27535-27540.Chart, H., Smith, H. R., La Ragione, R. M., & Woodward, M. J. (2000). An investigation into the pathogenic properties of Escherichia coli strains BLR, BL21, DH5α and EQ1. Journal of applied microbiology, 89(6), 1048-1058.Chauleau, M., Mora, L., Serba, J., & de Zamaroczy, M. (2011). FtsH-dependent processing of RNase colicins D and E3 means that only the cytotoxic domains are imported into the cytoplasm. Journal of Biological Chemistry, 286(33), 29397-29407.Chen, Y., Radford, S. E., & Brockwell, D. J. (2015). Force-induced remodelling of proteins and their complexes. Current opinion in structural biology, 30, 89-99.Chimento, D. P., Mohanty, A. K., Kadner, R. J., & Wiener, M. C. (2003a). Crystallization and initial X-ray diffraction of BtuB, the integral membrane cobalamin transporter of Escherichia coli. Acta Crystallographica Section D: Biological Crystallography, 59(3), 509-511.Chimento, D. P., Mohanty, A. K., Kadner, R. J., & Wiener, M. C. (2003b). Substrate-induced transmembrane signaling in the cobalamin transporter BtuB. Nature Structural & Molecular Biology, 10(5), 394-401.Choul-Li, S., Adams, H., Pattus, F., & Celia, H. (2008). Visualization of interactions between siderophore transporters and the energizing protein TonB by native PAGE. Electrophoresis, 29(6), 1333-1338.Cierpicki, T., Liang, B., Tamm, L. K., & Bushweller, J. H. (2006). Increasing the accuracy of solution NMR structures of membrane proteins by application of residual dipolar couplings. High-resolution structure of outer membrane protein A. Journal of the American Chemical Society, 128(21), 6947-6951.123Clavel, T., Germon, P., Vianney, A., Portalier, R., & Lazzaroni, J. C. (1998). TolB protein of Escherichia coli K-12 interacts with the outer membrane peptidoglycan-associated proteins Pal, Lpp and OmpA. Molecular microbiology, 29(1), 359-367.Clément, E., Mesini, P. J., Pattus, F., & Schalk, I. J. (2004). The binding mechanism of pyoverdin with the outer membrane receptor FpvA in Pseudomonas aeruginosa is dependent on its iron-loaded status. Biochemistry,43(24), 7954-7965.Clifton, L. A., Johnson, C. L., Solovyova, A. S., Callow, P., Weiss, K. L., Ridley, H., ... & Lakey, J. H. (2012). Low resolution structure and dynamics of a colicin-receptor complex determined by neutron scattering. Journal of Biological Chemistry, 287(1), 337-346.Cobessi, D., Meksem, A., & Brillet, K. (2010). Structure of the heme/hemoglobin outer membrane receptor ShuA from Shigella dysenteriae: heme binding by an induced fit mechanism. Proteins: Structure, Function, and Bioinformatics, 78(2), 286-294.Dalal, K., & Duong, F. (2010). Reconstitution of the SecY translocon in nanodiscs. Protein Secretion: Methods and Protocols, 145-156.Delcour, A. H. (2009). Outer membrane permeability and antibiotic resistance. Biochimica et Biophysica Acta (BBA)-Proteins and Proteomics,1794(5), 808-816.De Mot, R., & Vanderleyden, J. (1994). The C-terminal sequence conservation between OmpA-related outer membrane proteins and MotB suggests a common function in both gram-positive and gram-negative bacteria, possibly in the interaction of these domains with peptidoglycan.Molecular microbiology, 12(2), 333.Denisov, I. G., & Sligar, S. G. (2011). Cytochromes P450 in nanodiscs.Biochimica et Biophysica Acta (BBA)-Proteins and Proteomics, 1814(1), 223-229.Di Masi, D. R., White, J. C., Schnaitman, C. A., & Bradbeer, C. (1973). Transport of vitamin B12 in Escherichia coli: common receptor sites for vitamin B12 and the E colicins on the outer membrane of the cell envelope.Journal of bacteriology, 115(2), 506-513.Domingo Köhler, S., Weber, A., Howard, S. P., Welte, W., & Drescher, M. (2010). The proline-rich domain of TonB possesses an extended polyproline II-like conformation of sufficient length to span the periplasm of Gram-negative bacteria. Protein Science, 19(4), 625-630.124Dong, C., Beis, K., Nesper, J., Brunkan-LaMontagne, A. L., Clarke, B. R., Whitfield, C., & Naismith, J. H. (2006). Wza the translocon for E. coli capsular polysaccharides defines a new class of membrane protein. Nature,444(7116), 226-229.Dover, L. G., Evans, L. J., Fridd, S. L., Bainbridge, G., Raggett, E. M., & Lakey, J. H. (2000). Colicin pore-forming domains bind to Escherichia coli trimeric porins. Biochemistry, 39(29), 8632-8637.Drennan, C. L., Matthews, R. G., & Ludwig, M. L. (1994). Cobalamin-dependent methionine synthase: the structure of a methylcobalamin-binding fragment and implications for other B 12-dependent enzymes. Current opinion in structural biology, 4(6), 919-929.Duché, D. (2007). Colicin E2 is still in contact with its receptor and import machinery when its nuclease domain enters the cytoplasm. Journal of bacteriology, 189(11), 4217-4222.Efremov, R. G., & Sazanov, L. A. (2012). Structure of Escherichia coli OmpF porin from lipidic mesophase. Journal of structural biology, 178(3), 311-318.Eisenhauer, H. A., Shames, S., Pawelek, P. D., & Coulton, J. W. (2005). Siderophore transport through Escherichia coli outer membrane receptor FhuA with disulfide-tethered cork and barrel domains. Journal of Biological Chemistry, 280(34), 30574-30580.Evans, J. S., Levine, B. A., Trayer, I. P., Dorman, C. J., & Higgins, C. F. (1986). Sequence-imposed structural constraints in the TonB protein of E. coli. FEBS letters, 208(2), 211-216.Evans, L. J. A., Cooper, A., & Lakey, J. H. (1996). Direct measurement of the association of a protein with a family of membrane receptors. Journal of molecular biology, 255(4), 559-563.Fanucci, G. E., Cadieux, N., Piedmont, C. A., Kadner, R. J., & Cafiso, D. S. (2002). Structure and dynamics of the β-barrel of the membrane transporter BtuB by site-directed spin labeling. Biochemistry, 41(39), 11543-11551.Fanucci, G. E., Lee, J. Y., & Cafiso, D. S. (2003a). Membrane mimetic environments alter the conformation of the outer membrane protein BtuB.Journal of the American Chemical Society, 125(46), 13932-13933.Fanucci, G. E., Cadieux, N., Kadner, R. J., & Cafiso, D. S. (2003b). Competing ligands stabilize alternate conformations of the energy coupling motif of a TonB-dependent outer membrane transporter. Proceedings of the National Academy of Sciences, 100(20), 11382-11387.125Farrance, O. E., Hann, E., Kaminska, R., Housden, N. G., Derrington, S. R., Kleanthous, C., ... & Brockwell, D. J. (2013). A force-activated trip switch triggers rapid dissociation of a colicin from its immunity protein. PLoS Biol,11(2), e1001489.Ferguson, A. D., Hofmann, E., Coulton, J. W., Diederichs, K., & Welte, W. (1998). Siderophore-mediated iron transport: crystal structure of FhuA with bound lipopolysaccharide. Science, 282(5397), 2215-2220.Ferguson, A. D., Welte, W., Hofmann, E., Lindner, B., Holst, O., Coulton, J. W., & Diederichs, K. (2000). A conserved structural motif for lipopolysaccharide recognition by procaryotic and eucaryotic proteins.Structure, 8(6), 585-592.Ferguson, A. D., Chakraborty, R., Smith, B. S., Esser, L., van der Helm, D., & Deisenhofer, J. (2002). Structural basis of gating by the outer membrane transporter FecA. Science, 295(5560), 1715-1719.Ferguson, A. D., & Deisenhofer, J. (2004). Metal import through microbial membranes. Cell, 116(1), 15-24.Fraenkel-Conrat, H., & Olcott, H. S. (1948). Reaction of formaldehyde with proteins VI. cross-linking of amino groups with phenol, imidazole, or indole groups. Journal of Biological Chemistry, 174(3), 827-843.Freed, D. M., Horanyi, P. S., Wiener, M. C., & Cafiso, D. S. (2010). Conformational exchange in a membrane transport protein is altered in protein crystals. Biophysical journal, 99(5), 1604-1610.Freed, D. M., Lukasik, S. M., Sikora, A., Mokdad, A., & Cafiso, D. S. (2013). Monomeric TonB and the Ton box are required for the formation of a high-affinity transporter–TonB complex. Biochemistry, 52(15), 2638-2648.Galanos, C., Lüderitz, O., Rietschel, E. T., Westphal, O., Brade, H., Brade, L., ... & Kusumoto, S. (1985). Synthetic and natural Escherichia coli free lipid A express identical endotoxic activities. European journal of biochemistry, 148(1), 1-5.Garcia-Herrero, A., Peacock, R. S., Howard, S. P., & Vogel, H. J. (2007). The solution structure of the periplasmic domain of the TonB system ExbD protein reveals an unexpected structural homology with siderophore-binding proteins. Molecular microbiology, 66(4), 872-889.126Garidel, P., Howe, J., & Brandenburg, K. (2008). Thermodynamic analysis of the interaction of lipopolysaccharides with cationic compounds. Engineering in Life Sciences, 8(5), 523-529.Goluch, E. D., Shaw, A. W., Sligar, S. G., & Liu, C. (2008). Microfluidic patterning of nanodisc lipid bilayers and multiplexed analysis of protein interaction. Lab on a Chip, 8(10), 1723-1728.Grandy, D. K., Langen, H., & Monsma Jr, F. J. (1995). Orphanin FQ: a neuropeptide that activates an opioidlike G protein—coupled receptor. Rn,40, 60.Gresock, M. G., Kastead, K. A., & Postle, K. (2015). From Homodimer to Heterodimer and Back: Elucidating the TonB Energy Transduction Cycle.Journal of bacteriology, 197(21), 3433-3445.Gudmundsdottir, A., Bell, P. E., Lundrigan, M. D., Bradbeer, C., & Kadner, R. J. (1989). Point mutations in a conserved region (TonB box) of Escherichia coli outer membrane protein BtuB affect vitamin B12 transport. Journal of bacteriology, 171(12), 6526-6533.Gumbart, J., Wiener, M. C., & Tajkhorshid, E. (2007). Mechanics of force propagation in TonB-dependent outer membrane transport. Biophysical journal, 93(2), 496-504.Harder, D., & Fotiadis, D. (2012). Measuring substrate binding and affinity of purified membrane transport proteins using the scintillation proximity assay.Nature protocols, 7(9), 1569-1578.Harkness, R. E., & Braun, V. (1989). Colicin M inhibits peptidoglycan biosynthesis by interfering with lipid carrier recycling. Journal of Biological Chemistry, 264(11), 6177-6182.Hébrard, M., Viala, J. P., Méresse, S., Barras, F., & Aussel, L. (2009). Redundant hydrogen peroxide scavengers contribute to Salmonella virulence and oxidative stress resistance. Journal of bacteriology, 191(14), 4605-4614.Helaine, S., Cheverton, A. M., Watson, K. G., Faure, L. M., Matthews, S. A., & Holden, D. W. (2014). Internalization of Salmonella by macrophages induces formation of nonreplicating persisters. Science, 343(6167), 204-208.Henriksen, J. R., & Andresen, T. L. (2011). Thermodynamic profiling of peptide membrane interactions by isothermal titration calorimetry: a search for pores and micelles. Biophysical journal, 101(1), 100-109.127Hickman, S., Gantner, M., Housden, N., Paci, E., Warriner, S., & Brockwell, D. (2014) Pulling the plug on the TonB-dependent transport mechanism. Poster presented at Single biomolecules - in silico, in vitro and in vivo at the University of Hertfordshire, UK. Abstract received from http://www.biochemistry.org/Portals/0/Conferences/abstracts/SA157/SA157P026.pdfHiggs, P. I., Larsen, R. A., & Postle, K. (2002). Quantification of known components of the Escherichia coli TonB energy transduction system: TonB, ExbB, ExbD and FepA. Molecular microbiology, 44(1), 271-281.Housden, N. G., Loftus, S. R., Moore, G. R., James, R., & Kleanthous, C. (2005). Cell entry mechanism of enzymatic bacterial colicins: porin recruitment and the thermodynamics of receptor binding. Proceedings of the national academy of sciences of the united states of america, 102(39), 13849-13854.Housden, N. G., Wojdyla, J. A., Korczynska, J., Grishkovskaya, I., Kirkpatrick, N., Brzozowski, A. M., & Kleanthous, C. (2010). Directed epitope delivery across the Escherichia coli outer membrane through the porin OmpF.Proceedings of the National Academy of Sciences, 107(50), 21412-21417.Housden, N. G., Hopper, J. T., Lukoyanova, N., Rodriguez-Larrea, D., Wojdyla, J. A., Klein, A., ... & Kleanthous, C. (2013). Intrinsically disordered protein threads through the bacterial outer-membrane porin OmpF. Science,340(6140), 1570-1574.Imlay, J. A. (2013). The molecular mechanisms and physiological consequences of oxidative stress: lessons from a model bacterium. Nature Reviews Microbiology, 11(7), 443-454.Iyer, R., & Delcour, A. H. (1997). Complex inhibition of OmpF and OmpC bacterial porins by polyamines. Journal of Biological Chemistry, 272(30), 18595-18601.Jacob, F., & Siminovitch, L. et Wollman, E. 1952. Sur la biosynthèse d'une colicine et sur son mode d'action. Ann. Inst. Pasteur83, 295-315.Jakes, K. S. (2014). Daring to be different: colicin N finds another way.Molecular microbiology, 92(3), 435-439.James, K. J., Hancock, M. A., Moreau, V., Molina, F., & Coulton, J. W. (2008). TonB induces conformational changes in surface-exposed loops of FhuA, outer membrane receptor of Escherichia coli. Protein Science, 17(10), 1679-1688.128James, K. J., Hancock, M. A., Gagnon, J. N., & Coulton, J. W. (2009). TonB interacts with BtuF, the Escherichia coli periplasmic binding protein for cyanocobalamin. Biochemistry, 48(39), 9212-9220.Jetten, A. M., & Jetten, M. E. R. (1975). Energy requirement for the initiation of colicin action in Escherichia coli. Biochimica et Biophysica Acta (BBA)-Bioenergetics, 387(1), 12-22.Johnson, C. L., Ridley, H., Marchetti, R., Silipo, A., Griffin, D. C., Crawford, L., ... & Lakey, J. H. (2014). The antibacterial toxin colicin N binds to the inner core of lipopolysaccharide and close to its translocator protein. Molecular microbiology, 92(3), 440-452.Jordan, L. D., Zhou, Y., Smallwood, C. R., Lill, Y., Ritchie, K., Yip, W. T., ... & Klebba, P. E. (2013). Energy-dependent motion of TonB in the Gram-negative bacterial inner membrane. Proceedings of the National Academy of Sciences, 110(28), 11553-11558.Journet, L., Bouveret, E., Rigal, A., Lloubes, R., Lazdunski, C., & Benedetti, H. (2001). Import of colicins across the outer membrane of Escherichia coli involves multiple protein interactions in the periplasm. Molecular microbiology, 42(2), 331-344.Kadner, R. J., & Heller, K. J. (1995). Mutual inhibition of cobalamin and siderophore uptake systems suggests their competition for TonB function.Journal of bacteriology, 177(17), 4829-4835.Kaiser, P., Regoes, R. R., Dolowschiak, T., Wotzka, S. Y., Lengefeld, J., Slack, E., ... & Hardt, W. D. (2014). Cecum lymph node dendritic cells harbor slow-growing bacteria phenotypically tolerant to antibiotic treatment. PLoS Biol, 12(2), e1001793.Kampfenkel, K., & Braun, V. (1992). Membrane topology of the Escherichia coli ExbD protein. Journal of bacteriology, 174(16), 5485-5487.Kampfenkel, K., & Braun, V. (1993). Topology of the ExbB protein in the cytoplasmic membrane of Escherichia coli. Journal of Biological Chemistry,268(8), 6050-6057.Karlsson, M., Hannavy, K., & Higgins, C. F. (1993). A sequence-specific function for the N-terminal signal-like sequence of the TonB protein.Molecular microbiology, 8(2), 379-388.Karpowich, N. K., Huang, H. H., Smith, P. C., & Hunt, J. F. (2003). Crystal structures of the BtuF periplasmic-binding protein for vitamin B12 suggest a functionally important reduction in protein mobility upon ligand binding.Journal of Biological Chemistry, 278(10), 8429-8434.129Kester, J. C., & Fortune, S. M. (2014). Persisters and beyond: mechanisms of phenotypic drug resistance and drug tolerance in bacteria. Critical reviews in biochemistry and molecular biology, 49(2), 91-101.Khawaja, X., Dunlop, J., & Kowal, D. (2008). Scintillation proximity assay in lead discovery. Expert opinion on drug discovery, 3(11), 1267-1280.Khursigara, C. M., De Crescenzo, G., Pawelek, P. D., & Coulton, J. W. (2004). Enhanced Binding of TonB to a Ligand-loaded Outer Membrane Receptor ROLE OF THE OLIGOMERIC STATE OF TonB IN FORMATION OF A FUNCTIONAL FhuA· TonB COMPLEX. Journal of Biological Chemistry,279(9), 7405-7412.Khursigara, C. M., De Crescenzo, G., Pawelek, P. D., & Coulton, J. W. (2005). Kinetic analyses reveal multiple steps in forming TonB-FhuA complexes from Escherichia coli. Biochemistry, 44(9), 3441-3453.Kim, M., Fanucci, G. E., & Cafiso, D. S. (2007). Substrate-dependent transmembrane signaling in TonB-dependent transporters is not conserved.Proceedings of the National Academy of Sciences, 104(29), 11975-11980.Kitagawa, M., Matsumura, Y., & Tsuchido, T. (2000). Small heat shock proteins, IbpA and IbpB, are involved in resistances to heat and superoxide stresses in Escherichia coli. FEMS microbiology letters, 184(2), 165-171.Kitagawa, M., Miyakawa, M., Matsumura, Y., & Tsuchido, T. (2002). Escherichia coli small heat shock proteins, IbpA and IbpB, protect enzymes from inactivation by heat and oxidants. European Journal of Biochemistry,269(12), 2907-2917.Kleanthous, C., & Walker, D. (2001). Immunity proteins: enzyme inhibitors that avoid the active site. Trends in biochemical sciences, 26(10), 624-631.Kleanthous, C. (2010). Swimming against the tide: progress and challenges in our understanding of colicin translocation. Nature Reviews Microbiology,8(12), 843-848.Klebba, P. E. (2016). ROSET Model of TonB Action in Gram-Negative Bacterial Iron Acquisition. Journal of bacteriology, 198(7), 1013-1021.Klock, H. E., & Lesley, S. A. (2009). The Polymerase Incomplete Primer Extension (PIPE) method applied to high-throughput cloning and site-directed mutagenesis. High Throughput Protein Expression and Purification: Methods and Protocols, 91-103.130Koedding, J., Howard, P., Kaufmann, L., Polzer, P., Lustig, A., & Welte, W. (2004). Dimerization of TonB is not essential for its binding to the outer membrane siderophore receptor FhuA of Escherichia coli. Journal of Biological Chemistry, 279(11), 9978-9986.Ködding, J., Killig, F., Polzer, P., Howard, S. P., Diederichs, K., & Welte, W. (2005). Crystal structure of a 92-residue C-terminal fragment of TonB from Escherichia coli reveals significant conformational changes compared to structures of smaller TonB fragments. Journal of Biological Chemistry, 280(4), 3022-3028.Korkhov, V. M., Mireku, S. A., Veprintsev, D. B., & Locher, K. P. (2014). Structure of AMP-PNP–bound BtuCD and mechanism of ATP-powered vitamin B12 transport by BtuCD–F. Nature structural & molecular biology,21(12), 1097-1099.Kramer, R. A., Brandenburg, K., Vandeputte-Rutten, L., Werkhoven, M., Gros, P., Dekker, N., & Egmond, M. R. (2002). Lipopolysaccharide regions involved in the activation of Escherichia coli outer membrane protease OmpT.European journal of biochemistry, 269(6), 1746-1752.Kuczynska-Wisnik, D., Kçdzierska, S., Matuszewska, E., Lund, P., Taylor, A., Lipinska, B., & Laskowska, E. (2002). The Escherichia coli small heat-shock proteins IbpA and IbpB prevent the aggregation of endogenous proteins denatured in vivo during extreme heat shock. Microbiology, 148(6), 1757-1765.Kurisu, G., Zakharov, S. D., Zhalnina, M. V., Bano, S., Eroukova, V. Y., Rokitskaya, T. I., ... & Cramer, W. A. (2003). The structure of BtuB with bound colicin E3 R-domain implies a translocon. Nature Structural & Molecular Biology, 10(11), 948-954.Larsen, R. A., Thomas, M. G., & Postle, K. (1999). Protonmotive force, ExbB and ligand-bound FepA drive conformational changes in TonB. Molecular microbiology, 31(6), 1809-1824.Larsen, R. A., Deckert, G. E., Kastead, K. A., Devanathan, S., Keller, K. L., & Postle, K. (2007). His20 provides the sole functionally significant side chain in the essential TonB transmembrane domain. Journal of bacteriology, 189(7), 2825-2833.Lau, P. C., & Condie, J. A. (1989). Nucleotide sequences from the colicin E5, E6 and E9 operons: presence of a degenerate transposon-like structure in the ColE9-J plasmid. Molecular and General Genetics MGG, 217(2-3), 269-277.131Lavigne, J. P., Sotto, A., Nicolas-Chanoine, M. H., Bouziges, N., Bourg, G., & Davin-Regli, A. (2012). Membrane permeability, a pivotal function involved in antibiotic resistance and virulence in Enterobacter aerogenes clinical isolates.Clinical Microbiology and Infection, 18(6), 539-545.Law, C. J., Penfold, C. N., Walker, D. C., Moore, G. R., James, R., & Kleanthous, C. (2003). OmpF enhances the ability of BtuB to protect susceptible Escherichia coli cells from colicin E9 cytotoxicity. FEBS letters,545(2), 127-132.Lawrence, J. G., & Roth, J. R. (1996). Evolution of coenzyme B12 synthesis among enteric bacteria: evidence for loss and reacquisition of a multigene complex. Genetics, 142(1), 11-24.Lazzaroni, J. C., Dubuisson, J. F., & Vianney, A. (2002). The Tol proteins of Escherichia coli and their involvement in the translocation of group A colicins.Biochimie, 84(5), 391-397.Leitz, A. J., Bayburt, T. H., Barnakov, A. N., Springer, B. A., & Sligar, S. G. (2006). Functional reconstitution of beta~ 2-adrenergic receptors utilizing self-assembling Nanodisc technology. Biotechniques, 40(5), 601.le Maire, M., Champeil, P., & Møller, J. V. (2000). Interaction of membrane proteins and lipids with solubilizing detergents. Biochimica et Biophysica Acta (BBA)-Biomembranes, 1508(1), 86-111.Liao, C. C., Hsiao, K. C., Liu, Y. W., Leng, P. H., Yuen, H. S., & Chak, K. F. (2001). Processing of DNase domain during translocation of colicin E7 across the membrane of Escherichia coli. Biochemical and biophysical research communications, 284(3), 556-562.Locher, K. P., Rees, B., Koebnik, R., Mitschler, A., Moulinier, L., Rosenbusch, J. P., & Moras, D. (1998). Transmembrane signaling across the ligand-gated FhuA receptor: crystal structures of free and ferrichrome-bound states reveal allosteric changes. Cell, 95(6), 771-778.Locher, K. P., & Borths, E. (2004). ABC transporter architecture and mechanism: implications from the crystal structures of BtuCD and BtuF.FEBS letters, 564(3), 264-268.Loftus, S. R., Walker, D., Maté, M. J., Bonsor, D. A., James, R., Moore, G. R., & Kleanthous, C. (2006). Competitive recruitment of the periplasmic translocation portal TolB by a natively disordered domain of colicin E9.Proceedings of the National Academy of Sciences, 103(33), 12353-12358.132Lukacik, P., Barnard, T. J., Keller, P. W., Chaturvedi, K. S., Seddiki, N., Fairman, J. W., ... & Hinnebusch, B. J. (2012). Structural engineering of a phage lysin that targets Gram-negative pathogens. Proceedings of the National Academy of Sciences, 109(25), 9857-9862.Ma, L., Kaserer, W., Annamalai, R., Scott, D. C., Jin, B., Jiang, X., ... & Newton, S. M. (2007). Evidence of ball-and-chain transport of ferric enterobactin through FepA. Journal of Biological Chemistry, 282(1), 397-406.Masi, M. (2013). Structure, function and regulation of outer membrane proteins involved in drug transport in Enterobactericeae: the OmpF/C–TolC case. The open microbiology journal, 7(1).Matuszewska, E., Kwiatkowska, J., Kuczyńska-Wiśnik, D., & Laskowska, E. (2008). Escherichia coli heat-shock proteins IbpA/B are involved in resistance to oxidative stress induced by copper. Microbiology, 154(6), 1739-1747.Melcher, K. (2004). New chemical crosslinking methods for the identification of transient protein-protein interactions with multiprotein complexes. Current Protein and Peptide Science, 5(4), 287-296.Migneault, I., Dartiguenave, C., Bertrand, M. J., & Waldron, K. C. (2004). Glutaraldehyde: behavior in aqueous solution, reaction with proteins, and application to enzyme crosslinking. Biotechniques, 37(5), 790-806.Mills, A. (2013). Characterization of interactions of TonB and Colicin M with FhuA reconstituted into Nanodiscs.Mills, A., Le, H. T., Coulton, J. W., & Duong, F. (2014). FhuA interactions in a detergent-free nanodisc environment. Biochimica et Biophysica Acta (BBA)-Biomembranes, 1838(1), 364-371.Moeck, G. S., Tawa, P., Xiang, H., Ismail, A. A., Turnbull, J. L., & Coulton, J. W. (1996). Ligand-induced conformational change in the ferrichrome–iron receptor of Escherichia coli K-12. Molecular microbiology, 22(3), 459-471.Moeck, G. S., & Letellier, L. (2001). Characterization of In Vitro Interactions between a Truncated TonB Protein fromEscherichia coli and the Outer Membrane Receptors FhuA and FepA. Journal of bacteriology, 183(9), 2755-2764.Monera, O. D., Sereda, T. J., Zhou, N. E., Kay, C. M., & Hodges, R. S. (1995). Relationship of sidechain hydrophobicity and α-helical propensity on the stability of the single-stranded amphipathic α-helix. Journal of peptide science, 1(5), 319-329.133Mosbahi, K., Lemaître, C., Keeble, A. H., Mobasheri, H., Morel, B., James, R., ... & Kleanthous, C. (2002). The cytotoxic domain of colicin E9 is a channel-forming endonuclease. Nature Structural & Molecular Biology, 9(6), 476-484.Mosbahi, K., Walker, D., James, R., Moore, G. R., & Kleanthous, C. (2006). Global structural rearrangement of the cell penetrating ribonuclease colicin E3 on interaction with phospholipid membranes. Protein science, 15(3), 620-627.Müller, D. J., & Engel, A. (1999). Voltage and pH-induced channel closure of porin OmpF visualized by atomic force microscopy. Journal of molecular biology, 285(4), 1347-1351.Nakamoto, H., & Vigh, L. (2007). The small heat shock proteins and their clients. Cellular and Molecular Life Sciences, 64(3), 294-306.Nasr, M. L., & Singh, S. K. (2013). Radioligand binding to nanodisc-reconstituted membrane transporters assessed by the scintillation proximity assay. Biochemistry, 53(1), 4-6.Naumann, D., Schultz, C., Born, J., Labischinski, H., Brandenburg, K., Busse, G., ... & Seydel, U. (1987). Investigations into the polymorphism of lipid A from lipopolysaccharides of Escherichia coli and Salmonella minnesota by Fourier-transform infrared spectroscopy.European Journal of Biochemistry, 164(1), 159-169.Neugebauer, H., Herrmann, C., Kammer, W., Schwarz, G., Nordheim, A., & Braun, V. (2005). ExbBD-dependent transport of maltodextrins through the novel MalA protein across the outer membrane of Caulobacter crescentus.Journal of bacteriology, 187(24), 8300-8311.Ng, C. L., Lang, K., Meenan, N. A., Sharma, A., Kelley, A. C., Kleanthous, C., & Ramakrishnan, V. (2010). Structural basis for 16S ribosomal RNA cleavage by the cytotoxic domain of colicin E3. Nature structural & molecular biology,17(10), 1241-1246.Nikaido, H. (2003). Molecular basis of bacterial outer membrane permeability revisited. Microbiology and molecular biology reviews, 67(4), 593-656.Nikaido, H. (2009). Outer Membrane, Gram-negative Bacteria. In Schaechter, M. (Ed.), Encyclopedia of Microbiology, Third Edition (pp. 439-452). Noinaj, N., Guillier, M., Barnard, T. J., & Buchanan, S. K. (2010). TonB-dependent transporters: regulation, structure, and function. Annual review of microbiology, 64, 43.134Patel, D. S., Re, S., Wu, E. L., Qi, Y., Klebba, P. E., Widmalm, G., ... & Im, W. (2016). Dynamics and interactions of OmpF and LPS: Influence on pore accessibility and ion permeability. Biophysical journal, 110(4), 930-938.Pawelek, P. D., Croteau, N., Ng-Thow-Hing, C., Khursigara, C. M., Moiseeva, N., Allaire, M., & Coulton, J. W. (2006). Structure of TonB in complex with FhuA, E. coli outer membrane receptor. Science, 312(5778), 1399-1402.Payne, M. A., Igo, J. D., Cao, Z., Foster, S. B., Newton, S. M., & Klebba, P. E. (1997). Biphasic binding kinetics between FepA and its ligands. Journal of Biological Chemistry, 272(35), 21950-21955.Penfold, C. N., B. Healy, N. G. Housden, R. Boetzel, M. Vankemmelbeke, G. R. Moore, C. Kleanthous, & James, R. (2004). Flexibility in the receptor-binding domain of the enzymatic colicin E9 is required for toxicity against Escherichia coli cells. J. Bacteriol. 186,4520–4527. Postle, K., Kastead, K. A., Gresock, M. G., Ghosh, J., & Swayne, C. D. (2010). The TonB dimeric crystal structures do not exist in vivo. MBio, 1(5), e00307-10.Prehna, G., Zhang, G., Gong, X., Duszyk, M., Okon, M., McIntosh, L. P., ... & Strynadka, N. C. (2012). A protein export pathway involving Escherichia coli porins. Structure, 20(7), 1154-1166.Pugsley, A. P. (1988). The immunity and lysis genes of ColN plasmid pCHAP4. Molecular and General Genetics MGG, 211(2), 335-341.Rassam, P., Copeland, N. A., Birkholz, O., Tóth, C., Chavent, M., Duncan, A. L., ... & Quinn, D. M. (2015). Supramolecular assemblies underpin turnover of outer membrane proteins in bacteria. Nature, 523(7560), 333-336.Raymond, K. N., Dertz, E. A., & Kim, S. S. (2003). Enterobactin: an archetype for microbial iron transport. Proceedings of the National Academy of Sciences, 100(7), 3584-3588.Rigal, A., Bouveret, E., Lloubes, R., Lazdunski, C., & Benedetti, H. (1997). The TolB protein interacts with the porins of Escherichia coli.Journal of bacteriology, 179(23), 7274-7279.Rocque, W. J., Coughlin, R. T., & McGroarty, E. J. (1987). Lipopolysaccharide tightly bound to porin monomers and trimers from Escherichia coli K-12.Journal of bacteriology, 169(9), 4003-4010.135Rutz, J. M., Liu, J., Lyons, J. A., Goranson, J., Armstrong, S. K., McIntosh, M. A., ... & Klebba, P. E. (1992). Formation of a gated channel by a ligand-specific transport protein in the bacterial outer membrane. Science,258(5081), 471-475.Sauter, A., Howard, S. P., & Braun, V. (2003). In vivo evidence for TonB dimerization. Journal of bacteriology, 185(19), 5747-5754.Schletter, J., Heine, H., Ulmer, A. J., & Rietschel, E. T. (1995). Molecular mechanisms of endotoxin activity. Archives of microbiology, 164(6), 383-389.Senior, B. W., Kwasniak, J., & Holland, I. B. (1970). Colicin E3-directed changes in ribosome function and polyribosome metabolism in Escherichia coli K12. Journal of molecular biology, 53(2), 205-220.Senior, B. W., & Holland, I. B. (1971). Effect of colicin E3 upon the 30S ribosomal subunit of Escherichia coli. Proceedings of the National Academy of Sciences, 68(5), 959-963.Sharpless, N. E., & Flavin, M. (1966). The Reactions of Amines and Amino Acids with Maleimides. Structure of the Reaction Products Deduced from Infrared and Nuclear Magnetic Resonance Spectroscopy. Biochemistry,5(9), 2963-2971.Sherbet, G. V., & Lakshmi, M. S. (1973). Characterisation of Escherichia coli cell surface by isoelectric equilibrium analysis. Biochimica et Biophysica Acta (BBA)-Biomembranes, 298(1), 50-58.Shi, Z., Chak, K. F., & Yuan, H. S. (2005). Identification of an essential cleavage site in ColE7 required for import and killing of cells. Journal of Biological Chemistry, 280(26), 24663-24668.Shoham, M., & Djebli, A. (1992). Structural studies on colicin E3 and its immunity protein. In Bacteriocins, Microcins and Lantibiotics (pp. 203-214). Springer Berlin Heidelberg.Shultis, D. D., Purdy, M. D., Banchs, C. N., & Wiener, M. C. (2006). Outer membrane active transport: structure of the BtuB: TonB complex. Science,312(5778), 1396-1399.Sikora, A. K., & Cafiso, D. (2016). TonB Binding Produces Allosteric Changes in the Outer Loops and Substrate Binding Site of the TBDT BTUB.Biophysical Journal, 110(3), 514a.Soelaiman, S., Jakes, K., Wu, N., Li, C., & Shoham, M. (2001). Crystal structure of colicin E3: implications for cell entry and ribosome inactivation.Molecular cell, 8(5), 1053-1062.136Strub, C., Alies, C., Lougarre, A., Ladurantie, C., Czaplicki, J., & Fournier, D. (2004). Mutation of exposed hydrophobic amino acids to arginine to increase protein stability. BMC biochemistry, 5(1), 9.Sun, S., Almaden, J., Carlson, T. J., Barker, J., & Gehring, M. R. (2005). Assay development and data analysis of receptor–ligand binding based on scintillation proximity assay. Metabolic engineering, 7(1), 38-44.Sverzhinsky, A., Chung, J. W., Deme, J. C., Fabre, L., Levey, K. T., Plesa, M., Carter, D.M., Lypaczewski, P. & Coulton, J. W. (2015). Membrane protein complex ExbB4-ExbD1-TonB1 from Escherichia coli demonstrates conformational plasticity. Journal of bacteriology, 197(11), 1873-1885.Taylor, R., Burgner, J. W., Clifton, J., & Cramer, W. A. (1998). Purification and characterization of monomeric Escherichia coli vitamin B12 receptor with high affinity for colicin E3. Journal of Biological Chemistry, 273(47), 31113-31118.Tuckman, M., & Osburne, M. S. (1992). In vivo inhibition of TonB-dependent processes by a TonB box consensus pentapeptide. Journal of bacteriology, 174(1), 320-323.Udho, E., Jakes, K. S., & Finkelstein, A. (2012). TonB-dependent transporter FhuA in planar lipid bilayers: partial exit of its plug from the barrel.Biochemistry, 51(34), 6753-6759.van der Heijden, J., Bosman, E. S., Reynolds, L. A., & Finlay, B. B. (2015). Direct measurement of oxidative and nitrosative stress dynamics in Salmonella inside macrophages. Proceedings of the National Academy of Sciences, 112(2), 560-565.van der Heijden, J., Vogt, S. L., Reynolds, L. A., Peña-Díaz, J., Tupin, A., Aussel, L., & Finlay, B. B. (2016). Exploring the redox balance inside gram-negative bacteria with redox-sensitive GFP. Free Radical Biology and Medicine, 91, 34-44.van der Heijden, J., Reynolds, L. A., Scholz, R., Mills, A. T., Imami, K., Foster, L. J., Franck Duong, F., and B. Finlay. (2016) Salmonella rapidly regulates membrane permeability to survive oxidative stress and form persisters. mBio01238-16 In print.Van der Ley, P., De Graaff, P., & Tommassen, J. (1986). Shielding of Escherichia coli outer membrane proteins as receptors for bacteriophages and colicins by O-antigenic chains of lipopolysaccharide. Journal of bacteriology, 168(1), 449-451.137Vankemmelbeke, M., Zhang, Y., Moore, G. R., Kleanthous, C., Penfold, C. N., & James, R. (2009). Energy-dependent immunity protein release during tol-dependent nuclease colicin translocation. Journal of Biological Chemistry,284(28), 18932-18941.Vankemmelbeke, M., James, R., & Penfold, C. N. (2012). Membrane activities of colicin nuclease domains: analogies with antimicrobial peptides. Biochem. Soc. Trans.,40, 1517-1521.Vankemmelbeke, M., Paul, O., James, R., & Penfold, C. N. (2012b). Interaction of nuclease colicins with membranes: insertion depth correlates with bilayer perturbation. PloS one, 7(9), e46656.Walker, D., Moore, G. R., James, R., & Kleanthous, C. (2003). Thermodynamic consequences of bipartite immunity protein binding to the ribosomal ribonuclease colicin E3. Biochemistry, 42(14), 4161-4171.Walker, D., Mosbahi, K., Vankemmelbeke, M., James, R., & Kleanthous, C. (2007). The role of electrostatics in colicin nuclease domain translocation into bacterial cells. Journal of Biological Chemistry, 282(43), 31389-31397.Yajima, S., Inoue, S., Ogawa, T., Nonaka, T., Ohsawa, K., & Masaki, H. (2006). Structural basis for sequence-dependent recognition of colicin E5 tRNase by mimicking the mRNA–tRNA interaction. Nucleic acids research,34(21), 6074-6082.Yamashita, E., Zhalnina, M. V., Zakharov, S. D., Sharma, O., & Cramer, W. A. (2008). Crystal structures of the OmpF porin: function in a colicin translocon.The EMBO journal, 27(15), 2171-2180.Yue, W. W., Grizot, S., & Buchanan, S. K. (2003). Structural evidence for iron-free citrate and ferric citrate binding to the TonB-dependent outer membrane transporter FecA. Journal of molecular biology, 332(2), 353-368.Zakharov, S. D., Zhalnina, M. V., Sharma, O., & Cramer, W. A. (2006). The colicin E3 outer membrane translocon: immunity protein release allows interaction of the cytotoxic domain with OmpF porin. Biochemistry, 45(34), 10199-10207.Zakharov, S. D., Sharma, O., Zhalnina, M. V., & Cramer, W. A. (2008). Primary Events in the Colicin Translocon: FRET Analysis of Colicin Unfolding Initiated by Binding to BtuB and OmpF†. Biochemistry, 47(48), 12802-12809.138Zasloff, M. (2002). Antimicrobial peptides of multicellular organisms. nature,415(6870), 389-395.139

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0319168/manifest

Comment

Related Items