THE REGULATORY MECHANISMS OF THE MALTOSE TRANSPORTER by Huan Bao B.Sc., Wuhan University, 2005 M.Sc., Chinese Academy of Sciences, 2008 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) March 2014 ? Huan Bao, 2014 Abstract ATP-binding cassette (ABC) transporters couple ATP hydrolysis to import and export of a large array of substances across cell membranes in all kingdoms of life. Since the transport reaction consumes cellular energy, substrate translocation mediated by ABC transporters must be regulated according to the requirements of the cell. This thesis uses the Escherichia coli maltose transporter MalFGK2 to understand the regulatory mechanisms of ABC importers. Biochemical and biophysical approaches were employed to investigate how this transport process is modulated by maltose, the maltose-binding protein MalE and the glucose-specific enzyme EIIAGlc. First, I show that ATP facilitates MalE binding to MalFGK2, which forms the complex of MalE-MalFGK2 for efficient maltose transport. In addition, when the external maltose level exceeds that required, maltose is able to limit the maximal transport rate by promoting dissociation of MalE from MalFGK2. Finally, I find that the N-terminal tail of EIIAGlc and acidic phospholipids are essential for the binding of the protein to the MalK dimer, so that cleavage of ATP by MalFGK2 is inhibited. These results, combined with previous genetic, biochemical and structural work, provide valuable insights into our understanding of the regulatory mechanisms of the maltose transport system. ii Preface A version of chapter 2 has been published. [Bao, H.], Dalal, K., Wang, V., Rouiller, I. and Duong, F. (2013) The maltose ABC transporter: action of membrane lipids on the transporter stability, coupling and ATPase activity. Biochim Biophys Acta. 1828(8): 1723-30. Dalal, K. assisted with static light scattering measurements. I created Figure 2.2b and 2.3 based on electron microscopy analysis performed by Wang, V. and Rouiller, I. I performed all other experiments and wrote this manuscript jointly with my supervisor. A version of chapter 3 has been published. [Bao H] and Duong F. Discovery of an auto-regulation mechanism for the maltose transporter MalFGK2. (2012) Plos One. 7(4): e34836. I conducted all experiments and wrote the manuscript jointly with my supervisor. A version of chapter 4 has been published. [Bao H] and Duong F. ATP alone triggers the outward-facing conformation of the maltose ABC transporter. (2013) J Biol Chem. 288(5): 3439-48. I conducted all experiments and wrote the manuscript jointly with my supervisor. A version of chapter 5 has been published. [Bao H] and Duong F. Anionic lipid phosphatidylglycerol directs binding and inhibition of EIIAGlc onto the maltose transporter. (2013) J Biol Chem. 288(33): 23666-74. I conducted all experiments and wrote the manuscript jointly with my supervisor. I must acknowledge Sung Hoon Choi for site-directed mutagenesis of MalFGK2 and EIIAGlc. iii Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables ..................................................................................................................................x List of Figures ............................................................................................................................... xi List of Symbols and Abbreviations .......................................................................................... xiii Acknowledgements ......................................................................................................................xv Dedication ................................................................................................................................... xvi Chapter 1: Introduction ................................................................................................................1 1.1 Mechanism of membrane transport ................................................................................ 1 1.2 ABC transport systems ................................................................................................... 2 1.2.1 Structural organization of ABC transporters .............................................................. 3 1.2.2 Regulation of ABC transport systems......................................................................... 9 1.2.3 Periplasmic binding proteins..................................................................................... 10 1.3 Maltose transport system .............................................................................................. 13 1.3.1 Coupling of ATP hydrolysis to maltose transport .................................................... 14 1.3.2 Structural analyses of the maltose transporter .......................................................... 16 1.4 Function of the maltose binding protein (MalE)........................................................... 19 1.5 Inhibition of the maltose transporter by glucose specific enzyme EIIAGlc ................... 21 1.5.1 The Phosphoenolpyruvate carbohydrate phosphotransferase system ....................... 21 1.5.2 Action of EIIAGlc on the maltose transporter ............................................................ 23 1.6 Thesis investigation ...................................................................................................... 25 iv Chapter 2: Action of membrane lipids on the maltose transporter MalFGK2 ......................27 2.1 Introduction ................................................................................................................... 27 2.2 Material and methods .................................................................................................... 28 2.2.1 Materials ................................................................................................................... 28 2.2.2 Reconstitution of MalFGK2 ...................................................................................... 28 2.2.3 Maltose transport assay ............................................................................................. 29 2.2.4 Dynamic and static light scattering ........................................................................... 29 2.2.5 Lipid phosphorus analysis of nanodiscs ................................................................... 30 2.2.6 Electron microscopy ................................................................................................. 30 2.2.7 Circular dichroism spectroscopy ............................................................................... 30 2.2.8 Other methods ........................................................................................................... 31 2.3 Results ........................................................................................................................... 31 2.3.1 Reconstitution of MalFGK2 into nanodiscs .............................................................. 31 2.3.2 Characterization of the lipid-rich nanodiscs ............................................................. 33 2.3.3 Single-particle EM analysis of nanodiscs ................................................................. 35 2.3.4 Effect of the lipids on the transporter ATPase activity and coupling to maltose ..... 36 2.3.5 The length of the lipid acyl chain is important for the transporter activity .............. 37 2.3.6 Lipids do not affect the binding between MalE and MalFGK2 ................................ 38 2.3.7 Lipids increase the folding stability of the transporter ............................................. 39 2.4 Discussion ..................................................................................................................... 41 Chapter 3: Discovery of an auto-regulatory mechanism for the maltose ABC transporter MalFGK2 .......................................................................................................................................44 3.1 Introduction ................................................................................................................... 44 v 3.2 Materials and methods .................................................................................................. 46 3.2.1 Materials ................................................................................................................... 46 3.2.2 Sedimentation and pull-down assays ........................................................................ 46 3.2.3 Fluorescence labeling................................................................................................ 47 3.2.4 Fluorescence spectroscopy........................................................................................ 47 3.2.5 In vivo maltose transport assays ............................................................................... 48 3.2.6 Thin-layer chromatography (TLC) ........................................................................... 48 3.2.7 Other methods ........................................................................................................... 48 3.3 Results ........................................................................................................................... 49 3.3.1 Maltose-free MalE binds with a high-affinity to the outward-facing transporter ..... 49 3.3.2 MalE captures maltose and loses affinity for MalFGK2 ........................................... 52 3.3.3 MalE with low affinity for maltose has a high affinity for MalFGK2 ...................... 53 3.3.4 The results obtained with the nanodisc are confirmed in proteoliposomes .............. 55 3.3.5 Dual effect of the sugar on the transporter activity in proteoliposomes ................... 58 3.3.6 Maltose is both substrate and regulator of the transporter ........................................ 60 3.3.7 Consequence of an unregulated maltose transport in intact cells ............................. 61 3.3.8 Rate-limiting step of the ATP hydrolytic cycle is the release of Pi .......................... 62 3.4 Discussion ..................................................................................................................... 67 Chapter 4: Conformational changes of the maltose transporter.............................................73 4.1 Introduction ................................................................................................................... 73 4.2 Materials and methods .................................................................................................. 76 4.2.1 Material ..................................................................................................................... 76 4.2.2 Reconstitution of MalFGK2 ...................................................................................... 76 vi 4.2.3 Cysteine Cross-linking Experiments......................................................................... 76 4.2.4 Fluorescence Labeling and Spectroscopy ................................................................. 76 4.2.5 Other Methods .......................................................................................................... 77 4.3 Results ........................................................................................................................... 77 4.3.1 ATP controls the outward-facing conformation of the transporter ........................... 77 4.3.2 ATP controls binding of MalE to the transporter ..................................................... 80 4.3.3 MalE and maltose do not facilitate the transition of MalFGK2 to the outward-facing conformation ......................................................................................................................... 84 4.3.4 MalFG exists in an outward-facing conformation .................................................... 87 4.3.5 MalFG is converted to the inward-facing conformation upon binding of MalK ...... 91 4.3.6 Tightly bound MalK inhibits the transition of MalFG to the outward-facing conformation ......................................................................................................................... 93 4.3.7 Mutations that decrease interaction of MalK with MalFG restore transport ............ 94 4.4 Discussion ..................................................................................................................... 96 Chapter 5: Phosphatidylglycerol directs binding and inhibitory action of EIIAGlc protein on the maltose transporter .............................................................................................................101 5.1 Introduction ................................................................................................................. 101 5.2 Materials and methods ................................................................................................ 103 5.2.1 Materials ................................................................................................................. 103 5.2.2 Reconstitution of MalFGK2 .................................................................................... 103 5.2.3 Cross-linking reactions ........................................................................................... 103 5.2.4 Other methods ......................................................................................................... 103 5.3 Results ......................................................................................................................... 104 vii 5.3.1 The Inhibition by EIIAGlc depends on the N-terminal tail and PG lipids ............... 104 5.3.2 PG lipids are necessary for the binding of EIIAGlc to MalFGK2 ............................ 106 5.3.3 Identification of the binding interface between EIIAGlc and MalK ........................ 107 5.3.4 EIIAGlc does not inhibit the ninding of ATP to MalK ............................................ 111 5.3.5 EIIAGlc inhibits the cleavage of ATP by MalK ....................................................... 111 5.4 Discussion ................................................................................................................... 114 Chapter 6: Summary and Future directions ...........................................................................118 References ...................................................................................................................................125 Appendices ..................................................................................................................................139 Appendix A Protein purification ............................................................................................. 139 A.1 Purification of MalFGK2......................................................................................... 139 A.2 Purification of MalE ............................................................................................... 140 A.3 Purification of EIIAGlc ............................................................................................ 140 Appendix B Illustration of nanodiscs and proteoliposomes reconstitution ............................ 142 Appendix C Equations ............................................................................................................ 143 C.1 Equation 1 ............................................................................................................... 143 C.2 Equation 2 ............................................................................................................... 143 C.3 Equation 3 ............................................................................................................... 143 C.4 Equation 4 ............................................................................................................... 144 C.5 Equation 5 ............................................................................................................... 144 Appendix D Isolation of the Nd-FGK2-MalE complex .......................................................... 145 D.1 Isolation of the Nd-FGK2-MalE complex in the presence of ATP and Vi ............. 145 D.2 Isolation of the the MalE-Nd-FGK2 complex in the absence of nucleotides .......... 146 viii Appendix E Effect of maltooligosaccharides on MalE binding to MalFGK2 ........................ 148 Appendix F Orientation of MalFGK2 reconsituted in proteoliposomes ................................. 149 Appendix G Binding of MalE to MalFGK2 in proteoliposomes ............................................ 150 Appendix H Conformation of MalFG and re-assembled MalFGK2 ....................................... 151 Appendix I Stability of MalFGK2 ........................................................................................... 152 ix List of Tables Table 2.1 Mass, diameter and number of lipid molecules per nanodisc particles. ...................... 33 Table 2.2 Number of lipid molecules in nanodiscs reconstituted with different synthetic lipids?....................................................................................................................................................... 38 Table 3.1 The dissociation constants of MalE for Nd-MalFGK2 ................................................. 50 Table 3.2 Kinetic parameters of the transport ATPase and affinity of the MalE mutants for maltose. ......................................................................................................................................... 60 Table 4.1 ATPase activity of the pyrene-labeled transporters reconstituted in proteoliposomes. ??????????????????????????????????????79 Table 4.2 The affinity of MalE for the transporter in nanodiscs was determined by fluorescence quenching assay. ........................................................................................................................... 84 Table 5.1 Equilibrium constants of TNP-ATP and ATP for the maltose transporter MalFGK2?????????????????????????????????..113 x List of Figures Figure 1.1 Architechture of ABC transporters. ............................................................................... 3 Figure 1.2 Crystal structures of different ABC transporters. .......................................................... 5 Figure 1.3 Structures of nucleotide-free and ATP-bound MalK. ................................................... 7 Figure 1.4 Coupling ATP binding to conformational change of ABC transporters. ...................... 8 Figure 1.5 Substrate binding to PBPs. .......................................................................................... 12 Figure 1.6 Structures of PBPs in I-VI classes. .............................................................................. 13 Figure 1.7 The gate region in the maltose transporter. ................................................................. 18 Figure 1.8 The conventional model for maltose transport. ........................................................... 20 Figure 1.9 Inhibition of the maltose transporter by carbon catabolite repression. ....................... 23 Figure 1.10 A model of EIIAGlc binding to the membrane. .......................................................... 24 Figure 2.1 Isolation of MalFGK2 nanodiscs made at low and high lipid ratios............................ 32 Figure 2.2 Characterization of MalFGK2 nanodiscs made at high lipid ratio. ............................. 34 Figure 2.3 Electron microscopy analyses of MalFGK2 nanodiscs made at low lipid ratio. ......... 36 Figure 2.4 Effect of the amount of membrane lipids on the transporter ATPase activity. ........... 37 Figure 2.5 The length of the lipid acyl chain is important for the transporter activity ................. 38 Figure 2.6 Effect of lipids on MalE binding to MalFGK2 nanodiscs. .......................................... 39 Figure 2.7 Effect of lipids on the transporter stability. ................................................................. 40 Figure 2.8 The ATPase and transport activity of wild type and mutant MalF500. ...................... 41 Figure 3.1 High-affinity binding of MalE to the MalFGK2 complex. .......................................... 51 Figure 3.2 MalE exhibits low affinity for maltose when bound to MalFGK2. ............................. 53 Figure 3.3 Binding of the MalE mutants to MalFGK2. ................................................................ 55 Figure 3.4 Binding of MalE to the MalFGK2 complex in proteoliposomes. ................................ 57 xi Figure 3.5 Regulation of the MalK transport ATPase by maltose. ............................................... 59 Figure 3.6 Maltose transport in intact cells. .................................................................................. 62 Figure 3.7 ATP binding is not the rate-limiting step of the ATP hydrolytic cycle. ...................... 63 Figure 3.8 Pi release is the rate-limiting step of the ATP hydrolytic cycle of MalFGK2. ............ 66 Figure 4.1 Two opposed models for maltose transport................................................................. 75 Figure 4.2 ATP triggers the transporter outward-facing conformation. ....................................... 80 Figure 4.3 ATP controls the binding of MalE to the transporter. ................................................. 81 Figure 4.4 Equilibrium titration of MalE binding to MalFGK2 in nanodiscs. .............................. 83 Figure 4.5 Neither MalE nor maltose facilitates the transition to the outward-facing state. ........ 86 Figure 4.6 Transition kinetics toward the outward-facing conformation. .................................... 87 Figure 4.7 Expression and purification of MalFG ........................................................................ 88 Figure 4.8 Without MalK, MalFG resides in an open conformation ............................................ 89 Figure 4.9 Without MalK, MalFG binds MalE with an intermediate affinity .............................. 91 Figure 4.10 Apo-MalK triggers MalFG inward-facing conformation. ......................................... 92 Figure 4.11 The inactive mutant F6 binds MalK tightly and cannot reach the outward-facing state....................................................................................................................................................... 94 Figure 4.12 Suppressors of mutant F6 rescue ability to reach outward-facing conformation. ..... 95 Figure 5.1 The activity of EIIAGlc depends on intact N-terminal_-helix and PG lipids. ............ 105 Figure 5.2 PG lipids control for the binding of EIIAGlc to MalFGK2. ...................................... 107 Figure 5.3 Model of interaction between EIIAGlc and MalFGK2. .............................................. 108 Figure 5.4 The cross-linking analysis supports the docking analysis prediction. ....................... 110 Figure 5.5 EIIAGlc inhibits the cleavage of ATP but not the binding to MalK. .......................... 114 Figure 6.1 The comprehensive model for maltose transport. ..................................................... 120 xii List of Symbols and Abbreviations ?: Angstrom 10-10 meters ATP: Adenosine tri-phosphate ABC: ATP-binding cassette AMP-PNP: Adenylyl-imidodiphosphate ATP?S: Adenosine 5?-[?-thio]triphosphate BL21: E. coli strain suitable for protein over expression BMOE: Bis-maleimido-ethane, a bifunctional cysteine crosslinker CP3: Copper3 (Phenanthroline), an oxidizing agent Cryo-EM: Cryo-electron microscopy DDM: n-Dodecyl ?-D-maltoside, a non-ionic detergent DTT: Dithiothreitol, a reducing agent DOPC: 1,2-dioleoyl-sn-glycero-3-phosphocholine DOPG: 1,2-Dioleoyl-sn-glycero-3-phospho-rac-(1-glycerol) EIIAGlc: The glucose specfic enzyme that inhibits the maltose transporter. E. coli: Escherichia coli EDTA: Ethylene-di-amine-tetra-acetic acid EPR: Electron paramagnetic resonance IODOGEN: 1,3,4,6-tetrachloro-3?,6?-diphenylglucoluril IPTG: Isopropyl 1-thio-?-D-galactopyranoside kDa: Kilodalton = 1000 gram/mole MalFGK2: The maltose ABC transporter in bacteria. MalE: The periplasmic maltose binding protein. xiii MSP: Membrane scaffold protein NBD: Nucleotide binding domain NEM: N-ethylmaleimide, blocks disulphide bond formation PAGE: Poly-acrylamide gel electrophoresis PDB: Protein data bank TMD: Transmembrane domian TNP-ATP: 2?,3?-O-(2,4,6-Trinitrophenyl) adenosine 5?-triphosphate monolithium trisodium salt SDS: Sodium dodecyl sulfate Vi: Vanadate xiv Acknowledgements I would like to give my deepest thanks to my thesis advisor, Dr. Franck Duong. It is a great pleasure and honor to work with him on this exciting project. I am very grateful to have a mentor that not only spent so much effort and provided invaluable suggestion in my project, but also trained me in manuscript writing, presentation skills and all other aspects in scientific research. His dedication and passion for science inspired me to pursue my own career in academia. I am honored to have the opportunities to work with many brilliant members of the Duong lab, who greatly assisted me over the years, including: K. Dalal, X. Zhang, C.S. Chan, J.F. Montariol, A. Mills, M. Carlson, T. Le, J. Young and H. Won. I thank my supervisory committee, including Dr. Filip Van Petegem and Dr. Steve Withers, for the sparkling comments about my project over the course of my degree. In particular, I would like to thank Dr. Filip Van Petegem for excellent career advice. I would also like to express my gratitude to Dr. George Mackie for sharing his expertise during my Ph.D. studies. I would like to acknowledge the University of British Columbia and NSERC for financial support. Lastly, I would like to thank my family and friends for their love and support over the years. xv Dedication Dedicated to my parents, Mr. Xuehai Bao and Mrs. Zhou Li. To my partner, Ms. Qian Ren. ?????????????xvi Chapter 1: Introduction 1.1 Mechanism of membrane transport Movement of molecules across the cell membranes is vital for many of life?s processes. Since the cell membrane is composed of lipid bilayers with embedded proteins that form a hydrophobic barrier around the cell, most polar and charged compounds cannot cross the membrane unassisted (Stillwell, 2013). To regulate the passage of compounds needed for survival, cells have evolved a plethora of membrane proteins for transporting different substances. In some cases, a membrane protein simply facilitates the diffusion of a substance across membranes at very high rates (? 106 per second; Gouaux & Mackinnon, 2005). The membrane protein either forms a pore or a channel, catalyzing the flow of a substrate down its electrochemical gradient. Substrate specificity varies among these channels. For instance, porin channels in the outer membrane of Gram-negative bacteria are thought to be nonspecific channels (Welte et al, 1995). The passage of substrates is only dependent on their gross physicochemical properties such as size, hydrophobicity, and charges (Kojima & Nikaido, 2013). In contrast, some ion channels exhibit an ability to transport substrates in a selective manner. Such selectivity can be extremely precise, for instance, between ions as similar as Na+ and K+ (Gouaux & Mackinnon, 2005). Nevertheless, membrane transport mediated by pores and channels is entropy-favored diffusion, meaning that movement of a substance against its electrochemical gradient will not occur. In other cases, a substance is transported by a membrane protein against its electrochemical gradient in a process that requires the input of energy, a process known as active transport (Stillwell, 2013). During each transport cycle, the transporter experiences different 1 conformational states, leading to alternative exposure of the substrate binding site on each side of the membrane (Jardetzky, 1966). The different binding affinities of distinct conformations for the substrate determine the direction of transport. Active transporters can be further classified into primary active transporters and secondary active transporters. Secondary active transporters power substrate transport using energy from movement of ?another substance? down its electrochemical gradient (Forrest et al, 2011; Shi, 2013). This amazing mechanism of secondary transport can be appreciated in the case of LacY, a paradigm of the major facilitator superfamily (MFS) that utilizes the electrochemical gradient of protons across membranes for transmembrane movement of lactose (Kaback et al, 2011). This type of transport is secondary only in the sense that the electrochemical gradient of ?another substance? is originally generated by primary active transport, which employs energy from an exergonic chemical reaction such as hydrolysis of adenosine triphosphate (ATP). Examples of primary active transporters using ATP as the source of energy include: P-type ATPases (sodium potassium pump; Thogersen & Nissen, 2012), F-type ATPases (ATP synthase; Dimroth et al, 2003), V-type ATPases (vacuolar ATPase; Benlekbir et al, 2012) and ATP binding cassette (ABC) transporters (Rees et al, 2009). 1.2 ABC transport systems ABC transporters move many substances across the lipid bilayer (Rees et al, 2009). They are composed of two transmembrane domains (TMDs) that provide a substrate translocation pathway, and two nucleotide-binding domains (NBDs) that hydrolyze ATP (Figure 1.1). In general, the chemical events of ATP hydrolysis by NBDs must be in concert with conformational changes of TMDs that alternatively expose the substrate binding site to either side of the membrane, so that the transmembrane movement of substrate ensues (Locher, 2009). ABC transporters are associated with many medically important functions including multidrug 2 resistance, cholesterol and lipid trafficking, ion homeostasis and surface-antigen presentation (Parcej & Tampe, 2010). Such medical significance and various distinct conformations make them attractive targets for mechanistic and structural studies. Figure 1.1 Architechture of ABC transporters. ABC transporters consist of two transmembrane domains (TMDs) and two nucleotide binding domains (NBDs). ATP binding and hydrolysis lead to conformational alternations of ABC transporters between the inward- and outward-facing states. 1.2.1 Structural organization of ABC transporters Following the determination of the crystal structure of the vitamin B12 importer BtuCD in 2002 (Locher et al, 2002), structural studies of ABC transporters have greatly advanced over the past few years (Dawson & Locher, 2006; Gerber et al, 2008; Khare et al, 2009; Locher et al, 2002; Oldham et al, 2007; Pinkett et al, 2007). Based on the structures of TMDs, ABC transporters have been subdivided into type I, II, III importers and exporters (Figure 1.2 a-d). ABC exporters have a common core architecture that consists of 12 transmembrane helices (Rees et al, 2009). Furthermore, these transmembrane helices extend approximately 25 ? beyond the cytoplasmic 3 boundary of the membrane and into the cytoplasm. Type I importers generally contain 10-14 transmembrane helices and mediate the uptake of small molecules such as ions, sugars and amino acids (Locher, 2009). Type II importers are specific for the uptake of metal chelates that are generally larger than the substrates of type I importers, featuring up to 20 transmembrane helices (Locher, 2009). The importers of Type III, also known as ECF(energy coupling factor) transporters, are a recently discovered family of ABC transporters involved in the uptake of vitamins and micronutrients (Zhang, 2013). Each ECF transporter comprises a transmembrane substrate-binding protein (the S component), an energy-coupling module that consists of a pair of cytosolic ATPases (the A and A? component) and another transmembrane protein (the T component; (Wang et al, 2013; Xu et al, 2013). In many cases, the energy-coupling module is shared among different S components that are unrelated in sequence and bind distinct substrates (Erkens et al, 2012). 4 Figure 1.2 Crystal structures of different ABC transporters. TMDs are shown in orange and cyan, whereas NBDs are in green and blue. a, the maltose transporter MalFGK2 (Type I importer; PDB code: 3FH6). b, the vitamin B12 transporter BtuCD (type II importer; PDB code: 4FI3). c, the ECF type ABC transporter for hydroxymethyl pyrimidine (type III importer; PDB code: 4HZU ). d, the multidrug transporter P-glycoprotein (ABC exporter; PDB code: 4F4C). 5 In contrast to the diversity of TMD structures, the NBDs exhibit a characteristic set of highly conserved motifs. This incredible conservation allows interchangeability of NBDs between the sn-glycerol-3-phosphate and maltose ABC transporters (Hekstra & Tommassen, 1993). Each NBD can be further divided into two subdomains: a catalytic core domain and a structurally diverse ?-helical domain (Rees et al, 2009). The catalytic core domain contains the conserved Walker A motif, a Walker B motif, a Q-loop and an H-motif (Figure 1.3). The ?-helical domain contains the ABC signature motif LSGGQ. In an intact ABC transporter, the two NBDs assemble in a head-to-tail manner, such that the Walker A motif of one NBD is orientated towards the signature motif of the other. The arrangement generates two ATP binding and hydrolysis sites at the interface between two NBDs. It is also observed that there is a gap between the NBDs in the structure of nucleotide-free transporters, whereas ATP-bound state shows a closed interface with ATP being sandwiched between the NBDs (Chen et al, 2003). Thus, it is generally assumed that two molecules of ATP are consumed during a single transport cycle (Patzlaff et al, 2003). However, it is still under debate as to whether the two ATPase sites of ABC transporters hydrolyze ATP simultaneously or consecutively (Jones et al, 2009). Crystal structures of many nucleotide-bound transporters suggest that two ATP molecules are hydrolyzed simultaneously (Oldham & Chen, 2011b; Oldham et al, 2007); yet the opposing fact that only one ATPase site is occluded by vanadate and beryllium fluoride in a number of studies is consistent with consecutive hydrolysis (George & Jones, 2013). Clearly, this controversy remains to be resolved. 6 Figure 1.3 Structures of nucleotide-free and ATP-bound MalK. a, the MalK dimer in the absence of nucleotides (PDB code: 3FH6). b, the MalK dimer in the presence of ATP (PDB code: 2R6G). c-d, stereview of the MalK dimer by a 90? rotation of the structures shown in a and b about a horizontal axis. MalF and MalG are omitted for simplicity. MalK are shown in green and blue. The Q loop, D-loop, walker A motif, walker B motif and the signature motif LSGGQ motif are depicted in yellow, cyan, red, orange and purple, respectively. The ATP molecule is shown in stick models (gray). Nevertheless, structures of both ABC importers and exporters have revealed a general mechanism for how ATP binding and hydrolysis by NBDs is coupled to substrate transport by TMDs (Dawson et al, 2007). The TMDs all interact with the NBDs through architecturally 7 conserved ?coupling helices? (Figure 1.4), which form the interface between the TMDs and NBDs. In all reported structures, these coupling helices interact with grooves formed primarily by the Q-loop and the ?-helical domain. In the absence of ATP, the NBDs adopt an open conformation as observed in MalFGK2 and ModBCA2 (Gerber et al, 2008; Khare et al, 2009). ATP binding drives the closure of NBDs which substantially decreases the distance between the two coupling helices (~10-15 ?). As a consequence, the approaching coupling helices cause the flipping of TMDs from the inward-facing to the outward-facing conformation. After dissociation of ADP and inorganic phosphate, the NBDs return to the open conformation, which increases the distance of the coupling helices and flips the TMDs back to the inward-facing conformation. An exception is found in vitamin B12 transporter BtuCD, such that ATP-binding dependent closure of the NBDs causes the conformational change of TMDs from the outward- to the inward-facing state (Korkhov et al, 2012). Figure 1.4 Coupling ATP binding to conformational change of ABC transporters. ATP binding leads to the closure of NBDs, which is transmitted to the conformational change of TMDs through a distance change of the coupling helix. a, structure of MalFGK2 in the inward-facing conformation 8 (PDB code: 3FH6). b, structure of MalFGK2 in the outward-facing conformation (PDB code: 2R6G), obtained in the presence of MalE and nucleotides. The coupling helix is shown in red. 1.2.2 Regulation of ABC transport systems ABC transporters consume cellular energy (ATP) to move substrates across membranes. In order to optimize energy usage, transport processes might be regulated through various mechanisms, including modulation of the transporter activity or control of the quantity of transporters at the cell surface. Activities of transporters can be modulated by non-covalent associating subunits. The bacterial Type I and Type II importers employ periplasmic binding protein to capture substrate and increase transport activities (Davidson et al, 2008). To prevent use of energy for import of substrate beyond that required, substrate uptake into the cells could also be inhibited by the level of intracellular substrate pool as exemplified in the methionine and the molybdate ABC transporters, a process known as the ?trans-inhibition? mechanism (Gerber et al, 2008; Kadaba et al, 2008). Recent structural studies have suggested that this type of inhibition involves substrate binding to the C-terminal regulatory domains of NBDs, which prevent subsequent conformational changes of the ABC transporters during each transport cycle. However, the trans-inhibition mechanism is not found in all importers. Whether and how other inhibitory regulation functions in limiting transport activity remains unclear. In addition, covalent modification is another means to control transporter activity. For instance, phosphorylation of CFTR regulatory domain by cAMP-dependent protein kinase (PKA) has an essential role in tuning its channel activities (Bozoky et al, 2013). The quantity of transporters at the cell surface can be first regulated through genetic control of the expression levels. It is established that the intracellular substrate can serve as an 9 inducer to activate the synthesis of the corresponding transporters. For instance, MalT activates transcription of mal genes based on the level of maltose transport into the cell (Richet et al, 2012). A second means is via regulation of transporter trafficking through the endocytic pathway. Cycling of transporters between the plasma membrane and intracellular sites allows cells to rapidly respond to continuous changing surroundings. This mechanism has been well demonstrated for canalicular ABC transporters such as the bile acid transporter ABCB11 (BSEP) and the organic cation transporter ABCB (Wakabayashi et al, 2006). Finally, lipids contribute to the regulation of ABC transporter activity. These transporters undergo substantial conformational changes during each transport cycle. The energy cost associated with lipid bilayer deformation can directly affect the protein activity through hydrophobic interactions (Lundbaek et al, 2010). It is thus unsurprising that a number of ABC transporters exhibit different ATPase activities in lipid environment and detergent (Borths et al, 2005; Reich-Slotky et al, 2000). Moreover, the charge of lipid head-group also has an influence on the transport activities as in the case of the transporter associated with antigen processing (TAP) and the lolCDE complex involved in the release of lipoproteins from membranes (Miyamoto & Tokuda, 2007; Scholz et al, 2011). 1.2.3 Periplasmic binding proteins For prokaryotic ABC import systems, substrate transport requires an additional periplasmic binding protein (PBP) that specifically captures the substrate. The nomenclature of periplasmic binding protein reflects the fact that PBPs were first discovered in the periplasm of gram-negative bacteria. In microorganisms lacking an outer membrane and periplasm such as gram-positive bacteria and Archaea, PBPs are attached to the cytoplasmic membrane using either a lipid anchor or a membrane-embedded peptide (Berntsson et al, 2010). In some cases, single or 10 multiple PBPs are also fused to the TMDs of ABC transporters, giving rise to two, four or even six PBP domains per functional transporter complex (van der Heide & Poolman, 2002). The function of PBPs anchored to the membrane or fused to TMDs is postulated to increase the effective concentration of substrates near transporters (Fulyani et al, 2013). PBPs generally consist of two symmetrical domains, connected by a hinge region that flanks the central substrate-binding pocket (Berntsson et al, 2010); Figure 1.5). Substrate binding causes the rotation of two domains toward each other and stabilizes a closed conformation to capture substrate with high-affinity, a process called ?Venus Fly-trap?(Quiocho & Ledvina, 1996). It is noteworthy that although fast single-step kinetics of substrate binding is common among many PBPs, the binding of substrate to PBPs was assumed to be a two-step binding process (Ledvina et al, 1998); Figure 1.5). In the first step, a substrate binds to its cognate PBP in the open conformation. This binding causes the bending motion of the hinge region and corresponding closure of the two domains of PBP in the second step. This two-step binding mechanism has been supported by kinetic measurements of substrate binding for a number of PBPs such as phosphate-binding protein PhoS and ectoine-binding protein TeaA (Ledvina et al, 1998; Marinelli et al, 2011). However, the kinetics of the second step might be very fast as compared to the first step for many PBPs and the corresponding substrate-binding kinetics can therefore be interpreted by a one-step binding mechanism. 11 Figure 1.5 Substrate binding to PBPs. The two-step substrate binding process is exemplified by maltose binding to MalE. Structures of MalE in (left, PBD code: 1JW4) open, maltose-free, (middle, PDB code: 1JW5) open, maltose-bound and (right, PDB code: 1ANF) close, maltose-bound. The maltose molecule is shown in stick model (red). PBPs are divided into six classes based on structure (Berntsson et al, 2010). The main difference is the structure of the hinge region between the two symmetrical domains of PBPs (Figure 1.6). Class I contains PBPs having a rigid ?-helix as the hinge region. The PBPs in class II have three interconnecting segments between the two domains. The hinge region of PBPs in class III consists of an extra domain and they are significant larger in size than the others. The two domains of PBPs in class IV are connected by two relatively short hinges. The hinge region of PBPs in class V contains a large helix. The hinge region of PBPs in class VI is similar to that in class IV. However, the length of these hinges is increased by almost two-fold, creating more flexibility inside the PBPs in class VI. 12 Figure 1.6 Structures of PBPs in I-VI classes. Proteins shown in these classes are: a, the vitamine B12 binding protein BtuF (PDB code: 1N2Z) in class I; b, the ribose binding protein RBP (PDB code: 1DRJ) in class II; c, the oligopeptide binding protein OppA (PDB code: 3DRF) in class III; d, the maltose binding protein MalE (PDB code: 1JW4) in class IV ; e, the ectone binding protein UehA (PDB code: 3FXB) in class V; f, the histidine binding protein HisJ (PDB code: 1HSL) in class VI. The hinge region is colored in orange. 1.3 Maltose transport system The maltose transporter has been a prototype to study molecular mechanisms of ABC transporters. The transporter (MalFGK2) consists of two membrane-integral subunits, MalF and MalG, and two copies of the ATPase subunits, MalK (Figure 1.2a). As a type I ABC transporter, MalF and MalG contain 8 and 6 transmembrane helices, respectively. In addition, transport of 13 maltose across membranes requires a periplasmic maltose-binding protein (MalE), a class IV periplasmic binding protein. 1.3.1 Coupling of ATP hydrolysis to maltose transport Reconstitution of maltose transport in proteoliposomes demonstrates that ATP hydrolysis by MalK provides the energy required for transport reaction (Davidson & Nikaido, 1990; Dean et al, 1990a). The results have shown that MalE stimulates the ATPase activity of MalFGK2 from ~10 nmol/min/mg to ~40 nmol/min/mg (Davidson et al, 1992; Gould et al, 2009). In comparison, addition of MalE and maltose increases the ATPase activity of MalFGK2 to ~400 nmol/min/mg (Davidson et al, 1992; Gould et al, 2009). The stoichiometry is approximately two to ten molecules of ATP hydrolyzed per molecule of substrate transported, suggesting that uncoupled ATP hydrolysis to maltose transport might also occur (Davidson & Nikaido, 1990). Since then, a significant amount of effort has concentrated on how MalE stimulates ATPase activity of MalFGK2 and how ATP hydrolysis is coupled to maltose transport. To elucidate the details of the interaction between MalE and MalFGK2, Shuman and colleagues have isolated MalFGK2 mutants in which maltose transport occurs in the absence of MalE (Covitz et al, 1994). Among these mutants, there is a positive correlation between MalE-independent transport activity and ATPase activity (Covitz et al, 1994), indicating that high ATP hydrolysis rates can preclude the requirement for MalE. For the wild type transporter, the interaction between MalE and TMDs must thus cause a trans-membrane signal that stimulates ATP hydrolysis by MalK, which is bypassed for the MalE-independent mutants through the mutations in TMDs. Furthermore, the apparent Km value of MalE-independent mutants for maltose is over 1000-fold higher than that of wild type in the presence of MalE (Covitz et al, 14 1994; Shuman, 1982; Treptow & Shuman, 1985). Clearly, MalE also increases the transport efficiency by helping the transporter capture maltose. To reveal how ATP binding and hydrolysis are coupled to conformational changes of MalFGK2, initial studies using isolated MalK showed that ATP binding and hydrolysis cause the dimerization of MalK (Chen et al, 2003; Lu et al, 2005; Smith et al, 2002). This has led to the conclusion that conformational changes of MalK upon ATP binding and hydrolysis are coupled to alternation of TMDs conformations. However, the exact role of ATP in inducing the conformational change of TMDs remains controversial. Crosslinking and limited proteolysis experiments have indicated that ATP binding promotes closure of the MalK dimer, presumably by changing the transporter?s conformation from inward- to outward-facing (Daus et al, 2007b; Daus et al, 2006; Hunke et al, 2000b). This is consistent with biochemical studies on other ABC transporters such as Molybdate and Vitamin B12 ABC transporters (Gerber et al, 2008; Joseph et al, 2011). In contrast, electron spin resonance spectroscopy (EPR) studies suggest that ATP binding alone is not able to trigger the conformational change (Orelle et al, 2008), which was used to explain the low ATPase activity of MalFKG2 reconstituted in proteoliposomes. It is presently unclear whether this discrepancy is caused by differences in experimental conditions. In addition, it must be noted that the EPR studies are performed in detergent solution. Since MalFGK2 in detergent shows a high ATPase activity (1000 nmol/min/mg), it is difficult to explain the inability of ATP to change the conformation of MalFGK2. This controversy calls for a rigorous characterization of the conformational changes of the transporter in relationship to the ATP hydrolytic cycle. 15 1.3.2 Structural analyses of the maltose transporter The crystal structures of MalFGK2 have been solved in two conformations: the inward-facing and the outward-facing states (Khare et al, 2009; Oldham et al, 2007). In the inward-facing conformation, the maltose-binding site of the TMDs is exposed to the cytosol and the ATP-binding sites of the MalK dimer are separated (Figure 1.4a). The crystal structure of MalFGK2, obtained in the presence of MalE and nonhydrolyzable ATP analogs, reveals an outward-facing conformation (Figure 1.4b). The maltose binding site opens toward the periplasm and the ATP-binding sites are bound together at the closed MalK dimer interface. MalE in a maltose-free open conformation is docked onto the periplasmic surface of MalF and MalG. The large periplasmic P2 loop of MalF, not being resolved in the structure of the inward-facing state, binds and orients MalE for interactions with MalFG. More recently, a complex of maltose-bound closed MalE with MalFGK2 (Oldham & Chen, 2011a), captured in the absence of nucleotides, shows a semi-open conformation of MalK dimer and inward-facing of MalFG. Together, these structures provide strong evidence for the alternating access model and the concerted conformational changes between TMDs and NBDs. In addition, these structures have given valuable insights into understanding the mechanism of maltose transport. First, the substrate translocation pathway has been revealed. In the outward-facing conformation, a maltose molecule is placed in the bottom of a large cavity (Oldham et al, 2007), which is located at the interface of MalE, MalF and MalG. This cavity is shielded from the cytoplasm by TM helices of MalF and MalG and is further sealed by MalE at the periplasmic side. Since maltose makes complementary interactions with ten surrounding residues from MalF, this cavity presumably represents the substrate-translocation pathway. Consistent with this structural data, six of the surrounding residues have been shown to play 16 important roles in maltose transport through mutagenic studies (Ehrle et al, 1996; Steinke et al, 2001). Second, entry of maltose into the translocation pathway is controlled by a periplasmic gate region (Khare et al, 2009). In the inward-facing conformation, the substrate translocation pathway narrows toward the periplasm and a hydrophobic gate beneath the interface of TMDs with MalE prevents entry of maltose (Figure 1.7). This gate comprises four loops, two from MalF and two from MalG. The loops are located at the end of kinked TM helices of MalF and MalG (TM 5 and 7 of MalF; TM 3 and 5 of MalG). MalE-independent transporter variants all contain mutations in regions proximal to this gate (Covitz et al, 1994), corroborating its critical role in regulating conformational changes of the transporter. Perhaps these mutations destabilize the gate region and thus disrupt the conformational equilibrium of MalFGK2. Nevertheless, the results imply that the gate region is essential to couple ATP hydrolysis to substrate transport. Consistent with this hypothesis, mutations at the gate regions have either eliminated substrate transport or uncoupled transport from ATP hydrolysis in the amino acid ABC transporter (Weidlich et al, 2013). Lastly, comparison of structures in the inward- and the outward-facing conformations reveals that the alternating access mechanism is achieved through rigid-body rotations of TMDs (Khare et al, 2009). This conformational change features a 22? rotation of the MalF core (MalF TM 4-7) and a 23? rotation of the MalG core (MalG TM 2-5) during the conformational transition of the maltose transporter from the inward- to the outward-facing state, resulting in re-orientation of the substrate translocation pathway. To accommodate this motion of TMDs, the helical domain of MalK rotates around the coupling helices of MalFG, which is orchestrated by the interaction of MalK with the ?-phosphate of ATP. Thus, these structures demonstrate how ATP binding results in such a large conformational change over 80? from 17 intracellular to extracellular sides of the membrane, without disrupting the association between TMDs and NBDs through rigid body rotations. Figure 1.7 The gate region in the maltose transporter. The gate region (red) is viewed along the membrane from the periplasm in the inward-facing (a) and outward-facing state (b). Furthermore, crystal structural analyses reveal the mechanism of ATP binding and hydrolysis by MalK. In the outward-facing conformation, the ATP ?-phosphate tightly tethers the two MalK proteins through interaction with the Walker A motif, the Q-loop, and the switch region of one MalK and the LSGGQ motif of the other, promoting the ATP-dependent closure of the MalK dimer (Oldham & Chen, 2011b). Comparison of structures of maltose transporter, obtained in the presence of AMPPNP or ADP plus a ?-phosphate mimic [vanadate (VO33-)], shows that the glutamate following the Walker B motif acts as a general base to polarize the water nucleophile, which attacks the ATP ?-phosphate group via a pentacovalent transition-state (Oldham & Chen, 2011b). It is thus concluded that ATP hydrolysis occurs via a general base mechanism. 18 1.4 Function of the maltose binding protein (MalE) MalE is a class IV PBP consisting of two distinct globular domains (named domains N and C) connected by a hinge region (Figure 1.6d); maltose is bound and sequestered in the deep groove between the two domains (Sharff et al, 1992); Figure 1.5). In the absence of maltose, MalE takes on a more open conformation in which the two domains are far apart and the groove is accessible to solvents. Maltose binding promotes a bending motion of the hinge and closing of the groove. Bound maltose experiences rapid exchange with a Koff of ~ 90 s-1 (Miller et al, 1983), indicating that the two domains open and close readily even in the presence of maltose. This property has attracted great interest to understand the relationship between MalE conformations and the affinity of the protein for maltose. It is found that in the absence of maltose, the hinge region and the maltose-binding site of MalE cannot maintain the protein in the open conformation (Telmer & Shilton, 2003). Instead, an interface on the opposite side of the hinge seems to prevent the closure of the N and C domains (Telmer & Shilton, 2003), whereas this interface is broken in the closed conformation. Consistently, destabilization of this interface stabilizes MalE in the closed conformation and increases the affinity for maltose (Marvin & Hellinga, 2001; Telmer & Shilton, 2003). Thus, this interface plays an active role in regulation of MalE conformation, known as the ?balancing interface?. Previous studies have demonstrated that MalE stimulates the ATPase activity of MalFGK2 by ~ 4 fold, whereas addition of MalE and maltose increases the ATPase activity of MalFGK2 by ~40 fold (Davidson et al, 1992; Gould et al, 2009). Using the ATPase and transport assays, the apparent affinity of MalE for MalFGK2 is ~14-100 ?M (Dean et al, 1992; Gould et al, 2009). A model was thus proposed that maltose-bound, closed MalE interacts with low affinity to the transporter in the inward-facing conformation, which then triggers a concerted 19 motion of MalK closure upon ATP binding, formation of the outward-facing MalFG and opening of MalE (Oldham et al, 2007; Figure 1.8). Reorientation of MalFG transfers maltose from MalE to the substrate-binding site of the TMDs and positions ATP at the catalytic site for hydrolysis. Subsequent release of ADP and inorganic phosphate (Pi) resets the transporter to the inward-facing conformation and maltose is released into the cell. In agreement with this model, MalE binds to the transporter with high affinity in the presence of ATP and vanadate (Chen et al, 2001). In addition, the transporter-bound MalE is in the open conformation and exhibits much lower affinity for maltose (Austermuhle et al, 2004). However, an essential question of how maltose regulates the complex formation remains largely unaddressed. Nevertheless, this model is consistent with the idea that MalE shuttles back-and-forth to the transporter to deliver maltose because MalE is found soluble in the periplasm. Figure 1.8 The conventional model for maltose transport. Closed liganded MalE triggers the outward-facing conformation. MalE binds maltose (Kd ~ 2 ?M) and then associates with the inward-facing transporter (Kd ~ 100 ?M). The transition to the outward-facing conformation facilitates the opening of MalE and the release of maltose to the MalFG cavity. Upon ATP hydrolysis, the transporter returns to the inward-facing state and maltose is released in the cytosol. In this model, MalE facilitates the pairing of MalK and the ATP hydrolysis step. Despite the prevalence of the model, it is not in agreement with certain studies. First, maltose-free MalE interacts with the transporter. Both ATPase assays and EPR spectroscopy 20 have indicated the complex formation of MalFGK2 with MalE in the absence of maltose (Austermuhle et al, 2004; Gould et al, 2009). In line with this notion, excess MalE can inhibit transport when maltose is held at sub-stoichiometric level (Merino et al, 1995). Second, retention of maltose by MalE might not be required for transport to occur. MalE variants possessing high affinity for maltose show impaired ability to stimulate ATPase activities of MalFGK2 (Gould et al, 2009). In contrast, a MalFGK2 variant is able to transport lactose only in the presence of MalE (Merino & Shuman, 1997). Since MalE cannot bind lactose, the results indicate the activation of the maltose transporter by maltose-free MalE. Finally, if maltose-bound MalE shows a low affinity for the transporter (~100 ?M), the transport efficiency may be weak at low maltose concentration because the periplasmic concentration of MalE depends on maltose. This is inconsistent with the notion that MalE facilitates transport especially at limiting substrate concentration. Altogether, maltose-free MalE plays a critical role in the transport reaction, which remains to be elucidated. 1.5 Inhibition of the maltose transporter by glucose specific enzyme EIIAGlc Uptake of maltose into bacteria is subjected to carbon catabolite repression (CCR), which allows for the preferential utilization of carbohydrates that support fastest growth (Gorke & Stulke, 2008). For instance, the availability of preferred carbohydrates such as glucose represses the synthesis and activities of proteins required for the transport and metabolism of secondary carbon sources such as maltose, lactose and glycerol (Deutscher et al, 2006). This regulatory phenomenon is achieved through different pathways in different microorganisms. 1.5.1 The Phosphoenolpyruvate carbohydrate phosphotransferase system In enteric bacteria, CCR is mediated by the phosphoenolpyruvate (PEP) carbohydrate phosphotransferase system (PTS) (Deutscher et al, 2006; Postma et al, 1993; Figure 1.9). The 21 PTS is a multicomponent system that is involved in the transport and simultaneous phosphorylation of preferred carbohydrates (PTS carbohydrates) as well as the regulation of metabolic pathways for secondary carbohydrates. It consists of at least three distinct components: the enzyme I (EI), the phosphocarrier protein (HPr), and several sugar-specific enzymes (EIIs). In most cases, EI and HPr are ?general? cytoplasmic components that are responsible for the phosphorylation of PTS carbohydrates (Deutscher et al, 2006). The EII enzymes, by contrast, are carbohydrate specific and bacteria usually contain different EIIs. Each EII complex consists of at least one transmembrane domain and two cytoplasmic domains such as the EIICBGlc-EIIAGlc complex for glucose in E. coli. In a sense, the transport pathways constituted by the specific EII complexes are connected to a common PEP/EI/HPr phosphoryl transfer pathway. For instance, transport across the membrane of glucose leads to the transfer of a phosphoryl group from PEP to the glucose molecule via phospho intermediates of EI, HPr, and EIICBGlc-EIIAGlc. 22 Figure 1.9 Inhibition of the maltose transporter by carbon catabolite repression. Glucose is transported and phosphorylated by the PTS. This caused an increased level of unphosphorylated EIIAGlc, which inhibits the transport activity of MalFGK2 via a yet unclear mechanism. As a consequence, the glucose uptake and corresponding phosphoryl transfer reaction increases the level of unphosphorylated EIIAGlc, which is able to reduce the utilization of secondary carbohydrates through several means. Unphosphorylated EIIAGlc directly binds and inhibits several secondary-carbohydrate transporters including the maltose transporter (MalFGK2), the lactose permease (LacY) and the melibiose permease (MelB) (Bluschke et al, 2006; Mitchell et al, 1982; Sondej et al, 2002). It also prevents transport of glycerol by binding to glycerol kinase, which is essential for glycerol metabolism (van der Vlag et al, 1994). In addition, the secondary carbohydrates or their derivatives are also inducers to regulate the expression of corresponding transporters and metabolic enzymes (Deutscher et al, 2006). Inhibition of uptake of these secondary carbohydrates would thus downregulate the synthesis of the corresponding metabolic system. Furthermore, phosphorylated EIIAGlc stimulates cAMP synthesis, which activates transcription of genes for many metabolic enzymes of secondary carbohydrates. Increased levels of unphosphorylated EIIAGlc would thus decrease the transcription of these genes. 1.5.2 Action of EIIAGlc on the maltose transporter EIIAGlc is composed of a disordered N-terminal tail (1-18) and an antiparallel ?-sandwich (residues 19-168) (Worthylake et al, 1991; Figure 1.10). Previous studies have shown that addition of EIIAGlc inhibits the transport and ATPase activity of MalFGK2 in proteoliposomes (Landmesser et al, 2002). Mutagenic analyses and peptide arrays have identified two possible binding sites of EIIAGlc for MalFGK2. One encompasses residues on ?-strands 5-7, whereas the 23 other comprises residues located on ?11 and ?2 (Bluschke et al, 2006). In addition, it is reported that the N-terminal tail of EIIAGlc is essential for the inhibition of maltose transporter (Bluschke et al, 2006). Since a peptide corresponding to this N-terminal tail strongly binds to PG lipids (Wang et al, 2000b), it is postulated that this N-terminal tail serves as a lipid anchor for EIIAGlc binding to MalFGK2. Mutations that confer MalFGK2 resistance to EIIAGlc inhibition are located on the NBD and C-terminal domain of MalK. The fact that these mutations cluster on opposite faces of the MalK monomer, but lie on the same face of the dimer, leads to the proposal that the binding site of the maltose transporter for EIIAGlc spans the dimer interface of MalK (Bohm et al, 2002; Dean et al, 1990b; Kuhnau et al, 1991). Although these studies have provided great details on the interaction between EIIAGlc and MalFGK2, the mechanism of EIIAGlc inhibition remains largely unclear. Figure 1.10 A model of EIIAGlc binding to the membrane. The N-terminal domain (shown in red) of EIIAGlc is disordered in solution and forms a ?-helical structure when bound to acidic lipids. 24 1.6 Thesis investigation Despite extensive structural characterizations, many key questions remain regarding how MalE and EIIAGlc regulate the transport reaction. First, it is still under debate about whether ATP is able to trigger the conformational change of MalFGK2. Second, MalE interacts with MalFGK2 in the absence or presence of maltose, raising questions as to the specificity of MalFGK2 in recognizing the conformational state of MalE. Third, what ensures the effective coupling of ATP hydrolysis to substrate transport is unresolved. Finally, where and how EIIAGlc binds to the transporter remains unclear. To address these questions, we set out to perform a quantitative dissection of the functional consequences of MalE and EIIAGlc binding on the transport activity of MalFGK2. We purified the maltose transporter in detergent micelles and reconstituted it into proteoliposomes and nanodiscs for biochemical studies on the interactions of MalFGK2 with MalE and EIIAGlc (Appendix B). Although the compartmentalization of proteoliposomes suffices for determining transport activity, it precludes the characterization of membrane proteins by techniques for soluble proteins, due to many of its inherent properties such as water-insolubility and light scattering. Nanodiscs provide an ideal solution to these problems by placing membrane proteins in a defined lipid environment yet rendering them soluble in aqueous solution (Bayburt & Sligar, 2010). Thus, our experimental systems allow us to analyze the binding of MalE and EIIAGlc to MalFGK2 and the consequent effect on the transport activity. Nanodiscs containing MalFGK2 (Nd-FGK2) were analyzed by light scattering and electron microscopy to determine their size and diameter. ATPase assays and circular dichroism spectroscopy were used to monitor lipid effects on the activity and stability of Nd-FGK2. 25 Next, we developed a fluorescence quenching assay to determine the binding of MalE to MalFGK2. The effect of maltose and ATP on this complex formation was determined. Using MalE mutants stabilized in the open- or the closed- state, we explored whether MalFGK2 selectively binds MalE in certain conformations. In addition, we examined the ATP hydrolytic cycle to understand how the ATPase activity of MalFGK2 is stimulated by MalE and maltose. We then related these kinetic measurements to the transporter activity using both in vitro and in vivo transport assays. To monitor the conformational changes of MalFGK2, we employed crosslinking and fluorescence assays. The transporter needs to alternate between the inward- and outward-facing conformations during each transport process. The effect of maltose, MalE and ATP on these conformational changes is investigated. We also isolated the lone transmembrane domain MalFG and studied how reassembly of MalK regulates MalFG conformation. Finally, we studied the interactions between MalFGK2 and EIIAGlc. To understand the molecular mechanism of EIIAGlc inhibition, we reconstituted its inhibitory effect on the ATPase activity of MalFGK2 in proteoliposomes and nanodiscs. We studied how lipids are able to increase the binding and inhibition of EIIAGlc onto MalFGK2. Using chemical crosslinking and automatic protein docking, we provided a model of the MalFGK2- EIIAGlc complex. Analyses of the ATPase cycle revealed the mechanism of how EIIAGlc inhibits the ATPase activity of MalFGK2. Together, these results unraveled the mechanism of EIIAGlc inhibition on the maltose transporter. 26 Chapter 2: Action of membrane lipids on the maltose transporter MalFGK2 2.1 Introduction An essential question in ABC transporter research is how ATP hydrolysis is coupled to substrate transport (Chen, 2013). For the maltose transporter MalFGK2, the inward-facing conformation has low ATPase activity because the two NBDs on the MalK dimer are positioned apart (Khare et al, 2009). The outward-facing conformation is instead poised for ATP hydrolysis because the two NBDs contact each other (Oldham et al, 2007). The mechanism that ensures the communication between NBDs and TMDs, and therefore the effective coupling between ATP hydrolysis and substrate transport remains unclear. It is suspected that a pair of conserved helices located at the interface between the two domains (hence termed coupling helices) serves to communicate the conformational changes (Oldham et al, 2007). Yet, the coupling mechanism is not strict and varies greatly among transporters. It has been shown that the rate of ATP hydrolysis and the rate of substrate transport can differ by 1?3 orders of magnitude (Borths et al, 2005; Davidson & Nikaido, 1990; Patzlaff et al, 2003; Poolman et al, 2005), and the reasons for this remain unknown. Interestingly, when extracted from the membrane with detergent, the ATPase activity of MalFGK2 becomes uncoupled and increases by ~ 100 fold (Reich-Slotky et al, 2000). This observation would suggest that membrane lipids play an important role in restricting the transporter ATPase activity and maintaining dependence on the substrate. Detergents however, are also known to interfere with protein structures and protein interactions, so it is difficult to firmly conclude regarding the contribution of lipids to the coupling reaction. Furthermore, it is impossible to measure the effect of lipids and lipid bilayers in the heterogeneous and polydisperse lipid-27 detergent micelles. In contrast, the nanodisc supports native-like membrane bilayers that can be modified in their lipid content (Denisov et al, 2004). In this study, we employed the nanodisc to demonstrate the importance of membrane lipids for the ABC transporter activity and coupling mechanism. The result presented below shows that i) different types and amounts of lipids can be incorporated into the disc, ii) the lipids with long acyl chains increase the thermodynamic stability of the transporter, iii) the stability imposed by the lipids reduces the transporter basal ATPase activity and iv) the low basal ATPase activity of the transporter is essential for the efficient coupling to maltose. We illustrate the importance of this relationship with the mutant MalF500. The mutant is unstable, it hydrolyzes large amounts of ATP, and it transports maltose inefficiently. Coupled or not, we show that MalE is bound with a high affinity to the outward-facing transporter, and excess maltose promotes the dissociation of MalE from MalFGK2. 2.2 Material and methods 2.2.1 Materials Detergent n-dodecyl ?-D-maltoside (DDM) was purchase from Anatrace. All lipids were purchased from Avanti Polar Lipids. Superose 6 10/300 GL, Superdex 200 10/300 GL, Resource 15Q and Ni2 +-NTA chelating Sepharose columns were obtained from GE Healthcare. BioBeads were purchased from BioRad. MalFGK2 and MalE were purified as described (Appendix A.1 and A.2). Other chemicals were obtained from Sigma. 2.2.2 Reconstitution of MalFGK2 The membrane scaffold MSP1D1 was obtained from the Sligar laboratory (Denisov et al, 2004). Phospholipids were dissolved in chloroform and dried under a steam of nitrogen. The lipids were resuspended in 50 mM Tris?HCl, pH 8, 50 mM NaCl, and 0.1% DDM.A typical 28 reconstitution experiment consists in mixing together the MalFGK2 complex, the MSPs and the solubilized lipids at a protein:MSP:lipid ratio of 1:3:60 or 1:3:400 in TSG buffer (50mM Tris-HCl, pH 8; 50 mM NaCl; 10% glycerol) containing 0.08% DDM. Detergent was slowly removed with BioBeads (1/3 volume) and gentle shaking (overnight, 4 ?C). BioBeads were removed by sedimentation and the sample was frozen in liquid nitrogen before storage at ? 80 ?C. Proteoliposomes were prepared at a protein: lipid ratio of 1:2000. The proteoliposomes were harvested by centrifugation (100,000 ?g, 1 h, 4 ?C), resuspended in 20 mM Tris?HCl (pH 8) and frozen in liquid nitrogen before storage in ? 80 ?C. Proteoliposomes were extruded through a 400-nm polycarbonate filter before use. 2.2.3 Maltose transport assay Proteoliposomes were reconstituted as described above in buffer A (50 mM sodium phosphate, pH 6.5 and 1 mM MgCl2) containing 5 mM ATP, then resuspended in buffer A and kept on ice until use. Transport reaction was carried out with 10 ?M MalE, 0.5 ?M MalFGK2 proteoliposomes and 20 ?M [14C]-maltose (57 ?Ci/?mol). At the indicated time, samples (50 ?L) were removed and diluted into 1 mL ice-cold sodium phosphate buffer (50 mM) containing maltose (10 mM) followed by filtration through a 0.22 ?M nitrocellulose Millipore filter and scintillation counting. The total cpm values were converted into nmol using known amounts of [14C]-maltose. 2.2.4 Dynamic and static light scattering Dynamic light scattering analysis was performed on a DynaPro nanostar instrument (Wyatt Technology) using a 1 ?L inner volume quartz cuvette. Data were fitted using the DYNAMICS software (Wyatt Technology) to approximate particle diameters. Static light scattering was performed using a Superdex 200 HR 10/10 column connected to a miniDAWN 29 light scattering detector and interferometry refractometer (Wyatt Technologies). The data were recorded in real time and the molecular masses were calculated using the Debye fit method using the ASTRA software (Wyatt Technology). 2.2.5 Lipid phosphorus analysis of nanodiscs Lipids were extracted following the method of Folch et al. (1979). Briefly, nanodiscs were incubated in 10 volumes of chloroform/methanol (2:1) for 10 min at room temperature. The solvent was mixed with 0.9% NaCl (2 volumes) and centrifuged (2000 rpm, 1 min). The lipids in the lower phase were dried under nitrogen and hydrolyzed in sulfuric acid (9 N, 200 ?L) for 30 min at 200 ?C. The reaction was completed with hydrogen peroxide (30%, 20 min), and released Pi was determined using the malachite green method (Zhou & Arthur, 1992). 2.2.6 Electron microscopy The nanodisc particles were diluted and applied to a freshly glow-discharged carbon coated EM grid. Excess fluid was removed and the samples were stained with 2% urany1 acetate. Images were collected on a Tecnai F20 microscope equipped with a Gatan Ultrascan 4000 4 k ? 4 k CCD Camera System at a nominal magnification of 50,000 ?, given a pixel size of 2.3 ? per image. The average size and standard deviation obtained after measurement of ~ 100 spherical particles was calculated using Digital Micrograph or the E2display module of the Eman2 image processing suite (Tang et al, 2007). 2.2.7 Circular dichroism spectroscopy Circular dichroism (CD) spectroscopy was performed in 50 mM sodium phosphate buffer (pH 7.5) using a Jasco Model 810 spectropolarimeter equipped with a Peltier device. The measurements were repeated three times with baseline correction in the 215?300 nm range (1 nm bandwidth, 1 s respond time). 30 2.2.8 Other methods The MalFGK2 ATPase activity was determined by monitoring the release of inorganic phosphate using the malachite green method (Zhou & Arthur, 1992). Linear gradient blue-native (BN), colorless-native (CN) gel electrophoresis and Sucrose gradient analysis were performed as described (Dalal & Duong, 2010). MalE was iodinated using the Iodogen reagent (Pierce-Thermo Scientific). The specific activity of [125I]-MalE was ~ 2 ? 105 cpm/?g. The detection of [125I]-MalE and 14C-labeled maltose (57 ?Ci/?mol, Molecular Probes) was performed using a phosphorimager scanner. 2.3 Results 2.3.1 Reconstitution of MalFGK2 into nanodiscs The formation of a nanodisc is a self-assembly process initiated by mixing together the membrane scaffold protein (MSP), the target membrane protein and the phospholipids. The final protein ratio in the disc, which is formed upon removal of the detergent, is mostly invariable (two MSPs per disc; Denisov et al, 2004), but the lipid stoichiometry can be modified. For this analysis, MalFGK2 was reconstituted with the scaffold protein MSP1D1 and DOPC lipids, either at low lipid ratio or high lipid ratio (1:3:60 and 1:3:400, respectively). On native-gel (Figure 2.1a), the discs formed at low lipid ratio migrated above the discs formed without added lipids (compare lane 1 with lane 2). In contrast, the discs formed at the high lipid ratio migrated as a smear (lane 3). Dynamic light scattering confirmed the heterogeneity of the latter preparation, containing particles with diameter ranging from 14 nm to > 200 nm (Figure 2.1b). The smaller particles, also termed nanolipoparticles (NLPs; Chromy et al, 2007), were separated by gel filtration (Figure 2.1c). Each fraction was analyzed by native-PAGE and SDS-PAGE, and their lipid content was determined by total phosphorus analysis (Figure 2.1d; Table 2.1). The ratios of 31 MSP to MalFGK2 were identical across the different fractions, showing that the protein stoichiometry in the disc is the same (2 MSP per MalFGK2), as expected. In contrast, the phosphorus analysis indicated an increasing amount of phospholipids (Table 2.1). Figure 2.1 Isolation of MalFGK2 nanodiscs made at low and high lipid ratios. a, the reconstitutions were performed without added lipids (lane 1), at low ratio DOPC lipids (1:3:60; lane 2), or at high ratio DOPC lipids (1:3:400; lane 3). The material was analyzed by native-gel and Coomassie blue staining. The position of molecular weight makers (in kDa) is indicated. b, dynamic light scattering analysis of theMalFGK2 nanodisc preparation at low(Left) and high lipid ratios (Right). Samples were analyzed on a DynaPro nanostar instrument (Wyatt Technology) using a 1 ?L inner volume quartz cuvette. Data were fitted using the DYNAMICS software (Wyatt Technology). c, the MalFGK2 nanodisc preparations were applied onto a Superose 6 10/300 column equilibrated in TSG buffer. d, the collected fractions (F) were analyzed by SDS-PAGE (left panel) and native-PAGE (right panel). The nanodisc preparation (Nd) made at low lipid ratio is loaded on the right lane for visual reference. The position of molecular weight makers (in kDa) is indicated. 32 Table 2.1 Mass, diameter and number of lipid molecules per nanodisc particles. a Determined by static light scattering as in Figure 2.2a. b Determined by electron microscopy as in Figure 2.2b. c Based on the molecular weight obtained by light scattering and assuming a stoichiometry MSP:MalFGK2 of 2:1. The applied equation was MWNLP = MWFGK + 2* MWMSP1D1+ N* MWDOPC where N is the number of lipid molecules per NLP. DOPC ~ 0.8kDa, MalFGK2 ~169 kDa; MSP1D1 ~24.6 kDa. d Lipids were extracted following the method of Folch et al (Folch et al, 1957). The released Pi was determined using the malachite green method as described (Zhou & Arthur, 1992). The standard deviation was obtained from 3 independent assays. e Based on the diameter determined by electron microscopy and assuming a perfectly circular shape. The applied equation was ?R2NLP = ?R2FGK + N* SDOPC where RNLP represents the radius of the NLP, RFGK the radius of the MalFGK2 transmembrane domain (6.2 nm), and SDOPC the surface area per DOPC lipid (72 ?2). f Nanodisc diameter was determined by SAXS in Denisov et al. 2.3.2 Characterization of the lipid-rich nanodiscs To determine the size of the particles and their lipid content of lipid-rich nanodiscs, each fraction was re-injected onto a gel filtration column coupled to a light scattering detector (Figure 2.2a). The size measured ranged from 330 kDa to 640 kDa. For comparison, the size of the discs formed at low lipid ratio was ~ 220 kDa (Table 2.1). Each fraction was also analyzed by negative-stain electron microscopy (Figure 2.2b). The images revealed homogeneous particles with a diameter ranging from 11 nm (for fraction F17) to 17 nm (for fraction F13). Since the EM images are projection images, it was not possible to distinguish discoidal from spherical Molecular weight a (kDa) Measured diameter b (nm) Calculated lipid content c Measured lipid content d Calculated lipid content e nanodiscs 220 ? 20 9.7 f 0 ~ 26 14 ? 5 108 NLP F17 330 ? 20 11 ? 2 140 ? 30 120 ? 30 160 ? 80 NLP F16 360 ? 30 11 ? 1 170 ? 30 140 ? 30 190 ? 70 NLP F15 440 ? 30 14 ?1 290 ? 40 230 ? 40 350 ? 90 NLP F14 570 ? 30 16 ? 2 440 ? 40 370 ? 30 400 ? 100 NLP F13 640 ? 50 17 ? 2 530 ? 60 460 ? 40 600 ? 100 33 assemblies. The exact geometry of the NLPs is still under debate in the field (Cavigiolio et al, 2008; Silva et al, 2008; Wu et al, 2011). Here, the images allowed us to calculate the number of DOPC molecules per particles, taking into account the particle diameter and the cross-section of a lipid molecule (Table 2.1 and legend for details on calculus). There were ~ 120?150 molecules in the 11 nm particles (fraction F17; 330 kDa) and ~ 550 molecules in the larger 17 nm assemblies (fraction F13; 640 kDa). In contrast, the discs reconstituted at the low lipid ratio (9.7 nm; 220 kDa) contained just a few lipid molecules as determined by phosphorus analysis (Table 2.1). Together, the data from three independent methods, light scattering, electron microscopy and total phosphorus analysis, were consistent with each other (Table 2.1). They show that the membrane scaffold proteins are flexible enough to trap the maltose transporter with varied amount of lipids. Furthermore, homogeneous and soluble nanoparticles containing MalFGK2 with different amounts of lipids can be prepared and isolated. Figure 2.2 Characterization of MalFGK2 nanodiscs made at high lipid ratio. a, the molecular weight of the particles in the collected fractions (F13 to F17) was determined on a Superdex 200 HR 10/10 column coupled to a mini DAWN light scattering detector and interferometry refractometer (Wyatt Technologies). The data were analyzed by Debye fit method using the ASTRA software. The results are reported in Table 2.1 with mean and standard deviations out of three repeats. b, the diameter of the 34 particles in the collected fractions (F13 to F17) was determined by negative stain electron microscopy. The particles were applied to a carbon-coated EM grid and stained with 2% urany1 acetate. Images were collected on a Tecnai F20 microscope at a magnification of 50,000X. The standard deviation was on calculated on ~100 particles using the Eman2 image processing suite. The data are reported in Table 2.1. 2.3.3 Single-particle EM analysis of nanodiscs To provide structural insight into the conformational dynamics of MalFGK2 in nanodiscs (Nd-FGK2), we sought to employ single-particle electron microscopy (EM). I purified nucleotide-free and nucleotide-bound Nd-FGK2. A complex of nucleotide-bound Nd-FGK2 with MalE was obtained in the presence of ATP and vanadate (Appendix D.1). However, the complex of nucleotide-free Nd-FGK2 with MalE is not stable. I therefore stabilized this complex using disulphide crosslinking between MalF205C and MalE80C (Appendix D.2). The above complexes were separated from excess MalE by size-exclusion chromatography. Clear native and SDS-PAGE shows that these purified complexes migrate as a monodisperse species (Figure D.1 and Figure D.2). 35 Figure 2.3 Electron microscopy analyses of MalFGK2 nanodiscs made at low lipid ratio. a, EM micrograph of MalFGK2 nanodiscs. b-c, 64 class avarage images (b) were used to caculate a 3D model (c). In collaboration with Dr. Isabelle Rouiller?s laboratory, we prepared negatively stained specimens of nucleotide-free Nd-FGK2 and examined its structure by EM (Figure 2.3). We selected ~4500 particles from 160 images and classified them into 64 classes. The class averages showed that MalF and MalG are encircled by the nanodiscs and MalK dimer extend 70 ? from the surface of the nanodiscs. These 2D averages were used to construct an initial 3D model, which resembles the crystal structure of nucleotide-free MalFGK2 in the inward-facing conformation. 2.3.4 Effect of the lipids on the transporter ATPase activity and coupling to maltose We tested the effect of the phospholipids on the transporter ATPase activity (Figure 2.4). With the particles formed at a low lipid ratio, the endogenous ATPase was high (~ 650 nmol/min/mg). In sharp contrast, the ATPase activity measured with any of the lipid-rich particles was much lower (~ 10 nmol/min/mg); an activity approaching that of the transporter reconstituted in proteoliposomes (labeled pL on Figure 2.4). In all cases, the addition of MalE stimulated the ATPase activity by ~ 3-fold (Figure 2.4), indicating that MalE binds to the transporter independently from the surrounding lipids. We next measured the effect of maltose. When the transporter was reconstituted at a low lipid ratio, maltose reduced the ATPase activity by ~ 2 fold, as previously reported. In sharp contrast, maltose stimulated the ATPase activity by ~ 10-fold when the transporter was reconstituted at a high lipid ratio (Figure 2.4). Together, the results suggest that phospholipids diminish the endogenous ATPase activity of the transporter and increase coupling to maltose. 36 Figure 2.4 Effect of the amount of membrane lipids on the transporter ATPase activity. The ATPase activity supported by the gel filtration fractions F13 to F17 (0.5 ?M in all cases) was measured at 37 ?C in the presence of MalE (2 ?M) and maltose (2 mM) as indicated. For comparison, the ATPase activity of MalFGK2 in nanodiscs made at a low lipid ratio or reconstituted in proteoliposomes (pL) is presented. The y-axis is scaled to show the results on the same plot. The error bar is derived from three independent measurements 2.3.5 The length of the lipid acyl chain is important for the transporter activity We reconstituted MalFGK2 at a high lipid ratio (1:3:400), but using four different types of synthetic lipids. Following gel filtration chromatography and phosphorus analysis, we selected and compared the particles containing a similar amount of lipids (Table 2.2). The ATPase measurements showed that the lipid head group did not have any significant effect on the transporter activity (compare DOPC to DOPG, Figure 2.4 and 2.5 respectively). In contrast, the length of the acyl chain had a major effect. Compared to longer lipids, the short acyl chain DLPG (12:0) was unable to restrict the transporter ATPase activity, and the transporter was uncoupled to maltose (Figure 2.5). This effect was not because of the transition temperature of the lipids; DLPG and POPG have a very similar Tm (Figure 2.5 and Table 2.2). The difference was also not because DLPG is less incorporated into the nanodisc compared to the other lipids 37 (Table 2.2). Thus, the length of the acyl chain seems to be an important lipid feature that contributes to the transporter ATPase activity and its coupling to maltose. Figure 2.5 The length of the lipid acyl chain is important for the transporter activity The ATPase activity of MalFGK2 reconstituted in nanodiscs with the indicated phospholipids at a high lipid ratio (1/3/400) was measured as above. The nanodiscs were separated from aggregates on a Superose 6 10/300 column equilibrated in TSG buffer. Table 2.2 Number of lipid molecules in nanodiscs reconstituted with different synthetic lipids. Measured lipid contenta Lipid Tm (?C) Lipid Chain length NLP-DLPG 160 ? 70 -3 12:0 NLP-DMPG 150 ? 60 23 14:0 NLP-POPG 150 ? 60 -2 16:0-18:1 NLP-DOPG 170 ? 70 -18 18:1 a. The discs reconstituted at high lipid ratio with the indicated lipids were separated by gel filtration chromatography, as in Figure 2.2a. In all case, fraction #17 was used to determine the lipid content and protein activity. The lipid content was measured as described in Table 2.1. The standard deviation was derived from three independent assays. 2.3.6 Lipids do not affect the binding between MalE and MalFGK2 Next, we examined the effect of lipids on the complex formation between MalE and MalFGK2. On native-PAGE, [125I]-MalE was tightly bound to MalFGK2 in the presence of the non-hydrolysable ATP analog AMPPNP. However, addition of maltose significantly decreased 38 the binding of MalE to MalFGK2 (Figure 2.6). This negative effect of maltose can also be observed by ATPase measurements. The ATPase activity of the transporter was higher with the mutant protein MalE-254, because the protein does not capture maltose and therefore does not dissociate from the transporter (Figure 2.4). Together, with or without lipids, it is found that maltose triggers the dissociation of MalE from the transporter. Figure 2.6 Effect of lipids on MalE binding to MalFGK2 nanodiscs. The binding of 125I-labeled MalE (~2 ? 105 cpm/?g; 0.5 ?M) to MalFGK2 reconstituted at low and high lipid ratios (fraction F17 is shown here) was analyzed by native-PAGE and autoradiography. Note that the protein migration is smeary when the discs are enriched with lipids. 2.3.7 Lipids increase the folding stability of the transporter Finally, we used CD spectroscopy to determine the effect of lipids on the structural stability of the transporter. The action of lipids on membrane protein folding is difficult to assess in detergent solution; the nanodisc allowed us to bypass this limitation (Figure 2.7). Compared to the nanodiscs made at a low lipid ratio, the amounts of guanidine-HCl required to unfold MalFGK2 reconstituted at a high lipid ratio increased by 0.8 M (34% increase, Figure 2.7b). To exclude the possibility of an indirect effect caused by an interaction of the lipid molecules with the MSPs, and also to confirm that the transporter instability leads to an increased ATPase 39 activity, we employed the mutant MalF500 (mutations G338R and N505I in MalF TMS5 and TMS8, respectively). It is suspected that the mutations increase the spontaneous conformational transitions of MalFG, and therefore hydrolysis of ATP occurs in a constitutive manner (Covitz et al, 1994). In support of this assumption, the ATPase activity of the mutant MalF500 in proteoliposomes was ~ 200 fold higher than the wild type, and independent from MalE and maltose (Figure 2.8a). In addition, the mutant displayed much lower protein stability, as indicated by the decrease of its unfolding transition midpoint (26% decrease compared to the wild type, Figure 2.7b). Finally, a maltose transport assay revealed the rate of maltose transport with the mutant MalF500 was ~ 5-fold lower than the wild type (Figure 2.8b). Altogether, the results demonstrate that lowering the energy state of MalFGK2 is essential for coupling the MalK ATPase activity to substrate transport. Figure 2.7 Effect of lipids on the transporter stability. a, the folding stability of MalFGK2 (1 ?M) in the nanodiscs made at low or high lipid ratios was monitored by circular dichroism. Examples of CD spectra obtained for the maltose transporter at different denaturant concentrations. Left: Nanodiscs (low lipids); middle: nanodiscs (high lipids); right: nanodiscs-F500 (low 40 lipids). b, the unfolded fraction at each guanidine-HCl concentration was calculated using the ellipticity at 222 nm and plotted as described (Greenfield, 2006). The data were fitted to a Boltzmann sigmoidal model to determine the folding transition midpoint: mutantMalF500 (1.7 M), nanodiscs made at low lipid (2.3 M) and nanodiscs made at high lipid ratio (3.1 M). Figure 2.8 The ATPase and transport activity of wild type and mutant MalF500. a, the ATPase activity of the wild type complex (upper panel) and mutant MalF500 (bottom panel) reconstituted in proteoliposomes (0.5 ?M) was determined as above. The ATP hydrolysis was measured at room temperature in the presence or absence of MalE (10 ?M) and maltose (1 mM). The error bar is derived from three independent experiments. b, the transport of [14C]-maltose in the proteoliposomes containing the wild type (upper panel) or mutant MalF500 (bottom panel) complex is described in the Material and Methods. 2.4 Discussion The efficient operation of ABC transporters depends, at least in theory, on preventing futile cycles of ATP hydrolysis in the absence of a proper liganded state. This can be achieved by maintaining the transporter in a stable conformation, away from the transition state. For the maltose transporter, this low energy conformation corresponds to the inward-facing state, with the ATP binding sites on MalK separated from each other (Khare et al, 2009). Our results using the nanodisc demonstrate that lipids have a critical role in maintaining the transporter away from 41 the transition state. Without conformational constraints, such as in nanodiscs containing a small amount of lipids or nanodiscs containing lipids with short acyl chains, the transporter is unstable and constitutively hydrolyses ATP. MalE and maltose are then unable to stimulate the transporter significantly, most likely because the transporter cycles freely and rapidly between the inward- and outward-facing states. Alternatively, in the absence of long enough chain lipids, the preferred conformation of the transporter may be the outward-facing conformation that constitutively hydrolyzes ATP. Either way, our current data provide direct evidence that lipids are needed to maintain the transporter away from the transition state. This allosteric action is essential because it allows the ATPase activity of the transporter to remain coupled to the substrate. Our analysis with the mutant MalF500 illustrates this later point well: the high basal ATPase activity of the mutant leads to poor dependence on maltose and poor transport efficiency. How membrane lipids maintain the transporter in a low energy conformation remains to be determined. The maltose transporter, like other ABC transporters, undergoes important conformational changes during catalysis, which unsurprisingly can be influenced by the nature of the lipid surrounding the transmembrane domains. The energy costs associated with lipid bilayer deformation can directly affect the protein activity through hydrophobic interactions (Lundbaek et al, 2010). A number of recent studies suggest that the lateral pressure imposed by the lipid bilayer, including elasticity, thickness and curvature can affect membrane protein conformation and function (Andersen & Koeppe, 2007; Charalambous et al, 2008; Dowhan & Bogdanov, 2012; Fyfe et al, 2001). In our study, the incorporation of threshold amounts of lipid with long enough acyl chains provokes the dramatic (~ 100 fold) decrease of the transporter endogenous ATPase activity. This decrease is not related to lipid head group since DOPG and DOPC produce the same effect. 42 It is also not related to the lateral pressure imposed by the packing of the lipids in the discs since the number of DOPG and DOPC lipids in the particles is the same (Table 2.2). It is also unlikely that the microviscosity around the transporter accounts for the lipid effect since DLPG and POPG possess very similar melting temperatures (Table 2.2). Thus, the possibility remains that the length of the acyl chain is directly affecting the motion of the TMDs, and also perhaps the coupling helices located at the interface between NBDs and TMDs. These helices are normally located near the membrane surface, and the nanodisc containing long acyl chain lipids may correctly reproduce this. It opens up the possibility that the thickness of the lipid bilayer plays an unsuspected role in regulating the activity of certain ABC transporters in the cell context. Future studies are needed to understand the action of lipids at the structural and cellular levels. These studies will be greatly helped with parallel studies with the nanodisc because the system is simple, controlled and well suited for monitoring the protein structural and functional changes as a function of the membrane lipid environment (Alvarez et al, 2010; Kawai et al, 2011). We are currently using single-particle EM analysis to visualize the influence of lipids on the structural dynamics of the maltose transporter. 43 Chapter 3: Discovery of an auto-regulatory mechanism for the maltose ABC transporter MalFGK2 3.1 Introduction In the maltose transport system, certain mutations in MalFGK2 render transport independent from the maltose-binding protein MalE (Covitz et al, 1994). In that case, translocation of maltose is strongly reduced as the Km,app for maltose increases from 2 ?M to 1 mM (Treptow & Shuman, 1985). The function of MalE is therefore essential to increase the affinity of the transporter for the substrate, and therefore the efficiency of transport. Since MalE is found in soluble form in the periplasm, it has naturally been proposed that the protein shuttles back-and-forth to the membrane to deliver maltose. The reconstitution of the reaction in proteoliposome and the crystallographic analysis of the transporter have completed the model (Davidson et al, 2008; Khare et al, 2009; Oldham & Chen, 2011a; Oldham et al, 2007). Upon binding of closed-liganded MalE, MalFGK2 switches toward the outward-facing conformation. This structural change forces the opening of MalE and the subsequent release of maltose inside the transporter cavity. After ATP hydrolysis, MalFGK2 returns to the inward-facing state, maltose is released in the cytosol, and MalE returns to the periplasm to capture another sugar. This ATP-driven alternating access model has been strongly supported by biochemical and crystallographic analysis, on this and other ABC transporters (Bordignon et al, 2010; Hollenstein et al, 2007). An EPR spectroscopy study has also concluded that liganded-MalE is required for the closure of the nucleotide-binding interface (Orelle et al, 2008). Yet, despite the long-lasting prevalence of the model, the notion that MalE shuttles back and forth to the membrane to deliver maltose has not always been consistent with earlier genetic analysis. For 44 example, it was expected that the transport constant Kt for maltose would decrease rapidly when MalE concentration increases in the periplasm (Bohl et al, 1995). In reality, the Kt decreased only ~2-fold when the MalE concentration increased more than ~20-fold (Manson et al, 1985). Similarly, the activation of the MalK ATPase was expected to be strong with a MalE variant possessing high affinity for maltose, but instead the mutant showed impaired ability to stimulate transport (Gould et al, 2009). In addition, a MalFGK2 allele capable of transporting lactose was still dependent on MalE for activity, although MalE does not bind lactose (Merino & Shuman, 1997). It was then unexpected that excess MalE can inhibit transport when maltose is held at a sub-stoichiometric level (Merino et al, 1995). Finally, maltose-loaded MalE was reported to have low affinity for the transporter (Kd: 50?100 ?M; Austermuhle et al, 2004). Since MalE periplasmic concentration depends on maltose (Shuman, 1982), the transport efficiency may be weak at low maltose, whereas MalE is able to facilitate transport especially at limiting substrate concentration. Together, these earlier observations prompt us to examine the problem of substrate delivery by measuring the effect of maltose on the stability of the MalE-MalFGK2 complex and by addressing how maltose stimulates the ATPase activity of MalFGK2. We employed the nanodisc because the system is well adapted to membrane proteins and transporters (Bayburt & Sligar, 2010; Denisov et al, 2004). We report that maltose-free MalE binds the outward-facing transporter with high affinity (Kd ~79 nM), whereas at saturating maltose concentration, MalE captures the sugar (with Kd ~120 ?M) and dissociates from the transporter. The surprising behavior of liganded-MalE was not specific to the disc because the same observations were made in proteoliposomes. The consequence of this maltose-regulated interaction was evaluated in vitro and in vivo: maltose transport and maltose-dependent MalK ATPase were found maximal when 45 MalE had low affinity for maltose, and minimal when MalE had high affinity for maltose. Complementary to these analyses, we test whether maltose stimulates the turnover of ATP hydrolytic cycle by increasing ATP cleavage or product release. We conclude that the transporter activity depends on two opposite effects: the capture and transport of maltose by the MalE-MalFGK2 complex, and the capture of maltose by MalE leading to its dissociation from MalFGK2. Maltose is therefore both substrate and regulator of its own transporter (i.e. homotropic regulator). A similar allosteric mechanism may apply to all ABC importers dependent on a substrate-binding protein similar to MalE. 3.2 Materials and methods 3.2.1 Materials Radiolabeled nucleotides [?-32P]ATP (25 Ci/mmol) and [?-32P]ATP (800 Ci/mmol) were purchased from MP Biomedicals. TNP-ATP and polyethylenimine cellulose TLC plates were obtained from Sigma. 3.2.2 Sedimentation and pull-down assays For the sedimentation assays, the MalFGK2 proteoliposomes (2 ?M) were incubated with MalE in 20 mM Tris-HCl (pH 8) containing 10 mM MgCl2 for 10 min at 37?C. The sample was diluted 25-fold in 20 mM Tris-HCl (pH 8), collected by ultracentrifugation (100,000? g, 1 h) and resuspended in 20 mM Tris-HCl (pH 8). Samples were analyzed by SDS-PAGE followed by Coomassie blue staining and autoradiography. For the pull-down assays, His6-tagged Nd-MalFGK2 particles were immobilized onto Ni-NTA resin (10 ?L per sample) in TSG buffer containing 5 mM MgCl2. Samples were then incubated with the indicated amount of [125I]-labeled MalE (10 min at room temperature). Unbound MalE was removed by washing the resin 3 46 times in TSGM buffer. The proteins were eluted in TSGM buffer containing 500 mM imidazole and analyzed by SDS-PAGE and autoradiography. 3.2.3 Fluorescence labeling MalE (3 mg/ml) in 500 ?L TSG buffer was incubated with a five-fold molar excess of ATTO-655 (Atto-Tec, GmbH) for 12 h at room temperature in the dark. The labeled protein was separated from excess dye by Superose 6 gel filtration chromatography. The labeling efficiency was determined at different protein concentration by absorbance spectroscopy (663 nm) using the extinction coefficient of 1.25?105 M?1cm?1. The typical ratio of fluorophore to MalE was 0.8, indicating very efficient labeling. 3.2.4 Fluorescence spectroscopy Fluorescence was recorded on a Cary Eclipse spectrofluorometer at 25?C. The affinity of MalE for maltose was determined by intrinsic fluorescence quenching. Excitation and emission wavelength were 280 nm and 350 nm, respectively (10 nm slit widths). For monitoring the binding of MalE to Nd-FGK2, fluorescence quenching of ATTO655-labeled MalE was recorded with excitation and collection wavelengths at 640 nm and 681 nm, respectively (10 nm slit width). The fluorescence emission was monitored over time and the signal was allowed to equilibrate after each addition for 180 s. The fluorescence quenching efficiency (E) was calculated according to the following equation, NdEE1+?=FFE where FE and FE+Nd are the fluorescence intensities of ATTO 655-labeled MalE in the presence and absence of Nd-MalF177wGK2. For measuring TNP-ATP binding to MalFGK2, excitation and emission wavelengths were 405 and 535 nm, respectively (10-nm slit widths). To determine the 47 binding affinity of TNP-ATP, the lipid-rich MalFGK2 nanodiscs (2 ?M) were incubated with TNP-ATP, and the fluorescence signal was allowed to equilibrate for 3 min. The fluorescence data were analyzed as described in Appendix C. 3.2.5 In vivo maltose transport assays Cell cultures were harvested during the late exponential phase of growth, washed twice with M63 salts, and resuspended in the same medium containing 100 ?g/ml chloramphenicol to an OD600 ~ 0.5. Each transport assay contained 200 ?l of cells and 200 ?l of M63 medium supplemented with [14C]-maltose at a final concentration of 100 ?M (5.7 ?Ci/?mol) or 1 mM (0.57 ?Ci/?mol). At the indicated time after incubation at room temperature, 20 ?l aliquots of cells were loaded onto a Bio-dot apparatus (Bio-Rad) and washed with 500 ?l of M63 medium. The membrane filters were dried and analyzed by autoradiography. The density of each dot was determined by using ImageQuant (GE Healthcare). 3.2.6 Thin-layer chromatography (TLC) ATPase hydrolysis assays were performed in TM buffer at room temperature with the indicated amount of [?-32P]ATP or [?-32P]ATP. Reactions were either stopped at 4 ?C by the addition of ice-cold EDTA (20 mM) and proteinase K (1 mg/ml) or subjected to centrifugal gel filtration using a desalting G25 spin column at 4 ?C in TM buffer. The eluted protein samples (0.5 ?l) were loaded at the bottom of a 10-cm-long PEI cellulose plate. The TLC was developed for 45 min in 0.3 M potassium phosphate, pH 3.4. The radioactive spots were revealed by a PhosphorImager scanner, and their intensity was quantified using ImageQuant (GE Healthcare). 3.2.7 Other methods Production and purification of MalFGK2 and MalE were performed as described (Appendix A.1 and A.2). Reconstitution of MalFGK2 into nanodiscs and proteoliposomes were 48 performed according to Chapter 2, except using E. coli total lipids. The MalFGK2 ATPase activity was determined by monitoring the release of inorganic phosphate using photo-colorimetric method. Linear gradient native gel electrophoresis and protein iodination were performed as described in Chapter 2. The detection of [125I]-MalE and 14C-labeled maltose (57 ?Ci/?mol, Molecular probes) was performed using a phosphor-imager scanner. 3.3 Results 3.3.1 Maltose-free MalE binds with a high-affinity to the outward-facing transporter In chapter 2, we reported that maltose results in decreased complex formation between MalE and MalFGK2 in the presence of AMP-PNP (Figure 2.6). Here, we further confirmed this observation by native gel electrophoresis and pull-down experiments (Figure 3.1a-3.1d). A complex of MalE and MalFGK2 can be isolated with non-hydrolysable ATP analogs or in the presence of vanadate (Chen et al, 2001; Oldham et al, 2007). These conditions stabilize the outward-facing transporter. Accordingly, MalE and Nd-MalFGK2 migrated to different positions on native-PAGE but together in the presence of AMP-PNP or ATP plus vanadate (Figure 3.1a). However, in the presence of maltose, the binding of MalE to MalFGK2 was significantly reduced (Figure 3.1a, compare lane 7 to lane 8). The negative effect of maltose was further confirmed by titration analysis (Figure 3.1b and 3.1c) and pull-down experiments (Figure 3.1d). To determine the binding affinities, we employed an electron transfer-based quenching reaction (Frank et al, 2008; Marme et al, 2003). MalE cysteine residue position 31 was modified with the oxazine-derivative dye ATTO655, and incubated with Nd-MalFGK2 bearing a tryptophan at position MalF-177. These two amino acids are within ~5? distance in the MalE-MalFGK2 complex structure (Chen et al, 2001; Daus et al, 2009). In the absence of nucleotide, very little quenching occurred (Figure 3.1f, green curve), in agreement with an earlier EPR 49 spectroscopy analysis showing that MalE has no measurable affinity to the inward-facing transporter (i.e. >50?100 ?M; (Austermuhle et al, 2004). In contrast, rapid and strong fluorescence quenching occurred with AMP-PNP, confirming that maltose-free MalE binds with high-affinity to the outward-facing transporter (Figure 3.1f, black curve). The affinity of maltose-free MalE for the outward-facing MalFGK2 was determined to be ~79 nM (Figure 3.1g, Table 3.1, Appendix C.4). As above, maltose had a negative effect on the stability of the complex because the binding affinity dropped ~5-fold to 390 nM (Figure 3.1g, red curve and Table 4.1). Interestingly, the binding of MalE to the transporter was still happening at saturating maltose concentration (Figure 3.1c and 3.1g). Since MalE is a highly dynamic protein that constantly binds, captures and releases maltose (Miller et al, 1983), it is possible that any maltose-free MalE is immediately captured by the outward-facing transporter. In support of the model, the quenching data were best fitted to a competitive ligand binding equation in which maltose-loaded MalE does not bind the transporter at all, whereas transporter-bound MalE has an affinity for maltose around ~127 ?M (Appendix C.5 and section below). Table 3.1 The dissociation constants of MalE for Nd-MalFGK2 Conditions Kd (nM) AMP-PNP 79 ? 9 AMP-PNP +maltose 390 ? 50 No Nu NA No Nu + maltose NA 50 Figure 3.1 High-affinity binding of MalE to the MalFGK2 complex. a, Nd-MalFGK2 (4 ?M) was incubated with MalE (1 ?M) or [125I]-MalE (~10,000 c.p.m., 1 ?M) in TSGM buffer containing nucleotides (1 mM) and maltose (1 mM) as indicated. Samples were analyzed by CN-PAGE followed by Coomassie blue staining (bottom part) and autoradiography (upper part). b, the indicated amount of Nd-MalFGK2 was incubated with MalE or [125I]-MalE in the presence or absence of maltose (2 mM) in TSGM buffer containing AMP-PNP (1 mM). Samples were analyzed by CN-PAGE followed by Coomassie blue staining (bottom part) then autoradiography (upper part). c, Nd-MalFGK2 (4 ?M) was incubated with [125I]-MalE (~10,000 c.p.m., 1 ?M) in TSGM buffer containing AMP-PNP (1 mM) and the indicated amount of maltose. Samples were analyzed by CN-PAGE and autoradiography. d, MalE binding to Nd-FGK2 was analyzed by pull down assay. e, structure of the complex MalFGK2-(E159Q) with MalE. The position MalE-31 and MalF-177 are indicated in red. f, time course fluorescence quenching between MalE (20 nM) and Nd-FGK2 (90 nM) in the presence or absence of maltose (1mM) and AMP-51 PNP (1mM). g, equilibrium titration of MalE binding to Nd-FGK2. When the data were fitted to Eq.4 (Appendix C.4), the dissociation constant in the presence of AMP-PNP was ~79 nM. The dissociation constant in the presence of AMP-PNP and maltose was ~390 nM. When the data were fitted to Eq.5 (Appendix C.5), in which maltose-bound MalE does not bind to the transporter and maltose acts as a competitor, the dissociation constant of transporter-bound MalE for maltose is 127 ?M. 3.3.2 MalE captures maltose and loses affinity for MalFGK2 We employed 14C-maltose to localize the sugar when incubated with MalE and MalFGK2. Free MalE has a relatively high affinity for maltose (Kd ~2 ?M), and native-PAGE can detect this association (Figure 3.2a, right panel). In contrast, when MalE was bound to the outward-facing transporter (Figure 3.2b), 14C-maltose was not associated with the complex (Figure 3.2a). Thus, either the binding site on MalE is not accessible to the sugar, or MalE binds the sugar but dissociates from the transporter. To test the two possibilities, ATTO655-labeled MalE was bound to the transporter in the presence of AMP-PNP (Figure 3.2c, black curve). Upon addition of maltose, there was a rapid loss of fluorescence quenching, indicating the dissociation of MalE from the transporter (Figure 3.2c, black curve). Using this assay, the maltose affinity of transporter-bound MalE was determined to be ~120 ?M (Figure 3.2d); a value very similar to that derived from the competitive one-site ligand binding equation (~127 ?M; Appendix C.5). The result was surprising because the X-ray structure of the MalE-MalFGK2 complex did not reveal any accessibility pathway for maltose (Oldham et al, 2007). Here, oligosaccharides from three (maltotriose) to seven (maltoheptaose) units were able to promote dissociation of the complex (Figure 3.2c and Appendix E). The interaction between MalE and MalFGK2 in the presence of AMP-PNP may be more dynamic than expected, or a path at the protein interface may be large enough to allow maltose access MalE. Most important to this analysis, the results showed that in the absence of transport, MalE captures maltose and dissociates from the transporter. 52 Figure 3.2 MalE exhibits low affinity for maltose when bound to MalFGK2. The indicated amount of MalE was incubated with Nd-MalFGK2 (0.5 ?M) and [14C]-maltose (10 ?M, 57 mCi/mmol) in TSGM buffer containing AMP-PNP (1 mM). After incubation (10 min, 37?C), samples were analyzed by CNPAGE followed by (a) autoradiography or (b) Coomassie blue staining. c, the binding of MalE to the transporter was monitored by fluorescence quenching, using MalE (20 nM) and Nd-MalFGK2 (70 nM). At the indicated time (arrow), 1 mM maltooligosaccharides were added to the reaction mixture. d, equilibrium titration to determine the affinity of the MalE-FGK2 complex for maltose using MalE (20 nM) and Nd-MalFGK2 (70 nM). The derived dissociation constant was ~120 ?M. 3.3.3 MalE with low affinity for maltose has a high affinity for MalFGK2 Since the binding of MalE to the transporter was controlled by maltose, the conformational state of MalE perhaps determines the binding affinity to the transporter. To test this hypothesis, we employed the mutant MalE-A96W/I329W (hereafter termed MalE-DW), which has ~60 fold stronger affinity for maltose (Marvin & Hellinga, 2001). The two mutations, located at the ?balancing interface? opposed to the sugar binding site, favor the closed state of 53 MalE even in the absence of maltose, as shown by NMR and SAXS analyses (Gould et al, 2009; Marvin & Hellinga, 2001; Telmer & Shilton, 2003). We also employed the mutant MalE-254 (mutation D65N) that displays very low affinity for maltose (Kd >1 mM; (Hall et al, 1997; Wandersman et al, 1979). The side chain D65 normally creates hydrogen bonds with the sugar hydroxyls and previous fluorescence and UV spectra analysis suggested that MalE-254 does not acquire the characteristic closed-liganded conformation until at least 10 mM maltose (Hall et al, 1997). It is thus very likely that MalE-254 would remain in open state at the maltose concentrations used in our assays (Figure 3.2a). Native-PAGE and ATPase assays were employed to determine the capacity of the two mutants to bind and activate the transporter (Figure 3.3b and 3.3c). MalE-DW was mostly unable to associate with Nd-MalFGK2 and it supported very little ATPase activity, which was further reduced by maltose (Figure 3.3b and 3.3c). In contrast, MalE-254 formed a tight complex with Nd-MalFGK2 and the ATPase activity was maximal and independent from maltose, as expected since MalE-254 does not capture the sugar. We therefore concluded that (i) maltose-free MalE facilitates the conversion of MalFGK2 toward the ATPase active conformation, (ii) maltose-free MalE binds with high-affinity to the outward-facing transporter (Kd ~79 nM), (iii) maltose has access to transporter-bound MalE, and (iv) upon capture of maltose, MalE loses its affinity for the transporter (>50?100 ?M). 54 Figure 3.3 Binding of the MalE mutants to MalFGK2. a, wild type MalE and mutants (1 ?M each) were mixed with [14C]-maltose in TSG buffer. Samples were analyzed by CN-PAGE and autoradiography. b, [125I]-labeled MalE and variants were incubated with Nd-MalFGK2 (4 ?M) in TSGM buffer containing AMP-PNP and maltose (2 mM) as indicated. After incubation (10 min, 37 ?C), samples were analyzed by CN-PAGE and autoradiography. c, the ATPase activity supported by the MalE mutants (1 ?M each) was determined in the presence of Nd-MalFGK2 (2 ?M) and maltose (1 mM). The reported values were derived from 3 independent experiments. 3.3.4 The results obtained with the nanodisc are confirmed in proteoliposomes The binding of MalE to MalFGK2 in proteoliposomes was assessed by co-sedimentation assays (Figure 3.4a and 3.4b). In proteoliposomes, the conformational state of the transporter is 55 shifted toward the inward-facing state, which has low ATPase activity and low affinity for MalE (Chen et al, 2001; Khare et al, 2009). As expected, AMP-PNP stabilized the outward-facing state and increased the affinity for MalE over 30-fold (Figure 3.4a and 3.4b). However, as in nanodiscs, the addition of maltose reduced the MalE equilibrium binding affinity by at least ~3-fold (Figure 3.4a). Furthermore, the co-sedimentation efficiency of MalE-254 was strong and independent of maltose (Figure 3.4c and 3.4d), whereas the co-sedimentation of MalE-DW was poor (~5-fold less than MalE-wt), and even weaker with maltose. To confirm that maltose had access to MalE when bound to the outward-facing transporter, the MalE-MalFGK2 complex was formed with AMP-PNP, then loaded on a sucrose gradient containing maltodextrins (Figure 3.4e). In all cases, there was a very obvious dissociation of fluorescent-labeled MalE from the transporter (Figure 3.4f). The control experiments showed that MalE did not co-sediment with MalFGK2 in the absence of AMP-PNP (Figure 3.4e, sample 2), but did co-sediment very well with AMP-PNP alone (Figure 3.4e, sample 1). Thus, the binding characteristics of MalE and variants, and the negative effect of maltose, were the same both in nanodiscs and proteoliposomes. In either system, maltooligosaccharides reduced the equilibrium binding affinity of MalE to the transporter. 56 Figure 3.4 Binding of MalE to the MalFGK2 complex in proteoliposomes. a, [125I]-MalE was incubated with MalFGK2 proteoliposomes (2 ?M) in TM buffer (20 mM Tris-HCl pH 8.0, 10 mM MgCl2) with or without AMP-PNP (1 mM) and maltose (1 mM). The fraction of MalE bound to MalFGK2 was isolated by ultra-centrifugation. The samples were subjected to SDS-PAGE followed by Coomassie blue. b, Autoradiography of the same gel. c, the co-sedimentation assay was performed using 57 MalE and variants in the presence of AMP-PNP. d, autoradiography of the same gel. e, MalFGK2 in proteoliposomes (10 ?M) was incubated with ATTO655-labeled MalE-31C (0.5 ?M) in the presence of AMP-PNP. The samples were applied on a sucrose density gradient containing the indicated maltooligosaccharides (1 mM). Equal fractions were collected and analyzed by SDS-PAGE and fluorescence assay. The control experiments showed that MalE did not co-sediment with MalFGK2 in the absence of AMP-PNP (sample2), but very well in the absence of maltose (sample 1). f, quantification of MalE bound to MalFGK2 in proteoliposomes. The amount of MalE bound to MalFGK2 without maltooligosaccharides was set to 100%. 3.3.5 Dual effect of the sugar on the transporter activity in proteoliposomes In proteoliposomes, the basal MalK ATPase activity is low (~10 nmol/min/mg), most likely because the lipid bilayer stabilizes the inward-facing transporter (Figure 3.5a). The addition of MalE stimulates the ATPase activity by ~4-fold (~40 nmol/min/mg), and a further ~10-fold in the presence of maltose (Figure 3.5a and see Discussion on this point). The ATPase measurements were then performed using the two MalE mutants described above. The maltose-dependent ATPase was best served with MalE-254 (~20 fold stimulation; Figure 3.5a), even though this mutant did not bind maltose at the concentration used in this assay. In contrast, the mutant MalE-DW, which captured maltose with a high affinity (Kd: ~50 nM), was unable to trigger the transporter ATPase activity (Figure 3.5a). Thus, maltose produced two effects in proteoliposomes: it stimulated the MalK ATPase and it diminished the affinity of MalE for the transporter. This second effect is opposed to the first because it reduces the maltose-dependent ATPase activity. 58 Figure 3.5 Regulation of the MalK transport ATPase by maltose. a, ATP hydrolysis was measured with MalFGK2 proteoliposomes (37?C,10 min) in the absence or presence of maltose (2 ?M) using MalE and variants (2 ?M each). b, steady-state transport ATPase using MalE variants (2 ?M each) as a function of the maltose concentration. Upper panel and bottom panel are the same curve but fitted to different x-axis. The data were fitted to the Michaelis-Menten equation to determine the maximal velocity Vmax and Kt of the transport ATPase reaction. The calculated values derived from 3 independent experiments are presented in Table 3.2. 59 3.3.6 Maltose is both substrate and regulator of the transporter To show that maltose produces two opposed effects during transport, the MalK ATPase was determined at various maltose concentrations. In proteoliposomes, the ATPase is coupled to maltose transport in an apparent stoichiometric manner (Davidson & Nikaido, 1990; Dean et al, 1989). With MalE-wt, the transport constant was low (Kt ~2 ?M) and the maximal velocity was reached as soon as the maltose concentration reached ~25 ?M (Figure 3.5b). With MalE-254, the transport constant was high (~800 ?M) and the transport ATPase was quasi-linear until ~1 mM maltose (Figure 3.5b). At 5 mM maltose, the transport ATPase supported by MalE-254 was almost 3-fold higher than with the wild type (Figure 3.5b and Table 3.2). Since the mutant MalE-254 is unable to capture the sugar, the transporter ATPase activity was dictated by the maltose concentration. By contrast, the mutant MalE-DW could barely sustain any maltose-dependent ATPase activity, as expected since the mutant captures maltose and is unable to bind the transporter. We constructed two additional MalE mutants with intermediate affinity for maltose (MalE-D65E and MalE-A63E; Figure 3.5b and Table 3.2). The results obtained with these mutants confirmed that the transport ATPase rate is inversely correlated to the affinity of MalE for maltose. Table 3.2 Kinetic parameters of the transport ATPase and affinity of the MalE mutants for maltose. MalE Vmax (nmol/min/mg) Kt (?M) Kd of MalE for maltose (?M) MalE-wt 320 ? 40 2 ? 0.7 2.8 ? 0.6 MalE-D65E 400 ? 70 78 ? 11 70 ? 10 MalE-A63E 610 ? 30 210 ? 40 160 ? 40 MalE-254 840 ? 50 800 ? 140 3700 ? 400 60 3.3.7 Consequence of an unregulated maltose transport in intact cells The work above allowed us to predict that maltose transport will be highest in bacteria expressing MalE-254. In contrast, maltose transport will be severely compromised in bacteria expressing MalE-DW. To confirm the prediction, maltose utilization was tested on MacConkey media (Figure 3.6a) and maltose accumulation was measured using 14C-maltose (Figure 3.6b). At saturating maltose concentration (i.e. 1 mM, equivalent to 0.04%), MalE-254 was the most proficient mutant for the transport and the fermentation of the sugar. In contrast, MalE-DW was unable to support cellular growth. Clearly, the high-affinity capture of maltose by MalE-DW was inhibiting transport and inversely, maltose transport was most effective with MalE-254 because the protein is unable to capture the sugar. We also tested the transport activity at sub-saturating maltose concentration (e.g. 100 ?M). Previous microbiological work indicated that bacterial growth is slower at maltose concentration below 1 mM (Rizk et al, 2011; Wandersman et al, 1979). At this limiting sugar concentration, the maltose import was better with MalE-wt compared to MalE-254 (Figure 3.6b). The result was expected because transporter-bound MalE increases the affinity for maltose, and thus the efficiency of transport when the substrate is limiting in the environment. Accordingly, the transport constant Kt obtained with MalE-wt is ~2 ?M whereas the Kt for MalE-254 is ~800 ?M (Table 3.2). 61 Figure 3.6 Maltose transport in intact cells. a, strain HS3309 (?MalE) transformed with pLH1 encoding for the indicated MalE protein was plated on M9-maltose (left) or MacConkey-maltose (right) agar plate. The color on MacConkey plates reflects maltose transport and fermentation after 10 h. The plasmid vector was used as a negative control. b, the transport assay using [14C]-maltose and strain HS3309 was performed as described in materials and methods. At the indicated time, cells were spotted on PVDF membrane and maltose import was detected by autoradiography. The intensity of each dot was determined by using ImageQuant (GE healthcare). 3.3.8 Rate-limiting step of the ATP hydrolytic cycle is the release of Pi We addressed how MalE and maltose stimulate the ATPase activity of MalFGK2. To do so, we began by performing an Arrhenius analysis of ATPase activity (Figure 3.7a). The linearity of the plots indicates a single rate-limiting step in the ATP hydrolytic cycle, which comprises 62 four individual chemical steps: ATP-binding to MalFGK2 (1), ATP cleavage (2), ADP release (3) and phosphate release (4). Figure 3.7 ATP binding is not the rate-limiting step of the ATP hydrolytic cycle. a, Arrehenius analysis of MalFGK2 ATPase. The ATPase activities of MalFGK2 reconstituted in lipid-rich nanodiscs were assayed from 15 to 42 ?C using ATP (2mM), MalE (10 ?M) and maltose (1 mM).Log turnover numbers (mean ? S.D., n = 3 independent replicates) were plotted against the reciprocal of absolute temperatures. b, ATP hydrolysis rate were determined at the indicated amount of ATP, in the presence and absence of MalE (10 ?M) and maltose (1 mM). c, equilibrium titration of MalFGK2 with TNP-ATP. The MalFGK2 complex reconstituted in lipid-rich nanodiscs was incubated with the indicated amount of TNP-ATP in the presence and absence of MalE (10 ?M) and maltose (1 mM). The fluorescence data were fit to equation 1 to determine the affinity of TNP-ATP for MalFGK2 (Appendex B.1). d, equilibrium titration of MalFGK2 with ATP. The MalFGK2 complex reconstituted in lipid-rich nanodiscs was incubated with TNP-ATP (80 ?M) in the presence and absence of MalE (10 ?M) and maltose (1 mM). The fluorescence data (mean ? S.D., n = 3 independent replicates) were recorded after addition of the indicated amount of ATP. The data were fit to Equation 2 and 3. To determine the rate-limiting step during the ATP hydrolytic cycle, we employed nucleotide labeled with the fluorescent moiety TNP (trinitrophenyl). The quantum yield of TNP-ATP increases significantly upon binding to a nucleotide-binding pocket (Poolman et al, 2005). However, the measurements with MalFGK2 could not be reliably performed in proteoliposomes 63 because the fluorescence emission of TNP-ATP increases in the lipid environment (data not shown). We therefore employed the MalFGK2 complex reconstituted into lipid rich nanodiscs. These lipid-rich nanodiscs reproduce the maltose transporter ATPase activity and its dependence on MalE and maltose, as in proteoliposomes (Chapter 2). In addition, the amount of lipids in the particles is sufficiently low to enable fluorescence measurements. The equilibrium titrations revealed a dissociation constant (Kd) for TNP-ATP of ~ 9.4 ?M (Figure 3.7c). The apparent Kd for ATP, measured by competitive replacement of TNP-ATP, was estimated to be ~220 ?M (Figure 3.7d). This value is similar to the Km derived from the ATPase measurements in lipid-rich nanodiscs and proteoliposomes (from ~200 to ~280 ?M, respectively; Figure 3.7b). When these measurements were repeated in the presence of MalE and maltose, no significant change in the affinity values was detected, thus indicating that MalE and maltose do not change the binding of ATP to the transporter. We then tested whether MalE and maltose accelerate ATP cleavage by measuring the ATPase activity of MalFGK2 under two different conditions: (i) with MalFGK2 present in 20-fold molar excess over the nucleotide, so that only a single round of ATP hydrolysis can occur (Figure 3.8a); and (ii) under steady-state conditions, where the nucleotide is present in 1000-fold excess over MalFGK2 so that multiple rounds of ATP hydrolysis are possible (Figure 3.8b). The MalFGK2 complex reconstituted in lipid-rich nanodiscs were incubated with [?-32P]-ATP in the presence and absence of MalE and maltose. ATP hydrolysis was detected by thin layer chromatography and autoradiography. Under single turn over conditions at 4 ?C, MalFGK2 hydrolyzes ~50% of total ATP within 1 min, which is only 20% less than at 37 ?C (Figure 3.8a). Addition of MalE increases ATP hydrolysis by ~ 2 fold, whereas maltose exhibits an adverse effect on this stimulation. This negative effect of maltose is exaggerated using the mutant MalE-64 DW and not observed using the mutant MalE-254 (Figure 3.8a). Together, these results indicate that binding of maltose-free, open-state MalE to the transporter stimulates the ATP cleavage step. Under multiple turn over conditions, MalFGK2 hydrolyzes ATP inefficiently but much more in the presence of MalE and maltose. Since ATP cleavage occurs efficiently under single turn-over conditions, the result implies that a rate-limiting step limits ATP hydrolysis under multiple turnover conditions. Maltose greatly stimulates this rate-limiting step, because ATP hydrolysis is increased by ~ 4-fold under multiple turnover conditions (Figure 3.8b). To determine whether this rate-limiting step corresponds to the release of Pi or ADP from the nucleotide binding pocket, we determined the amount of ADP and Pi that is co-purified with MalFGK2 using [?-32P]-ATP and [?-32P]-ATP (Figure 4.8c and 4.8d). We performed a control experiment using SecA, which possesses a higher affinity for ADP than Pi and ATP (Robson et al, 2009; Sianidis et al, 2001). Accordingly, a significant increase in the amount of ADP that is co-purified with SecA can be detected (Figure 3.8c). In contrast, MalFGK2 shows an enrichment of Pi (Figure 3.8d). Thus, Pi release appears to be the rate limiting step in the ATP hydrolytic cycle of MalFGK2 in the presence of MalE and maltose. 65 Figure 3.8 Pi release is the rate-limiting step of the ATP hydrolytic cycle of MalFGK2. a-b, effect of MalE and maltose on ATP hydrolysis in single (a) and multiple turnover (b) conditions. The lipid-rich MalFGK2 nanodiscs (a, 20 ?M; b, 1?M) were incubated with MalE (75 ?M) and maltose (1mM) at indicated temperature with [?-32P]-ATP (a, 1?M for 1 min; b, 1 mM for 5 min). The reactions were stopped by the addition of ice-cold EDTA (20 mM) and proteinase K (1 mg/ml). Samples were analyzed by TLC and autoradiography. The radioactive spots (mean ? S.D., n = 3 independent replicates) were quantified by densitometry to determine the fraction of ADP produced. c-d, release of ADP and Pi from MalFGK2. The lipid-rich MalFGK2 nanodiscs (10 ?M) were incubated with MalE (75 ?M) and maltose 66 (1mM) for 3 min at indicated temperature with 200 ?M [?-32P]-ATP (c) or 200 ?M [?-32P]-ATP (d). The control experiment using SecA (10 ?M) was performed in parallel. The total amount of ADP and Pi produced during the reaction was determined by TLC before centrifugal gel filtration (CGF). The total amount of ADP and Pi remaining bound to the transporter was determined by TLC after centrifugal gel filtration. 3.4 Discussion In the conventional model, closed-liganded MalE binds and activates the inward-facing transporter. The binding triggers a series of ATP-driven conformational changes that eventually leads to the opening of MalE, release of maltose and transport across the membrane (Figure 4.1a). Over the last twenty years, the different steps of the model have been analyzed in details at the biochemical, biophysical and structural levels (Davidson et al, 2008). Yet, the binding affinity of maltose-free MalE has never been characterized and the effect of maltose on the stability of the MalE-MalFGK2 complex has never been reported. Here, we confirm that closed-liganded MalE has weak affinity for MalFGK2 (Kd >50?100 ?M), but we show that open-unliganded MalE in contrast possesses nanomolar affinity for the outward-facing transporter (Kd ~79 nM). In addition, we show that maltose can access transporter-bound MalE (Kd ~120 ?M) whereas, in the absence of maltose uptake, MalE captures the sugar (Kd ~2 ?M) and dissociates from the transporter (Kd >50 ?M). The knowledge of the binding affinities leads us to propose a different model, in which MalE is bound to the transporter to create a low-affinity maltose-binding site. Maltose transport increases the ATPase cycle, probably via facilitating a conformational switch of the transporter from the outward- to inward-facing states. If maltose is not immediately transported, MalE acquires a closed-liganded conformation and dissociates from the transporter (Figure 4.1b). We justify below the reasons for this model and the mechanistic and physiological implications. 67 First, our results show that the maximal ATPase activity and transport velocity are inversely correlated to the affinity of MalE for maltose in proteoliposomes. If maltose was to cause MalE to activate the transporter, the MalK ATPase activity at saturating maltose concentration (Vmax) should be independent of the affinity of MalE for maltose. For example, a variant with low affinity would support as much ATPase activity, provided the sugar concentration is sufficiently high. Conversely, a variant with high affinity would support the same maximal ATPase but at low concentration of maltose. The results from ATPase assays (Figure 3.5) and the maltose transport in vivo (Figure 3.6) are not consistent with such a model. Furthermore, the binding assays in proteoliposomes show that maltose decreases the equilibrium affinity of MalE to the transporter, whether in the outward- or inward-facing state (Figure 3.4). The current model is therefore insufficient to explain these data. Instead, we believe that a model in which MalE is bound to the transporter to facilitate the capture of maltose, but dissociates from the transporter when the substrate concentration increases, can explain why the maximal velocity depends on MalE affinity for maltose. A variant with low affinity would display a lower transport rate (i.e. higher Kt) but remain bound to the transporter when the substrate concentration increases, allowing for a higher maximal rate of transport. In contrast, a variant with high affinity for maltose would capture the sugar and dissociate from the transporter, hence lowering the maximal rate of transport. Our results (Figure 3.5 and Figure 3.6), as well as those in Wandersman et al. (1979) and Gould et al. (2009) which describe the behavior of the mutant MalE-254 and MalE-DW respectively, concur with this analysis. The model is also consistent with the observation that excess MalE can inhibit transport when maltose is held at a sub-stoichiometry level (Merino et al, 1995). In the latter case, all maltose molecules would be captured by excess MalE and away from the transporter. 68 Second, we find that maltose decreases the affinity of MalE for MalFGK2 in nanodiscs and the MalE-dependent MalK ATPase activity. According to the former model (Figure 4.1a), maltose should instead stimulate the MalK ATPase, or at least leave it unchanged. We believe that the observed decrease of MalK ATPase activity can be explained by the negative effect of maltose on the MalE-MalFGK2 interactions. In support of this model, using a fluorescence-based binding assay (Figure 3.1), we were able to show that maltose shifts the binding equilibrium toward the dissociation of MalE from the transporter (Figure 3.2), and therefore toward the diminution of the MalK ATPase. The binding assays also revealed that transporter-bound MalE is accessible to maltose and longer maltooligosaccharides (Figure 3.2 and Figure 3.4). This last observation was surprising because the atomic structure of the MalE-MalFGK2 complex did not reveal a sugar accessibility pathway at the protein interfaces. It cannot be excluded that protein crystallography may have suppressed some otherwise transient interactions, such as those detected by molecular dynamics simulations (Oliveira et al, 2011). Here, the affinity of the complex MalE-MalFGK2 for maltose was estimated at ~120 ?M. In the absence of transport, this maltose concentration would lead to 50% dissociation of MalE from the transporter. It is important to note that our results do not exclude the possibility that transporter-bound MalE binds maltose just before MalE associates tightly with the outward-facing transporter. However, if MalE was to capture maltose and thus acquire a closed-state conformation, MalE would dissociate from the transporter and return to the periplasm (Figure 4.1b). The interplay between MalE, maltose and MalFGK2 is complicated by the dynamic nature of MalE that constantly binds, captures and releases maltose (Ledvina et al, 1998; Miller et al, 1983). Based on ATPase measurements, the affinity of liganded-MalE for the transporter would be considered significantly high (Km,app ~14 ?M; Gould et al, 2009). When interpreting 69 the value however, one should remember that liganded-MalE spontaneously releases maltose (Miller et al, 1983), whereas maltose-free MalE binds the transporter with high-affinity. The spontaneous release of the ligand would explain why saturating maltose does not abolish the binding of MalE to the outward-facing transporter (Figure 3.1 and Figure 3.4). The same phenomena may occur during histidine transport because the binding of HisJ to HisQM is reduced, but only 3-fold in the presence of saturating amount of histidine (Ames et al, 1996). In fact, the modest affinity of MalE and HisJ for their ligands (?M range) may be essential to allow sufficient influx of these nutrients even at saturating environmental concentration. In contrast, for the vitamin B12 transporter, the substrate binding protein BtuF binds its ligand with very high affinity (Kd ~15 nM). In that case, the interaction between BtuF and BtuCD is dramatically reduced by saturating amounts of vitamin B12 (~105 fold; Lewinson et al, 2010). Thus, even though type I and type II ABC importers (i.e. MalFGK2 and BtuCDF) have different membrane domain and substrate binding protein structures, the regulatory effect of the substrate may be similar. In this context, it is particularly interesting that a low-affinity and a high-affinity transporter for the same substrate -molybdate- can exist in the same cell (George & Jones, 2011; Tirado-Lee et al, 2011). Perhaps the high-affinity molybdate transporter decreases activity as the molybdate concentration increases. The cell would then use the low-affinity transporter system in order to maintain constant molybdate uptake. Another essential question addressed in this study is how maltose stimulates ATPase activity of MalFGK2. It has been proposed that only closed, maltose-bound MalE initiates the transport cycle, thereby stimulating ATP hydrolysis by the transporter (Oldham et al, 2007). This indicates that maltose facilitates conformational transition of the transporter from the inward- to outward-facing states. Since only the latter state is able to hydrolyze ATP, it follows that maltose 70 stimulates the ATP cleavage step of the MalFGK2 ATPase cycle. However, our data instead indicate that maltose accelerates Pi release from MalFGK2, which is also the rate-limiting step in the ATP hydrolytic cycle. This could arise, for example, if the ATP ?-phosphate exhibits higher affinity for the outward-facing transporter than ADP and ATP. Consistently, crystal structures of nucleotide-bound MalFGK2 suggest that the ATP ?-phosphate forms over 10 hydrogen bonds with the Walker A, the Q-loop, and the switch histidine of one NBD and the LSGGQ motif of the other, thus tightly tethering the NBD dimer to stabilize the outward-facing conformation (Oldham & Chen, 2011b). Assuming Pi release is concurrent with the return of transporter from the outward- to inward-facing conformations, we postulate that maltose stimulates MalFGK2 ATPase activities by promoting this conformational change. Although this prediction awaits further experimental validation, it is consistent with my following studies in Chapter 4, showing that maltose does not facilitate conformation change of the transporter from the inward- to outward-facing states. In conclusion, the activity of a membrane transporter usually depends on two factors: the affinity of the transporter for the substrate and the velocity of the transport reaction. For the maltose transport system, transporter-bound MalE controls the affinity and ATP the velocity. We show here that maltose also contributes to the transport kinetics. This negative regulation may be crucial for the cell because ABC importers are unidirectional and can achieve (at least in theory) very high concentration gradients, either toxic or consuming unnecessarily the metabolic energy. Homotropic allosteric regulation represents a simple way to limit transport when the environmental substrate concentration is high. Why MalE is twenty-fold more abundant than the transporter in maltose-induced cells is still not entirely clear (Dietzel et al, 1978; Manson et al, 1985). It has been proposed that excess MalE ensures that the periplasmic maltose concentration 71 varies more slowly than the outside, a mechanism termed ?retention effect?, especially important during bacterial chemotaxis (Ames et al, 1996; Silhavy et al, 1975). It has also been proposed that the high MalE concentration may ensure the hopping of maltose from MalE to MalE to facilitate transport across the gel-like environment of the periplasm (Ames et al, 1996; Brass et al, 1986). It is also possible that large pool of MalE may serve to buffer the negative effect of maltose because a fraction of ligand-free MalE would always be available to bind the transporter even at saturating maltose concentration. All these possibilities remain to be tested. 72 Chapter 4: Conformational changes of the maltose transporter 4.1 Introduction Major progress in the description of ABC transporters has been achieved by x-ray crystallography (Hollenstein et al, 2007; Procko et al, 2009). In the structure of MalFGK2 without ligands, the sites for ATP binding on the MalK dimer are separated, and the MalFG membrane domain forms a cavity that is exposed to the cytosol (inward-facing state; Khare et al, 2009). In the structure obtained in the presence of MalE and nonhydrolyzable ATP analogs, the ATP-binding sites are bound together, whereas the MalFG cavity opens toward the periplasm (outward-facing state; Oldham & Chen, 2011b; Oldham et al, 2007). Together, these snapshots of the transporter structure have provided the molecular framework to explain how ATP binding and hydrolysis is coupled to membrane transport, the so-called alternating access mechanism. However, what triggers the reaction and how maltose is transferred from MalE to MalFG remain unclear. MalE consists of two lobes with the maltose-binding site located at the interface. When the sugar binds, the two lobes close and eventually capture maltose with a high affinity (Sharff et al, 1992; Spurlino et al, 1991). Because the ATPase activity of the transporter is maximal only in the presence of MalE and maltose, it has been proposed that closed liganded MalE binds the transporter and triggers ATP hydrolysis (Davidson, 2002; Davidson et al, 1992). The x-ray structures were interpreted with this assumption in mind, leading to the model in figure 4.1a (Davidson et al, 2008). In that model, the binding of closed liganded MalE to the transporter induces the pairing of MalK. The pairing of MalK triggers the MalFG outward-facing conformation, which in turn forces the opening of MalE and the release of maltose. Upon ATP hydrolysis, the transporter returns to the inward-facing state, and maltose enters the cytosol. In 73 support of the model, an EPR spectroscopy study reported that the MalK nucleotide-binding interface closes only in the presence of MalE and ATP (Orelle et al, 2008). A partial closure of the MalK interface was also observed when MalFGK2 was co-crystallized with closed liganded MalE (Oldham & Chen, 2011a). It was concluded that MalE facilitates the pairing of MalK and therefore hydrolysis of ATP. However, despite the evidence and prevalence of the model, I showed in Chapter 3 that closed liganded MalE does not bind the transporter but dissociates when MalE captures maltose. I also reported that open state MalE binds with nanomolar affinity to the outward-facing transporter and creates a low affinity receptor for maltose. These data, difficult to reconciliate with the former model, led me to propose a new model (Figure 4.1b and legend for details about the binding affinities). Here, we further investigate the effect of MalE, maltose, and ATP on the conformational changes of MalFGK2. Indeed, an important and central question is the mechanism leading to the outward-facing conformation. The transition depends on MalE and maltose in the former model, but not in the newer model. Using cross-linking, co-sedimentation analysis, native gels, and fluorescence assays, we show that MalE and maltose do not facilitate the closure of MalK dimer, nor the transition of MalFG to the outward-facing state. Instead, we find that ATP alone is sufficient. Because MalE and maltose do not facilitate the binding of ATP to the transporter, nor the conversion to the outward-facing conformation, the results suggests that MalE and maltose stimulate a step occurring during the return of the transporter to the inward-facing state. In addition, the model in figure 4.1a also suggests that ATP binding is not sufficient to change the conformation of MalFGK2, because the inward-facing membrane domain MalFG imposes constraints that prevent the closure of MalK (Orelle et al, 2008). Although this has not been experimentally examined, this hypothesis is not consistent with the new model in figure 74 4.1b. Clearly, how the conformations of MalFG and MalK are reciprocally regulated is essential for the understanding of the transport reaction. To address this question, we isolated the MalFG transmembrane domain and showed that MalK dimer stabilizes MalFG in the inward-facing state. Figure 4.1 Two opposed models for maltose transport. a, in the conventional model, closed liganded MalE triggers the outward-facing conformation. MalE binds maltose (Kd = ~2 ?M) and then associates with the inward-facing transporter (Kd = ~100 ?M). The transition to the outward-facing conformation facilitates the opening of MalE and the release of maltose to the MalFG cavity. Upon ATP hydrolysis, the transporter returns to the inward-facing state and maltose is released in the cytosol. In this model, MalE facilitates the pairing of MalK and the ATP hydrolysis step. b, in the proposed new model, ATP alone triggers the outward-facing conformation. The outward-facing transporter binds unliganded MalE with a high affinity (Kd = ~50-80 nM). Maltose then binds to the MalE-MalFGK2 assembly (Kd = ~120 ?M). Upon ATP hydrolysis, the transporter returns to the inward-facing conformation, and maltose is released in the cytosol. In this model, MalE stimulates the return of the transporter to the inward-facing conformation. If ATP hydrolysis does not take place immediately, or if maltose is present in excess, MalE acquires its closed liganded conformation and dissociates from the transporter (negative autoregulation). 75 4.2 Materials and methods 4.2.1 Material Sephadex G-25 Sepharose column was obtained from GE Healthcare. N-(1-Pyrene)-maleimide was obtained from AnaSpec. Bis-maleimidoethane (BMOE) and N-ethylmaleimide were obtained from Sigma. 4.2.2 Reconstitution of MalFGK2 The lipids 1,2-dioleoyl-sn-glycero-3-phosphocholine and 1,2-dioleoyl-sn-glycero-3-phospho-(1?-rac-glycerol) were mixed at a ratio of 7:3, dissolved in chloroform, and dried under a stream of nitrogen. The lipids were resuspended in TSG buffer (50 mM Tris-HCl, pH 8.0, 100 mM NaCl, 10% glycerol) containing 0.2% DDM, which were then used in the reconstitution of MalFGK2 in nanodiscs and proteoliposomes, following the procedures described in Chapter 2. Proteoliposomes were extruded through a 100-nm polycarbonate filter before use. 4.2.3 Cysteine Cross-linking Experiments Cross-linking reactions were performed in 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2 containing 0.5 ?M MalFGK2 proteoliposomes mixed with 1 mM ATP, 10 ?M MalE, and 1 mM maltose as indicated. The samples were incubated with 50 ?M BMOE (10 min at room temperature) and then treated with N-ethylmaleimide (5 mM) before analysis by SDS-PAGE. 4.2.4 Fluorescence Labeling and Spectroscopy The dye N-(1-pyrene)-maleimide was used to label the MalFGK2 cysteine mutants. Briefly, the purified MalFGK2 complex was incubated with 0.5 mM DTT (30 min on ice) and passed through a G25 column. The labeling reaction was then performed by mixing MalFGK2 and N-(1-pyrene)-maleimide at a molar ratio of 1:10 in TSG buffer containing 0.01% DDM and 1 mM TCEP for 2 h at room temperature. The unreacted N-(1-pyrene)-maleimide was removed 76 by Superdex 200 10/30 gel filtration. The purified and labeled complex was reconstituted in proteoliposomes as described above. The labeling efficiency was determined by absorbance spectroscopy (343 nm) using an extinction coefficient of 40,000 cm?1 M?1. A typical ratio was between 1.2 and 1.4. Because there are two cysteine substitutions per MalFGK2 variant, this is equal to ?60?70% labeling efficiency. The pyrene-labeled MalFGK2 variants retained ?80% ATPase activity and 70% transport activity of wild type MalFGK2 (see Table 4.1). The fluorescent measurements were carried out at 25 ?C in FL buffer (50 mM Tris-HCl, pH 8.0, 5 mM MgCl2) on a Cary Eclipse spectrofluorometer. The fluorescence spectra were recorded from 365 to 520 nm with excitation wavelength at 345 nm (5-nm slide width). The binding of MalE to MalFGK2 was measured by fluorescence quenching assay as descrbied in Chapter 3. 4.2.5 Other Methods The transporter ATPase activity was determined by measuring the release of inorganic phosphate using photo colorimetric method (Zhou et al, 1992). Maltose transport assay, Linear gradient native gel electrophoresis and protein iodination were performed as described in Chapter 2. The sedimentation assay was carried out as described in Chapter 3. 4.3 Results 4.3.1 ATP controls the outward-facing conformation of the transporter Cysteine residues were introduced at diagnostic positions on MalK, MalF, and MalG, so that sulfhydryl-based cross-linking methods could be employed to monitor the conformation of the transporter in the lipid bilayer (Figure 4.2a). The cysteine residues were introduced into a complex mutated at the endogenous Cys-40 of MalK because this position naturally reacts with maleimide reagents (data not shown). According to the crystallographic data (Khare et al, 2009; Oldham et al, 2007), the distance between the periplasmic positions MalF446/MalG234 increases 77 from ?9 to ?18 ? during the conformational transition from the inward-facing state to the outward-facing state. Meanwhile, the distance between the cytosolic positions MalK83/MalK83 and MalF396/MalG185 decreases from ? 21 to ?8 ?. We thus employed the homobifunctional cross-linker BMOE with a spacer arm of 8 ? to monitor these conformational changes (Figure 4.2). We note that the C-terminal tail of MalG in the crystal structure is located near the MalK dimer interface and may potentially interfere with the formation of the cross-link MalK83/MalK83. However, as reported by others, an efficient ATP-dependent cross-linking occurs with this pair (and also with MalK85/MalK85), showing that the C-terminal tail of MalG is sufficiently mobile in solution to allow cross-link formation (Daus et al, 2007b; Grote et al, 2008). We also considered the possibility that MalFGK2 becomes inaccessible to MalE when the transporter is reconstituted in liposomes. This was ruled out because (i) the transporter was functional (Table 4.1), and its activity was dependent on MalE and maltose, (ii) the sedimentation of MalE with the proteoliposome was dependent on MalFGK2 and nucleotides (Figure 4.3a), and (iii) the thiol-reactive membrane-impermeable reagent 5-IAF labeled the periplasmic side of the complex with at least 50% efficiency (Appendix F). We then probed what the transporter conformation is in the membrane. In the absence of ligands (apo state), there was only a little cross-link formation with the pairs MalK83/MalK83 and MalF396/MalG185 (pairs located in the cytosolic side of the membrane), whereas cross-links readily formed with the pair MalF446/MalG234 (located on the periplasmic side) (Figure 4.2 b?d). These results were expected because the transporter resides naturally in the inward-facing state in the absence of ligands (Khare et al, 2009). In the presence of ATP, however, a predominant cross-link appeared with the pairs MalK83/MalK83 and MalF396/MalG185, whereas the cross-link 78 of MalF446/MalG234 was decreased by ?90%, as determined by densitometry analysis (data not shown). The binding of ATP, and not its hydrolysis, was responsible for this conformational change because ATP?S produced the same effect (Figure 4.2e). Strikingly, the addition of MalE with or without maltose did not cause any differences in the cross-linking efficiencies (Figure 4.2e). Together, these data provided strong preliminary evidence that ATP on its own controls the transition of the transporter from inward- to outward-facing state. Table 4.1 ATPase activity of the pyrene-labeled transporters reconstituted in proteoliposomes ATPase activity (nmol/min/mg) Transport activity (nmol/min/mg) Transporter -MalE and maltose + MalE and maltose -MalE + MalE Wild type 11 ? 4 890 ? 90 0~0.1 2.3 ? 0.4 MalK83-MalK83 9 ? 2 730 ? 60 0~0.1 1.8 ? 0.3 MalF396-MalG185 12 ? 3 790 ? 80 0~0.1 1.7 ? 0.4 MalF446-MalG234 7 ? 2 700 ? 70 0~0.1 1.5 ? 0.2 79 Figure 4.2 ATP triggers the transporter outward-facing conformation. a, the red spheres indicate the location of the cysteine residues introduced in MalF, MalG, and MalK and their relative positions in the inward- (left) and outward-facing (right) conformations. b-d, the indicated transporters reconstituted in proteoliposomes were incubated with the homobifunctional cross-linker BMOE (50 ?M) in the indicated conditions. The samples were treated with N-ethylmaleimide (5mM) prior to analysis by 15% SDS-PAGE and Coomassie Blue staining. e, cross-linking results for the pairs MalF396/MalG185 and MalF446/MalG234 in the presence of ATP?S. 4.3.2 ATP controls binding of MalE to the transporter We have shown that MalE binds with high affinity to the outward-facing transporter in Chapter 3. Because ATP alone appears sufficient for the transition to the outward-facing 80 conformation, we hypothesized that ATP alone should also control the binding of MalE to the transporter. Accordingly, the stabilization of MalFGK2 in the outward-facing state with ATP plus vanadate allowed an efficient co-sedimentation of MalE with the transporter (Figure 4.3a, lanes 2). In the absence of vanadate, the binding of MalE was weak because the transporter hydrolyzes ATP and returns to the inward-facing conformation, which has little affinity for MalE. Accordingly, when the mutants MalK83/MalK83 and MalF396/MalG185 were treated with ATP plus BMOE to stabilize the transporter in the outward-facing state, the binding of MalE was as efficient as with ATP plus vanadate (Figure 4.3a, compare lanes 2 with lanes 4). In contrast, when the transporter was locked in the inward-facing state with the pair MalF446/MalG234, the binding of MalE was at background levels, even in the presence of ATP and vanadate (Figure 4.3a, right panel, lane 4). The results therefore confirm that ATP controls the transition from the inward- to outward-facing state, which in turn controls the binding of MalE to the transporter. Figure 4.3 ATP controls the binding of MalE to the transporter. a, the binding of MalE to MalFGK2 in proteoliposomes was analyzed by co-sedimentation assay. [125I]MalE (~10,000 cpm, 10 ?M) was incubated with MalFGK2 proteoliposomes (2 ?M) in the presence of ATP, 81 vanadate, and BMOE as indicated (room temperature, 10 min). The samples were diluted 25-fold in 20mM Tris-HCl, pH 8.0, before ultracentrifugation (100,000 ? g, 1 h). The fraction of MalE bound to MalFGK2 was analyzed by 15% SDS-PAGE followed by Coomassie Blue staining (top panels) or autoradiography (bottom panels). b, the binding of [125I]MalE to MalFGK2 in nanodiscs was analyzed by native gel electrophoresis. [125I]MalE (~5,000 cpm, 0.5 ?M) was incubated with MalFGK2 nanodiscs (1 ?M) in the presence of ATP, vanadate, and BMOE as indicated (room temperature, 10 min). The samples were analyzed by native gel electrophoresis and autoradiography. We next employed a fluorescence-based assay to determine the binding affinity of MalE to MalFGK2 in the inward- and outward-facing states. We reconstituted the transporter in nanodiscs because the system is well suited for spectroscopic analysis. First, we confirmed on native gel that MalE binds to the transporter with the same characteristics in nanodiscs as in proteoliposomes (Figure 4.3, compare b with a; ATP dependence). We then titrated the fluorescence-labeled MalE with an increasing amount of transporter (Figure 4.4). The data showed that MalE binds with a high affinity (Kd = ?50?80 nM) only when the transporter is stabilized in the outward-facing conformation with ATP-vanadate or ATP-BMOE (binding data and standard deviations reported in Table 4.2). In contrast, in the absence of nucleotides, the binding affinity of MalE for the transporter was ?7 ?M and increased furthermore to ?45 ?M when MalE was liganded with maltose (Table 4.2 and Appendix G). Because the transporter might spontaneously acquire its outward-facing conformation even in the absence of ATP, the affinity of liganded MalE to the inward-facing state may in fact be lower than 45 ?M. Indeed, the interaction of maltose-bound MalE was virtually null when the transporter was stabilized in the inward-facing state using the cysteine pair MalF446/MalG234 (Figure 4.4d). The reaction was not saturable, and therefore the affinity could not be determined, in agreement with earlier binding assays in proteoliposomes (Austermuhle et al, 2004). We concluded that ATP controls the transporter outward-facing conformation and thereby controls the high affinity binding of MalE to the transporter. 82 Figure 4.4 Equilibrium titration of MalE binding to MalFGK2 in nanodiscs. The MalFGK2 complex in nanodiscs was incubated with ATP (1mM), BMOE(50 ?M), and vanadate (10 ?M) as indicated (10 min at room temperature). The nanodiscs were repurified on a desalting G25 column 83 before fluorescence quenching assay using ATTO-655-labeled MalE (20 nM). Three independent measurements were performed, and the data points were fitted to a one-site binding equation (mean and standard deviation results are presented in Table 1). a, MalFGK2 wild type. b, MalFGK2 with mutations MalKS83C/MalKC40S (MalK83-MalK83). c, MalFGK2 with mutations MalFP396C/MalGD185C/MalKC40S (MalF396-MalG185). d, MalFGK2 with mutations MalFL446C/MalGS234C/MalKC40S (MalF446-MalG234). The left and right panels are the same curves but fitted to a different x axis. Table 4.2 The affinity of MalE for the transporter in nanodiscs was determined by fluorescence quenching assay. a. The data are derived from three independent measurements and the fit is using equation 1. b. The measurement was performed in the presence of Vi. c. The measurement was performed using crosslinked MalF446-MalG234. 4.3.3 MalE and maltose do not facilitate the transition of MalFGK2 to the outward-facing conformation As an additional way to probe the transporter conformational state and to explore the transition kinetics, we employed the fluorescent dye N-(1-pyrene) maleimide. When attached to a protein thiol, the dye produces two distinct emission peaks at 375 and 395 nm. When two pyrene-thiols interact within 6?10 ? distance, they form excited state dimers (excimers) that emit at longer wavelengths than the lone excited fluorophore (Lehrer, 1995; Sahoo et al, 2000). The transporter labeled with the pyrene probe at the cysteine positions described above retained ?80% ATPase activity and 70?80% maltose transport activity as compared with the wild type Affinity for MalE (?M)a Transporter ATP + Vi ATP + BMOE No Nu No Nu + BMOE No Nu + maltose Wild type 0.079 ? 0.009 1.41 ? 0.07 7 ? 1 5 ? 1 40 ? 20 MalK83-MalK83 0.046 ? 0.007 0.057 ? 0.006 6 ? 1 5 ? 0.7 NA MalF396-MalG185 0.06 ? 0.01 0.070 ? 0.010 6 ? 1 4 ? 0.9 NA MalF446-MalG234 0.085 ? 0.008 1.18 ? 0.13b 7 ? 2 14 ? 3 >45c 84 complex (Table 4.1). The fluorescence spectra were recorded between 365 and 520 nm to monitor the conformational changes in the presence of ATP, MalE, and maltose (Figure 4.5). With the pairs MalK83/MalK83 and MalF396/MalG185, which monitor the conformational changes on the cytosolic side of the membrane, ATP decreased the monomeric emissions but increased the excimer emissions (Figure 4.5a and 4.5b). The reverse was observed with the pair MalF446/MalG234 (Figure 4.5c). This was expected because the pair monitors the conformational change on the opposite side of the membrane. In contrast, MalE and maltose had no effect on the monomeric and excimer emissions, in agreement with the conclusion that ATP alone supports the outward-facing conformation of the transporter. Furthermore, we observed that MalE and maltose had no effect on the kinetics of the conformational transition (Figure 4.6). These time-resolved experiments showed that liganded MalE is unable to modify the kinetics of MalK closure, even though we employed a cysteine pair that produced the highest fluorescence signal difference (i.e., MalK83/MalK83). It therefore seems that liganded-MalE is not required, nor does it facilitate the closure of the nucleotide-binding domain in the intact and functional maltose ABC transporter. 85 Figure 4.5 Neither MalE nor maltose facilitates the transition to the outward-facing state. The steady state fluorescence emission spectra of pyrene-labeled MalFGK2 recorded in the absence (black traces) or presence of ATP, MalE, and maltose (red traces) using the following pyrene-labeled MalFGK2 transporters: MalK83/MalK83 (a), MalF396/MalG185 (b), and MalF446/MalG234 (c). The fluorescence spectra were recorded from 365 to 520nm with excitation wavelength at 345 nm. The fluorescence spectra recorded around 465 nm are magnified in the insets. 86 Figure 4.6 Transition kinetics toward the outward-facing conformation. The time course fluorescence emission spectra for the pair MalK83/MalK83 were monitored after the addition of ATP, MalE, and maltose as indicated. The spectra were recorded at 375 nm to monitor pyrene monomers (a) and 460 nm to monitor exited state dimers (b). The fluorescence is expressed in arbitrary units as in Figure 4.5. 4.3.4 MalFG exists in an outward-facing conformation In a distinct study (Bao H and Duong F, unpublished data), we analyzed the effect of MalK on the conformation of the isolated MalFG membrane domain. MalFGK2 forms a tight complex in the membrane and, as is often the case for these membrane assemblies, it is difficult to separate the subunits or express them individually. It has been reported however that MalK modified with a N-terminal 23 amino-acid histidine tag extension (termed HTMalK) can be released from the membrane with urea treatment (Sharma et al, 2005). We therefore expressed HTMalK together with N-terminal His6-tagged MalFG (hereafter termed HTMalFG). This 87 transporter (termed HTMalFG- HTMalK) supports maltose transport in vivo and in vitro with kinetics similar to the wild type (Figure 4.7a and Figure 4.10a). After treatment of the membrane with urea, HTMalFG was purified by Ni2+-affinity and gel filtration chromatography. The complex HTMalFG has an apparent molecular weight of ~95 kDa in detergent solution and is completely devoid of HTMalK (Figure 4.7b). Figure 4.7 Expression and purification of MalFG a, Strain ED169 (?malB) containing the indicated plasmids were plated on MacConkey agar plate containing 0.4 % maltose for 12 hours at 37?C. The red color shows the capacity of the strain to transport and ferment maltose. (-): parental pTrc99a plasmid; (WT): untagged MalFGK2; (hisMalFG + hisMalK): N-terminal His6-tagged MalFG coexpressed with N-terminal His6-tagged MalK. b, Analysis of the purified MalFG and MalFGK2 complexes by size exclusion chromatography (Superdex S200) in TSG buffer plus 0.01% DDM. The fractions 6 to 12 ml are analyzed by 12% SDS-PAGE and Coomassie blue staining. 88 We measured the conformation of reconstituted MalFG in the membrane bilayer using the cysteine pair MalF396C-MalG185C and MalF446C-MalG234C. In contrast to the intact MalFGK2, HTMalFG was found in an outward-facing conformation. As expected, the equilibrium conformation of HTMalFG was not modified by ATP (lane 6; Figure 4.8a and 4.8b). Thus, in the absence of MalK, the membrane domain can spontaneously reach an outward-facing state. MalK therefore seems essential to stabilize MalFG in the inward-facing state. Figure 4.8 Without MalK, MalFG resides in an open conformation a-b, the conformation of MalFG and MalFGK2 reconstitituted in proteoliposomes (2?M) is detected using the homobifunctional cross-linker BMOE (50?M). The cartoon represents the inward- and outward-facing conformation of MalFG with the cysteine residues depicted as red spheres. Proteoliposomes were incubated with or without ATP (1mM) for 10 min at room temperature. Unreacted cysteine were modified with NEM (5mM) prior to protein analysis by SDS-PAGE (12%) and immunoblot against MalF. Describe how you calculate the % of cross-linked material. The quantification was carried out by densitometry analysis using ImageQuant software. 89 To confirm this observation, we determined the affinity of MalE for MalFG. We have shown that MalE binds with much higher affinity to MalFGK2 in the outward-facing state (Kd ~79 nM) than in the inward-facing state (Kd > 50 ?M). Here, the affinity of MalE to MalFG devoid of MalK was found intermediate, around 1.5 ?M (Figure 4.9a and 4.9b). This result supports the notion that MalFG is oriented toward the periplasm in the absence of MalK. The same result was obtained by native gel electrophoresis (Figure 4.9c). The binding of MalE to MalFGK2 occurred only when transporter had been converted to the outward-facing state with AMPPNP (Figure 4.9c, left panel). In contrast, in the absence of MalK, a large fraction of MalE (>50%) was already bound to MalFG (Figure 4.9c, right panel). We then tested whether the conformational state of MalE affects its binding to MalFG, as previously observed with MalFGK2 in Chapter 3. A MalE variant stabilized in the open state because of a low affinity for maltose (MalE-254) formed a tight complex with MalFG (Figure 4.9c). In contrast, a MalE variant stabilized in the closed state (MalE-DW) was unable to associate with MalFG (Figure 4.9c). Thus, MalFG binds preferentially open state MalE. The results also demonstrate that MalK stabilizes MalFG in the inward-facing conformation, which has very low affinity for MalE. 90 Figure 4.9 Without MalK, MalFG binds MalE with an intermediate affinity a, [125I]-labeled MalE and variants (~5000 cpm, 0.5 ?M each) were incubated at room temperature for 10 min with MalFG nanodiscs (Nd-FG; 1 ?M), in the presence of maltose (1mM) and AMPPNP (1 mM), as indicated. The complex between MalE and MalFG was detected by native gel electrophoresis and autoradiography. In comparison, the experiment using MalFGK2 and MalE was performed in parallel. b, Kinetics of MalE binding to MalFG and MalFGK2. This fluorescence quenching experiment was performed using maltose-free MalE (20 nM), MalFG and MalFGK2 nanodiscs (100 nM), and AMPPNP (1 mM), as indicated. c, Equilibrium titrations of MalE binding (20 nM) to MalFG and MalFGK2 nanodiscs (0-2.5 ?M). Experiments were performed with AMPPNP (1mM) and maltose (1mM), as indicated. 4.3.5 MalFG is converted to the inward-facing conformation upon binding of MalK We tested whether MalK can reverse the conformation of MalFG to the inward-facing state. We confirmed first that the reassembled complex regains maltose transport activity (Figure 4.10a). Next, we monitored the conformation of MalFG using the cysteine pairs MalF396C-MalG185C and MaF446C-MalG234C described above. Compared to MalFG alone (Appendix H, lane 2), a significant fraction of the reassembled complex had reverted to an inward-facing conformation, as judged by decreased crosslink efficiency between MalF396C-MalG185C and increased crosslinking between MaF446C-MalG234C (Appendix H, lane 4). Furthermore, the reassembled complex regained sensitivity to nucleotides since it switched back to the outward-facing conformation in the presence of ATP (lane 5). The capacity of MalK to revert the 91 conformation of MalFG was also tested in nanodiscs (Figure 4.10b and 4.10c). The addition of MalK abolished the binding of MalE to MalFG (Figure 4.10d, compare lane 2 to lane 5), whereas the binding of MalE was restored upon addition of non-hydrolysable ATP (lane 7). Together, these results show that apo-MalK binds to MalFG and reverts the conformation of the membrane domain to the inward-facing state. Figure 4.10 Apo-MalK triggers MalFG inward-facing conformation. a, Maltose transport activity of re-assembled MalFGK2. Wild type MalFGK2 and re-assembled MalFGK2 (MalFG incubated with his-MalK) were incorporated into proteoliposomes. Maltose transport activity was determined in the presence of MalE (10 ?M) and [3H]-maltose (50 ?Ci/?mol, 20 ?M). b, re-assembly of MalK onto MalFG. MalFG nanodiscs (1?M) were incubated with MalK (0-5 ?M) for 5 min at room temperature. The proteins were analyzed by native electrophoresis and immunoblot against MalF. c, binding MalE to MalFG with or without MalK. MalFG nanodiscs (1?M) were incubated with MalK (10 ?M) for 5min at room temperature. Samples were then incubated with [125I]-MalE (~5,000 cpm, 0.5 ?M) in the presence of ATP or ATP + vanadate (1mM) before analyze by native gel electrophoresis and autoradiography. 92 4.3.6 Tightly bound MalK inhibits the transition of MalFG to the outward-facing conformation While the transporter made by HTMalFG and HTMalK is fully functional (Figure 4.7a and Figure 4.10a), we observed that the transporter made by HTMalFG and WTMalK is inactive (Figure 4.11a and 3.11b). The mutant (labeled HTMalFG-WTMalK) was unable to form a stable complex with MalE in the presence of AMPPNP or ATP plus vanadate (Figure 4.11c) because of a 3 fold decrease in the MalE binding affinity (Figure 4.11d). Since our results above show that HTMalK is loosely bound to HTMalFG (Figure 4.11), we reasoned that WTMalK is perhaps very tightly bound to HTMalFG, and thereby inhibits transition to the outward-facing state. The association of HTMalFG with MalK was tested in the membrane and in detergent. The results showed that WTMalK co-expressed with HTMalFG was highly resistant to membrane urea extraction compared to HTMalK (Figure 4.12a). In detergent solution, the purified complex HTMalFG-WTMalK was also more stable than the wild type (Appendix I). Thus, the results showed that WTMalK is very tightly bound to HTMalFG. This increased association is likely to hinder the transition of HTMalFG to the outward-facing state in the presence of ATP. 93 Figure 4.11 The inactive mutant F6 binds MalK tightly and cannot reach the outward-facing state a, strain ED169 (?MalB) expressing N-terminal His6-tagged MalFG together with wild type MalK (termed F6 mutant) is unable to transport maltose. b, ATPase activities of F6 mutant and wild type MalFGK2 in proteoliposomes. ATPase activities were determined at 37?C in the presence of MalE (2?M) and maltose (1mM), as indicated. c, mutant F6 is unable to reach the outward-facing conformation. Mutant F6 and wild type MalFGK2 reconstituted in nanodiscs (2 ?M) were incubated with [125I]-MalE (~10,000 cpm, 2 ?M) in the presence of AMPPNP or ATP + vanadate (room temperature, 10min), as indicated. The samples were analyzed by native gel electrophoresis followed by Coomassie blue staining and autoradiography.d, equilibrium titration of MalE binding to mutant F6 in the presence of AMPPNP (1mM) and maltose (1mM), as indicated. 4.3.7 Mutations that decrease interaction of MalK with MalFG restore transport To corroborate the hypothesis above, we isolated suppressor mutations into WTMalK. These suppressors were expected to decrease the stability of the HTMalFG-WTMalK complex, and thereby restore transport activity. The malK gene was mutated using error-prone PCR and the active transporters were identified on MacConkey agar plus maltose (Figure 4.12b). Five mutants that formed bright red colonies were isolated and characterized. The results showed that the mutants MalK#1, MalK#2 and MalK#4 were too unstable for purification (Figure 4.12c). The mutants MalK#3 and MalK#5 produced MalFGK2 to a level comparable to the wild type. In 94 agreement with the prediction, the protein MalK#3 and MalK#5 were loosely bound to HTMalFG (Figure 4.12a). As a consequence of this decreased interaction, the protein MalK#3 and MalK#5 were able to promote the transition of HTMalFG to the outward-facing state in the presence of ATP (Figure 4.12d). Figure 4.12 Suppressors of mutant F6 rescue ability to reach outward-facing conformation. a, isolation of F6 suppressor mutations into MalK. Mutagenesis in malK was performed using error prone PCR (see M&M). The mutagensized plasmid was introduced into strain ED170 (?MalF) and cells were plated on MacConkey-maltose plates for 12 hours at 37?C.b, stability of the F6 suppressor mutants. 95 Membranes were prepared from cells expressing F6 or F6 suppressors. After treatment with 6M urea, membranes were solubilized with 1% DDM (4?C, 1h) and subjected to pull down using Ni-NTA Sepharosse beads in TSG buffer containing 0.01% DDM and 5mM MgCl2. Proteins were eluted with TSG buffer containing 0.01% DDM and 500 mM imidazole. The Samples were analyzed by blue-native electrophoresis. c, binding of MalE to F6 suppressors. F6, MalFGK2 wild type and F6 suppressors in nanodiscs (2 ?M ) were incubated with [125I]-MalE wild type (~5,000 cpm, 1 ?M) in the presence of nucleotides as indicated (room temperature, 10min). The samples were analyzed by native electrophoresis and autoradiography. 4.4 Discussion We report that ATP alone is sufficient for the closure of the MalK interface and for the conversion of MalFG to the outward-facing conformation. This conclusion is based on cysteine cross-links and pyrene-based fluorescence of probes placed at diagnostic positions on the transporter. The cysteine substitutions and the pyrene label modestly affect the ATPase activity of the transporter, and do not affect its capacity to import maltose (i.e., 70?80% activity of the wild type; Table 4.1). The conclusion we reach is in fact not surprising but rather in agreement with other studies reporting that ATP drives the dimerization of MalK in the membrane or in solution (Daus et al, 2007b; Hunke et al, 2000a; Smith et al, 2002). The conclusion is also consistent with the x-ray structures that showed the rigid body rotation of MalFG upon closure of MalK (Khare et al, 2009). That ATP alone regulates the conformation of the transporter has also been reported for other type I and type II importers (ModBC and BtuCD, respectively), as well as for ABC exporters (Gerber et al, 2008; Goetz et al, 2009; Hollenstein et al, 2007; Joseph et al, 2011; Locher, 2009). Our results, however, disagree with the accepted notion that closed liganded MalE facilitates the closure of the MalK dimer. This mechanism was proposed (and perhaps intuitively accepted) because rapid ATP hydrolysis activity occurs only in the presence of MalE and maltose. To the best of our knowledge, this model has been directly tested by EPR spectroscopy only (Orelle 96 et al, 2008). This study concluded that MalK closure strictly depends on MalE. The spin label reduced the MalK ATPase activity by 2-fold (from 2,100 to 1,150 nmol/min/mg), yet the transporter was still active. The spin labels were, however, placed at two different positions on the same MalK subunit, giving rise to four different possible ways of spin interactions. Most importantly, the EPR distance measurements were performed in detergent solution, whereas the transporter ATPase is ?100-fold higher than in the membrane and poorly dependent on MalE and maltose (Bao & Duong, 2012). The sensitivity of EPR in the membrane environment is indeed problematic because of the large spin decay inherent to intermolecular dipolar interactions (McHaourab et al, 2011), and accordingly, the authors could not report the interspin distances with the transporter reconstituted in liposomes. Nevertheless, this pioneering work showed that ATP is essential for the pairing of MalK, and this conclusion is consistent with our results. Our results also contrast with the interpretation of the crystal structure that captured closed liganded MalE with the transporter in a semi-closed conformation (Oldham & Chen, 2011a). It was concluded that closed liganded MalE binds to MalFG and induces the preclosure of MalK and thus ATP hydrolysis. Our pyrene-based fluorescence experiments rather show that the kinetics of MalK pairing is the same, whether MalE and maltose are present or not (Fig. 4.6). In addition, we find that the binding of closed liganded MalE to the inward-facing transporter is very low in solution (Kd > 45 ?M; Table 4.2). In contrast, the binding of MalE to the outward-facing transporter occurs with nanomolar affinity (Kd = ?60 nM; Table 4.2). Clearly, a far more dominant interaction is taking place between ATP-bound outward-facing MalFGK2 and unliganded open state MalE. Given the high concentration of ATP in the cytosol (?1 mM), we think this mode of interaction represents the resting state of the transporter. This view is supported by an in vivo cross-linking analysis, which reported that MalE is bound to the transporter even in the absence of maltose (Daus 97 et al, 2007a). This view is also consistent with our conclusion that transporter-bound MalE forms a receptor for maltose (Bao & Duong, 2012). In contrast, a model involving the low affinity interaction between closed liganded MalE and MalFGK2 (Kd > 45 ?M), as a way to deliver the sugar and activate the transporter, seems unsustainable because the concentration of periplasmic MalE decreases further as the environmental maltose concentration decreases. Our data provide evidence that nucleotide-free MalK stabilize MalFG in the inward-facing state. This information is important because how the transporter switches to the inward-facing state after hydrolysis of ATP is unknown. This post-hydrolytic state has not yet been crystallized. In addition, why the transporter rests in the inward-facing conformation in the absence of nucleotide is unclear. At least two models can be proposed. In the first (Figure 4.1a), MalFG is naturally stable in the inward-facing state, perhaps due to constraining action of phospholipids on the ?-helical transmembrane segments (Orelle et al, 2008). This resting conformation of MalFG would therefore oppose the closure of MalK in the absence of nucleotide, and also facilitates the return of the transporter to the inward-facing conformation after hydroysis of ATP. In the second model (Figure 4.1b), nucleotide-free MalK plays an important role in maintaining MalFG in the inward-facing state. Using crosslinking and MalE-binding assays, we find that MalFG alone exists rather in an outward-facing conformation. For MalFGK2, the conversion from the inward- to outward-facing conformation is thought to create a significant energetic barrier so that ATP binding and liganded MalE are required for this conformational change to occur (Davidson, 2002). However, we find that the conformation of MalFG alone is closer to the higher energetic state. This observation is consistent with previous genetics studies which show that two single mutations in the transmembrane domains are sufficient to increase of transporter ATPase activity by 1000 fold (Covitz et al, 1994). Since 98 only the outward-facing conformation of MalFGK2 is able to hydrolyze ATP, these results suggest that the transmembrane domains are able to change the conformational equilibrium of MalFGK2 to the high energetic state. Indeed, crosslinking and accessibility assays confirm that these mutants rest in the outward-facing conformation (Daus et al, 2007b; Daus et al, 2006; Mannering et al, 2001). Since MalFGK2 rests in the inward-facing conformation in the absence of ATP, it is thus expected that binding of MalK onto MalFG shifts the transmembrane domains into the inward-facing state. This implies that the open MalK dimer stabilizes MalFG in the inward-facing conformation, which argues against the hypothesis that the inward-facing conformation of MalFG restrains MalK dimer in the open state (Orelle et al, 2008). Together, these results indicate that MalK exploits ATP binding but also hydrolysis to stabilize the transporter in two distinct conformations. We also examine the effect of the association of NBDs with TMDs on the activities of the transporter. In comparison to the wild type, a mutant (HTMalFG-WTMalK) exhibits increased association of MalK with MalFG, whereby it becomes inactive and unable to reach the outward-facing conformation. This defect of HTMalFG-wtMalK can be rescued by mutations of MalK that decrease the association of MalK with MalFG. It is thus reasonable to propose that the affinity between TMDs and NBDs contributes to the regulation of transporter activities. This might occur due to an essential rotational movement of the ?-helical subdomain of MalK around the interface between TMDs and NBDs during the conformational change of the transporter from the inward- to outward-facing conformation (Khare et al, 2009; Orelle et al, 2010). In line with this hypothesis, it is also reported that the conformational changes of histidine ABC transporter (HisQMP2) are in concert with changes in the affinity between NBDs and TMDs (Liu et al, 1999). 99 The results that MalE and maltose do not facilitate the conversion of the transporter from the inward- to outward-facing state, prompt a reconsideration regarding the function of MalE and maltose in the conformational changes of MalFGK2. In chapter 3, we show that MalE and maltose might stimulate Pi release in the ATP hydrolytic cycle of MalFGK2. Earlier studies suggested that ATP product release may coincide with the reorientation of MalFGK2 to the inward-facing conformation (Orelle et al, 2008). For the mammalian P-glycoprotein, it was proposed that phosphate release is coupled to transport (Urbatsch et al, 1995). Therefore, we suspect that MalE and maltose perhaps stimulate a step that occurs when the transporter returns to the inward-facing conformation. This hypothesis is compatible with the auto-regulation model presented in Figure 4.1b. 100 Chapter 5: Phosphatidylglycerol directs binding and inhibitory action of EIIAGlc protein on the maltose transporter 5.1 Introduction Bacteria selectively metabolize certain sugars through a mechanism termed carbon catabolite repression (Gorke & Stulke, 2008). In enteric bacteria, the phosphoenolpyruvate carbohydrate phosphotransferase system regulates the selective utilization of these carbon sources (Postma et al, 1993). The phosphotransferase system consists of a sugar transporter and a phosphorylation system that is composed of at least three distinct components: the enzyme EI, the phosphocarrier protein HPr, and several sugar-specific enzymes called EII. Transport of a preferred sugar across the membrane leads to the transfer of a phosphoryl group from phosphoenolpyruvate to the different EII proteins, whose action is to reduce utilization of nonpreferred carbon sources (Deutscher et al, 2006). Among the different EII proteins, the role of the glucose-specific EIIAGlc has been particularly well studied. The dephosphorylated form of EIIAGlc, present during glucose transport, is responsible for the allosteric inhibition of several permeases and kinases involved in the import of maltose, lactose, melibiose, and glycerol (Postma et al, 1993). The astonishing capacity of EIIAGlc to regulate the activity of numerous enzymes, located both in the cytosol and within the membrane, has raised some interesting questions regarding the mechanism of recognition and interaction (Cai et al, 2003; Deutscher et al, 2006; Hurley et al, 1993; Wang et al, 2000a; Worthylake et al, 1991). On its own, the protein consists of an unstructured N-terminal tail (residues 1?18) attached to a globular core (residue 19?168) made by an antiparallel ?-sheet sandwich (Wang et al, 2000b). Structural analyses of EIIAGlc in complex 101 with some of its cytosolic effectors, such as the phosphocarrier protein HPr, the glycerol kinase, and the subunit EIIBGlc, have revealed a common binding surface on the globular core of EIIAGlc (Cai et al, 2003; Hurley et al, 1993; Wang et al, 2000a). For the membrane permease, a limited number of studies based on peptide mapping and site-directed mutagenesis concluded that the same binding surface is also involved in the recognition of the maltose and lactose permeases (Bluschke et al, 2006; Sondej et al, 2000). However, the affinity of EIIAGlc for these permeases is weak, and the modality of inhibition remains obscure, in part due to the difficulty of isolating complexes suitable for structural analysis. Interestingly, it was reported that the N-terminal tail of EIIAGlc is essential for the inhibition of the lactose and maltose permeases, but not for inhibition of cytosolic proteins such as HPr (Bluschke et al, 2006; Meadow & Roseman, 1982; Misko et al, 1987; Wang et al, 2000b). It was also found that a synthetic peptide corresponding to the N-terminal tail of EIIAGlc could adopt an amphipathic helical structure in the presence of phosphatidylglycerol (PG) lipids (Wang et al, 2003; Wang et al, 2000b). Together, these earlier observations hint at a possible mechanism to increase the binding of EIIAGlc to the membrane permeases. In this work, we investigate the association of EIIAGlc with the maltose transporter MalFGK2. The transporter consists of two membrane-integral subunits, MalF and MalG, and two copies of the ATPase subunit, MalK. We show that phosphatidylglycerol and the N-terminal amphipathic tail of EIIAGlc are essential for the inhibition of the ATPase activity of the transporter. Using site-directed cross-linking experiments, we map the interaction of EIIAGlc to the nucleotide-binding domain and the C-terminal regulatory domain of the MalK dimer. Analysis of the ATPase cycle under single and multiple turnover conditions shows that EIIAGlc does not change the affinity of MalK for nucleotide but instead inhibits its capacity to cleave ATP. 102 5.2 Materials and methods 5.2.1 Materials Cross-linkers disuccinimidyl suberate and succinimidyl 3-(2-pyridyldithio)-propionate (SPDP) were from Thermo Scientific, and 1,3-propanediyl bismethanethiosulfonate (MTS-3-MTS) was from Toronto Research Chemicals. EIIAGlc was purified as described (Appendix A.3). 5.2.2 Reconstitution of MalFGK2 The reconstitutions of the maltose transporter in nanodiscs and proteoliposomes were performed as described in Chapter 2. 5.2.3 Cross-linking reactions The cross-linking reactions using disuccinimidyl suberate and SPDP were performed in HM buffer (50 mM K-HEPES, pH 7.5; 10 mM MgCl2). The cross-linking reactions using MTS-3-MTS were in TM buffer (50 mM Tris-HCl, pH 8.0; 10 mM MgCl2). The MalFGK2 proteoliposomes (2 ?M) and EIIAGlc (10 ?M) were mixed together and incubated with 100 ?M of the indicated cross-linker for 20 min at room temperature. The reactions were stopped with Tris-HCl (100 mM) or N-ethylmaleimide (5 mM) where appropriate. Proteins were dissolved in sample buffer and analyzed by SDS-PAGE and Western blotting against MalK. 5.2.4 Other methods The rate of ATP hydrolysis (production of Pi) was determined by the malachite green method. Fluorescence assay for TNP-ATP binding and Thin-layer Chromatography analyses were carried out as described in Chapter 3. For the co-sedimentation assays, the MalFGK2 proteoliposomes (5 ?M) and the indicated amount of EIIAGlc were incubated in TM buffer for 5 min at room temperature. The samples were diluted 25-fold into Tris-HCl (20 mM, pH 8), collected by ultracentrifugation (100,000 ? g, 1 h), and resuspended in Tris-HCl (20 mM, pH 8) 103 followed by SDS-PAGE analysis. The automatic protein docking analysis was performed on the ClusPro 2.0 Web server, using the crystal structures of EIIAGlc (PDB code: 1F3G) and MalFGK2 (PDB code: 3FH6 and 2R6G). 5.3 Results 5.3.1 The Inhibition by EIIAGlc depends on the N-terminal tail and PG lipids EIIAGlc inhibits the ATPase activity of the maltose transporter reconstituted in proteoliposomes by ?4-fold, as reported previously (Bluschke et al, 2006; Figure 5.1b). We show here that the ATPase activity of the transporter is virtually unaffected by EIIAGlc when MalFGK2 is maintained in detergent solution or reconstituted in nanodiscs at a low lipid ratio (Figure 5.1b). These last observations raise the possibility that membrane lipids are necessary for inhibition. Accordingly, an ?4-fold inhibition similar to that seen in proteoliposomes was obtained when the transporter was reconstituted in lipid-rich nanodiscs (Figure 5.1b). Significantly, the inhibition of MalFGK2 ATPase showed a cooperative dependence on EIIAGlc concentration, with a Hill coefficient of 1.8 (Figure 5.1c). This last result suggests a 2:1 stoichiometry of interaction between EIIAGlc and MalFGK2. Because a peptide corresponding to the N-terminal tail of EIIAGlc (residues 1?18) possesses affinity for phosphatidylglycerol (Wang et al, 2000b), we hypothesized that the inhibition of the maltose transporter in proteoliposomes also depends on the presence of the lipid in the membrane or in the disc. EIIAGlc was therefore incubated with the transporter reconstituted in proteoliposomes made with DOPG lipids, DOPC lipids, or a mixture of DOPC (70%) and DOPG (30%). The results only showed a strong inhibition of the MalK ATPase activity when DOPG was present in the membrane (?75% reduction, Figure 5.1d). The transporter ATPase activity was barely reduced when the transporter was reconstituted with only DOPC lipids 104 (?15% reduction, Figure 5.1c). To show that the N-terminal tail of EIIAGlc is necessary for inhibition, we employed the mutant EIIAGlc ?1?18 (Figure 5.1d). As expected, this mutant protein was unable to inhibit the transporter ATPase activity (less than ?10% reduction). Figure 5.1 The activity of EIIAGlc depends on intact N-terminal_-helix and PG lipids. a, working model for EIIAGlc binding to the membrane. The N-terminal amphipathic ?-helix (residues 1?18) of EIIAGlc is colored in red. b, the maltose transporter was reconstituted in proteoliposomes (pL) and lipid-rich nanodiscs (high lipid ratio) using total E. coli lipids. The ATPase activity of the transporter (2 ?M each) was determined at 37 ?C in the presence of EIIAGlc (96 ?M), MalE (10 ?M), and maltose (1mM), as indicated. The activity of the maltose transporter in detergent solution or reconstituted in nanodisc at low lipid ratio is presented for comparison. Three independent experiments were analyzed (mean and S.D.). c, the inhibition of MalFGK2 ATPase activity is cooperative. The maltose transporter was reconstituted in proteoliposomes (2 ?M) made with DOPG. ATPase activities were determined in the presence of MalE (10 ?M) and maltose (1 mM). Three independent experiments were analyzed (mean and S.D.). The data (black squares) were fit to a one-site binding equation (dashed line) or to its expanded version, which includes a term for the Hill coefficient (solid line). The inset shows a magnification of the data up to 20 ?M EIIAGlc. d, the maltose transporter was reconstituted in proteoliposomes (2 ?M MalFGK2) made of E. coli total lipids (pLE.c), DOPG (pLPG), DOPC (pLPC), or a mixture of DOPC and DOPG (70 and 30%, respectively; 105 pLPG + PC). The transporter ATPase activity in the presence of MalE (10 ?M) and maltose (1 mM) was determined at 37 ?C, supplemented with EIIAGlc (96 ?M) and EIIAGlc?1?18 (98 ?M), as indicated. Three independent experiments were analyzed (mean and S.D.) 5.3.2 PG lipids are necessary for the binding of EIIAGlc to MalFGK2 We employed co-sedimentation assays to monitor the binding of EIIAGlc to MalFGK2. The results show that sedimentation of EIIAGlc occurs only when the MalFGK2 proteoliposomes were made with DOPG lipids. Very little co-sedimentation of EIIAGlc occurred with proteoliposomes made with DOPC (Figure 5.2a). The sedimentation of EIIAGlc was also reduced to background level when the N-terminal amphipathic helix of EIIAGlc was deleted (Figure 5.2b). We note that a significant level of binding of EIIAGlc occurs at the surface of the liposomes made with DOPG lipids (Figure 5.2c). Thus, to demonstrate that EIIAGlc binds to MalFGK2 and not merely to acidic lipids, we employed the amine-reactive homobifunctional cross-linker disuccinimidyl suberate. In that case, a prominent cross-link was formed between MalK and EIIAGlc, but only when the proteoliposomes contained DOPG lipids (Figure 5.2d). Together, these results demonstrate that phosphatidyl glycerol lipids direct the binding of EIIAGlc to the maltose transporter. The binding depends on the N-terminal amphipathic tail of EIIAGlc. 106 Figure 5.2 PG lipids control for the binding of EIIAGlc to MalFGK2. a, the transporter, reconstituted in proteoliposomes (pL) with DOPC (pLPC) or DOPG (pLPG) lipids, was incubated at room temperature for 5 min with the indicated concentration of EIIAGlc. The proteoliposomes were isolated by ultracentrifugation, and the amount of EIIAGlc bound to MalFGK2 was visualized by SDS-PAGE and Coomassie Blue staining of the gel. b, same as a but using EIIAGlc?1?18. The protein does not co-sediment and therefore does not appear on the SDS-PAGE analysis. c, same as a but using liposomes devoid of MalFGK2. d, the proteoliposomes in a were incubated with EIIAGlc (10 ?M) and the amine-reactive cross-linker disuccinimidyl suberate (100 ?M) for 20 min at room temperature. A control experiment with EIIAGlc alone was performed in parallel (right lane). The reactions were stopped with Tris-HCl (100 mM, pH 8). The cross-link products were identified by Western blot using antibodies against MalK. The bands of high molecular weight represent cross-links between MalK and MalFG. An arrow indicates the cross-link between MalK and EIIAGlc. 5.3.3 Identification of the binding interface between EIIAGlc and MalK First we employed a molecular modeling approach to identify the potential binding interface between EIIAGlc and MalFGK2. The structure of EIIAGlc (PDB code: 1F3G) was docked onto the crystal structure of the maltose transporter (PDB code: 2R6G and 3FH6) using the protein-protein docking server ClusPro (Kozakov et al, 2010). This protein docking algorithm uses the fast Fourier transform correlation approach combined with an automatic clustering method to propose interactive surfaces with favorable free energies (Kozakov et al, 107 2010). From the different models proposed (data not shown), we retained the models where EIIAGlc is interacting with the MalK part of the transporter (Figure 5.3). Indeed, the mutations that render the transporter resistant to the inhibitory action of EIIAGlc are located in the nucleotide-binding domain (NBD) and the C-terminal domain of the MalK ATPase unit (Bohm et al, 2002; Dean et al, 1990b; Kuhnau et al, 1991). It was not possible to determine the stoichiometry of EIIAGlc binding using this automatic docking analysis because the computer program uses a 1:1 mode of interaction. However, because MalK is a symmetric dimer, it can be deduced that two binding sites for EIIAGlc exist on the maltose transporter, in agreement with the Hill coefficient determined above (Figure 5.1c). Figure 5.3 Model of interaction between EIIAGlc and MalFGK2. The model of interaction was generated with the automatic protein docking server ClusPro using the crystallography structures of EIIAGlc (PDB code: 1F3G) and MalFGK2 (PDB code: 3FH6). a, lateral view of the complex MalFGK2-EIIAGlc shown with the membrane plane. Cyan, MalF; orange, MalG; blue and green, MalK dimer; yellow, EIIAGlc. Because the MalK dimer is symmetric, two EIIAGlc molecules are 108 bound per MalFGK2 complex. This is not shown in the docking analysis because the computer program uses a 1:1 mode of interaction. b, magnified view of EIIAGlc interface with MalK. The colors are the same as in a. In the fully closed ATP-bound conformation, the ATP molecule is contacting residues in the Walker A motif (red) from one MalK and the LSGGQ motif (purple) of the other. Next, based on the proposed models, we introduced a series of unique cysteine residues into the NBD and C-terminal domain of MalK (Figure 5.4). The protein complexes were purified, reconstituted into proteoliposomes with DOPG lipids, and incubated with EIIAGlc in the presence of an amine-to-sulfhydryl cross-linking reagent (SPDP; spacer arm ?7 ?). The protein cross-links were detected by nonreducing SDS-PAGE and immunoblot against MalK. The cysteine positions that formed a covalent bond with EIIAGlc were the following: Cys15, Cys40, Cys128, Cys276, and Cys324 (Figure 5.4a). Interestingly, these residues are located on the opposite sides of the MalK monomer but cluster together on the same side when MalK forms a dimer (Figure 5.4c). This pattern of cross-linking is consistent with the working model presented in Figure 5.3. Finally, to confirm the orientation of EIIAGlc when it is bound to MalFGK2, two unique cysteine residues were introduced at positions EIIAGlc-97C and EIIAGlc-147C, respectively (Figure 5.4c). We then employed a sulfhydryl-to-sulfhydryl cross-linking reagent (MTS-3-MTS; spacer arm of 5 ?) to identify the neighboring cysteine residues on MalK. This cross-link analysis showed that EIIAGlc-97C is proximal to MalK-276C and MalK-324C, whereas EIIAGlc-147C is proximal to MalK-15C and MalK-40C (Figure 5.4b). This cross-link pattern is consistent with the working model above, where each EIIAGlc binds simultaneously the NBD domain of one MalK and the C-terminal domain of another MalK. Furthermore, this mode of interaction places the N-terminal tail EIIAGlc in proximity to the phospholipid bilayer. 109 Figure 5.4 The cross-linking analysis supports the docking analysis prediction. a, the MalFGK2 complex carrying the indicated cysteine residue was reconstituted in proteoliposome. The proteoliposomes (2 ?M each) were incubated with wild type EIIAGlc (10 ?M) in the presence of the sulfhydryl- and amino-reactive cross-linker SPDP (100 ?M). The reactions were stopped with Tris-HCl (100 mM, pH 8). The proteins were dissolved in sample buffer, and the cross-link products were detected by immunoblot against MalK. The cross-links corresponding to a complex between MalK and EIIAGlc are 110 annotated by a red asterisk. The cross-link products MalK-EIIAGlc have two distinct mobilities during electrophoresis. b, the proteins EIIAGlc-97C and EIIAGlc-147C were incubated with the indicated MalFGK2 proteoliposomes in the presence of sulfhydryl-reactive homobifunctional cross-linker MTS-3-MTS (100 ?M, 20 min, room temperature). The reactions were stopped by N-ethylmaleimide (5 mM). The cross-link products were detected by immunoblot against MalK. The red asterisk denotes the cross-links between MalK and EIIAGlc. The cross-linking pattern obtained with wild type EIIAGlc is presented for comparison. c, the cysteine residues introduced into MalK and EIIAGlc are indicated by the red spheres. 5.3.4 EIIAGlc does not inhibit the ninding of ATP to MalK How EIIAGlc inhibits the ATPase activity of MalK is unknown. EIIAGlc may prevent ATP binding, ATP hydrolysis, or the release of hydrolysis products (Figure 5.5f). To address this question, we employed the fluorescence analog TNP-ATP. We employed the MalFGK2 complex reconstituted into lipid-rich nanodiscs. The equilibrium titrations revealed a dissociation constant (Kd) for TNP-ATP of ?9.4 ?M (Figure 5.5a). The apparent Kd for ATP, measured by competitive replacement of TNP-ATP, was estimated to be ?220 ?M (Figure 5.5b). This value is similar to the Km derived from the ATPase measurements in lipid-rich nanodiscs and proteoliposomes (from ?200 to ?280 ?M, respectively; Figure 5.5c and (Davidson et al, 1996)). When these measurements were repeated in the presence of EIIAGlc, no significant change in the affinity values was detected, thus indicating that EIIAGlc does not inhibit the binding of ATP to the transporter (Table 5.1). This conclusion is consistent with the docking analysis in Figure 5.3b as the nucleotide binding site remains accessible to ATP. 5.3.5 EIIAGlc inhibits the cleavage of ATP by MalK We then tested whether EIIAGlc inhibits the hydrolysis of ATP and/or the release of ADP and Pi. The MalFGK2 complex reconstituted in lipid-rich nanodiscs was incubated with [?-32P]ATP and [?-32P]ATP in the presence of MalE and maltose. After incubation, the free nucleotides were removed by centrifugal gel filtration. The nucleotides remaining bound to the transporter were detected by thin layer chromatography and autoradiography. As a control, the 111 MalFGK2 complex was incubated with sodium vanadate. With this phosphate analog, ADP remains trapped in the nucleotide-binding pocket of MalK (Chen et al, 2001). Accordingly, there was a significant increase in the amount of ADP that was co-purified with the transporter in the presence of vanadate (Figure 5.5d, compare lane 4 with lane 6). In contrast, whether EIIAGlc was present or not, the amount of ADP and Pi that co-purified with MalFGK2 was unchanged (Figure 5.5d, compare lane 4 with lane 5). Together, these results indicate that EIIAGlc does not increase or decrease the affinity of MalFGK2 for ADP and Pi. Finally, we assessed whether EIIAGlc inhibits the ATP cleavage step by testing the ATPase activity of MalFGK2 under two different conditions: (i) with MalFGK2 present in 20-fold molar excess over the nucleotide, so that only a single round of ATP hydrolysis can occur; and (ii) under steady-state conditions, where the nucleotide is present in 1000-fold excess over MalFGK2 so that multiple rounds of ATP hydrolysis are possible. A control experiment showed that sodium vanadate affects the cleavage of ATP only in steady-state conditions, as expected; vanadate does not inhibit the chemistry of ATP hydrolysis, but rather the release of ADP (Figure 5.5e, compare lane 3 with lane 6). In the presence of EIIAGlc, however, the number of ATP molecules hydrolyzed was decreased by more than 50%, under both single and multiple turnover conditions (Figure 5.5e). Together, these results show that the binding of EIIAGlc to the transporter inhibits the cleavage of ATP, and not the binding or the release of the nucleotides from MalK. 112 Table 5.1 Equilibrium constants of TNP-ATP and ATP for the maltose transporter MalFGK2 Kd of TNP-ATP (?M) Kd,app of ATP (?M) Km of ATP (?M) MalFGK2 9.4 ? 1.2 222 ? 31 279 ? 47 MalFGK2 + EIIAGlc 10.1 ? 1.3 236 ? 29 294 ? 64 113 Figure 5.5 EIIAGlc inhibits the cleavage of ATP but not the binding to MalK. a, equilibrium titration of MalFGK2 with TNP-ATP.The MalFGK2 complex reconstituted in lipid-rich nanodiscs was incubated with TNP-ATP (3 min, 25 ?C) in the presence of MalE (10 ?M), maltose (1mM), and EIIAGlc (96 ?M) as indicated. The fluorescent data (mean ? S.D., n=3 independent replicates) were fit to Equation 1 to determine the affinity of TNP-ATP for MalK.. b, equilibrium titration of MalFGK2 with ATP. The MalFGK2 complex reconstituted in lipid-rich nanodiscs was incubated with TNP-ATP (80 ?M) in the presence of MalE (10 ?M), maltose (1mM), and EIIAGlc (96 ?M) as indicated. The fluorescence was measured after the addition of the indicated amount of ATP.The data (mean ? S.D., n =3 independent replicates) were fit to Equations 2 and 3. c, rate of ATP hydrolysis. The rate was determined with the lipid-rich MalFGK2 nanodiscs (2 ?M) at the indicated ATP concentration, in the presence of MalE (10 _M), maltose (1 mM), and EIIAGlc (96 ?M). The data (mean ? S.D., n = 3 independent replicates) were fit to the Michaelis-Menten equation to determine the Km. d, effect of EIIAGlc on the release of ADP and Pi. The lipid-rich MalFGK2 nanodiscs (20 ?M), MalE (75 ?M), and maltose (1mM) were incubated for 10 min at room temperature with 300 ?M [?-32P]ATP (lower panel) or 300 ?M [?-32P]ATP (upper panel), with or without EIIAGlc (96 ?M) as specified. The control experiment with 200 ?M vanadate (Vi) was performed in parallel. The total amount of ADP and Pi produced during the reaction was determined by TLC before centrifugal gel filtration (CGF). The total amount of ADP and Pi remaining bound to the transporter was determined by TLC after centrifugal gel filtration. e, effect of EIIAGlc on ATP hydrolysis in single and multiple turnovers. The lipid-rich MalFGK2 nanodiscs (left panel, 20 ?M; right panel, 1 ?M) were incubated with MalE (75 ?M) and maltose (1mM) at room temperature with [?-32P]ATP (left panel, 1 ?M ATP for 1 min; right panel, 1 mM ATP for 10 min), in the presence of EIIAGlc (96 ?M) or vanadate (200 ?M) as indicated. The reactions were stopped by the addition of ice-cold EDTA (20 mM) and proteinase K (1 mg/ml). Samples were analyzed by TLC and autoradiography. The radioactive spots (mean ? S.D., n = 3 independent replicates) were quantified by densitometry to determine the fraction of ADP produced. f, the ATPase cycle is a succession of four chemical steps: ATP binding, ATP cleavage, phosphate dissociation, and ADP dissociation. MalFGK2 is denoted as a tailed circle. 5.4 Discussion EIIAGlc regulates the activity of at least 10 distinct proteins in the context of glucose uptake and catabolic repression (Gorke & Stulke, 2008; Postma et al, 1993). Furthermore, the regulated proteins are located both in the cytosol and in the membrane and have little or no obvious structural homology with one another (Deutscher et al, 2006; Wang et al, 2000a). Not surprisingly therefore, the regulatory interactions of EIIAGlc with these diverse proteins are generally weak and transient (Cai et al, 2003; Dean et al, 1990b; Wang et al, 2000b). This seems particularly true for the membrane permeases. In vivo, the inhibition of the maltose transporter requires ?5-fold more EIIAGlc than the glycerol kinase (van der Vlag et al, 1994). Because the affinity of EIIAGlc for the glycerol kinase is only ?4 ?M (Novotny et al, 1985), it is likely that the 114 affinity of EIIAGlc for the maltose transporter is even lower. As a direct consequence, the biochemical and structural analysis of the MalFGK2-EIIAGlc complex is difficult, and the molecular basis of the inhibition remains obscure. In the work reported here, we provide direct evidence that the N-terminal tail of EIIAGlc together with PG lipids is essential for high affinity binding to the maltose transporter. It was previously reported that deletion of the N-terminal tail of EIIAGlc relieves the inhibitory activity on the lactose and maltose transporter, but not on the cytosolic effectors (Meadow & Roseman, 1982; Misko et al, 1987; Wang et al, 2000a; Wang et al, 2000b). It was also shown that a peptide corresponding to the N-terminal tail of EIIAGlc adopts an amphipathic ?-helix structure in the presence of PG lipids, but remains in a random coil with PC lipids (Wang et al, 2003; Wang et al, 2000b). Our work thus links these observations together and demonstrates the importance of acidic lipids for directing EIIAGlc to the maltose transporter. We propose that the affinity of EIIAGlc for acidic lipids, combined with a restricted diffusion of the protein bound to the lipid bilayer, serves to increase the otherwise low affinity interaction of EIIAGlc with the maltose transporter. Accordingly, a previous study indicated that EIIAGlc weakly inhibits the ATPase activity of the MalK subunit in solution (Landmesser et al, 2002). It is tempting to speculate that the membrane lipid composition might be an important parameter for the regulation of MalFGK2 by EIIAGlc. In this context, we recently reported that the length of the lipid acyl chain is a strong determinant of the maltose transporter activity (Bao et al, 2013). In addition to its targeting function, the N-terminal tail probably allows the optimal positioning of the C-terminal inhibitory domain of EIIAGlc on the transporter catalytic site. In the absence of co-crystals between EIIAGlc and MalFGK2, we employed a molecular docking method to identify potential binding surfaces. We then introduced a series of unique cysteine 115 residues to perform amine-to-sulfhydryl and sulfhydryl-to-sulfhydryl cross-linking analysis. By combining these approaches, we were able to identify a binding interface comprising the NBD domain and C-terminal domain on the MalK dimer and the ?-strands 5 and 7 on EIIAGlc (Figure 5.4). In our working model (Figure 5.3), the N-terminal tail of EIIAGlc is facing the lipid membrane, whereas the globular domain of EIIAGlc spans the MalK dimer interface, in direct contact with the NBD and C-terminal regulatory domains. Because the MalK dimer is symmetric, two EIIAGlc molecules are bound per MalFGK2 complex. This model is consistent with the cooperative inhibition we observe (Figure 5.1c) and the location of the mutations that cause resistance to EIIAGlc (Bohm et al, 2002; Dean et al, 1990b; Kuhnau et al, 1991). These mutations (i.e. A124T, F241I, G278P, G284S, E119K, R228C, G302D, and S322F) are located on opposite sides on the monomer, but on the same side when MalK forms a dimer. Interestingly, in our working model, the tip of the regulatory domain on MalK remains accessible for binding and segregation of the transcriptional activator MalT (Bohm et al, 2002; Richet et al, 2012). Should this be the case, the action of EIIAGlc and MalT would be complementary, meaning that the cell would be able to inhibit maltose transport and keep mal gene expression to basal level by acting on the same complex simultaneously. How EIIAGlc inhibits the activity of the transporter was an important unanswered question. The dissection of the MalFGK2 ATPase cycle allows us to conclude that EIIAGlc inhibits the cleavage of ATP by MalK, not the binding of the molecule or the release of the hydrolysis products. Because the binding of ATP leads to the closure of the MalK dimer, and because the closure of the MalK dimer is necessary for ATP hydrolysis (Chen et al, 2003), we conclude that EIIAGlc interferes at the closure step. In the magnified view of the modeled EIIAGlc-MalFGK2 complex, part of EIIAGlc is inserted between the NBD domains of MalK 116 (Figure 5.3b). Blocking the closure of the MalK dimer would also stabilize the MalFG membrane domain in its inward-facing conformation (i.e. transporter closed on the periplasmic side). Because MalE binds with a high affinity to only the outward-facing conformation (Bao & Duong, 2013), it is predicted that EIIAGlc also prevents the binding of MalE to the transporter. Finally, how the phosphorylation of one residue (i.e. His90) prevents the action of EIIAGlc on the maltose transporter is another important question. The co-crystal structure of EIIAGlc with the glycerol kinase (PDB code: 1GLA) indicates that the phosphorylation of His90 causes charge repulsion with the residue Glu478 in the glycerol kinase, thereby reducing the interaction and inhibitory capacity of EIIAGlc (Hurley et al, 1993; Wang et al, 2000a). A similar mechanism is perhaps occurring for the maltose transporter. In our working model, His90 on EIIAGlc is within ?10 ? distance to Asp119 on MalK. Further biochemical and structural analysis will tell how phosphorylation of a single residue can suppress the inhibitory activity of EIIAGlc on the maltose transporter. Coincidentally, while a manuscript of this work was under revision at JBC, the crystal structure of MalFGK2-EIIAGlc is reported (Chen et al, 2013), which is similar to the model presented in Figure 5.3. 117 Chapter 6: Summary and Future directions ATP-binding cassette (ABC) transporters are highly conserved in all organisms. Over the past few years, structural analyses have advanced our understanding of this superfamily of membrane proteins, revealing architectural principles that define ABC transporters, unlocking secrets of translocation pathways, and elucidating the mechanism of the ?alternating-access? model (Chen, 2013; Rees et al, 2009). Although crystal structures have revealed static snapshots of transporters in a few conformational states, much less is known about how dynamic each conformational state is, how cofactors and associated proteins affect each conformational state, and how many other conformational states exist. Moreover, crystallization is achieved in detergent micelles, which differ in their physicochemical properties from the lipid bilayer, the natural milieu of membrane proteins. Together these problems prompted us to study the dynamic properties of ABC transporters in a native-like environment, using the maltose transporter MalFGK2 as a prototype. In Chapter 2, we report a general methodology for the reconstitution of MalFGK2 into nanodiscs to achieve our objectives. Nanodiscs place membrane proteins in a defined lipid environment, and render them soluble in aqueous solution in the absence of detergent. Native gel electrophoresis and light scattering analyses of the nanodiscs-embedded MalFGK2 indicate one transporter being incorporated per nanodisc. Electron microscopy analyses show that nanodiscs containing MalFGK2 average a diameter of ~10 nm, consistent with previous characterization of nanodiscs (Denisov et al, 2004). We then investigated the effect of membrane lipids on the transporter activities using nanodiscs. The results show that the acyl-chain length of lipids is critical to reduce the basal ATPase activity of the transporter. Since ABC transporters experience large conformational changes during each transport and ATPase cycle, the movement of TMDs 118 will alter lipid packing in the surrounding bilayer. The energy cost associated with lipid bilayer deformation can directly affect the transporter activity through hydrophobic interactions. Thus, our data reveal the possibility that the thickness of the lipid bilayer plays an important role in regulating ABC transporter activities. Future biophysical experiments are needed to elucidate the action of lipids in further details (Hardie & Franze, 2012). We also demonstrated that the functional interaction of MalFGK2 with MalE can be captured in nanodiscs. Native gel electrophoresis shows that the affinity of the transporter for MalE is regulated by nucleotides. The complex of MalE and the outward-facing MalFGK2 is stablized by ATP and Vi, whereas the complex of MalE and the inward-facing MalFGK2 is isolated by crosslinking (Appendix D). Electron microscopic analyses of these complexes show that the conformational states of the transporter are consistent with previous biochemical data and crystal structures (Bao H, Fabre L, Rouiller I and Duong F, chapter 2 and unpublished results). We believe that further cryo-EM studies of the transporter might allow visualization of other conformational states that have escaped from crystal structural studies. In Chapter 3, the functional consequence of MalE binding on the transport activity of MalFGK2 is examined. We develop a fluorescence quenching assay to study the complex formation between MalFGK2 and MalE. Our data show that the dissociation constant (Kd) of the complex formed by open-state MalE and outward-facing MalFGK2 is ~80 nM, and maltose can cause the disassembly of the complex with a Kd,app[maltose] of ~120 ?M. Using MalE mutants stabilized either in the open or the closed state, we show that closed-state MalE has low affinity for outward-facing MalFGK2. Thus, the decreased affinity of MalFGK2 for MalE in the presence of maltose is due to the conformational change of MalE upon maltose binding from the open- to the closed-state. To examine the corresponding consequence of decreased MalE binding on the 119 transport activity, we employed maltose transport assays, showing that closed-state MalE is not able to stimulate the transporter activity. We also found that the maximal transport rate is inversely correlated to the affinity of MalE for maltose, although MalE significantly decreases the Km,app of maltose. Collectively, we propose an auto-regulation mechanism, in which maltose is able to restrain the maximal transport activity by tuning the affinity of MalE for MalFGK2 (Figure 6.1). Figure 6.1 The comprehensive model for maltose transport. The conventional (black) and auto-regulation (red) model are integrated into this comprehensive model. Comparing ATP hydrolysis under single and multiple turnover conditions, our data implies that maltose-free MalE stimulates ATP cleavage, whereas transport of maltose accelerates the Pi release. This hypothesis will be further examined by analysis of the pre-steady-state ATPase activity using transient kinetic methods. In addition, since Pi release is postulated to coincide with the return of transporter to the inward-facing conformation (Orelle et al, 2008), 120 these data also indicate that transport of maltose stimulates the ATPase activity by accelerating the transition of transporter from the outward- to inward-facing state. To test this prediction, future experiments need to determine the timescale of conformational changes of the transporter. Recent advances in single molecule measurements should allow such analyses (Erkens et al, 2013; Joo et al, 2008; Roy et al, 2008). Based on the above discussion, we expect that the kinetics of Pi release should match the timescale of the conformational change of the transporter from the outward- to inward-facing state. Nanodiscs will be an excellent tool for these studies because a single transporter is reconstituted in a defined lipid environment. How maltose enters into the translocation pathway remains unclear in the auto-regulation model (Figure 6.1). Some molecular dynamics simulations of the ATP-bound MalE-MalFGK2 complex suggest a large amplitude rotation of MalF-P2 loop and MalE on the periplasmic side of the membrane (Oliveira et al, 2011). This movement pulls away the C-lobe of MalE from the transporter, creating sufficient space for maltose to access the C-lobe of MalE and the translocation pathway (Oliveira A and Soares C, personal communication). In the conventional model (Figure 6.1), maltose-bound, closed-state MalE interacts with the inward-facing conformation of MalFGK2, forming a pre-translocation complex. ATP binding to this pre-translocation complex results in the outward-facing conformation of MalFGK2 and opening of MalE, which transfers maltose from MalE to the substrate binding site of the transporter. Clearly, elucidation of the interaction between maltose and the MalE-MalFGK2 complex is essential to understand the transport mechanism. In addition, it also remains to be determined why transporter is specific for maltose. It is established that MalFGK2 is able to transport linear maltooligosaccharides up to seven glucose units long. These substrates, when reduced or oxidized, are not transported but bind tightly to MalE. Complicating matters further is the fact 121 that a number of MalE mutants exhibit high affinity for these substrates, yet they do not support substrate transport. Clearly, these data suggest that binding of a substrate to MalE does not always lead to its transport. Further studies are necessary to explain how the functional interactions among substrates, MalE and MalFGK2 result in the decision making step that routes substrate to the transport process. We are currently using unnatural amino acid and photocrosslinking to identify sites where maltose interacts with MalE and MalFGK2. In Chapter 4, we centered our analyses on the conformational changes of maltose transporter. We have addressed the enigma of how the conformational changes of MalFGK2 between inward- and outward-facing are regulated. We first employed a crosslinking approach to show that ATP alone triggers the outward-facing conformation of MalFGK2. This result was further supported by a fluorescence assay showing that the kinetics of this ATP-dependent conformational change is not changed by MalE and maltose. Furthermore, we showed that the ATP-triggered outward-facing MalFGK2 exhibits high affinity for maltose-free MalE. Together, these results led us to conclude that ATP binding results in the outward-facing MalFGK2, which is able to bind MalE with high affinity and perhaps be primed to transport maltose. Remarkably, the lone transmembrane domain MalFG was found to exist in the outward-facing conformation, whereas reassembly of MalK reverses MalFG to the inward-facing conformation. Clearly, MalK exploits the energy from ATP binding but also from ATP hydrolysis to alternatively stabilize the transporter in the inward- and outward-facing conformations. In Chapter 5, interaction of the maltose transporter with EIIAGlc is investigated. We reconstituted the inhibitory effect of EIIAGlc on the ATPase activity of MalFGK2 in proteoliposomes and nanodiscs. Crosslinking and co-sedimentation assays demonstrate that acidic lipid phosphatidylglycerol and the N-terminal tail of EIIAGlc are essential for the binding 122 and inhibition of EIIAGlc to the maltose transporter. Using site-specific crosslinking and automatic protein docking, we are able to generate a 3D model of the MalFGK2-EIIAGlc complex, showing that EIIAGlc binds to the NBD of one MalK and the C-terminal domain of the other, which is consistent with a recent crystal structure of MalFGK2-EIIAGlc (Chen et al, 2013). Dissection of the ATPase cycle revealed that EIIAGlc does not alter ATP binding to MalK but rather inhibits the ATP cleavage step. Together, these results not only provide a model for the inhibition of the maltose transporter by EIIAGlc, but also imply a simple mechanism for targeting EIIAGlc to different membrane permeases. Future work should focus on how the phosphorylation of one residue (i.e. His90) prevents the action of EIIAGlc on the maltose transporter. In our model of the MalFGK2-EIIAGlc complex, His90 on EIIAGlc is within ?10 ? distance to Asp119 on MalK. Phosphorylation of His90 might cause charge repulsion and thus reduce the inhibitory capacity of EIIAGlc. Further biochemical and structural analyses are necessary to test this hypothesis. In contrast to extensive studies on the transport mechanism, much remains unclear regarding the biogenesis of the maltose transporter. In bacteria, most membrane proteins are integrated co-translationally into the plasma membrane (Park & Rapoport, 2012). This process begins with recognition of the transmembrane sequence of a growing polypeptide chain from ribosome by the signal-recognition-particle (SRP). The ribosome-nascent-chain-SRP complex then engages the SecY translocation channel for the movement of transmembrane segments into the lipid phase. In this context, it is not surprising that in vivo targeting of MalF to the membrane is strongly dependent on the SRP component 4.5S RNA (Wagner et al, 2008). In vitro crosslinking experiments also show that TM1 of nascent MalF interacts with ribosome, SRP and SecY. However, the fact that MalF and MalG aggregate upon separate overexpression indicates that a concerted post-translational mechanism of assembly of MalFGK2 is involved in the 123 biosynthesis of the transporter (Panagiotidis et al, 1993; Sharma et al, 2005). In support of this prediction, our work in Chapter 4 shows that the interface between MalFG and MalK contributes greatly to the integrity of the transporter. Intriguingly, accumulating evidence shows that an assembly-dependent quality control mechanism exists for the biogenesis of membrane proteins in eukaryotic cells (Feigel & Hendershot, 2013; Rabeh et al, 2012). It remains unknown whether a similar mechanism exists in prokaryotes. It is possible that other translocation components such as the YidC and SecCDF proteins might be involved in this process. Our work on the maltose transporter provides an ideal system for such future studies. Summarizing, we have established a new model for maltose transport and elucidated the molecular mechanism for EIIAGlc inhibition. We should within the next decade witness significant progress in our understanding and application of ABC transporters. It might be possible to observe in real time substrate transport and ATP hydrolysis by ABC transporters at the single-molecule level. Maybe substrate specificity could be modified to engineer microorganisms that are able to produce biofuels. Perhaps diseases associated with folding and trafficking of ABC transporters can be overcome by designing new small molecules that restore the defect of their biogenesis pathways. The findings presented in this thesis provide an excellent foundation for such future prospects. 124 References Alvarez FJ, Orelle C, Davidson AL (2010) Functional reconstitution of an ABC transporter in nanodiscs for use in electron paramagnetic resonance spectroscopy. J Am Chem Soc 132: 9513-9515 Ames GF, Liu CE, Joshi AK, Nikaido K (1996) Liganded and unliganded receptors interact with equal affinity with the membrane complex of periplasmic permeases, a subfamily of traffic ATPases. J Biol Chem 271: 14264-14270 Andersen OS, Koeppe RE (2007) Bilayer thickness and membrane protein function: An energetic perspective. Annu Rev Biophys Biomol Struct 36: 107-130 Austermuhle MI, Hall JA, Klug CS, Davidson AL (2004) Maltose-binding protein is open in the catalytic transition state for ATP hydrolysis during maltose transport. J Biol Chem 279: 28243-28250 Bao H, Dalal K, Wang V, Rouiller I, Duong F (2013) The maltose ABC transporter: Action of membrane lipids on the transporter stability, coupling and ATPase activity. Biochim Biophys Acta 1828: 1723-1730 Bao H, Duong F (2012) Discovery of an auto-regulation mechanism for the maltose ABC transporter MalFGK2. PLoS One 7: e34836 Bao H, Duong F (2013) ATP Alone Triggers the Outward Facing Conformation of the Maltose ATP-binding Cassette Transporter. J Biol Chem 288: 3439-3448 Bayburt TH, Sligar SG (2010) Membrane protein assembly into Nanodiscs. FEBS Lett 584: 1721-1727 Benlekbir S, Bueler SA, Rubinstein JL (2012) Structure of the vacuolar-type ATPase from Saccharomyces cerevisiae at 11-A resolution. Nat Struct Mol Biol 19: 1356-1362 Berntsson RP, Smits SH, Schmitt L, Slotboom DJ, Poolman B (2010) A structural classification of substrate-binding proteins. FEBS Lett 584: 2606-2617 Bluschke B, Volkmer-Engert R, Schneider E (2006) Topography of the surface of the signal-transducing protein EIIA(Glc) that interacts with the MalK subunits of the maltose ATP-binding cassette transporter (MalFGK(2)) of Salmonella typhimurium. J Biol Chem 281: 12833-12840 Bohl E, Shuman HA, Boos W (1995) Mathematical treatment of the kinetics of binding protein dependent transport systems reveals that both the substrate loaded and unloaded binding proteins interact with the membrane components. J Theor Biol 172: 83-94 125 Bohm A, Diez J, Diederichs K, Welte W, Boos W (2002) Structural model of MalK, the ABC subunit of the maltose transporter of Escherichia coli: implications for mal gene regulation, inducer exclusion, and subunit assembly. J Biol Chem 277: 3708-3717 Bordignon E, Grote M, Schneider E (2010) The maltose ATP-binding cassette transporter in the 21st century - towards a structural dynamic perspective on its mode of action. Mol Microbiol 77: 1354-1366 Borths EL, Poolman B, Hvorup RN, Locher KP, Rees DC (2005) In vitro functional characterization of BtuCD-F, the Escherichia coli ABC transporter for vitamin B12 uptake. Biochemistry 44: 16301-16309 Bozoky Z, Krzeminski M, Chong PA, Forman-Kay JD (2013) Structural changes of CFTR R region upon phosphorylation: a plastic platform for intramolecular and intermolecular interactions. FEBS J 280: 4407-4416 Brass JM, Higgins CF, Foley M, Rugman PA, Birmingham J, Garland PB (1986) Lateral diffusion of proteins in the periplasm of Escherichia coli. J Bacteriol 165: 787-795 Cai M, Williams DC, Jr., Wang G, Lee BR, Peterkofsky A, Clore GM (2003) Solution structure of the phosphoryl transfer complex between the signal-transducing protein IIAGlucose and the cytoplasmic domain of the glucose transporter IICBGlucose of the Escherichia coli glucose phosphotransferase system. J Biol Chem 278: 25191-25206 Cavigiolio G, Shao B, Geier EG, Ren G, Heinecke JW, Oda MN (2008) The interplay between size, morphology, stability, and functionality of high-density lipoprotein subclasses. Biochemistry 47: 4770-4779 Charalambous K, Miller D, Curnow P, Booth PJ (2008) Lipid bilayer composition influences small multidrug transporters. BMC Biochem 9: 31 Chen J (2013) Molecular mechanism of the Escherichia coli maltose transporter. Curr Opin Struct Biol Chen J, Lu G, Lin J, Davidson AL, Quiocho FA (2003) A tweezers-like motion of the ATP-binding cassette dimer in an ABC transport cycle. Mol Cell 12: 651-661 Chen J, Sharma S, Quiocho FA, Davidson AL (2001) Trapping the transition state of an ATP-binding cassette transporter: evidence for a concerted mechanism of maltose transport. Proc Natl Acad Sci U S A 98: 1525-1530 Chen S, Oldham ML, Davidson AL, Chen J (2013) Carbon catabolite repression of the maltose transporter revealed by X-ray crystallography. Nature 499: 364-368 126 Chromy BA, Arroyo E, Blanchette CD, Bench G, Benner H, Cappuccio JA, Coleman MA, Henderson PT, Hinz AK, Kuhn EA, Pesavento JB, Segelke BW, Sulchek TA, Tarasow T, Walsworth VL, Hoeprich PD (2007) Different apolipoproteins impact nanolipoprotein particle formation. J Am Chem Soc 129: 14348-14354 Covitz KMY, Panagiotidis CH, Hor LI, Reyes M, Treptow NA, Shuman HA (1994) Mutations That Alter the Transmembrane Signaling Pathway in an Atp Binding Cassette (Abc) Transporter. Embo Journal 13: 1752-1759 Dalal K, Duong F (2010) Reconstitution of the SecY translocon in nanodiscs. Methods Mol Biol 619: 145-156 Daus ML, Berendt S, Wuttge S, Schneider E (2007a) Maltose binding protein (MalE) interacts with periplasmic loops P2 and P1 respectively of the MalFG subunits of the maltose ATP binding cassette transporter (MalFGK(2)) from Escherichia coli/Salmonella during the transport cycle. Mol Microbiol 66: 1107-1122 Daus ML, Grote M, Muller P, Doebber M, Herrmann A, Steinhoff HJ, Dassa E, Schneider E (2007b) ATP-driven MalK dimer closure and reopening and conformational changes of the "EAA" motifs are crucial for function of the maltose ATP-binding cassette transporter (MalFGK2). J Biol Chem 282: 22387-22396 Daus ML, Grote M, Schneider E (2009) The MalF P2 Loop of the ATP-Binding Cassette Transporter MalFGK(2) from Escherichia coli and Salmonella enterica Serovar Typhimurium Interacts with Maltose Binding Protein (MalE) throughout the Catalytic Cycle. J Bacteriol 191: 754-761 Daus ML, Landmesser H, Schlosser A, Muller P, Herrmann A, Schneider E (2006) ATP induces conformational changes of periplasmic loop regions of the maltose ATP-binding cassette transporter. Journal of Biological Chemistry 281: 3856-3865 Davidson AL (2002) Mechanism of coupling of transport to hydrolysis in bacterial ATP-binding cassette transporters. J Bacteriol 184: 1225-1233 Davidson AL, Dassa E, Orelle C, Chen J (2008) Structure, function, and evolution of bacterial ATP-binding cassette systems. Microbiol Mol Biol Rev 72: 317-364, table of contents Davidson AL, Laghaeian SS, Mannering DE (1996) The maltose transport system of Escherichia coli displays positive cooperativity in ATP hydrolysis. J Biol Chem 271: 4858-4863 Davidson AL, Nikaido H (1990) Overproduction, solubilization, and reconstitution of the maltose transport system from Escherichia coli. J Biol Chem 265: 4254-4260 127 Davidson AL, Shuman HA, Nikaido H (1992) Mechanism of maltose transport in Escherichia coli: transmembrane signaling by periplasmic binding proteins. Proc Natl Acad Sci U S A 89: 2360-2364 Dawson RJ, Hollenstein K, Locher KP (2007) Uptake or extrusion: crystal structures of full ABC transporters suggest a common mechanism. Mol Microbiol 65: 250-257 Dawson RJ, Locher KP (2006) Structure of a bacterial multidrug ABC transporter. Nature 443: 180-185 Dean DA, Davidson AL, Nikaido H (1989) Maltose transport in membrane vesicles of Escherichia coli is linked to ATP hydrolysis. Proc Natl Acad Sci U S A 86: 9134-9138 Dean DA, Davidson AL, Nikaido H (1990a) The role of ATP as the energy source for maltose transport in Escherichia coli. Res Microbiol 141: 348-352 Dean DA, Hor LI, Shuman HA, Nikaido H (1992) Interaction between maltose-binding protein and the membrane-associated maltose transporter complex in Escherichia coli. Mol Microbiol 6: 2033-2040 Dean DA, Reizer J, Nikaido H, Saier MH, Jr. (1990b) Regulation of the maltose transport system of Escherichia coli by the glucose-specific enzyme III of the phosphoenolpyruvate-sugar phosphotransferase system. Characterization of inducer exclusion-resistant mutants and reconstitution of inducer exclusion in proteoliposomes. J Biol Chem 265: 21005-21010 Denisov IG, Grinkova YV, Lazarides AA, Sligar SG (2004) Directed self-assembly of monodisperse phospholipid bilayer Nanodiscs with controlled size. J Am Chem Soc 126: 3477-3487 Deutscher J, Francke C, Postma PW (2006) How phosphotransferase system-related protein phosphorylation regulates carbohydrate metabolism in bacteria. Microbiol Mol Biol Rev 70: 939-1031 Dietzel I, Kolb V, Boos W (1978) Pole cap formation in Escherichia coli following induction of the maltose-binding protein. Arch Microbiol 118: 207-218 Dimroth P, von Ballmoos C, Meier T, Kaim G (2003) Electrical power fuels rotary ATP synthase. Structure 11: 1469-1473 Dowhan W, Bogdanov M (2012) Molecular genetic and biochemical approaches for defining lipid-dependent membrane protein folding. Biochim Biophys Acta 1818: 1097-1107 Ehrle R, Pick C, Ulrich R, Hofmann E, Ehrmann M (1996) Characterization of transmembrane domains 6, 7, and 8 of MalF by mutational analysis. J Bacteriol 178: 2255-2262 128 Erkens GB, Hanelt I, Goudsmits JM, Slotboom DJ, van Oijen AM (2013) Unsynchronised subunit motion in single trimeric sodium-coupled aspartate transporters. Nature 502: 119-123 Erkens GB, Majsnerowska M, ter Beek J, Slotboom DJ (2012) Energy coupling factor-type ABC transporters for vitamin uptake in prokaryotes. Biochemistry 51: 4390-4396 Feigel MJ, Hendershot LM (2013) Quality Control of Integral Membrane Proteins by Assembly-Dependent Membrane Integration. Mol Cell 51: 297-309 Folch J, Lees M, Stanley GHS (1957) A Simple Method for the Isolation and Purification of Total Lipides from Animal Tissues. Journal of Biological Chemistry 226: 497-509 Forrest LR, Kramer R, Ziegler C (2011) The structural basis of secondary active transport mechanisms. Biochim Biophys Acta 1807: 167-188 Frank GA, Kipnis Y, Smolensky E, Daube SS, Horovitz A, Haran G (2008) Design of an optical switch for studying conformational dynamics in individual molecules of GroEL. Bioconjug Chem 19: 1339-1341 Fulyani F, Schuurman-Wolters GK, Zagar AV, Guskov A, Slotboom DJ, Poolman B (2013) Functional Diversity of Tandem Substrate-Binding Domains in ABC Transporters from Pathogenic Bacteria. Structure 21: 1879-1888 Fyfe PK, McAuley KE, Roszak AW, Isaacs NW, Cogdell RJ, Jones MR (2001) Probing the interface between membrane proteins and membrane lipids by X-ray crystallography. Trends Biochem Sci 26: 106-112 George AM, Jones PM (2011) Type II ABC permeases: are they really so different? Structure 19: 1540-1542 George AM, Jones PM (2013) An asymmetric post-hydrolysis state of the ABC transporter ATPase dimer. PLoS One 8: e59854 Gerber S, Comellas-Bigler M, Goetz BA, Locher KP (2008) Structural basis of trans-inhibition in a molybdate/tungstate ABC transporter. Science 321: 246-250 Goetz BA, Perozo E, Locher KP (2009) Distinct gate conformations of the ABC transporter BtuCD revealed by electron spin resonance spectroscopy and chemical cross-linking. FEBS Lett 583: 266-270 Gorke B, Stulke J (2008) Carbon catabolite repression in bacteria: many ways to make the most out of nutrients. Nat Rev Microbiol 6: 613-624 Gouaux E, Mackinnon R (2005) Principles of selective ion transport in channels and pumps. Science 310: 1461-1465 129 Gould AD, Telmer PG, Shilton BH (2009) Stimulation of the maltose transporter ATPase by unliganded maltose binding protein. Biochemistry 48: 8051-8061 Greenfield NJ (2006) Determination of the folding of proteins as a function of denaturants, osmolytes or ligands using circular dichroism. Nature Protocols 1: 2733-2741 Grote M, Bordignon E, Polyhach Y, Jeschke G, Steinhoff HJ, Schneider E (2008) A comparative electron paramagnetic resonance study of the nucleotide-binding domains' catalytic cycle in the assembled maltose ATP-binding cassette importer. Biophys J 95: 2924-2938 Hall JA, Gehring K, Nikaido H (1997) Two modes of ligand binding in maltose-binding protein of Escherichia coli. Correlation with the structure of ligands and the structure of binding protein. J Biol Chem 272: 17605-17609 Hardie RC, Franze K (2012) Photomechanical responses in Drosophila photoreceptors. Science 338: 260-263 Hekstra D, Tommassen J (1993) Functional exchangeability of the ABC proteins of the periplasmic binding protein-dependent transport systems Ugp and Mal of Escherichia coli. J Bacteriol 175: 6546-6552 Hollenstein K, Dawson RJ, Locher KP (2007) Structure and mechanism of ABC transporter proteins. Curr Opin Struct Biol 17: 412-418 Hunke S, Landmesser H, Schneider E (2000a) Novel missense mutations that affect the transport function of MalK, the ATP-binding-cassette subunit of the Salmonella enterica serovar typhimurium maltose transport system. J Bacteriol 182: 1432-1436 Hunke S, Mourez M, Jehanno M, Dassa E, Schneider E (2000b) ATP modulates subunit-subunit interactions in an ATP-binding cassette transporter (MalFGK(2)) determined by site-directed chemical cross-linking. Journal of Biological Chemistry 275: 15526-15534 Hurley JH, Faber HR, Worthylake D, Meadow ND, Roseman S, Pettigrew DW, Remington SJ (1993) Structure of the regulatory complex of Escherichia coli IIIGlc with glycerol kinase. Science 259: 673-677 Jardetzky O (1966) Simple allosteric model for membrane pumps. Nature 211: 969-970 Jones PM, O'Mara ML, George AM (2009) ABC transporters: a riddle wrapped in a mystery inside an enigma. Trends Biochem Sci 34: 520-531 Joo C, Balci H, Ishitsuka Y, Buranachai C, Ha T (2008) Advances in single-molecule fluorescence methods for molecular biology. Annu Rev Biochem 77: 51-76 130 Joseph B, Jeschke G, Goetz BA, Locher KP, Bordignon E (2011) Transmembrane gate movements in the type II ATP-binding cassette (ABC) importer BtuCD-F during nucleotide cycle. J Biol Chem 286: 41008-41017 Kaback HR, Smirnova I, Kasho V, Nie Y, Zhou Y (2011) The alternating access transport mechanism in LacY. J Membr Biol 239: 85-93 Kadaba NS, Kaiser JT, Johnson E, Lee A, Rees DC (2008) The high-affinity E. coli methionine ABC transporter: structure and allosteric regulation. Science 321: 250-253 Kawai T, Caaveiro JMM, Abe R, Katagiri T, Tsumoto K (2011) Catalytic activity of MsbA reconstituted in nanodisc particles is modulated by remote interactions with the bilayer. FEBS Lett 585: 3533-3537 Khare D, Oldham ML, Orelle C, Davidson AL, Chen J (2009) Alternating access in maltose transporter mediated by rigid-body rotations. Mol Cell 33: 528-536 Kojima S, Nikaido H (2013) Permeation rates of penicillins indicate that Escherichia coli porins function principally as nonspecific channels. Proc Natl Acad Sci U S A 110: E2629-2634 Korkhov VM, Mireku SA, Locher KP (2012) Structure of AMP-PNP-bound vitamin B12 transporter BtuCD-F. Nature 490: 367-372 Kozakov D, Hall DR, Beglov D, Brenke R, Comeau SR, Shen Y, Li K, Zheng J, Vakili P, Paschalidis I, Vajda S (2010) Achieving reliability and high accuracy in automated protein docking: ClusPro, PIPER, SDU, and stability analysis in CAPRI rounds 13-19. Proteins 78: 3124-3130 Kuhnau S, Reyes M, Sievertsen A, Shuman HA, Boos W (1991) The activities of the Escherichia coli MalK protein in maltose transport, regulation, and inducer exclusion can be separated by mutations. J Bacteriol 173: 2180-2186 Landmesser H, Stein A, Bluschke B, Brinkmann M, Hunke S, Schneider E (2002) Large-scale purification, dissociation and functional reassembly of the maltose ATP-binding cassette transporter (MalFGK(2)) of Salmonella typhimurium. Biochimica Et Biophysica Acta-Biomembranes 1565: 64-72 Ledvina PS, Tsai AL, Wang Z, Koehl E, Quiocho FA (1998) Dominant role of local dipolar interactions in phosphate binding to a receptor cleft with an electronegative charge surface: equilibrium, kinetic, and crystallographic studies. Protein Sci 7: 2550-2559 Lehrer SS (1995) Pyrene excimer fluorescence as a probe of protein conformational change. Subcell Biochem 24: 115-132 131 Lewinson O, Lee AT, Locher KP, Rees DC (2010) A distinct mechanism for the ABC transporter BtuCD-BtuF revealed by the dynamics of complex formation. Nat Struct Mol Biol 17: 332-338 Liu PQ, Liu CE, Ames GF (1999) Modulation of ATPase activity by physical disengagement of the ATP-binding domains of an ABC transporter, the histidine permease. J Biol Chem 274: 18310-18318 Locher KP (2009) Review. Structure and mechanism of ATP-binding cassette transporters. Philos Trans R Soc Lond B Biol Sci 364: 239-245 Locher KP, Lee AT, Rees DC (2002) The E. coli BtuCD structure: a framework for ABC transporter architecture and mechanism. Science 296: 1091-1098 Lu G, Westbrooks JM, Davidson AL, Chen J (2005) ATP hydrolysis is required to reset the ATP-binding cassette dimer into the resting-state conformation. Proc Natl Acad Sci U S A 102: 17969-17974 Lundbaek JA, Collingwood SA, Ingolfsson HI, Kapoor R, Andersen OS (2010) Lipid bilayer regulation of membrane protein function: gramicidin channels as molecular force probes. J R Soc Interface 7: 373-395 Mannering DE, Sharma S, Davidson AL (2001) Demonstration of conformational changes associated with activation of the maltose transport complex. J Biol Chem 276: 12362-12368 Manson MD, Boos W, Bassford PJ, Jr., Rasmussen BA (1985) Dependence of maltose transport and chemotaxis on the amount of maltose-binding protein. J Biol Chem 260: 9727-9733 Marinelli F, Kuhlmann SI, Grell E, Kunte HJ, Ziegler C, Faraldo-Gomez JD (2011) Evidence for an allosteric mechanism of substrate release from membrane-transporter accessory binding proteins. Proc Natl Acad Sci U S A 108: E1285-1292 Marme N, Knemeyer JP, Sauer M, Wolfrum J (2003) Inter- and intramolecular fluorescence quenching of organic dyes by tryptophan. Bioconjug Chem 14: 1133-1139 Marvin JS, Hellinga HW (2001) Manipulation of ligand binding affinity by exploitation of conformational coupling. Nat Struct Biol 8: 795-798 McHaourab HS, Steed PR, Kazmier K (2011) Toward the fourth dimension of membrane protein structure: insight into dynamics from spin-labeling EPR spectroscopy. Structure 19: 1549-1561 Meadow ND, Roseman S (1982) Sugar transport by the bacterial phosphotransferase system. Isolation and characterization of a glucose-specific phosphocarrier protein (IIIGlc) from Salmonella typhimurium. J Biol Chem 257: 14526-14537 132 Merino G, Boos W, Shuman HA, Bohl E (1995) The inhibition of maltose transport by the unliganded form of the maltose-binding protein of Escherichia coli: experimental findings and mathematical treatment. J Theor Biol 177: 171-179 Merino G, Shuman HA (1997) Unliganded maltose-binding protein triggers lactose transport in an Escherichia coli mutant with an alteration in the maltose transport system. J Bacteriol 179: 7687-7694 Miller DM, 3rd, Olson JS, Pflugrath JW, Quiocho FA (1983) Rates of ligand binding to periplasmic proteins involved in bacterial transport and chemotaxis. J Biol Chem 258: 13665-13672 Misko TP, Mitchell WJ, Meadow ND, Roseman S (1987) Sugar transport by the bacterial phosphotransferase system. Reconstitution of inducer exclusion in Salmonella typhimurium membrane vesicles. J Biol Chem 262: 16261-16266 Mitchell WJ, Misko TP, Roseman S (1982) Sugar transport by the bacterial phosphotransferase system. Regulation of other transport systems (lactose and melibiose). J Biol Chem 257: 14553-14564 Miyamoto S, Tokuda H (2007) Diverse effects of phospholipids on lipoprotein sorting and ATP hydrolysis by the ABC transporter LolCDE complex. Biochim Biophys Acta 1768: 1848-1854 Novotny MJ, Frederickson WL, Waygood EB, Saier MH, Jr. (1985) Allosteric regulation of glycerol kinase by enzyme IIIglc of the phosphotransferase system in Escherichia coli and Salmonella typhimurium. J Bacteriol 162: 810-816 Oldham ML, Chen J (2011a) Crystal structure of the maltose transporter in a pretranslocation intermediate state. Science 332: 1202-1205 Oldham ML, Chen J (2011b) Snapshots of the maltose transporter during ATP hydrolysis. Proc Natl Acad Sci U S A 108: 15152-15156 Oldham ML, Khare D, Quiocho FA, Davidson AL, Chen J (2007) Crystal structure of a catalytic intermediate of the maltose transporter. Nature 450: 515-521 Oliveira AS, Baptista AM, Soares CM (2011) Inter-domain communication mechanisms in an ABC importer: a molecular dynamics study of the MalFGK2E complex. PLoS Comput Biol 7: e1002128 Orelle C, Alvarez FJ, Oldham ML, Orelle A, Wiley TE, Chen J, Davidson AL (2010) Dynamics of alpha-helical subdomain rotation in the intact maltose ATP-binding cassette transporter. Proc Natl Acad Sci U S A 107: 20293-20298 133 Orelle C, Ayvaz T, Everly RM, Klug CS, Davidson AL (2008) Both maltose-binding protein and ATP are required for nucleotide-binding domain closure in the intact maltose ABC transporter. Proc Natl Acad Sci U S A 105: 12837-12842 Panagiotidis CH, Reyes M, Sievertsen A, Boos W, Shuman HA (1993) Characterization of the structural requirements for assembly and nucleotide binding of an ATP-binding cassette transporter. The maltose transport system of Escherichia coli. J Biol Chem 268: 23685-23696 Parcej D, Tampe R (2010) ABC proteins in antigen translocation and viral inhibition. Nat Chem Biol 6: 572-580 Park E, Rapoport TA (2012) Mechanisms of Sec61/SecY-mediated protein translocation across membranes. Annu Rev Biophys 41: 21-40 Patzlaff JS, van der Heide T, Poolman B (2003) The ATP/substrate stoichiometry of the ATP-binding cassette (ABC) transporter OpuA. J Biol Chem 278: 29546-29551 Pinkett HW, Lee AT, Lum P, Locher KP, Rees DC (2007) An inward-facing conformation of a putative metal-chelate-type ABC transporter. Science 315: 373-377 Poolman B, Doeven MK, Geertsma ER, Biemans-Oldehinkel E, Konings WN, Rees DC (2005) Functional analysis of detergent-solubilized and membrane-reconstituted ATP-binding cassette transporters. Methods Enzymol 400: 429-459 Postma PW, Lengeler JW, Jacobson GR (1993) Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiol Rev 57: 543-594 Procko E, O'Mara ML, Bennett WFD, Tieleman DP, Gaudet R (2009) The mechanism of ABC transporters: general lessons from structural and functional studies of an antigenic peptide transporter. FASEB J 23: 1287-1302 Quiocho FA, Ledvina PS (1996) Atomic structure and specificity of bacterial periplasmic receptors for active transport and chemotaxis: variation of common themes. Mol Microbiol 20: 17-25 Rabeh WM, Bossard F, Xu H, Okiyoneda T, Bagdany M, Mulvihill CM, Du K, di Bernardo S, Liu Y, Konermann L, Roldan A, Lukacs GL (2012) Correction of both NBD1 energetics and domain interface is required to restore DeltaF508 CFTR folding and function. Cell 148: 150-163 Rees DC, Johnson E, Lewinson O (2009) ABC transporters: the power to change. Nat Rev Mol Cell Biol 10: 218-227 Reich-Slotky R, Panagiotidis C, Reyes M, Shuman HA (2000) The detergent-soluble maltose transporter is activated by maltose binding protein and verapamil. J Bacteriol 182: 993-1000 134 Richet E, Davidson AL, Joly N (2012) The ABC transporter MalFGK(2) sequesters the MalT transcription factor at the membrane in the absence of cognate substrate. Mol Microbiol 85: 632-647 Rizk SS, Paduch M, Heithaus JH, Duguid EM, Sandstrom A, Kossiakoff AA (2011) Allosteric control of ligand-binding affinity using engineered conformation-specific effector proteins. Nat Struct Mol Biol 18: 437-442 Robson A, Gold VAM, Hodson S, Clarke AR, Collinson I (2009) Energy transduction in protein transport and the ATP hydrolytic cycle of SecA. Proc Natl Acad Sci U S A 106: 5111-5116 Roy R, Hohng S, Ha T (2008) A practical guide to single-molecule FRET. Nat Methods 5: 507-516 Sahoo D, Narayanaswami V, Kay CM, Ryan RO (2000) Pyrene excimer fluorescence: a spatially sensitive probe to monitor lipid-induced helical rearrangement of apolipophorin III. Biochemistry 39: 6594-6601 Scholz C, Parcej D, Ejsing CS, Robenek H, Urbatsch IL, Tampe R (2011) Specific lipids modulate the transporter associated with antigen processing (TAP). J Biol Chem 286: 13346-13356 Sharff AJ, Rodseth LE, Spurlino JC, Quiocho FA (1992) Crystallographic evidence of a large ligand-induced hinge-twist motion between the two domains of the maltodextrin binding protein involved in active transport and chemotaxis. Biochemistry 31: 10657-10663 Sharma S, Davis JA, Ayvaz T, Traxler B, Davidson AL (2005) Functional reassembly of the Escherichia coli maltose transporter following purification of a MalF-MalG subassembly. J Bacteriol 187: 2908-2911 Shi Y (2013) Common folds and transport mechanisms of secondary active transporters. Annu Rev Biophys 42: 51-72 Shuman HA (1982) Active transport of maltose in Escherichia coli K12. Role of the periplasmic maltose-binding protein and evidence for a substrate recognition site in the cytoplasmic membrane. J Biol Chem 257: 5455-5461 Sianidis G, Karamanou S, Vrontou E, Boulias K, Repanas K, Kyrpides N, Politou AS, Economou A (2001) Cross-talk between catalytic and regulatory elements in a DEAD motor domain is essential for SecA function. Embo Journal 20: 961-970 Silhavy TJ, Szmelcman S, Boos W, Schwartz M (1975) On the significance of the retention of ligand by protein. Proc Natl Acad Sci U S A 72: 2120-2124 135 Silva RAGD, Huang R, Morris J, Fang J, Gracheva EO, Ren G, Kontush A, Jerome WG, Rye KA, Davidson WS (2008) Structure of apolipoprotein A-I in spherical high density lipoproteins of different sizes. Proc Natl Acad Sci U S A 105: 12176-12181 Smith PC, Karpowich N, Millen L, Moody JE, Rosen J, Thomas PJ, Hunt JF (2002) ATP binding to the motor domain from an ABC transporter drives formation of a nucleotide sandwich dimer. Mol Cell 10: 139-149 Sondej M, Seok YJ, Badawi P, Koo BM, Nam TW, Peterkofsky A (2000) Topography of the surface of the Escherichia coli phosphotransferase system protein enzyme IIAglc that interacts with lactose permease. Biochemistry 39: 2931-2939 Sondej M, Weinglass AB, Peterkofsky A, Kaback HR (2002) Binding of enzyme IIAGlc, a component of the phosphoenolpyruvate:sugar phosphotransferase system, to the Escherichia coli lactose permease. Biochemistry 41: 5556-5565 Spurlino JC, Lu GY, Quiocho FA (1991) The 2.3-A resolution structure of the maltose- or maltodextrin-binding protein, a primary receptor of bacterial active transport and chemotaxis. J Biol Chem 266: 5202-5219 Steinke A, Grau S, Davidson A, Hofmann E, Ehrmann M (2001) Characterization of transmembrane segments 3, 4, and 5 of MalF by mutational analysis. J Bacteriol 183: 375-381 Stillwell W (2013) An introduction to biological membranes : from bilayers to rafts, London ; Waltham, MA: Elsevier/Academic Press. Tang G, Peng L, Baldwin PR, Mann DS, Jiang W, Rees I, Ludtke SJ (2007) EMAN2: An extensible image processing suite for electron microscopy. Journal of Structural Biology 157: 38-46 Telmer PG, Shilton BH (2003) Insights into the conformational equilibria of maltose-binding protein by analysis of high affinity mutants. J Biol Chem 278: 34555-34567 Thogersen L, Nissen P (2012) Flexible P-type ATPases interacting with the membrane. Curr Opin Struct Biol 22: 491-499 Tirado-Lee L, Lee A, Rees DC, Pinkett HW (2011) Classification of a Haemophilus influenzae ABC transporter HI1470/71 through its cognate molybdate periplasmic binding protein, MolA. Structure 19: 1701-1710 Treptow NA, Shuman HA (1985) Genetic evidence for substrate and periplasmic-binding-protein recognition by the MalF and MalG proteins, cytoplasmic membrane components of the Escherichia coli maltose transport system. J Bacteriol 163: 654-660 136 Urbatsch IL, Sankaran B, Bhagat S, Senior AE (1995) Both P-Glycoprotein Nucleotide-Binding Sites Are Catalytically Active. Journal of Biological Chemistry 270: 26956-26961 van der Heide T, Poolman B (2002) ABC transporters: one, two or four extracytoplasmic substrate-binding sites? EMBO Rep 3: 938-943 van der Vlag J, van Dam K, Postma PW (1994) Quantification of the regulation of glycerol and maltose metabolism by IIAGlc of the phosphoenolpyruvate-dependent glucose phosphotransferase system in Salmonella typhimurium. J Bacteriol 176: 3518-3526 Wagner S, Pop OI, Haan GJ, Baars L, Koningstein G, Klepsch MM, Genevaux P, Luirink J, de Gier JW (2008) Biogenesis of MalF and the MalFGK(2) maltose transport complex in Escherichia coli requires YidC. J Biol Chem 283: 17881-17890 Wakabayashi Y, Kipp H, Arias IM (2006) Transporters on demand: intracellular reservoirs and cycling of bile canalicular ABC transporters. J Biol Chem 281: 27669-27673 Wandersman C, Schwartz M, Ferenci T (1979) Escherichia coli mutants impaired in maltodextrin transport. J Bacteriol 140: 1-13 Wang G, Keifer PA, Peterkofsky A (2003) Solution structure of the N-terminal amphitropic domain of Escherichia coli glucose-specific enzyme IIA in membrane-mimetic micelles. Protein Sci 12: 1087-1096 Wang G, Louis JM, Sondej M, Seok YJ, Peterkofsky A, Clore GM (2000a) Solution structure of the phosphoryl transfer complex between the signal transducing proteins HPr and IIA(glucose) of the Escherichia coli phosphoenolpyruvate:sugar phosphotransferase system. Embo Journal 19: 5635-5649 Wang G, Peterkofsky A, Clore GM (2000b) A novel membrane anchor function for the N-terminal amphipathic sequence of the signal-transducing protein IIAGlucose of the Escherichia coli phosphotransferase system. J Biol Chem 275: 39811-39814 Wang T, Fu G, Pan X, Wu J, Gong X, Wang J, Shi Y (2013) Structure of a bacterial energy-coupling factor transporter. Nature 497: 272-276 Weidlich D, Wiesemann N, Heuveling J, Wardelmann K, Landmesser H, Sani KB, Worth CL, Preissner R, Schneider E (2013) Residues of a proposed gate region in type I ATP-binding cassette import systems are crucial for function as revealed by mutational analysis. Biochim Biophys Acta 1828: 2164-2172 Welte W, Nestel U, Wacker T, Diederichs K (1995) Structure and function of the porin channel. Kidney Int 48: 930-940 137 Worthylake D, Meadow ND, Roseman S, Liao DI, Herzberg O, Remington SJ (1991) Three-dimensional structure of the Escherichia coli phosphocarrier protein IIIglc. Proc Natl Acad Sci U S A 88: 10382-10386 Wu ZP, Gogonea V, Lee X, May RP, Pipich V, Wagner MA, Undurti A, Tallant TC, Baleanu-Gogonea C, Charlton F, Ioffe A, DiDonato JA, Rye KA, Hazen SL (2011) The Low Resolution Structure of ApoA1 in Spherical High Density Lipoprotein Revealed by Small Angle Neutron Scattering. J Biol Chem 286: 12495-12508 Xu K, Zhang M, Zhao Q, Yu F, Guo H, Wang C, He F, Ding J, Zhang P (2013) Crystal structure of a folate energy-coupling factor transporter from Lactobacillus brevis. Nature 497: 268-271 Zhang P (2013) Structure and mechanism of energy-coupling factor transporters. Trends Microbiol 21: 652-659 Zhou X, Arthur G (1992) Improved Procedures for the Determination of Lipid Phosphorus by Malachite Green. J Lipid Res 33: 1233-1236 138 Appendices Appendix A Protein purification A.1 Purification of MalFGK2 The genes MalF, MalG and MalK were separately amplified from the E. coli K12 genome by PCR, and placed in tandem in pBAD22 plasmid. A His6-tag was inserted at the C-terminus of MalK (yielding p22-FGKhis). Mutations were introduced by PCR-site directed mutagenesis and all constructs were verified by DNA sequencing. Overproduction of MalFGK2 was performed using E. coli strain BL21. Briefly, 12L of LB medium containing ampicillin (100 ?g/ml) were inoculated with an overnight culture. At OD600 ~ 0.5, plasmid expression was induced with 0.2% (w/v) arabinose. After 3 h, cells were collected in TSG buffer (50 mM Tris-HCl, pH 8; 100 mM NaCl; 10% glycerol) containing 0.01% PMSF, and lysed through a French Press (8,000 psi, twice). After low speed centrifugation (5,000?g, 10 min), the membrane fraction was isolated by ultracentrifugation (100,000, 1 h, 4?C) and resuspended in buffer B (50 mM Tris-HCl, pH 8; 5 mM MgCl2, 20% glycerol). Membranes (5 mg/ml) were incubated with 1% n-dodecyl ?-D-maltoside (DDM) with gentle shaking (3 h, 4?C). The solubilized proteins were isolated by ultracentrifugation (100,000? g, 1 h, 4?C) and applied onto a Ni-NTA Sepharose column (10 ml resin) equilibrated in buffer B containing 0.01% DDM (buffer C). After intensive washes (10 column volume in buffer C), proteins were eluted with a gradient of imidazole (0?600 mM) in buffer C. The isolated MalFGK2 complex was further purified by gel filtration (Superdex 200 HR10/30) in buffer C. 139 A.2 Purification of MalE The gene encoding for the mature part of MalE was cloned into pBAD33, yielding plasmid p33-MalE. Mutations were introduced by PCR-site directed mutagenesis and all constructs were verified by DNA sequencing. Overproduction of MalE was as described for MalFGK2, excepted the antibiotic was chloramphenicol (50 ?g/ml). Cells were collected in buffer D (50 mM Tris-HCl pH 8.8; 10% glycerol) and lysed through a French Press (8,000 psi, twice). After ultracentrifugation (100,000?g, 1 h, 4?C), the supernatant was applied onto a Resource Q column (1 ml) equilibrated in buffer D. Proteins were eluted with a gradient of NaCl (0?1000 mM). The protein fractions containing MalE were pooled and denatured with 6M Guanidine-HCl. Protein refolding was performed by dialysis with 3 changes of TSG buffer (50 mM Tris-HCl, pH 8; 100 mM NaCl; 10% glycerol). A.3 Purification of EIIAGlc The chromosomal gene crr (encoding for EIIAGlc) was cloned into pBAD33 with a His6 tag sequence introduced at the 3? end of the gene, yielding plasmid p33-EIIAhis. Overproduction of EIIAGlc was achieved in E.coli strain BL21 grown in 6 liters of M9 medium supplemented with chloramphenicol (50 ?g/ml) and glucose (0.8%). At A600 ?0.5, EIIAGlc synthesis was induced with 0.2% arabinose. After 3 h, cells were collected in TSG buffer (50 mM Tris-HCl, pH 8; 100 mM NaCl; 10% glycerol) containing 0.01% PMSF and lysed through a French press (8,000 p.s.i., twice). After ultracentrifugation (100,000 ? g, 1 h, 4 ?C), the supernatant was applied onto a Ni2+-nitrilotriacetic acid-Sepharose column (5 ml) equilibrated in TSG buffer. The washing step was in TSG buffer plus 30 mM imidazole (10 column volumes), and the elution was across a gradient of TSG buffer plus 0?600 mM imidazole. The eluted EIIAGlc protein was further purified on a Superdex 200 GL 10/300 in TSG buffer. fer plus 30 mM imidazole (10 140 column volumes), and the elution was across a gradient of TSG buffer plus 0?600 mM imidazole. The eluted EIIAGlc protein was further purified on a Superdex 200 GL 10/300 in TSG buffer. Purified EIIAGlc and EIIAGlc?1?18 were homogeneous, forming a single elution peak in size-exclusion chromatography and migrating at their expected position on SDS-PAGE analysis (?19 and ?17 kDa, respectively). 141 Appendix B Illustration of nanodiscs and proteoliposomes reconstitution Figure B.1 a, the nanodiscs reconstitution involved mixing together the MalFGK2 complex purified in detergent, the membrane scaffold protein (MSP) and the solubilized lipids at a molecular ratio of 1:3:60 in TSG buffer ( Tris-HCl (50mM, pH 8.0), NaCl (100mM), Glycerol (5%)) containing 0.04% DDM. Detergent was slowly removed by Biobeads and the reconstituted nanodiscs were purified by Superdex 200 10/30 gel filtration equilibrated in TSG buffer. b, the proteoliposomes was prepared at a protein: lipid ratio of 1:2000 in TSG buffer. Detergent was removed using Biobeads as for nanodiscs reconstitutions. The proteoliposomes were harvested by centrifugation (100,000 ? g, 1h, 4 ?C) and resuspened in Tris buffer (20 mM, pH 8.0). 142 Appendix C Equations C.1 Equation 1 To determine the affinity of TNP-ATP for MalFGK2, the fluorescence measured in the presence of MalFGK2 nanodiscs was subtracted from that measured in the absence of MalFGK2 nanodiscs, yielding the subtracted fluorescence value (Fs). The data were then plotted as a function of TNP-ATP concentration (L) and fit to Equation 1, dmaxs][][KLLFF+?= , (Eq.1) where Fmax is the maximal subtracted fluorescence at saturating amount of TNP-ATP, and Kd is the equilibrium dissociation constant of TNP-ATP for MalFGK2. C.2 Equation 2 For measuring the apparent affinity of ATP for MalFGK2, the lipid-rich MalFGK2 nanodiscs (2 ?M) were mixed with TNP-ATP (80 ?M) for 5 min at room temperature and then titrated with the indicated amount of ATP. The fluorescence data were fit to Equation 2, appi,1appi,appi,0s][][][ KIIFKIKFF+?++?= , (Eq.2) in which F0 is the subtracted fluorescence in the absence of ATP, F1 is the subtracted fluorescence in the presence of saturating amount of ATP, [I] is the ATP concentration, and Ki,app is the apparent inhibition constant of ATP at the specified amount of TNP-ATP. C.3 Equation 3 The Kd,app of ATP for MalFGK2 is calculated from Ki,app using Equation 3, )][1(dappd,appi,KLKK +?= , (Eq.3) 143 in which [L] is the TNP-ATP concentration, and Kd is the dissociation constant of TNP-ATP for MalFGK2. C.4 Equation 4 To determine the affinity of Nd-FGK2 for MalE by the fluorescence quenching assay, the data were fit to Equation 4, dmax][][KLLEE+?= (Eq.4) where E is the fluorescence quenching efficiency, Emax is the maximal quenching efficiency at saturating amount of Nd-FGK2 and [L] is the Nd-FGK2 concentration. C.5 Equation 5 Maltose and Nd-FGK2 are competing for the binding to MalE. The data (Fig. 4.1g, red curve) were fitted to equation 5, defined as: ?????? ++=Idmax][1][1][KILKEE (Eq.5) where [E] is the fluorescence quenching efficiency, Emax is the maximal quenching efficiency at saturating amount of Nd-FGK2, [L] the Nd-MalFGK2 concentration, [I] the maltose concentration, Kd the affinity of MalE for Nd-FGK2 and KI the competitive affinity of MalE for maltose. When the Kd value is 79 nM, fitting the data (Figure 4.1g, red curve) into the equation give the KI value of 127?M. 144 Appendix D Isolation of the Nd-FGK2-MalE complex D.1 Isolation of the Nd-FGK2-MalE complex in the presence of ATP and Vi Nd-FGK2 (10 ?M) and MalE (15 ?M) were incubated at room temperature for 10 min in the presence ATP (1mM) and Vanadate (10 ?M). The Nd-FGK2-MalE complex was separated from MalE by size exclusion chromatography (SEC) using a superdex s200 10/100 column equilibrated in TM buffer (50 mM Tris-HCl, pH 8.0; 5 mM MgCl2) (Figure C.1a). The fractions from SEC analyses were analyzed by SDS-PAGE and CN-PAGE (Figure C.1b and C.1c). Control experiments were performed using mere MalE or Nd-FGK2. Complex formation was evident from the coelution of MalE and Nd-FGK2. The fractions corresponding to the Nd-FGK2-MalE complex were subject to electron microscopy analyses. 145 Figure D.1 isolation of the NdFGK2-MalE complex in the presence of ATP and Vi a, superdex-200 gel filtration profiles of MalE (red), Nd-FGK2 (blue), and Nd-FGK2 with MalE (black). The left panel is the expanded profile of the right panel from 0 to 6.5 ml. The elution peak between 6.5 to 10 ml corresponds to ATP. b, SDS-PAGE gel of fractions from the gel filtration analysis. c, fractions corresponding to Nd-FGK2 alone and with MalE (F2 to F4 from b) were analyzed by CN-PAGE. For comparison, MalE and Nd-FGK2 were analyzed in paralle after incubation at room temperature for 10 min in the presence of ATP and Vi (lane 1-3). D.2 Isolation of the the MalE-Nd-FGK2 complex in the absence of nucleotides Nd-FGK2 with mutation MalFS205C (Nd-205C, 10 ?M) and MalET80C (30 ?M) were incubated at room temperature for 10 min in the presence CP3 (50 ?M). The Nd-FGK2-MalE 146 complex was separated from MalE by size exclusion chromatography (SEC) using a superdex s200 10/200 column equilibrated in TM buffer (50 mM Tris-HCl, pH 8.0; 5 mM MgCl2) (Figure C.2). The crosslinking between MalET80C and MalFS205C was observed by SDS-PAGE (Figure C.2). The fractions corresponding to the Nd-FGK2-MalE complex were subject to electron microscopy analyses. Figure D.2 Isolation of Nd-MalFGK2-MalE in the absence of nucleotides. Upper panel: superdex-200 gel filtration profiles of the Nd-FGK2-MalE complex formed by disulphide crosslinking. Bottom panel: SDS-PAGE gel of fractions from the gel filtration analysis. 147 Appendix E Effect of maltooligosaccharides on MalE binding to MalFGK2 Figure E.1 Nd-MalFGK2 (4 ?M) was incubated with [125I]-MalE (~10,000 c.p.m., 1 mM) in TSGM buffer containing AMP-PNP (1 mM) and the indicated maltooligosaccharides (1 mM). After incubation (10 min, 37?C), samples were analyzed by CN-PAGE and autoradiography. 148 Appendix F Orientation of MalFGK2 reconsituted in proteoliposomes Figure F.1 a, the labeling efficiency of residue MalFT177C (in mutant background MalKC40S) was assayed using the membrane-impermeable reagent 5-IAF in the presence or absence of DDM. b, the samples were analyzed by 15% SDS-PAGE followed by fluorescence scanning (left panel) or Coomassie Blue staining (right panel). c, the fluorescence of MalFT177C when labeled in the presence of DDM was normalized to 100%. The labeling efficiency of MalFT177C was determined by absorbance spectroscopy (498 nm), using an extinction co-efficient of 75, 500 cm-1 M-1. 149 Appendix G Binding of MalE to MalFGK2 in proteoliposomes Figure G.1 maltose promotes the dissociation of MalE when MalFGK2 is stabilized in the outward facing state. The binding of MalE to MalFGK2 in proteoliposomes was analyzed by sedimentation assay. The indicated amount of [125I]MalE was incubated with MalFGK2 (pair MalK83-MalK83) in proteoliposomes in the presence of ATP or ATP and BMOE (room temperature, 10 min). The samples were diluted 25-fold in 20mM Tris-HCl, pH 8, with or without 1mM maltose. The fraction of MalE bound to MalFGK2 was isolated by ultracentrifugation (100,000 ? g, 1 h). The amount of bound MalE was analyzed by 12% SDS-PAGE followed by Coomassie Blue staining (top panels) or autoradiography (bottom panels). Note that MalF and MalK are migrating at the same position on 12% SDS-PAGE. 150 Appendix H Conformation of MalFG and re-assembled MalFGK2 Figure H.1 MalFG and re-assembled MalFGK2 carrying the indicated cystein residues pairs were reconstituted in proteoliposomes. Proteoliposomes (2 ?M) were incubated for 10 min at room temperature with the homobifunctional cross-linker BMOE (50?M). Reactions were stopped by NEM (5mM) prior to analysis by SDS-PAGE (4-12%) and immunoblot against MalF. 151 Appendix I Stability of MalFGK2 Figure I.1 purified MalFGK2 variants were incubated with the indicated amount of urea and analyzed by blue-native PAGE. 152