UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural analysis of beta-lactamase and resistant transpeptidase inhibition Gretes, Michael 2009

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

Item Metadata

Download

Media
24-ubc_2009_fall_gretes_michael.pdf [ 53.42MB ]
Metadata
JSON: 24-1.0228826.json
JSON-LD: 24-1.0228826-ld.json
RDF/XML (Pretty): 24-1.0228826-rdf.xml
RDF/JSON: 24-1.0228826-rdf.json
Turtle: 24-1.0228826-turtle.txt
N-Triples: 24-1.0228826-rdf-ntriples.txt
Original Record: 24-1.0228826-source.json
Full Text
24-1.0228826-fulltext.txt
Citation
24-1.0228826.ris

Full Text

STRUCTURAL ANALYSIS OF BETA-LACTAMASE AND RESISTANT TRANSPEPTIDASE INHIBITION  by MICHAEL GRETES B.Sc., University of Western Ontario, 2003  A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in The Faculty of Graduate Studies (Biochemistry and Molecular Biology)  The University of British Columbia (Vancouver)  June, 2009 © Michael Gretes, 2009  ABSTRACT Beta-lactam antibiotics have achieved phenomenal success in the treatment of infections by inhibiting the transpeptidase enzymes that cross-link the bacterial cell wall. Beta-lactamase-producing pathogenic bacteria and multi-drug-resistant “superbugs” such as methicillin-resistant Staphylococcus aureus (MRSA) have emerged, however. Overcoming resistance factors is thus a research priority. BLIP (Beta-Lactamase Inhibitory Protein) from Streptomyces clavuligerus binds a variety of beta-lactamase enzymes with widely ranging specificity. Its interaction with Escherichia coli beta-lactamase TEM-1 is a well-established model system for protein-protein interaction studies. Presented in Chapter 2 are crystal structures of two BLIP relatives: BLIP-I (a highaffinity inhibitor, alone and in complex with TEM-1) and BLP (which appears not to inhibit beta-lactamases). Substantial variation appears possible in the sub-nanomolar binding of TEM-1 by two homologous proteinaceous inhibitors and such favorable interactions can be negated by a few, strongly unfavorable interactions. OXA-10 is a Pseudomonas aeruginosa beta-lactamase that is resistant to inhibitors in clinical use. Cyclobutanone beta-lactam mimics could be used instead. Chapter 3 reports the crystal structure of OXA-10 covalently modified at its catalytic serine nucleophile with a cyclobutanone inhibitor to form a hemiketal. Favorable and unfavorable contacts made at the active site are examined with a view to improved inhibitor design. PBP2a is the resistant transpeptidase that allows MRSA to maintain the bacterial cell wall in the presence of beta-lactam antibiotics. Ceftobiprole is the most clinically-advanced among a new generation of beta-lactams designed to treat MRSA by targeting PBP2a itself. Chapter 4 uses the crystal structure of a truncated, soluble form of PBP2a solved in complex ii  with ceftobiprole to explain its inhibitory power and evaluate current anti-MRSA drug design hypotheses. Its efficacy appears to arise from improved binding affinity that overcomes the disfavored energetics of acylation. Ceftobiprole clinical trials reported no bacterial resistance, yet fully ceftobiproleresistant MRSA (MIC 128 !g/ml) were generated by passage through subinhibitory concentrations of ceftobiprole, discussed in Chapter 5. Resistance emerges in most cases via mutations to the gene encoding PBP2a. Computational modeling predicts that ceftobiprole resistance may be mediated in PBP2a by alteration of binding affinity, acylation efficiency, or by influencing interactions with other proteins.  iii  TABLE OF CONTENTS ABSTRACT ...................................................................................................................ii TABLE OF CONTENTS ............................................................................................ iv LIST OF TABLES.........................................................................................................x LIST OF FIGURES......................................................................................................xi LIST OF EQUATIONS .............................................................................................xiii LIST OF ABBREVIATIONS .................................................................................... xiv ACKNOWLEDGEMENTS ...................................................................................... xvii DEDICATION...........................................................................................................xix STATEMENT OF CO-AUTHORSHIP .................................................................... xx CHAPTER 1: INTRODUCTION ................................................................................ 1 1.1  Overview of Antibiotics and Antibiotic Resistance ............................................ 1  1.1.1 Context: The Antibiotic Age.................................................................................................. 1 1.1.2 Overview of Antibiotic Mechanisms of Action.................................................................. 2 1.1.3 Overview of Antibiotic Resistance Strategies...................................................................... 4 1.2  ß-Lactams: Antibiotics Targeting the Cell Wall ................................................. 5  1.2.1 Cell Wall Structure ................................................................................................................... 5 1.2.2 Cell Wall Biosynthesis: Penicillin-Binding Proteins............................................................ 9 1.2.3 ß-Lactam Antibiotics and Mechanism of Action.............................................................. 21 1.3  ß-Lactamase Enzymes...................................................................................... 24  1.3.1 Origins ..................................................................................................................................... 25 1.3.2 Class A ß-Lactamases: TEM-1............................................................................................. 28 1.3.3 Class D ß-Lactamases: OXA-10.......................................................................................... 32 1.3.4 Overcoming Resistance: ß-Lactamase Inhibitors ............................................................. 38  iv  1.4  ß-Lactam Resistant PBPs ................................................................................. 48  1.4.1 MRSA PBP2a ......................................................................................................................... 48 1.4.2 S. pneumoniae PBP2x............................................................................................................... 53 1.4.3 Overcoming Resistance: Ceftobiprole, a Next-Generation ß-Lactam .......................... 53 1.5  Objectives of the Present Thesis ...................................................................... 57  1.6  Postscript: A Brief Protein Crystallography Primer.......................................... 59  1.7  References ......................................................................................................... 61  CHAPTER 2: INSIGHTS INTO POSITIVE AND NEGATIVE REQUIREMENTS FOR PROTEIN-PROTEIN INTERACTIONS BY CRYSTALLOGRAPHIC ANALYSIS OF THE ß-LACTAMASE INHIBITORY PROTEINS BLIP, BLIP-I AND BLP ................................... 77 2.1  Introduction ...................................................................................................... 77  2.2  Results and Discussion..................................................................................... 81  2.2.1 Preliminary Investigation of BLP Function....................................................................... 81 2.2.2 Structure Determination ....................................................................................................... 84 2.2.3 Sequence and Structure Comparisons ................................................................................ 86 2.2.4 Surface Charge (APBS) and Long-Range Interactions .................................................... 90 2.2.5 Overview of TEM-1 Interaction Interfaces....................................................................... 94 2.2.6 Examination of BLIP, BLIP-I and BLP at TEM-1 Interaction Hotspots ................... 96 2.2.6.1  C1 – favorable interactions by all BLIPs ................................................................... 99  2.2.6.2  C2 – favorable BLIP-I interactions; limited BLP interactions.............................102  2.2.6.3  C3 – no BLIP-I nor BLP interactions .....................................................................103  2.2.6.4  C4 – no favorable BLIP-I nor BLP interactions....................................................104  2.2.6.5  C5 – favorable BLIP-I interactions; unfavorable BLP interactions ....................104  2.2.6.6  C6 – favorable interactions by all BLIPs .................................................................105  2.2.7 Binding Energetics and Energetic Penalty.......................................................................105 2.2.8 Concluding Summary ..........................................................................................................110 v  2.3  Materials and Methods .................................................................................... 112  2.3.1 Molecular Biology ................................................................................................................112 2.3.1.1  Construction of a recombinant plasmid for expressing bliA in E. coli................112  2.3.1.2  Construction of a recombinant plasmid for expressing blp in E. coli ..................112  2.3.1.3  Mutation of blp.............................................................................................................113  2.3.1.4  Mutation of bli .............................................................................................................113  2.3.1.5  Use of protoplast fusion to prepare bli::tsr-blp::apr double mutants.....................113  2.3.2 Protein Production and Purification.................................................................................115 2.3.2.1  BLIP-I ...........................................................................................................................115  2.3.2.2  BLP ..............................................................................................................................116  2.3.3 Enzyme Inhibition Kinetics and Electromobility Shift Assays ....................................116 2.3.4 Crystallization, Data Processing and Structure Refinement..........................................117 2.3.4.1  BLIP-I ...........................................................................................................................118  2.3.4.2  BLP ..............................................................................................................................118  2.3.5 Accession Numbers.............................................................................................................119 2.3.6 Computational Analysis ......................................................................................................120 2.4  Acknowledgements..........................................................................................122  2.5  References ........................................................................................................123  CHAPTER 3: CRYSTALLOGRAPHIC ANALYSIS OF THE OXA-10 ß-LACTAMASE COVALENTLY MODIFIED AT ITS SERINE NUCLEOPHILE BY A NON-ß-LACTAM CYCLOBUTANONE INHIBITOR (JJ05-850) TO FORM A TETRAHEDRAL HEMIKETAL ....127 3.1  Introduction .....................................................................................................127  3.2  Results and Discussion....................................................................................129  3.3  Conclusions: Directions for Inhibitor Design .................................................139  3.4  Experimental Section.......................................................................................140  3.5  References ........................................................................................................142 vi  CHAPTER 4: STRUCTURAL BASIS FOR THE ACTIVITY OF CEFTOBIPROLE AGAINST METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS .................................................................145 4.1  Introduction .....................................................................................................145  4.2  Results and Discussion....................................................................................152  4.2.1 Examination of Current Anti-MRSA Drug Design Strategy in Light of the PBP2a*-Ceftobiprole Complex .........................................................................................152 4.2.2 Conformational Change in PBP2a Upon Acylation.......................................................158 4.2.3 Understanding Resistance from a Susceptible/Resistant Protein Comparison .........165 4.3  Conclusion .......................................................................................................168  4.4  Materials and Methods ....................................................................................170  4.4.1 Protein Expression, Purification and Crystallization......................................................170 4.4.2 Derivatization, Data Collection and Structure Determination .....................................170 4.4.3 Data Processing and Refinement ......................................................................................171 4.4.4 Circular Dichroism Spectroscopy......................................................................................172 4.4.5 Differential Scanning Calorimetry.....................................................................................172 4.5  Acknowledgements and Funding....................................................................174  4.6  References ........................................................................................................175  CHAPTER 5: IN VITRO SELECTION AND CHARACTERIZATION OF CEFTOBIPROLE-RESISTANT METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS .................................................................180 5.1  Introduction .....................................................................................................180  5.2  Results..............................................................................................................182  5.2.1 Generation of BPR-Resistant Mutants.............................................................................182 5.2.2 BPR-Resistant Strains Developed PBP2a Mutations.....................................................184 5.2.3 PBP2a Mutations Confer BPR-Resistance in mecA-Containing Strains......................186 5.2.4 Molecular Modeling of PBP2a Mutants ...........................................................................187  vii  5.2.5 Resistance in mecA-Negative COLnex(pAW8) Appears to be Mediated by Chromosomal Genes...........................................................................................................191 5.3  Discussion........................................................................................................192  5.4  Materials and Methods ....................................................................................196  5.4.1 Reagents.................................................................................................................................196 5.4.2 Bacterial Strains ....................................................................................................................196 5.4.3 Multipassage Selection in BPR...........................................................................................198 5.4.4 Plasmid Curing .....................................................................................................................198 5.4.5 Spectrophotometric ß-Lactamase Assay ..........................................................................198 5.4.6 Population Analysis .............................................................................................................199 5.4.7 Growth Curves.....................................................................................................................199 5.4.8 DNA Manipulations............................................................................................................199 5.4.9 Molecular Modeling.............................................................................................................200 5.5  Acknowledgments............................................................................................201  5.6  References ....................................................................................................... 202  CHAPTER 6: IMPLICATIONS, APPLICATIONS, DIRECTIONS, AND CONCLUSIONS ............................................................................................ 204 6.1  Major Contributions and Implications of Research ....................................... 204  6.1.1 Form and Function of ß-Lactamase Inhibitory Protein Family Members .................204 6.1.2 Cyclobutanone Analogues of ß-Lactams .........................................................................206 6.1.3 Features of Ceftobiprole Binding to PBP2a....................................................................208 6.1.4 Mechanisms of Ceftobiprole Resistance ..........................................................................209 6.2  Applications and Recommendations for Related Research............................212  6.2.1 ß-Lactamase Inhibitory Protein Family: Protein Engineering and Derived Inhibitors...............................................................................................................................212 6.2.2 Further Development of ß-Lactam Analogues of Cyclobutanones, and Engineering of an OXA-10 Variant for Drug Lead Screening.....................................215 viii  6.2.3 Further PBP2a Drug Development Based on Ceftobiprole Complex........................217 6.2.4 Elucidation of Specific Mechanisms of PBP2a- and Non-PBP2a-Based Ceftobiprole Resistance ......................................................................................................220 6.3  References ....................................................................................................... 224  APPENDIX I: PEER-REVIEWED PUBLICATIONS AND COPYRIGHT ....... 227  ix  LIST OF TABLES Table 1.1: PBP and BLA catalytic residues and proposed mechanistic roles............................ 15! Table 1.2: Proteins shown to interact with PBP family members .............................................. 19! Table 1.3: BLIP inhibition constants (Ki) for Class A BLAs ...................................................... 43! Table 2.1: BLIP, BLIP-I and BLP data collection and refinement statistics............................. 85! Table 2.S1: RMSD values among observations of BLIP homologues....................................... 89! Table 2.2: BLIP, BLIP-I, and BLP interaction residues and ""G(mut) values .....................100! Table 3.1: OXA-10 data collection and refinement statistics ....................................................141! Table 4.1. Changes in UV-CD spectrum upon acylation of nPBP2a*.....................................162! Table 4.2: Thermal stability determined for nPBP2a* upon acylation.....................................164! Table 4.3: PBP2a*-ceftobiprole data collection and refinement statistics...............................171! Table 5.1: Parental strains and phenotypes used in ceftobiprole passaging ............................196! Table 5.2: Derivatives of MRSA COLnex with plasmid-carried mecA ...................................197!  x  LIST OF FIGURES Figure 1.1: The four major classes of ß-lactam antibiotics ............................................................ 3! Figure 1.2: Peptidoglycan polymer structure in Staphylococcus aureus ............................................. 7! Figure 1.3: Likely orientation of S. aureus PBP2 at the cell membrane ...................................... 11! Figure 1.4: Transpeptidase catalytic and inhibition reaction mechanisms................................. 13! Figure 1.5: Active site comparison of PBPs and ß-lactamases.................................................... 27! Figure 1.6: Proposed mechanisms for TEM-1 and OXA-10 ß-lactam hydrolysis................... 30! Figure 1.7: Clinical ß-lactamase inhibitors ...................................................................................... 38! Figure 1.8: BLIP-II structure alone and in complex with TEM-1.............................................. 40! Figure 1.9: BLIP structure alone and in complex with TEM-1 .................................................. 41! Figure 1.10: Inhibition of ß-lactamase TEM-1 by BLIP and penG ........................................... 44! Figure 1.11: Chemical structure of ceftobiprole ............................................................................ 55! Figure 2.1: Pair-wise structural alignments of BLIP, BLIP-I, and BLP..................................... 82! Figure 2.2: Fold of BLIP, BLIP-I, and BLP alone and in complex with TEM-1 .................... 87! Figure 2.S1: Surface charge potential of BLIP, BLIP-I, and BLP .............................................. 92! Figure 2.3: Binding surfaces of BLIP, BLIP-I, and BLP.............................................................. 97! Figure 2.4: Binding hotspots C1, C2 and C5 of BLIP, BLIP-I, and BLP ...............................101! Figure 2.5: Scatterplot of "G(Ki) vs. "G(K) for BLIP-TEM-1 interaction ...........................107! Figure 3.1: Structures of cyclobutanone ß-lactam analogues.....................................................128! Figure 3.2: Electron-density maps of JJ05-850 bound to OXA-10..........................................129! Figure 3.3: OXA-10 residues positioned to interact with JJ05-850 ..........................................131! Figure 3.4: Diagram of favorable OXA-10•JJ05-850 interactions............................................132! Figure 3.5: Alternative conformations of JJ05-850 .....................................................................134! xi  Figure 3.S1: Structure of moxalactam ...........................................................................................135! Figure 3.6: Crystal contacts made by Lys-95 ................................................................................137! Figure 4.1: Chemical structures of ß-lactams and PBP2a substrates........................................147! Figure 4.2: Electron density of ceftobiprole PBP2a* adduct.....................................................152! Figure 4.3: Comparison of ceftobiprole and nitrocefin PBP2a* adducts ................................ 156! Figure 4.4: nPBP2a* UV-CD spectra during acylation by !-lactams .......................................161! Figure 4.5: Differential scanning calorimetry of nPBP2a*-ß-lactam complexes ....................163! Figure 4.6: PBP2a* serine nucleophile availability.......................................................................166! Figure 5.1: Ceftobiprole tolerance of MRSA strains during serial passage..............................183! Figure 5.2: Ceftobiprole susceptibilities of prepassage strains ..................................................183! Figure 5.3: Growth curves of strains before and after BPR passage........................................184! Figure 5.4: PBP2a amino acid substitutions in ceftobiprole-passaged strains ........................185! Figure 5.5: Population analyses of strains +/- ceftobiprole-resistance plasmid.....................186! Figure 5.6: Predicted structural models of ceftobiprole-resistant PBP2a mutants.................188! Figure 5.S1: Comparative structures of !-lactam antibiotics .....................................................190! Figure 6.1: Serine-BLA and MBL inhibition by cyclobutanone ß-lactam mimic ...................216! Figure 6.2: Structure of investigational anti-MRSA cephalosporin OPC-20011 ....................218! Figure 6.3: Ceftobiprole-resistance mutations modeled in PBP2a-ceftobiprole structure....220! Figure 6.4: S. aureus PBP4 with ceftobiprole-resistance associated mutations........................222!  xii  LIST OF EQUATIONS Equation 1.1: ß-lactam inactivation of PBPs ................................................................................. 51! Equation 1.2: Electron-density function......................................................................................... 59!  xiii  LIST OF ABBREVIATIONS !G  Gibbs’ free energy change  !H  binding enthalpy change  !S  binding entropy change  3-D  three-dimensional  6-APA  6-aminopenicillanic acid  7-ACA  7-aminocephalosporanic acid  aa  amino acid(s)  ALS  Advanced Light Source  APBS  Adaptive Poisson-Boltzmann Solver  apr  apramycin  ASU  (crystallographic) asymmetric unit  BLIP  ß-Lactamase Inhibitory Protein  BLP  ß-lactamase inhibitory protein-Like Protein  BORSA -  borderline-(methicillin-)resistant Staphylococcus aureus  BPR  ceftobiprole  BSA  bovine serum albumin  C1-C6  hotspot interaction clusters C1-C6  CCD  charge-coupled device  CFU  colony-forming unit(s)  CM  cytoplasmic membrane  CNS  Crystallography and NMR System  DMSO  dimethyl sulfoxide  DNA  deoxyribonucleic acid  DTT  dithiothreitol  EC  Enzyme Commission class  EDTA  ethylene-diaminetetraacetic acid  EMSA  electromobility shift assay  ESBL  extended-spectrum ß-lactamase  GDa  gigaDalton(s)  GlcNAc  N-acetylglucosamine  HEPES  4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid xiv  HPLC  high-performance liquid chromatography  HMW  high molecular weight  IPTG  isopropyl-!-D-thiogalactopyranoside  ITC  isothermal titration calorimetry  Kd  dissociation equilibrium constant  Kcx  carbamylated lysine  kbp  kilobase pair(s)  kDa  kiloDalton(s)  LMW  low molecular weight  LB  Luria Bertani medium  MAD  multiwavelength anomalous diffraction  MALDI  matrix-assisted laser desorption ionization (mass spectrometry)  MBL  metallo-ß-lactamase  MBI  mechanism-based (enzyme) inhibitor  mc  (protein) main chain (i.e. backbone)  MPD  2-methyl-2,4-pentanediol  MRSA  methicillin-resistant Staphylococcus aureus  MS  mass spectrometry  MurNAc  N-acetyl muramic acid  NMR  nuclear magnetic resonance  NPD  non-penicillin-binding domain  OD600  optical density at 600 nm  OM  outer membrane  PAGE  polyacrylamide gel electrophoresis  PBP  penicillin-binding protein  PCR  polymerase chain reaction  PDB  Protein Data Bank  PEG  polyethylene glycol  PenG  penicillin G (benzylpenicillin)  PG  peptidoglycan  QM/MM  quantum mechanical/molecular mechanical  RMSD  root mean squared difference  SDS  sodium dodecylsulfate xv  SeMet  selenomethionine  sc  (amino acid) side chain  sp.  species (singular)  spp.  species (plural, i.e. multiple species within a particular genus)  thio  thiostrepton  Tris  2-amino-2-hydroxymethyl-1,3-propanediol  TS  transition state  TSA  transition state analogue  UDP  undecaprenyl  UV  ultraviolet  VDW  van der Waals  v/v  unit volume (ml) per unit volume (ml)  w/v  unit weight (g) per unit volume (ml)  xvi  ACKNOWLEDGEMENTS Many thanks to my thesis advisor Natalie Strynadka for welcoming me into her excellent laboratory and research group, for her persistent belief in my abilities, and for entrusting me with challenging and interesting projects. I have always appreciated the considerable freedom you allowed me in choosing approaches and priorities, and your support in my choice of future career. Thanks also to my thesis committee for their advice, guidance, and support: Robert Hancock, Lawrence McIntosh, and Christopher Overall. Best regards and thanks to all the members and alumni of the Strynadka lab who have been so generous with their time and expertise over the years. You have all contributed immensely to my training and thinking. I want to specifically mention Andrew Lovering for being so delightful and brilliant in all our work together, I truly could not have done it without you, my friend; Trevor Moraes for all your morale-boosting efforts and sage advice; in addition, Liam Worrall, Gerd Prehna, Frank Rao, and Tom Spreter for years of fabulous twice-daily coffee sessions where all problems and achievements in science and life were up for discussion and much progress was made; as well as for this, Raz Zarivach and Haizhong Zhu for teaching me all about synchrotron work at the ALS; Mark Wilke, for being an excellent bench mate, for stimulating conversation and memorable shared graduate student (mis)adventures; Liza de Castro, Gunnar Olovsson, Marija Vuckovic, Leo Lin, Igor D'Angelo, and Paula Lario for their patience, knowledge, and indispensible technical help. I wish also to acknowledge the efforts and contributions of all my external scientific collaborators, in particular the interactions I greatly enjoyed with Susan Jensen (U Alberta), Ritu Banerjee (UCSF), Dan Lim (MIT), Jarrod Johnson and Gary Dmitrienko (U Waterloo). In keeping me from death due to starvation or exposure, I wish to thank the Natural xvii  Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research for generous graduate fellowships, as well as the families of Dorothy Helmer, SH Zbarsky, and Richard A. Robertson for their deeply appreciated awards in memoriam of their loved ones. I am at least as deeply indebted to many Universities Allied for Essential Medicines leaders, volunteers, and supporters at UBC and beyond—not only for all your time and effort, but for building this movement that has given me so much renewed purpose. I want to single out Patricia Kretz, for being especially reasonable, resourceful, and persuasive. Thanks to my fine and loving friends (if I’ve forgotten one of you here, possibly it is because you wronged me in some small way, who knows): Andrew Gray for awesome hugs; Freya Kristensen for patience, loyalty, and waffles; John Coyle for making me laugh so hard and on purpose; Claire Heslop for The Lorax and all the wine, music, and spicy food; Steph Gatto for believing in me and urging me on to a new career; David Kent for letting me follow you out to BC, being always up for fun and giving me things to say about sports; Lindsay Galbraith for also being a great roommate and reminding me about the outdoors; Kaeli Stark for all of our bicycle adventures, awesome talks, and being so caring; Matt Reimer for being my favourite hobo and for sharing all your top-secret plans; Maja Stachura for all the good times and closeness of being housemates and for not dying on a mountain; Kevin Forbes for drawings of sad robots that make me happy; Shannon LaBelle for keeping the Beaufort expatriates together and reminding me how excellent terrible things can be. And a super-special thanks to Prof. Glencora SK Borradaile, for all the advice, fondness, and encouragement I can’t imagine having done without in this final stretch. Thanks for saying I could keep a bicycle down in Corvallis. xviii  DEDICATION To my wonderful parents, James and Mary, and my dear sister Jennifer, for many continuing years of love, kindness, and unwavering support. And thanks for teaching me to read when I was so little. Who knew how many years of education would follow?  xix  STATEMENT OF CO-AUTHORSHIP Chapter 1 contains brief sections on the glycosyltransferase reaction, a table, and a figure adapted from a published review [AL Lovering, M Gretes, and NCJ Strynadka. (2008) Structural Details of the Glycosyltransferase Step of Peptidoglycan Assembly. Curr Opin Struct Biol, 8:534-543]. In Chapter 2, I planned and performed native and Selenomethionine derivative protein expression, osmotic shock, protein purification, crystallization, data collection, phasing and structure refinement for BLP. I carried out all data analysis, including final structure refinement and repository deposition of all structures. I prepared the manuscript including all text and figures with revisions and advice from Dr. Lim, Dr. Jensen, and Dr. Strynadka. This comparative project was conceived of by myself, Dr. Jensen and Dr. Strynadka. BLIP structure determination was initially done by Dr. Strynadka. BLIP-I protein was provided by Dr. Kang and Dr. Lee. BLIP-I and BLIP-I•TEM-1 complexes were crystallized by Dr. Lim and Liza De Castro, and their initial structure determination was done by Dr. Lim. The plasmid bearing blp was provided by Dr. Jensen. For Chapter 3, I determined, optimized, and carried out protocols for all crystalinhibitor soaking and cryoprotectant screening, data collection and analysis, structure refinement and inhibitor modeling. I prepared all manuscript text and figures, except for Fig. 3.5 (by Jarrod Johnson). Inhibitor synthesis, ab initio conformational optimization and initial characterization (Ki, MIC values) were provided by Jarrod Johnson, Valerie Goodfellow and Dr. Dmitrienko. Dr. Danel provided OXA-10 protein and Liza De Castro performed OXA10 crystallization. The project was initiated by a collaboration between Dr. Dmitrienko and Dr. Strynadka. xx  Chapter 4 describes a project in which myself and Dr. Lovering analyzed data and prepared a manuscript, with revisions by Dr. Page and Dr. Strynadka. Dr. Lovering, Dr. Page and Dr. Strynadka designed the research project. Liza De Castro crystallized PBP2a and performed crystal-inhibitor soaking experiments. Dr. Lovering carried out x-ray diffraction experiments. Dr. Danel and Dr. Page contributed new reagents and carried out thermal stability and CD experiments. For the study reported in Chapter 5, I performed all molecular modeling and structural anaylsis. Dr. Banerjee and I together prepared the manuscript, with revisions by Dr. Strynadka and Dr. Chambers. Dr. Banerjee and Dr. Chambers initiated the project, which became a collaboration with myself and Dr. Strynadka. Dr. Banerjee and L Basuino performed the ceftobiprole passaging experiments, MIC determination, and genome sequencing.  xxi  CHAPTER 1: INTRODUCTION 1.1  Overview of Antibiotics and Antibiotic Resistance  1.1.1 Context: The Antibiotic Age For nearly all of history, deaths due to bacterial infection have been an inescapable part of the human condition. The discovery of penicillin in 1928 (Fleming, 1945) would change this reality. Just over a decade later, a new age—in which formerly deadly infections are easily treated—was ushered in by the validation of penicillin as a life-saving antibiotic (Florey, 1945, Chain, 1945) and the development of methods of its mass production (see Queener, 1986 for a review). Indeed, the effectiveness of modern medicine rests upon the availability of essential medicines including antibiotics (WHO, 2007). This is made plain by the extreme morbidity and mortality that result from the deterioration or absence of modern health care systems, for example in the world’s conflict zones and lowest-income countries; see Brown et al., 2008 for an illustrative study. Helplessness in the face of bacterial pathogens is becoming increasingly unrestricted to societies experiencing acute poverty, natural disaster, or war. Antibiotic-resistant bacteria are presently endemic in hospitals and communities even in wealthy and politically stable countries (Doshi et al., 2009). Just as a large natural stockpile of antibiotics is found in highly competitive soil ecosystems (Baltz, 2008), so too has a vast reservoir of antibiotic resistance determinants lain in wait (Allen et al., 2009). Penicillin-resistant strains of Staphylococcus aureus emerged within three years of the clinical introduction of penicillin in the 1941 (Neu, 1992). Widespread medical, agricultural, and industrial use of antibiotics exerts massive selective pressure on the development, spread, and maintenance of antibiotic resistance in pathogenic bacteria (Levy, 2001, Palumbi, 2001). Thus the age of antibiotics rapidly became 1  synonymous with the age of antibiotic resistance. 1.1.2 Overview of Antibiotic Mechanisms of Action There are a great many antibiotics in clinical use, targeting a wide array of vital bacterial cell processes, including cell wall biosynthesis (ß-lactams, glycopeptides), protein synthesis (tetracyclines, aminoglycosides), DNA gyrase/topoisomerase (quinolones), and cell membrane polarity (e.g. daptomycin); for a comprehensive, concise review see Bumann, 2008. The focus of the research described herein is on the ß-lactam antibiotics, which are used to target the transpeptidation reaction that crosslinks the polymers comprising the bacterial cell wall. Among the many advantages of this group of antibiotics is that the target is accessible (located at the surface of the cytoplasmic membrane, which is exposed in gram positive bacteria and thus ß-lactams are particularly effective against these), has no human counterpart (hence ß-lactams often exhibit low toxicity), and is essential for the survival of the organism. This latter property renders ß-lactams generally bactericidal rather than bacteriostatic; this activity is generally considered advantageous in the treatment of immunecompromised individuals, important for rapid treatment of central nervous system infections, and essential for treatment of endocarditis (Finberg et al., 2004). Along with these key advantages are cheap production and easily-varied chemical moieties, resulting in a popular, large, diverse, and therefore long-lived family of antibiotics (Fig. 1.1). However, the tremendous success of the ß-lactams has led to proportionally widespread resistance in bacteria, given their large numbers, short generation times, and promiscuous sharing of genetic information (Neu, 1992).  2  Figure 1.1: The four major classes of ß-lactam antibiotics (Martin and Kaye, 2004, 3  Bryskier, 2000, Kattan et al., 2008, Singh, 2004), showing generalized core structures with positions of variable groups R(1,2) indicated; (a) the penams (penicillins) with the precursor structure 6-aminopenicillanic acid (6-APA) indicated with bold bonds; (b) the cephalosporins with the precursor structure 7-aminocephalosporanic acid (7-APA) in bold, sample structures from the various generations (1G, 2G, etc.) are given; (c) the carbapenems; (d) the monobactams.  1.1.3 Overview of Antibiotic Resistance Strategies There are, generally speaking, three distinct mechanisms that drive resistance to antibiotics (Walsh, 2000): degradation or modification of the antibiotic itself (Page, 2008), exclusion or efflux of the antibiotic from the bacterial cell (Martínez-Martínez, 2008), and modification of the antibiotic target (Livermore, 2006). As effective as ß-lactams have proven, bacteria have developed diverse means of resistance within the scope of all three strategies. The two mechanisms explored in detail here are (1) the enzymes able to hydrolyze ßlactams, in particular the Class A and Class D serine ß-lactamases; and (2) the resistant ßlactam target enzyme PBP2a. The ß-lactamases are the most widespread factor conferring bacterial resistance to ß-lactam antibiotics. ß-lactamase-related resistance to latter-generation, broad-spectrum ß-lactams as well as to small-molecule inhibitors is a well-established and widespread phenomenon. Likewise, altered PBPs are able to confer resistance to a variety of ß-lactams and may achieve levels of tolerance up to 3 orders of magnitude greater than clinically-achievable antibiotic concentrations (Kondo et al., 2001). Combined, these phenomena have serious implications for the treatment of bacterial infections.  4  1.2  ß-Lactams: Antibiotics Targeting the Cell Wall  1.2.1 Cell Wall Structure The cell wall (CW) envelops the bacterial cytoplasmic membrane (CM) like a mesh sac, conferring strength and rigidity while remaining flexible, defining cell shape (Zapun et al., 2008) and providing resilience against osmotic pressure (Giesbrecht et al., 1998, Vollmer and Bertsche, 2008). The sacculus, as it is indeed termed, is for the most part comprised of a massive (several GDa) macromolecule called murein, which is a polymer of peptidoglycan (PG): long sugar chains crosslinked with peptide bridges. PG also provides an attachment point for proteins (especially important in Gram positive bacteria) and other cell wall polymers, as well as the outer membrane in Gram negative bacteria (Dramsi et al., 2008). In sum, the PG layer is a porous, elastic structure that is dynamically remodeled in response to cell growth and alters its deposition to form septa for cell division (Park and Uehara, 2008, Vollmer et al., 2008a). The organization and orientation of the sugar chains and peptide crosslinks has been a matter of much debate. Some models postulate an essentially 2-dimensional structure, with chains and crosslinks laid out parallel to the CM, either in an orderly (Holtje, 1998) or random (Koch, 1998) fashion. In Gram negative bacteria, a single, thin (~2.5 nm; Labischinski et al., 1991) layer is sandwiched between the CM and the outer membrane (OM). In Gram positive bacteria, which lack an OM, the thick (9±1 nm; Hayhurst et al., 2008) PG wall is proposed to consist of multiple, stacked layers. An alternative model supposes that glycan strands project perpendicularly from the cell, with crosslinks running parallel to the CM (Dmitriev et al., 2004). In an extended computational model of this predicted manner of murein organization, the highly crosslinked cell wall of the Gram 5  positive bacterium S. aureus is explained by staggered, zigzag zippering of adjacent glycan strands by peptide crosslinks at multiple positions along their length (Dmitriev et al., 2004). This latter model may be falling out of favour. In the Gram positive bacterium Bacillus subtilis, atomic force microscopy appears to rule out the scaffold model and support a coiled-coil organization of the PG layer (Hayhurst et al., 2008). The scaffold model appears also to be at odds with the reported thickness of the PG layer in Gram negative bacteria given the observed average length of glycan chains; along with recent electron cryotomography evidence many investigators presently favour a disordered, single-layered model (Vollmer and Bertsche, 2008, Gan et al., 2008). Variability among species, growth state and cellular location complicates the generalizability of such studies. It is noteworthy, however, that a remarkable homogeneity was observed in NMR studies of isolated PG sacculi (Kern et al., 2008). Thus there is good reason to expect that a general model of CW organization will eventually emerge. At the atomic scale the view is clearer. PG strands are linear sugar chains of alternating N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) residues (Fig. 1.2). MurNAc is attached via its 3-OH to the crosslinking peptide, L-Ala-!-D-Glu-L3xx-D-Ala (where D-Glu is "-amidated and often termed D-iGlx for its !-linkage; L-3xx stands for L-Lys in Gram positives like S. aureus and for L,L-diaminopimelic acid in Gram negatives and a few other Gram positives). Peptide bridges are crosslinked from the terminal D-Ala peptide carboxylate to the free amino group on the L-3xx residue; additional bridging amino acids are featured in certain species, e.g. pentaglycine in Staphylococci is substituted at the "-amino group of the L-Lys residue (Vollmer et al., 2008a).  6  Figure 1.2: Peptidoglycan polymer structure in Staphylococcus aureus. Glycan chains of alternating N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) units linked by a [ß-1,4] glycosidic bond extend in either direction for variable lengths n(1-4) (which range from 3 to 26 and average 6 units; Boneca et al., 2000). Strands are crosslinked at opposing strand MurNAc residues by stem peptides (L-Ala-D-iGlx-L-Lys-D-Ala-D-Ala) covalently bonded via a pentaglycine (Gly5) bridge joined to opposing stem peptide residues L-Lys-!-NH2 and penultimate D-Ala (replacing the terminal D-Ala in the course of the crosslinking transpeptidation reaction).  7  8  In addition to such differences between organisms, the composition of PG stem peptides of a given organism may change and diversify under selective pressure by glycopeptide antibiotics (e.g. vancomycin), which act by binding and sequestering the stem peptide substrate (Mainardi et al., 2008). Recent studies of the composition of the sacculus have described CW glycopolymers other than PG, such as teichoic acids (Weidenmaier and Peschel, 2008). 1.2.2 Cell Wall Biosynthesis: Penicillin-Binding Proteins PG is formed via polymerization and cross-linking of a lipid II monomer. Lipid II (undecaprenyl-P-P-MurNAc-[pentapeptide]-GlcNAc) is comprised of the PG stem peptide linked to a soluble UDP-MurNAc substrate. The stem peptide is synthesized by sequential addition of amino acids to a MurNAc sugar during a series of cytoplasmic reactions (Bavsar and Brown, 2006). Localization to the inner membrane via transfer to a C55 UDP lipid forms lipid I; attachment of the GlcNAc sugar to MurNAc completes lipid II synthesis (Barreteau et al., 2008). Translocation to the outer leaflet (likely by the so-called “flippase” MviN, recently renamed MurJ; Ruiz, 2008) marks the beginning of the extracellular phase of PG synthesis. A glycosyltransferase (GT) reaction polymerizes lipid II molecules into a nascent chain (immature PG). Once two lipid II molecules have been joined with the loss of one UDP to form the tetrasaccharide lipid IV, the growing chain is extended by sequential addition of the lipid II monomer. The final step in murein biogenesis and the focus of the present thesis is the transpeptidase (TP) reaction, the intended target of inhibition by ßlactams. The enzymes responsible for the TP reaction (and, incidentally, the bulk of GT activity) are called penicillin-binding proteins (PBPs), since they were initially identified as 9  the targets of penicillin and other ß-lactam antibiotics via radiolabelling studies (Blumberg and Strominger, 1972, Spratt and Pardee, 1975). A given bacterial species' PBPs are classified as either high molecular weight (HMW; ~60-120 kDa) or low molecular weight (LMW; ~2050kDa) and are numbered in order of descending molecular weight. LMW PBPs may be soluble or membrane-associated, and are believed to function in CW remodeling and PG turnover via DD-transpeptidase, DD-carboxypeptidase, and DD-endopeptidase activities and are respectively placed in classes A, B and C. The HMW PBPs are essential, transmembrane enzymes whose inhibition can result in cell death. These are also divisible into three classes (as described by Massova and Mobashery, 1998, Goffin and Ghuysen, 1998). Class A enzymes are bifunctional TP and GT enzymes, Class B PBPs are TPs that also bear an N-terminal, non-penicillin-binding domain (NPD) of uncertain function, and Class C PBPs are ß-lactam sensor-transducer proteins involved in the signaling cascade that regulates antibiotic resistance gene expression (Hardt et al., 1997, Zhang et al., 2001). We will examine the TP reaction in detail and the HMW PBPs responsible (classes A and B) from the perspective of the clinically-important Gram positive bacterium Staphylococcus aureus. There are four well-documented PBPs endogenously present in S. aureus (Georgopapadakou and Liu, 1980): PBP1 (85kDa), PBP2 (81kDa), PBP3 (75kDa) and PBP4 (45kDa). Of these, PBP1, PBP2, and PBP3 are believed to be essential (Wada and Watanabe, 1998, Georgopapadakou et al., 1982). PBP4, while apparently dispensable (Katayama et al., 2003), appears to be responsible for tight crosslinking of PG, so is likely to be required under certain environmental conditions. In addition, PBP4 makes important contributions to the ß-lactam resistance conferred by the exogenous PBP2a (Memmi et al., 2008, Banerjee et al., 2008). Recently, FmtA was identified as a fifth native S. aureus PBP and appears to be 10  required for the formation of highly crosslinked PG and to play a part in the PBP2adetermining ß-lactam resistance phenotype (Fan et al., 2007). The organization of the PG biosynthesis machinery in S. aureus has been recently reviewed (Zapun et al., 2008). The bifunctional (Class A) enzymes segregate GT activity from the TP reaction in individual domains, clearly delineated by a linker region, placing the TP functionality in all known enzymes at the C-terminus (Fig 1.3). This arrangement is intuitive given the positioning of the transmembrane helices at the N-terminus and the requirement that the TP activity occur distally from the membrane. Modularity of function has been proven by the generation of isolated TP domain structures from proteolyzed bifunctional constructs (Macheboeuf et al., 2005, Lovering et al., 2006).  Figure 1.3: Likely orientation of S. aureus PBP2 at the cell membrane, shown in complex with moenomycin (PDB ID: 2olv; Lovering et al., 2007), which likely mimics the disaccharide units involved in the TG reaction; TP and GT domains, linker region, as well as N- and C- termini are indicated.  Exploration of the final two steps of peptidoglycan synthesis has historically been heavily weighted towards the TP proteins, largely due to the clinical importance of the inhibition of this reaction by ß-lactams. The disparity in knowledge between GT and TP functionalities also arises, however, from the intrinsic difficulty of studying a membraneassociated process: the GT reaction that couples lipid II and releases UDP is believed to occur at the membrane. 11  The catalytic mechanism of the TP reaction has been described in detail by a wealth of PBP crystal structures. Representatives of HMW PBP Class A enzymes include those solved by Lovering et al., 2006, Lovering et al., 2007, Contreras-Martel et al., 2006, Macheboeuf et al., 2005, and Yuan et al., 2007. Work by Gordon et al., 2000, Lim and Strynadka, 2002, Sauvage et al., 2002, and Powell et al., 2009 describes those of Class B. LMW PBP structures are described by McDonough et al., 2002, Rhazi et al., 2003, and Sauvage et al., 2005 (class A); Morlot et al., 2005 and Nicola et al., 2005 (class B). These structures reveal a remarkable conservation of fold and active site architecture. Three conserved motifs (Fisher et al., 2005) involved in catalysis are: Ser-Xxx-Xxx-Lys (motif I), [Ser/Tyr]-Xxx-Asn (motif II), and [His/Lys]-[Thr/Ser]-Gly (motif III). The consensus opinion, illustrated in Fig 1.4a using Neisseria gonorrhoeae PBP2 (by analogy with mechanisms from related soluble, LMW PBPs; Rhazi et al., 2003, Silvaggi et al., 2003), is that catalysis begins with general base activation of the Motif I Ser (usually by the Motif I Lys) for nucleophilic attack of the carbonyl carbon of the amide bond joining the terminal D-Ala-DAla of the donor peptide stem. The tetrahedral transition state collapses with the leaving of the terminal D-Ala residue, generating a stable acyl-enzyme intermediate (Rhazi et al., 2003). Activation of the acceptor peptide amino group (of the L-3xx residue side-chain or Gly5 bridge, as discussed in 1.2.1) for nucleophilic attack on this acyl-enzyme may be achieved via the side chain hydroxyl of the Motif ii Ser/Tyr acting as general base (Silvaggi et al., 2003). The net result is deacylation of the enzyme and amide bond formation between the L-3xx side chain of one stem peptide and the penultimate D-Ala of another. Residues and motifs involved in the reaction coordinates are summarized in Table 1.1 and compared to those of selected homologous ß-lactam-resistance enzymes discussed in detail in sections 1.3 and 1.4. 12  Figure 1.4: Simplified catalytic and inhibition reaction mechanisms of ß-lactam susceptible PBPs, using N. gonorrhoeae PBP2 as an example; (a) substrate turnover, illustrated using donor and acceptor peptides from Staphylococcus aureus for ease of comparison with reaction schemes used elsewhere in Chapter 1; (b) mechanism-based inhibiton with a ß-lactam, in this case a prototypical penam (penicillin), coloured to emphasize similarities with substrate D-Ala-D-Ala in (a).  13  14  Table 1.1: Catalytic residues and their proposed mechanistic roles for the transpeptidases N. gonorrhoeae PBP2 and S. aureus PBP2a as well as the ß-lactamases E. coli TEM-1 and P. aeruginosa OXA-10. PBP2 (susceptible transpeptidase)a  PBP2a (resistant transpeptidase)  TEM-1 (Class A ß-Lactamase)  OXA-10 (Class D ß-Lactamase)  Motif Ib  S310-x-x-K313  S403-x-x-K406  S70-x-x-K73  S67-x-x-K70  Motif II  S362-x-N364  S462-x-N464  S130-x-N132  S115-x-V117  Motif III  K497-T498-G499  K597-S598-G599  K234-S235-G236  K205-T206-G207  Distinct Feature  -  Rearrangement of œ2 helix and ß3 strand required for acylationc  Omega loop (E166) involved in catalysisd  Carbamylated K70 (Kcx-70)  Representative PDB Structures  3EQUe  1VQQ (apo), 1MWS (nitrocefin complex)c  1AXBf - complex with transition state analogue; 1FQGg – acyl-enzyme intermediate h 1M40 – ultra-high resolution (0.85Å) structure  1E4Di – carbamylated Kcx-70; 1FOFj - uncarbamylated Lys70 1K6Sj - in complex with phenylboronic acid inhibitor  Ser-310-O-:  Ser-403-O-:  Ser-70-O-:  Ser-67-O-:  Nucleophile (Acylation)  15  PBP2 (susceptible transpeptidase)a  PBP2a (resistant transpeptidase)  TEM-1 (Class A ß-Lactamase)  OXA-10 (Class D ß-Lactamase)  General Base (Activate Nu for Acylation)  Lys-313-H2N:  Lys-406-H2N: or ß-lactam COO-:c  Glu-166COO-:k,l  Lys-73H2N:m,q  Kcx-70COO-:n  Lys-70-H2N:j  Proton Shuttlingo Path (Acylation)  Lys-313 ! ( uncertain ! ) Lv N  Lys-406 ! Ser-462 ! Lv N  Glu-166 ! H2 O ! Lys-73 ! Ser-130 ! Lv Np  Lys-73 ! Ser-130 ! Lv N  Kcx-70 ! Ser-115q ! Lv Nr  Lys-70 ! Ser-115! Lv Ns  Oxyanion Hole  Ser-310, Thr-500 (backbone N atoms)  Ser-403, Thr-600 (backbone N atoms)  Ser-70, Ala-237t (backbone N atoms)  Ser-67, Phe-208 (backbone N atoms)  Donor Identity (Preferred)  Stem peptide L-3xx amine  Stem peptide L-3xx amine  H2 O  H2 O  General Base (Activate Donor for Deacylation)  Ser-362u  (not determined)  Glu-166  Kcx-70 COO-:  Lys-70-H2N:  Proton Shuttling Path (Deacylation)  (not determined)  (not determined)  Glu-166 ! Lys-73 ! Ser-70v  Kcx70! Ser-115w ! Ser-67  Lys-70! Ser-115! Ser-67  16  a  b c d e f g h i j k  l  m  n o  p q r s  t u v w  Generally, mechanism was initially defined by analogy to Serine protease mechanism (Henderson, 1970); mechanism supported by studies with, e.g., soluble LMW PBPs from Streptomyces K15 (Rhazi et al., 2003) and R61 (Kelly et al., 1985, McDonough et al., 2002, Silvaggi et al., 2003) This shared minimal motif is common amongst all related enzymes (Massova and Mobashery, 1998) Lim and Strynadka, 2002 Strynadka et al., 1996b Powell et al., 2009 Maveyraud et al., 1998 Paetzel et al., 2000 Minasov et al., 2000 Golemi et al., 2001 Kerff et al., 2003 On the basis of impairment of acylation rates of E166 mutants (Guillaume et al., 1997) as well as phosphonate TSA bound to protonated E166 (ultra-high resolution structure described by Minasov et al., 2002); also supported by QM/MM calculations (Hermann et al, 2003) It must be noted that the latest published QM/MM simulation appears to support the existence of both Glu-166- and Lys-73-mediated activation of the Serine nucleaophile (Meroueh et al., 2005) Based on acyl-enzyme studies with E166N deacylation-defective mutant (Adachi et al., 1991, Strynadka et al., 1992); Lys-234 probably electrostatically assists Maveyraud et al., 2000 Proton shuttling here refers to path of the proton abstracted by the general base during nucleophile activation from the nucleophile to the leaving group (this is a N atom during the acylation and the initial Serine nucleophile at deacylation) Nukaga et al. 2003 By extrapolation only from postulated role of Ser-115 in deacylation proton shuttle (Maveyraud et al., 2000) Favoured in model with Kcx as the general base (Maveyraud et al., 2002) Favoured by the structure reported by Paetzel et al., 2000; in which the Lys-70 carboxylate is not observed. In this mechanism, the protonated Lys205 would also serve to electrostatically push the shuttled proton toward the intermediate From Kuzin et al., 1999 By analogy with R61 DD-peptidase (Silvaggi et al., 2003) From boronate TSA (Strynadka et al., 1996b) Suggested by Maveyraud et al., 2000 because Ser-115 is at hydrogen-bonding distance to Ser-67  17  The class B HMW PBPs' N-terminal NPD domain bears no apparent structural relation to the GT domain of class A HMM PBPs, yet appears to be homologous to YoeB, a protein thought to enhance cell survival in the presence of cell wall targeted antibiotics by mediating autolysin activity via protein-protein interactions (Salzberg and Helmann, 2007). Thus it may affect co-operative interactions with other S. aureus proteins, including PBPs (Leski and Tomasz, 2005). Indeed, it has been suggested that this domain serves to direct class B HMM PBPs to the divisome, where murein biosynthesis enzymes have long been postulated to exist as part of multi-enzyme CW synthase complexes. Many PBPs have been shown to interact with a variety of proteins (various examples are provided in Table 1.2). Rod-shaped bacteria appear to maintain their shape via inclusion of these enzymes into specific “elongase” (peripherally located) and “divisome” (septally located) complexes (reviewed recently by den Blaauwen et al., 2008).  18  Table 1.2: Enzymes and structural proteins shown to interact with PBP family members. Interaction Partner / Domain MreC  Interacting PBP(s) / Domain PBP2, PBP2a, PBP2b, PBP1a, PBP1b  Species Caulobacter crescentus  Method Used / Evidence for Interaction Affinity chromatography using stationary MreC Colocalization under immunofluorescence microscopy  Representative Structure of Partner (PDB)  Reference(s)  2J5U  (Divakaruni et al., 2005)  ¨  ¨  (Slovak et al., 2006)  MreC  PBP2  Rhodobacter sphaeroides  MreC / ßdomain  PBP1, PBP2a, PBP3, PBP4, PBP2c, PBP4b  Bacillus subtilis  Bacterial two-hybrid assay  ¨  ¨  (van den Ent et al., 2006)  MreC / Nterminal domain (periplasmic, transmembrane)  PBP2d, PbpH, PBP2b  B. subtilis  Bacterial two-hybrid assay  ¨  ¨  (van den Ent et al., 2006)  MreC  PBP1 / transmembrane  B. subtilis  ¨  ¨  (Claessen et al., 2008)  MreB  PBP2  R. sphaeroides  MreD  PBP1, PBP2 (only via complex with MreB, MreC)  E. coli  1JCE, 1JCF, 1JCG  (Slovak et al., 2006)  N/A  (Kruse et al., 2005)  1O6Y  (Dasgupta et al., 2006)  1TXO  (Dasgupta et al., 2006)  N/A  (Claessen et al., 2008)  B. subtilis  Bacterial two-hybrid assay; repositioned GpsB is sufficient to reposition PBP1 in vivo  N/A  (Claessen et al., 2008) (Mohammadi et al., 2007)  PknB  PBPA  Mycobacterium tuberculosis  PstP  PBPA  M. tuberculosis  EzrA  GpsB  PBP1 / transmembrane PBP1 / transmembrane +cytoplasmic tail (intact, together only)  Bacterial two-hybrid assay Colocalization under immunofluorescence microscopy (certain stages of cell cycle)  B. subtilis  Bacterial two-hybrid assay In vitro kinase assay and coexpression (in E. coli) combined with sitedirected mutagenesis shows PknB phosphorylates PBPA In vitro dephosphorylation assay Bacterial two-hybrid assay  MurG  PBP3  E. coli  PBP3 is required for localization of MurG to division site  1F0K  MraY  PBP3 (by assn with MurG)  E. coli  coimmunprecipitation  N/A  RodA  PBP2  E. coli  RodA is required for PBP2 activity  N/A  (Mohammadi et al., 2007) (Ishino et al., 1986)  19  Interaction Partner / Domain FtsN / cytoplasmic domain (aa58-125 are critical) FtsN / parts of cytoplasmic domain (aa58-125 or 166319) are sufficient for binding; include N-term, TM domain for highest affinity  Interacting PBP(s) / Domain  Species  Method Used / Evidence for Interaction  Structure of Partner (PDB)  Reference(s)  PBP1a, PBP3  E. coli  Affinity chromatography using stationary FtsN  1UTA  (Muller et al., 2007)  E. coli  Affinity chromatography, pulldown assay and surface plasmon resonance using stationary FtsN; bacterial two-hybrid assay  E. coli  Bacterial two-hybrid assay  E. coli  Bacterial two-hybrid assay  E. coli  Bacterial two-hybrid assay  N/A  (Karimova et al., 2005)  E. coli  Bacterial two-hybrid assay  FtsN – 1UTA  (Derouaux et al., 2008)  Pulldown assay  FtsH – 2DHR HflK/C – N/A  (von Rechenberg et al., 2005)  N/A  (Vollmer et al., 1999)  2AE0  (Vollmer et al., 1999)  PBP1b  FtsA, FtsN  PBP3 (“FtsI”)  FtsW, FtsQ / Cterminal tail (aa246-276)  PBP3 (“FtsI”) / periplasmic (noncatalytic, aa42-70) PBP3 (“FtsI”) / periplasmic (noncatalytic, aa70250) Monofunctional GT  FtsL, YmgF FtsN, FtsW FtsH:HflK:HflC  PBP3 (“FtsI”)  E. coli  Affinity chromatography with stationary MipA; Simultaneous binding of MipA and MltA to PBP1a immobilized on BIAcore SPR sensor chip Affinity chromatography using stationary MltA (may be mediated by MipA. MltA fails to bind SPR-PBP1b in its absence)  ¨  ¨  FtsA – 1E4F, 1E4G FtsQ – 2VH1, 2VH2 FtsW – N/A  MipA  PBP1b  E. coli  MltA†  (PBP1a) PBP1b, PBP1c, PBP2, PBP3  E. coli  MltB  PBP1b, PBP1c, PBP3  E. coli  Affinity chromatography using stationary MltB  1LTM  Slt70  PBP3, PBP7, PBP8  E. coli  Affinity chromatography using stationary Slt70  1SLY  Slt70  PBP1b, PBP1c, PBP2, PBP3  E. coli  Affinity chromatography using stationary Slt70  ¨  B. subtilis  Localization of PBP2b to cell division sites as observed by immunofluorescence microscopy is lost under LytE depletion  LytE  PBP2b  ¨  2K1G (homologous domain)  (Muller et al., 2007)  (Karimova et al., 2005) (Karimova et al., 2005)  (von Rechenberg et al., 1996) (Romeis and Holtje, 1994) (von Rechenberg et al., 1996) (CarballidoLopez et al., 2006)  20  It is unknown at present how important the relative positions of the GT and TP active sites are in converting immature peptidoglycan to its mature form. It is certain that transpeptidation will only occur on the transglycosylation product (Terrak et al., 1999, Born et al., 2006, Bertsche et al., 2005), and that new peptidoglycan is attached to the old wall via the TP reaction alone (Born et al., 2006). Thus the Class B NPD may serve only as an extension to position the TP activity at an appropriate distance from the CM (Macheboeuf et al., 2006). An additional interesting PBP feature is the insertion of a ~100 aa domain into some of the bifunctional enzymes (e.g. E. coli PBP1b). Certain PBPs may functionally dimerize via the GT domain (Zijderveld et al., 1995, Bertsche et al., 2005), and the S. aureus PBP2 and Aquifex aeolicus PBP1a structures show different interfaces within their crystal forms (Lovering et al., 2008). With only two members of the family analyzed thus far, putative dimerization patterns are still uncertain–placement of the larger PBP2 structure over the PBP1a interface results in steric clashes, yet it is possible that the less extensive PBP2 interface is an artefact of crystallization. It is not inconceivable that family members exhibit distinct dimerization patterns, but if dimerization has a functional role, it would be expected to present common faces of the GT domain to binding partners, especially the lipid IItranslocating “flippase”. 1.2.3 ß-Lactam Antibiotics and Mechanism of Action The essential nature of peptidoglycan, its ubiquity across the kingdom of eubacteria (and absence in humans), and extracellular assembly make its synthesis an advantageous antibiotic target. ß-lactams are mechanism-based inhibitors of PBPs that mimic the D-AlaD-Ala peptide bond with respect to shape and charge distribution (Tipper and Strominger, 21  1965) and thus target the TP crosslinking reaction (Fig 1.4b). The lactam ring bearing the peptide-like bond is cleaved via the usual mechanism. Unlike reaction with the native substrate, however, the leaving group is retained in the active site via the remaining intact carbon-carbon bond from the four-membered lactam ring. Thus the acceptor peptide position in the active site is occupied, precluding the approach of a deacylating peptide or, alternatively, a hydrolytic water (Ghuysen et al., 1986). The only nucleophilic centre suitably positioned for deacylation is the former lactam nitrogen; cyclization to reform the strained four-membered ring is, of course, highly disfavoured. Thus this acyl-enzyme species is stable, and the cell is rendered unable to perform PG crosslinking. In the case of ß-lactams that are bactericidal, this inhibition persists for a time period sufficient to ensure the death of the cell. The bactericidal effects of irreversible PBP inhibition specifically arise from the resultant imbalance in cell wall biosynthesis and cell wall degradation activity. Autolysins and other cell wall remodelling enzymes (e.g. exo- and endo-peptidases) that function in normal cell growth and division appear to be the proximal cause of cell death (Holtje, 1998, Vollmer et al., 2008b). Indeed, repression of autolysin activity appears able to supplement ß-lactam resistance (Salzberg and Helmann, 2007). Electron microscopy confirms that pore formation along the plane of cell division allows the escape of cytoplasmic contents, driven by osmotic pressure in typically hypotonic environments (Giesbrecht et al., 1998). The turning point in the clinical development of penicillin, the first ß-lactam in use, was its crystallization in 1943 followed by its structure determination in 1945 (reviewed in Hodgkin, 1949, Glusker, 1994). Once its molecular core featuring the highly unusual fourmembered ring (6-aminopenicillanic acid, 6-APA, Fig 1.1) could be isolated (Sheehan and Logan, 1959) and produced in large quantities (Batchelor et al., 1959), a broad array of 22  derivative penicillins was made available by substitution at the variable group R (see Sheehan et al., 1980 for a review of this work). Discovery of the natural product cephalosporin C permitted the analogous production of the 7-aminocephalosporanic acid (7-ACA) pharmacore by hydrolysis, permitting the eventual production of a variety of R1- and R2substituted cephalosporins (reported in Loder et al., 1961, reviewed in Queener, 1986). Diverse organisms have yielded natural products that now form entire classes of ßlactam antibiotics. The penicillins were discovered in Penicillium notatum (Fleming, 1944); cephalosporin C, from which 7-ACA and its semisynthetic derivatives were produced, was isolated from Cephalosporium acremonium (Brotzu, 1948); the carbapenems were identified in Streptomyces spp. (initially S. cattleya; Box et al., 1979); the monobactams are found in Pseudomonas spp., Gluconobacter spp., and Chromobacterium spp. and were first found in C. violaceum (Cooper et al., 1985). Cephalosporins have undergone extensive development, through multiple “generations” of improvement (Bryskier, 2000). Successive generations were developed in part to expand the spectrum of target pathogens, but in many cases for greater resistance to hydrolysis by extended-spectrum ß-lactamases (ESBLs), discussed in detail below (Essack, 2000). The essential molecular structures and clinically-relevant examples of the major ßlactam classes are also given in Fig. 1.1. Other ß-lactams include clavams, a class of natural ßlactamase inhibitors (discussed below) isolated from Streptomyces clavuligerus (Brown et al., 1976); Actinomycetes spp. were found to yield cephamycins (related to and often categorized along with the cephalosporins as cephems)–for a review of these and other minor classes, refer to Liras and Martín, 2006.  23  1.3  ß-Lactamase Enzymes The most common means of resistance against ß-lactam antibiotics is their hydrolysis  by secreted ß-lactamase enzymes (EC 3.5.2.6). These have been assigned, according to amino acid (aa) sequence, to four classes (A-D; Ambler, 1980). A more detailed functional classification system exists (Bush et al., 1995) but precision will be sacrificed for convenience and the Ambler classifications used exclusively here, as commonly done in the literature. As enzymes which hydrolyze the D-Ala-D-Ala peptide bond, they are closely comparable in function to proteases, and indeed are listed in the MEROPS peptidase database† (Clan SE). Class B ß-lactamases are metalloenzymes and, like metalloproteases, employ one or two divalent cations (commonly zinc) to generate an hydroxide ion from water for direct attack and hydrolysis of the peptide bond without formation of an acyl-enzyme intermediate (for reviews on the mononuclear and binuclear enzymes, refer to Garau et al., 2005, Heinz and Adolph, 2004). These metallo-ß-lactamases (MBLs) are of greatest clinical concern for their ability to hydrolyze the carbapenems (e.g. meropenem, imipenem), used as a last resort in treatment of cephalosporin-resistant infections. They are widespread among bacterial pathogens, most importantly Pseudomonas aeruginosa, Serratia marcescens, and Acinetobacter spp. Geographically, the IMP family of enzymes are widely distributed, but of minor importance compared with the SPM-1 family (which has been rapidly disseminated in Latin America) and the VIM family (Jones et al., 2005). VIM-2, predominantly found in P. aeruginosa, is the preeminent MBL world wide, found in 37 countries in 5 continents. No clinical inhibitors of these enzymes are available and none are likely to become so for several years (Walsh, 2008).  †  http://merops.sanger.ac.uk  24  All other ß-lactamase classes (A, C, and D) employ a serine nucleophile. The Class C ß-lactamases (MEROPS family S12) are widely distributed amongst bacterial species and are of great clinical importance. The AmpC cephalosporinases, for example, are able to efficiently hydrolyze even 3rd-generation cephalosporins and are widespread (Jacoby 2009). Enterobacter spp. in particular are of major concern. Following exposure to broad-spectrum cephalosporins, E. cloacae bearing chromosomal ampC is reported to stably overproduce AmpC; prevalence in British hospitals is 30-40% (Livermore and Woodford, 2006) and in the United states, 15-25% (Jones et al., 2008). In Canada, 232 unique, cefoxitin-resistant, AmpC-bearing strains of Escherichia coli were identified among 29,323 isolates from 12 hospitals (Mulvey et al., 2005). These enzymes are also disseminated on plasmids: 44 Salmonella strains encoding plasmid-mediated CMY enzymes (an AmpC subgroup) were identified from 1,378 strains in the United States (Whichard et al., 2007). Klebsiella pneumoniae is an important AmpC-positive blood-borne infection in hospitals, especially affecting cancer patients and organ-transplant recipients, and has been found in the US and across Southeast Asia (reviewed in Jacoby 2009). ß-lactamases of Classes A and D (MEROPS family S11) are the subject of the present thesis, and are each treated in detail below. 1.3.1 Origins The most widely stated hypothesis concerning the origin and evolution of antibiotic biosynthesis and secretion is that microorganisms employ these to gain advantage against competitors for scarce resources, for example in a crowded soil ecosystem. If this is true, then the repurposing of existing genes to confer antibiotic resistance logically follows for tolerance of self and defense against others. This is consistent with the reported production 25  of antibiotics and antibiotic-resistance determinants by the same organism, at times within the same gene cluster (e.g. that encoding the biosynthetic enzymes S. clavuligerus cephamycin C and a class A ß-lactamase; Alexander and Jensen, 1998, Perez-Llarena et al., 1997). ß-lactamases appear to be naturally plentiful: numerous and diverse ß-lactamases were identified in soil from a remote, undisturbed site (Allen et al., 2009). Indeed, that study identified a class A ß-lactamase (EU408346) from an unculturable soil microbe that strongly resembles the ß-lactamase E. coli TEM-1 (43% identical sequence identity, E-value‡ = 3 x 10-54). As functional metagenomics and similar studies progress, the wealth of ß-lactam biosynthetic and hydrolytic functionalities to be found in nature should become clear. While it must be noted that this ontological outlook is hardly uncontested (a number of alternate views are offered in Yim et al., 2006), it is clear that penicillin-binding proteins and all classes of serine ß-lactamases share common ancestral origin (Massova and Mobashery, 1998). Indeed, despite modest variation in overall sequence and fold, the catalytic motifs are remarkably conserved among themselves and the PBPs (Table 1.1); the residues comprising the active site architecture may also be superimposed quite closely in space (Fig. 1.5).  ‡  i.e. the number of equivalently-scoring matches from a BLAST search that would be expected to occur by chance (Altschul et al., 1990)  26  Figure 1.5: Active site structure of homologous enzymes involved in making or breaking the transpeptide bond (residues involved in binding or catalysis are shown): PBP2 (top left), PBP2a (top right), TEM-1 (bottom left), and OXA-10 (bottom right). Adjacent enzymes were chosen as those most closely related by structure and these are superimposed at the edges of the figure as shown. In the centre, all active site residues are compared with the PBP2 backbone shown in white. Superimposing entire enzymes by iterative, secondary structure matching gives RMSDs (with number of superposable C! atom 27  pairs) as shown on the figure. OXA-10 superposes on PBP2 with RMSD 1.2Å (97 atom pairs) and TEM-1 on PBP2a with RMSD 1.1Å (67 pairs). PDB files used are 3equ (PBP2), TEM-1 from its complex with BLIP-I (solved in Chapter 2), OXA-10 as solved in Chapter 3, and PBP2a as solved in complex with ceftobiprole in Chapter 4.  The ubiquity and commonality of ß-lactamase enzymes is due in part to the location of their genes on plasmids and mobile genetic elements, easily shared within and among bacterial species (Weldhagen, 2004). Furthermore, treatment with antibiotics can activate the general SOS bacterial stress response, which in turn has been found to increase rates of transcriptional error as well as frequency of sharing of genetic material–both driving forces of evolution (Maiques et al., 2006). These mechanisms may have been responsible for the rapid rise and widespread distribution of the extended-spectrum ß-lactamases (ESBL), capable of hydrolyzing latter-generation ß-lactams (Livermore, 2008, Bush, 2008). While the major ESBL families include the TEM, SHV, IMP, CTX-M, KPC and OXA enzymes, the emergence of a great many others is anticipated (Naas et al., 2008). A detailed discussion of representatives of the TEM and OXA families follows. 1.3.2 Class A ß-Lactamases: TEM-1 TEM-1 was discovered in the early 1960s and was the first plasmid-borne, transposon-mediated ß-lactamase to be described. It spread rapidly among bacterial species and had been disseminated world-wide by the 1980s, to become the progenitor of a large number of ß-lactamases (reviewed in Harada et al., 2008). One of the largest ESBL groupings is now the Class A TEM ß-lactamase family from E. coli, comprising, to date,  28  some 167 inhibitor-resistant and extended-spectrum variants† reported from clinical isolates. Despite ranking among the best-studied enzymes of any description, certain aspects of the TEM-1 mechanism remain controversial. The general features of its catalyzed reaction will be familiar from the previous discussion of the TP enzymes: a serine nucleophile (Ser70), activated by a general base (uncertain; either Lys-73 or Glu-166), first attacks a carbonyl electrophilic centre, leading to the formation of an unstable tetrahedral intermediate. The collapse of this intermediate results in a stable acyl-enzyme species. A second nucleophile (a molecule of water) is then activated by a general base (Glu-166) and, via another tetrahedral intermediate, achieves deacylation of the enzyme. The key distinction between the serine ßlactamases and the TP enzymes (PBPs) is that the former uses an oxygen atom of a water molecule as the deacylating nucleophile to carry out efficient hydrolysis of ß-lactam substrates, rather than using a donor peptide to form a peptide bond. TEM-1 is further distinguished by an omega loop at its active site groove, which bears the deacylating general base Glu-166. Table 1.1 summarizes the residues and motifs which play these and various other roles in catalysis; see Fig. 1.6 for the electron-pushing reaction scheme.  †  see the curated database at www.lahey.org/Studies/temtable.asp  29  Figure 1.6: Proposed mechanisms for TEM-1 and OXA-10 ß-lactam hydrolysis. A simplified ß-lactam backbone (showing only the fused ring structure and reactive functional groups) is used for clarity. The distinguishing feature of the alternate reaction mechanisms 30  for each enzyme are emphasized in bold black font: Glu-166 vs. Lys-73 as the general base for acylation in the TEM-1 scheme, (a) and (b), respectively; the Kcx-70 vs. Lys-70 as the general base in the OXA-10 scheme, (c) and (d), respectively. Also in bold are the initial substrate (intact ß-lactam) and final product (cleaved ß-lactam). Electrons (bonded or lone pairs, plus their associated atom) involved in nucleophilic attack (as nucleophile or general base) are shown in bold blue font. Collapse of each unstable tetrahedral intermediate involves proton shuttling to the leaving group. To highlight this, hydrogen atoms and electrons (bonded or lone pairs) involved in proton shuttling are shown in crimson font, with endpoints–the initially abstracted hydrogen and acceptor electrons (bonded or lone pairs, plus the associated atom)–in bold crimson font.  For TEM-1, a large number of enzyme-inhibitor complexes have been solved in conjunction with detailed kinetic studies (Galleni et al., 1995, Matagne et al., 1998). There is thus a large evidence base from which to evaluate hypotheses concerning its catalytic mechanism. Even so, competition persists between models of the identity of the general base in the acylation step. The case for Glu-166 as this base (in a water-mediated protonshuttling mechanism; Fig. 1.6a), is supported by: lower acylation rates of Glu-166 mutants (Guillaume et al., 1997); the observation of a protonated Glu-166 in complex with a phosphonate transition-state analogue (TSA), thought to represent its state following abstraction of a proton from Ser-70 (Minasov et al., 2002); and quantum mechanical/molecular mechanical (QM/MM) simulations (Hermann et al., 2003). Alternatively, it has been argued that the ability of E166N mutants to undergo even modestly efficient rates of acylation must exclude Glu-166 as the general base (Adachi et al., 1991); the only other available general base from the crystallographic evidence is Lys-73 (Fig. 1.6b; Strynadka et al., 1992). In the end, however, further ab initio QM/MM simulations appear to support the parallel existence of both pathways in competition with each other (Meroueh et al., 2005). This is consistent with all available experimental evidence, wherein the Lys-73 general base is able to compensate to some degree for the loss of the Glu-166 31  pathway in the E166N mutants. The candidate for the general base in deacylation appears to be unambiguous: the crystal structure of the boronate TSA bound to TEM-1 clearly implicates Glu-166 (Strynadka et al., 1996b). Further insights into catalytic mechanism and residues important for binding various substrates are provided by examination of ESBL-defining residues. The putative functions and effects of substitution of amino acids at these positions are comprehensively reviewed and discussed in Page, 2008. 1.3.3 Class D ß-Lactamases: OXA-10 The prototypical Class D ß-lactamases are of the OXA family and were first distinguished by an ability to hydrolyze oxacillin (Matthew, 1979) with the enzyme PSE-2 from Pseudomonas aeruginosa (Huovinen et al., 1988). They are a common defining feature of extended-spectrum ß-lactam resistant pathogens (Hall et al., 1993). They appear to be only distantly related, with ~8% and 16% amino acid sequence identity, respectively, to the other serine ß-lactamases of classes A and C (Naas and Nordmann, 1999), although there is certainly tertiary structure resemblance to these classes (Paetzel et al., 2000). They appear most closely related to the Class C HMW PBPs (BlaR/MecR signaling PBPs; Maveyraud et al., 2000). There are only 11 known variants in the largest curated database of ESBLs‡, although there are numerous others§ not registered there. Most clinically significant aa substitutions are adjacent to the enzyme active site and thus likely affect substrate profile indirectly by influencing catalytic residues (Paetzel et al., 2000). However, there is greater  ‡ §  www.lahey.org/Studies/webt.asp#OXA one large grouping (Gasteiger et al., 2003) clusters around OXA-27, and represents a group of probable homologues that (given <30% identity to any members of the list on www.lahey.org/Studies/webt.asp) most likely are not clinical variants of the canonical group.  32  than 80% sequence variability among so called OXA enzymes, because newly-identified enzymes are frequently assigned to this group by substrate profile rather than probable sequence homology (Jacoby, 2006). The expanded repertoire of OXA-hydrolysable substrates includes aztreonam (Fig. 1.1) as well as the third-generation cephalosporins cefotaxime (Fig. 1.1) and ceftriaxone, which is identical to cefotaxime except for an aromatic, much bulkier R2 group (Naas and Nordmann, 1999). Indeed, such broad-spectrum, ß-lactamase-resistant ß-lactams are generally ineffective against pathogens producing Class D enzymes (Keith et al., 2005, Brown and Amyes, 2005). Among these are the carbapenems, often considered drugs of last resort. These are hydrolyzed, for example, by OXA-51 (Brown et al., 2005; for a discussion of the resultant clinical impact see Walsh, 2008). Of further concern is that they are also often resistant to the ß-lactamase inhibitors in clinical use: clavulanate, sulbactam, and tazobactam (Page, 2000, Payne et al., 1994, Zhou et al., 1994, Stapleton et al., 1995). Finally, the OXA enzymes also appear to be rapidly spread due to their distribution on plasmids and integrons (Lafond et al., 1989, Naas and Nordmann, 1999, Rossolini et al., 2008). In terms of structure, let the discussion focus on the family member under study in the present thesis: OXA-10. This enzyme was the first dimeric ß-lactamase identified. Dimerization was confirmed by analytical centrifugation and found to be dependent upon concentration and the presence of EDTA (Maveyraud et al., 2000). Certain other family members, OXA-1 for example, are monomeric (Sun et al., 2001). Each OXA-10 subunit is comprised of two domains: a ß-domain (7-stranded antiparallel sheet, plus the 2 terminal alpha helices distally packed), and an !-domain (6 helices), with the active site located at the domains' interface. A divalent metal cation has been observed to bridge the OXA-10 33  monomers in several structures (Paetzel et al., 2000, Danel et al., 2001), and so is thought to explain EDTA-inhibition of dimerization. Conversely, other investigators (including Maveyraud et al., 2000) do not observe this cation at the dimerization interface (which is extensive, >1,000 Å2). These divergent reports may nevertheless be mutually consistent since the latter studies report a higher value for the dimerization dissociation constant, Kd ~1!M, on the same order observed by Paetzel et al (Kd ~1-20!M) in the absence of added cations. Another dimeric OXA family member, the variant OXA-29, is probably of distinct origin. In addition to a high degree of sequence variance (it shares not even 50% aa identity with any other OXA), its dimeric structure is stable under dilution as well as in the presence of EDTA (Franceschini et al., 2001). See also Helfand and Bonomo, 2003 for a concise review of the canonical OXA family. The class D ß-lactamase mechanism is considered to be similar to that of the other serine hydrolases discussed here (Golemi et al., 2000); the Ser-67 nucleophile, for instance, has long been known from labeling studies (Ledent et al., 1993). The exceptional feature of OXA-10 catalysis is the carbamylated active site lysine residue (Kcx-70), which is rarely observed in protein structures, particularly with a catalytic role. The list of other known enzymes featuring such a functional (as opposed to structural) Kcx residue includes the critical photosynthesis enzyme RuBisCO, in which it acts to initiate the CO2-fixing reaction via enolization (Kannappan and Gready, 2008). Others include phosphodiesterase (Benning et al., 1995), urease (Jabri et al., 1995), and the BlaR/MecR ß-lactam-sensing domain (Wilke et al., 2004). It is thought that protein carbamylation may more often be non-specific and is implicated in various human disease states (Wang et al., 2007b). The Kcx-70 of OXA-10 and of other Class D ß-lactamases is believed to serve as the 34  general base in both acylation and deacylation steps. This is the most dramatic difference between the Class D and Class A enzymes, which otherwise share close topological resemblance. In addition, the two classes share the invariant residues Ser-67/-70 and Kcx70/Lys-73 (motif I), Ser-115/-130 (motif II), Lys-205/-234, and Gly-207/-236 with a RMSD for these residues ~0.5 Å (Golemi et al., 2000). The oxyanion hole is also formed by the backbone N atoms of residues (Ser-67, Phe-208) aligning with those of the Class A enzymes (Ser-70, Ala-237). Table 1.1 catalogues these residues and others according to their proposed roles, with reference to the experimental evidence available. The acylation mechanism employing Kcx-70 as the general base, suggested by several crystal structures and supported by a 300ps molecular dynamics simulation (Maveyraud et al., 2002) is illustrated in Fig. 1.6c. The region in OXA-10 otherwise occupied by the TEM-1 omega loop (bearing the general base Glu-166) is hydrophobic (Val-117, Trp-154, Leu-155) and thus excludes the possibility of a general base candidate analogous to Glu-166 at this location. The Kcx-70 carboxylate is stabilized by Ser-67-O", the Trp-154 side-chain N, and a buried water molecule. The physiological relevance of the Kcx is supported by activity dependence on HCO3 concentration at pH >7; the observation of saturation kinetics behaviour with respect to the addition of HCO3; the !M HCO3 dissociation constant (Kd), given that the cytoplasmic concentration of dissolved CO2 in bacteria has been estimated to be ~1.3 mM (Tien et al., 1999); finally, a decrease in enzyme activity (and loss of biphasic kinetics, discussed below) is observed in OXA-17 bearing the Asn-73-Ser mutation–Asn73 coordinates a water molecule that is positioned to stabilize one of the Kcx-70 carbamylate oxygen atoms. The function of Kcx-70 as the general base in the deacylation step is consistent with the inhibition of OXA-10 by 6!-substituted penicillinate derivatives, which 35  would sterically block the approach of the Kcx-70-bound hydrolytic water molecule (Golemi et al., 2000). Complicating this well-established view of the OXA-10 mechanism is the possibility that the non-carbamylated Lys-70 may also act as the general base (Fig. 1.6d). Its deprotonated state (and hence its general base readiness) is indeed favoured by the hydrophobic pocket it occupies. Biphasic enzyme kinetics are observed for OXA-10: a fast initial hydrolytic step, followed by a slow steady state (Danel et al., 1998). This is likely not due to cooperativity between dimers, since such kinetics are observed for the monomer (i.e. at enzyme concentrations far below that of the dissociation constant of dimerization). Thus a loss of catalytic efficiency due to the uncarbamylated Lys-70 may explain the slow step of the biphasic kinetics. The existence of the biphasic kinetics is likely related to the carbamylation of Lys-70, due to concentration-dependent behaviour observed upon supplementation with 50mM NaHCO3 at pH 7.0. The detailed picture of this phenomenon is complex, however; the hydrolysis of certain substrates exhibits biphasic kinetics in the absence of additional bicarbonate (Golemi et al., 2001). Indeed, enzyme activity, judged both by assay in solution and acylation observed in crystal structures, appears to be strongly pH dependent (OXA-10 has no activity and an active site in disarray at pH 6.0, full activity at pH 8.0, and moderate activity at intermediate pH values); this is attributed to Lys-70 carbamylation, which increases with rising pH (Golemi et al., 2001). In addition, loss of activity in degassed buffer may be restored by the addition of bicarbonate. From this perspective, Lys-70 facilitates catalysis more or less efficiently depending on its degree of covalent modification. It should be noted that the 36  structures of the related ß-lactamases OXA-1 (PDB ID: 1m6k) and OXA-2 (1k38) have been solved and both feature carbamylated Kcx-70 residues. Interestingly, although OXA-10 is resistant to clinical inhibitors, it appears to be readily inactivated by NaCl (Naas and Nordmann, 1999), perhaps by facilitating the decarbamylation of Kcx-70 (Maveyraud et al., 2000). This suggestion is supported by a series of crystal structures determined at increasing salt concentrations (Sun et al., 2001). Although OXA-13 is most likely related to OXA-10, early studies of this enzyme made no reference to carbamylation of Lys-70 (Pernot et al., 2001) and this residue is found to be unmodified in all observations of its structure (PDB IDs: 1h8y, 1h8y, 1h5x). Likewise, the more distantly related OXA-24 (2cj7) is uncarbamylated (Santillana et al., 2007). Structures of other variants are unfortunately not yet available, and others (OXA-48, OXA85) have been withdrawn from the PDB (for reasons not given). In any case, Lys-70 certainly plays a critical role in catalysis, as the K70A mutant is inactive (Golemi et al., 2001). The broad substrate specificity of OXA-10 is likely based in part on its open active site, where Ser-115 (motif II) is found on a flexible loop. Ser-115 may be a proton relay group, given the similarity of its position to that of the Class A proton relay group Ser-130 (Maveyraud et al., 2000; see also Fig. 1.5). One further distinguishing feature among the various ß-lactamase classes is the direction of approach of the hydrolytic water, either from the ! or ß face of the ß-lactam. This has been probed via the use of inhibitor sets diastereomeric at the R substituent (reviewed in Golemi et al., 2000). Both Class A and Class D ß-lactamases are inhibited by the !-substituted inhibitor, indicating the approach of the deacylating water from the ! side (i.e. from behind the page in Fig. 1.6). This confirms that the Class D ß-lactamases are truly 37  distinct: they lack a general base of the Glu-166 type associated with the Class A enzymes, yet mechanistically differ from the Class C enzymes, which exhibit a ß-face approach of the hydrolytic water (Golemi et al., 2000). 1.3.4 Overcoming Resistance: ß-Lactamase Inhibitors A variety of ß-lactamase inhibitors are known, yet there are only three in clinical use: the clavams (clavulanate) and the penam sulfones (sulbactam and tazobactam), depicted in Fig. 1.7. These are poorly-hydrolyzed structural analogues of ß-lactams, which act via mechanism-based inhibition. Clavulanate is first cleaved in the typical fashion by the ßlactamase serine nucleophile to form an acyl enzyme intermediate. Spontaneous opening of the five-membered oxazolidine ring occurs afterward to form, via branched reaction pathways, various interconvertible iniminum, enamine, and aldehyde intermediates (Therrien and Levesque, 2000). The net effect is the prolonged occupation of the active site, as well as possible irreversible inactivation by crosslinking at Ser-130.  Figure 1.7: Clinical ß-lactamase inhibitors, with ß-lactam core structure emphasized in bold. The inhibitor behaviour at the active site is thus complex: a variety of major and minor intermediates have been described that are of uncertain relative importance, with often poorly-defined conformations (discussed in Totir et al., 2006). Crystallographic and 38  spectroscopic studies have concluded that all three inhibitors act in similar fashion (for clavulanate and sulbactam, refer to Padayatti et al., 2005; for tazobactam see Padayatti et al., 2006). Generally lacking antibiotic potency themselves, these inhibitors are administered clinically in combination with an antibiotic to enhance its activity, for example amoxicillinclavulanate; a few other combination therapies are in use and several others are under investigation (Livermore et al., 2008). The clinical impact of these inhibitors has been tremendous; for a review of this, refer to Akova, 2008. In addition to inhibition by small molecules, ß-lactamases are also known to be susceptible to macromolecular inhibitors. Clavulanate was initially isolated from Streptomyces clavuligerus, a producer of related clavam metabolites as well as penicillin and cephamycin antibiotics (Doran et al., 1990). In a serendipitous assay of S. clavuligerus culture filtrate, high levels of ß-lactamase inhibition were noted, yet the low concentration of clavulanate could account for only 1% of this. Heating the sample to 98ºC or treating it with methanol abolished the additional inhibition, suggesting that the unidentified inhibitory factor was proteinaceous. An 18 kDa ß-lactamase inhibitor protein (BLIP) was subsequently isolated from the growth media. ß-lactamase inhibitory activity was likewise detected in Streptomyces exfoliatus, a related but non-ß-lactam-producing species. Two distinct BLIPs (BLIP-I and BLIP-II) were isolated and characterized (Kim and Lee, 1994, Park and Lee, 1998, Kang et al., 2000). BLIP-I shows 38% identity to BLIP, indicating a very strong likelihood of common evolutionary origin (Altschul et al., 1990) as well as probable structural and functional similarity. BLIP-II, on the other hand, is distinct from the other BLIPs described thus far: it lacks the cysteine residues that typify the other BLIPs and indeed appears to be unrelated to BLIP 39  or BLIP-I. Comparison with other proteins in public sequence and structure databases yields only proteins of unknown function or those associated with chromosome condensation in eukaryotes. BLIP-II features seven tandem repeats of a short ~40 aa domain (Fig. 1.8a). All BLIPs show 1:1, reversible, non-covalent inhibition of ß-lactamases (Fig.1.8b). Given the focus of this thesis, further discussion will be limited to BLIP and BLIP-I.  Figure 1.8: BLIP-II (a) drawn as a cartoon inside its molecular surface, viewed facing its concave binding surface and showing each of its seven tandemly-repeated domains in a different colour; (b) drawn in blue, with side chains Asn-50 and Asp-52 drawn as white stick models, in its inhibitory complex with TEM-1 ß-lactamase (yellow surface). BLIP has been solved both alone (Strynadka et al., 1994) and in complex with TEM-1 ßlactamase (Fig. 1.9; Strynadka et al., 1996a). This particular complex has become widely used as a model system for the study of protein-protein interactions (Reichmann et al., 2008, Wang et al., 2009, Harel et al., 2007, Kozer et al., 2007, Wang et al., 2007a, Reynolds et al., 2006, Joughin et al., 2005, Kozer and Schreiber, 2004, Zhang and Palzkill, 2004, Rudgers et al., 2001, Selzer and Schreiber, 2001, Albeck et al., 2000, Huang et al., 2000). 40  Figure 1.9: BLIP (a) drawn as a cartoon inside its molecular surface, viewed facing its concave binding surface and exhibiting a two-domain, tandemly-repeated structure with key residues labeled. Domain 1 is rendered in blue (also showing a stick model of side chain Lys74), ß-hairpin Ala-46–Tyr-51 (showing Asp-49) in green, domain 2 in pink, extended ß-loop (showing Phe-142) in purple; (b) coloured as in (a), in its inhibitory complex with TEM-1 ßlactamase, drawn in yellow as a van der Waals surface. Shown in red as side-chain stick models are TEM-1 active site residues Ser-130, Lys-234, Ser-235, Arg-243, and Tyr-105.  BLIP is a 165 aa protein comprised of two homologous, 76 aa tandemly repeated segments suggesting evolution by gene duplication and fusion (Doran et al., 1990). It adopts an overall flat, mildly concave conformation, with dual helix-loop-helix motifs packed behind an extended 8-stranded ß-sheet to form a slightly concave face (Fig. 1.9a; Strynadka et al., 1994). Inhibition arises from three principal features of the BLIP•ß-lactamase interaction (Fig. 1.9b). First, the concavity and great size of the binding surface (2636Å2 buried at the BLIP•TEM-1 intermolecular interface; Strynadka et al., 1996a) gives rise to high-affinity binding: inhibitory constants (Ki) were determined to be 0.11-0.6 nM (Table 1.3). Second, a layer of water molecules at the interface mediates numerous hydrogen bond contacts (Strynadka et al., 1996a), making these contacts flexible and adaptable; it has been 41  argued that this ‘water adapter’ (which allows tolerance of mutations of either BLIP or TEM-1) is largely responsible for the ability of BLIP to bind and inhibit a broad array of ßlactamase enzymes (Zhang and Palzkill, 2003). Lastly, two loops insert into the active site pocket of TEM-1, making numerous favourable contacts there (Fig. 1.10a; Rudgers and Palzkill, 1999). Most notably, BLIP residue D49 forms hydrogen bonds with four critical TEM-1 active site residues, involved in the catalytic mechanism or in substrate binding: Ser130, Lys-234, Ser-235, and Arg-243 (Strynadka et al., 1996a). Also, TEM-1-Tyr-105, otherwise involved in aromatic ring stacking with a bound ß-lactam thiazolidine ring, is in the complex rotated 90 degrees and stacked between BLIP-Phe-142 and -Lys-74. The analogous domains also show an elegant form of tandem functionality. In the inhibitory complex, both the ß-hairpin A46-Y51 from BLIP domain 1 (specifically, Asp-49) and the analogous extended ß-loop from domain 2 (Phe-142) insert into the TEM-1 active site and mimic the carboxylate and benzyl groups, respectively, of bound penicillin-G (Fig. 1.10b; Petrosino et al., 1999).  42  Table 1.3: Inhibition Constants (Ki) for BLIPs with Various Class A B-lactamases. BLIP BLIP BLIP  !-lactamase Proteus vulgaris 1028 E. coli TEM-3  !-lactam nitrocefin nitrocefin  BLIP  E. coli TEM-1  BLIP  E. coli TEM-1  BLIP  E. coli TEM-1  BLIP  Klebsiella oxytoca 1082  BLIP  Serratia marcescens SME-1  BLIP  Bacillus anthracis Bla1  BLIP BLIP  Pseudomonas aeruginosa PSE-1 B. cereus ATCC1-27348  BLIP  Klebsiella pneumoniae SHV-1  BLIP BLIP BLIP BLIP BLIP BLIP  S. aureus 853 B. licheniformis 749/C Streptomyces albus G Actinomadura R39 S. aureus 157 Lysobacter enzymogenes  BLIP  B. cereus NCIB 8933  nitrocefin  Ki (nM) Rxn Reference 0.018 W (Strynadka et al., 1994) 0.3 W (Strynadka et al., 1994) (Petrosino et al., cephaloridine 0.11 X 1999) cephalosporin (Zhang and Palzkill, 0.5 Y C 2003) (Strynadka et al., nitrocefin 0.6 W 1994, Strynadka et al., 1996a) (Strynadka et al., nitrocefin 1.0 W 1994) (Zhang and Palzkill, cephalosporin C 2.4 Y 2003) (Zhang and Palzkill, nitrocefin 2.5 Y 2004) nitrocefin 3 W (Strynadka et al., 1994) nitrocefin 1,000 W (Strynadka et al., 1994) (Zhang and Palzkill, cephalosporin C 1,100 Y 2004) 3 nitrocefin 3 x 10 W (Strynadka et al., 1994) 3 nitrocefin 3 x 10 W (Strynadka et al., 1994) 3 nitrocefin 3 x 10 W (Strynadka et al., 1994) 3 nitrocefin 11 x 10 W (Strynadka et al., 1994) nitrocefin +* W (Strynadka et al., 1994) nitrocefin -* W (Strynadka et al., 1994) no effect W  (Strynadka et al., 1994)  BLIPI BLIP-I BLIP-I BLIP-I BLIP-I BLIP-I BLIP-I  E. coli TEM-1  cephaloridine  0.047  X  (Kang et al., 2000)  Bactopenase B. subtilis Type I B. subtilis Type II E. cloacae Type III E. cloacae Type IV B. cereus penicillinase  penicillin G cephaloridine cephaloridine cephaloridine cephaloridine penicillin G  0.062 70 x103 70x103 >106 >106 no effect  Z X X X X Z  (Kim and Lee, 1994) (Kang et al., 2000) (Kang et al., 2000) (Kang et al., 2000) (Kang et al., 2000) (Kim and Lee, 1994)  BLIP-I  E. cloacae cephalosporinase  penicillin G  no effect  Z  (Kim and Lee, 1994)  * “+” or “-“: activity respectively enhanced or decreased and Ki not determined; W50mM phosphate pH 7.0 30°C, preincubated 1h; X50mM phosphate pH7.0 25°C, preincubated 2h +1mg/ml BSA; Y50mM phosphate pH7.0 25°C, preincubated "1h +1mg/ml BSA; Z100mM Phosphate pH 7.0 30°C  43  Figure 1.10: Inhibition of ß-lactamase TEM-1: (a) close view of BLIP loops inserted into the TEM-1 active site. BLIP Asp-49 (green) interacts via hydrogen bonds with TEM-1 residues (shown in red) Ser-130, Ser-235, and Arg-243, and with Lys-234 via a salt bridge (dashed white lines); BLIP Phe-142 (purple) and Lys-74 (blue) interacts with TEM-1 Tyr-105 (red); (b) TEM-1 ß-lactamase (coloured as above) with bound substrate penicillin-G (penG). To emphasize comparison with (a) the penicillin-G carboxylate functional group is drawn in green (mimicked by BLIP Asp-49) and its phenyl group is drawn in purple (mimicked by BLIP Phe-142). In the TEM-1 ß-lactamase active site with bound penicillin-G, its carboxylate forms hydrogen bonds (dashed white lines) with TEM-1 residues Lys-234, Ser235 and Arg-243, and its 5-membered penem ring stacks with TEM-1-Tyr-105. BLIP resembles clavulanic acid and other small-molecule inhibitors to the extent that it is a weak inhibitor of PBPs; BLIP binds Enterococcus faecalis PBP 5 with a Ki of 12 !M but otherwise fails to inhibit two of the best-studied exocellular cell wall biosynthetic enzymes produced by actinomycetes (Streptomyces R61 or Actinomadura R39 exocellular carboxypeptidases/transpeptidases; Strynadka et al., 1994). Table 1.3 summarizes several studies that have reported the Ki for BLIP and BLIP-I inhibition of various ß-lactamases versus a given ß-lactam. All tested Class C enzymes (cephalosporinases) are either unaffected or stimulated by the presence of either BLIP or BLIP-I. Likewise, no Class B ß-lactamases (MBLs) have been reportedly inhibited. However, all Class A enzymes tested were strongly 44  inhibited by BLIP with !M to nM and sub-nM Ki, with the exception of S. aureus 157, B. cereus NCBI 8933 and Salmonella typhimurium OXA-2. This variation in the degree of BLIP inhibition of distinct ß-lactamases is somewhat puzzling given their highly similar fold. The most likely explanation is that specific BLIP surface features either accommodate or clash with corresponding features on the inhibited enzyme (Zhang and Palzkill, 2003, Zhang and Palzkill, 2004). From an extensive study comparing the effects of dozens of BLIP binding surface mutations on its ability to inhibit two different ß-lactamases (TEM-1 and SME-1) it was first shown that an aromatic patch on each BLIP domain largely confers specificity (Zhang and Palzkill, 2003). A subsequent study, again employing alanine-scanning mutagenesis, identified distinct but overlapping functional epitopes of BLIP that bind to four different ß-lactamases from various bacterial species: TEM-1 (Escherichia coli), SHV-1 (Klebsiella pneumoniae), SME-1 (Serratia marcescens) and Bla1 (Bacillus anthracis) (Zhang and Palzkill, 2004). Of interest is the identification of residues that explain why BLIP inhibits SHV-1 (Kuzin et al., 1999) far more weakly (Ki = 1.1 !M) than it does TEM-1 (Ki = 0.5 nM), despite the fact that the two ß-lactamases have 68% amino acid identity and themselves hydrolyze a similar spectrum of substrates. The authors also report the key residues that permit BLIP to strongly inhibit the divergent ß-lactamases SME-1 and Bla1, which have no more than 30% identity with TEM-1. Six clusters of surface residues that act in a modular, independent, and additive fashion to bind TEM-1 have been identified (Reichmann et al., 2007). Additional features of BLIP that affect its binding affinity for TEM-1 have been identified beyond the binding interface (Selzer and Schreiber, 2001, Joughin et al., 2005). These act via ‘long-range’ charge-charge interactions; certain charged residues exert 45  electrostatic attractive forces on one another to promote the properly-oriented association of otherwise randomly-diffusing proteins. Since such charge properties are relatively independent of protein conformation, they may help to explain the ability of BLIP to selectively inhibit structurally similar ß-lactamases. BLIP-I has been far less studied than BLIP. However, there is experimental evidence for at least one shared functional epitope between the two proteins: the mutation D49A in BLIP-I increases the Ki for TEM-1 from ~0.5 nm to 10 nM, diminishing binding affinity to the near-identical value for the corresponding BLIP D49A mutant (Kang et al., 2000). Otherwise, BLIP-I appears to be quite specific for inhibiting TEM-1 and Bactopenase (a commercial preparation of B. cereus ß-lactamase); it has little effect on the limited number of other ß-lactamases tested (Table 1.3). Efforts to circumvent ß-lactamase activity in the clinic have thus far taken two major approaches. The first is to develop ß-lactams that are resistant to ß-lactamase hydrolysis. The alternative approach involves co-administering small-molecule inhibitors of ß-lactamase along with traditional ß-lactams. Both of these approaches, however, have been circumvented by mutants of the widespread and clinically important ß-lactamase TEM-1 (Schroeder et al., 2002). Specific mutations in this enzyme allow cleavage of extendedspectrum (i.e. ß-lactamase resistant) cephalosporins (including the aminothiazole oxime ßlactam derivatives ceftazidime, cefotaxime, and aztreonam; Arg-164-Ser) and confer resistance to inhibitors like clavulanic acid (Arg-244-Ser). For a time, one study raised the spectre of a double mutant (Arg-164-Ser/Arg-244-Ser) that gains both advantages (Imtiaz et al., 1994). While it had been suggested that TEM enzymes may be incapable of dual promiscuity and resistance (Schroeder et al., 2002), a clinical isolate later yielded TEM-121, 46  which exhibits dual Arg-164-Ser/Arg-244-Ser mutations and, as expected, the corresponding extended hydrolysis profile and inhibitor resistance (Poirel et al., 2004). Furthermore, the Ser-130-Gly TEM mutant also appears able to resist combined extended-spectrum/inhibitor antibiotic therapies (Helfand et al., 2007). The BLIP mechanism of ß-lactamase inhibition is distinct from that of the clinical small-molecule inhibitors. Because inhibitor-resistant mutations to the common ß-lactamase TEM-1 confer no resistance to BLIP, it is hoped that BLIP derivatives may one day be used in the clinic. Presently, several such synthetic peptides are known to bind and inhibit TEM-1, although with affinities still several of orders of magnitude too poor for prospective use.  47  1.4  ß-Lactam Resistant PBPs  1.4.1 MRSA PBP2a PBP2a, also called PBP2', is the protein responsible for the broad-spectrum ß-lactam resistance exhibited by methicillin-resistant Staphylococcus aureus (MRSA; Brown and Reynolds, 1980, de Lencastre et al., 1994). It is able to maintain crosslinking of the PG layer in staphylococcal cells even when all other essential PBPs are inhibited by ß-lactams. Thus PBP2a and the GT function of PBP2 together provide for the synthesis of the cell wall under these conditions, with perhaps some supplemental crosslinking activity by other PBPs, e.g. PBP4, which is essential for ß-lactam resistance in certain community-acquired MRSA strains (Memmi et al., 2008). Crosslinking is not unaffected, however. Survival in the presence of ß-lactam antibiotics is accompanied by a change in the chemical composition of the peptidoglycan sacculus (Severin et al., 1995). Like the ß-lactamases, the gene encoding PBP2a appears to have spread horizontally among bacteria. It is highly unlikely to have originated by gene duplication and modification of any native S. aureus PBP, having <21% amino acid identity to these and is presently thought to have originated in another species of the same genus, S. sciuri (Fuda et al., 2007). PBP2a is structurally most closely related† to Streptococcus pneumoniae PBP1b (PDB IDs: 2bg1, 2uwx), PBP1a (2c5w), PBP2x (1qme, 1pyy, 1rp5, 1k25, 1pmd;). The DALI‡ server (Sinha Roy et al., 2001) adds the following structural near neighbours: Neisseria gonorrhoeae PBP2 (3equ, 3eqv), S. pneumoniae PBP2b (2wad, 2wae, 2waf) and PBP2 (2olu, 2olv), S. aureus PBP2 (3dwk) and BlaR1 (1xkz, 1xa1), as well as the Class D ß-lactamases OXA-1 (1m6k), OXA-24  † ‡  According to PHYRE Protein Fold Recognition Server (www.sbg.bio.ic.ac.uk/phyre; Bennett-Lovsey et al., 2008) http://ekhidna.biocenter.helsinki.fi/dali_server  48  (2jc7), OXA-10 (1k55), Class A ß-lactamases TEM-1 (1fqg), SHV-1 (1vm1), CTX-M-9 (1yly) and KPC-2 (2ods). The PBP2a resistance DNA cassette SCCmec (Staphylococcal Cassette Chromosomal mec), 32-50 kbp in size and bearing the PBP2a-encoding gene mecA in addition to its regulon (Zhang et al., 2001), has 5 major variants, differing in size and regulatory genes (Robinson and Enright, 2004). It appears, however, to be a poorly mobile genetic element. It is inserted consistently at a certain point in the S. aureus genome (Katayama et al., 2000) and few discrete MRSA lineages are known to exist, suggesting that gene transfer may be exceedingly rare. SCCmec appears, however, to have been transferred relatively unaltered to coagulase negative Staphylococci. It has been suggested that a limited number of staphylococcal strains are of compatible genetic background with the mecA gene product (Katayama et al., 2005). Although possession of PBP2a in S. aureus is widely considered synonymous with the MRSA phenotype, there is great variability in resistance levels among different strains (oxacillin MICs can range from !g/ml to mg/ml levels) and degree of homogeneity within cultures (from total resistance to as few as 1:107 resistant to susceptible colonies; Ikonomidis et al., 2008 and references therein). So-called borderline-resistant S. aureus (BORSA) may arise in the absence of mecA by variation and overexpression of native PBPs, which may achieve MICs of up to 16 !g/ml (Tomasz et al., 1989, Hiramatsu, 1992, Suzuki et al., 1992, Chambers et al., 1994, Hackbarth et al., 1995, Henze and Berger-Bächi, 1995). The formal requirements for positive identification of MRSA are thus both the possession of the mecA gene and an oxacillin* MIC " 4 !g/ml (CLSI, 2007).  *  methicillin is no longer commercially available in the United States  49  It is clear that PBP2a is a formidable resistance determinant: MIC values of 2000 !g/ml for MRSA have been reported (Kondo et al., 2001). In addition, SCCmec frequently carries one or more additional resistance genes and so may be functionally similar to DNA cassettes called “pathogenicity islands”, multidrug-resistant cassettes common in Gram negative bacteria (Ito and Hiramatsu, 1998, Hiramatsu et al., 2001). Of further concern is that, for more than a decade, MRSA strains have been described with decreased susceptibility (MIC = 8-16 !g/ml) and even full resistance (MIC " 32 !g/ml) to vancomycin, the traditional MRSA drug of choice§. The mechanism by which PBP2a may be able to maintain the crosslinking of the PG layer in the presence of ß-lactam antibiotics was explained by the determination of its crystal structure alone and in complex with several antibiotics, which were able to acylate a soluble PBP2a construct following prolonged exposure to high antibiotic concentrations (Lim and Strynadka, 2002). The overall fold and topology of its TP domain resembles the corresponding domain from other PBPs and serine ß-lactamases, especially but not exclusively Class Ds. However, three-dimensional superposition of the PBP2a transpeptidase domain reveals key differences: only ~10 to 20% of amino acids are identical when superposed and substantial RMSD values are measured in comparison with E. coli PBP5 (3.0 Å for 198 Ca pairs; Nicola et al., 2005), Streptomyces R61 DD-carboxypeptidase (2.9 Å for 219 Ca pairs; Kelly et al., 1985), Streptomyces K15 DD-transpeptidase (2.7 Å for 203 Ca pairs; Fonzé et al., 1999), and S. pneumoniae R6 PBP2x (2.2 Å for 288 Ca pairs; Gordon et al., 2000). Despite this variation, it is clear that motifs I-III are strongly conserved, and it is presumed  §  www.cdc.gov/ncidod/dhqp/ar_lab_mrsa.html#3  50  that the catalytic mechanism is generally similar to that of susceptible PBPs (Lim and Strynadka, 2002); see Table 1.1 and Fig 1.4a. How then, with essentially the same catalytic machinery, is PBP2a so resistant to inactivation by ß-lactams? Recall that ß-lactam antibiotics inactivate bacterial cell wall transpeptidases (PBPs) by mimicking their natural substrate. According to Equation 1.1: K  k  k  d 2 3 PBP + "-lactam #% % $ PBP • "-lactam %% $ PBP&"-lactam % % % $ PBP + "-lactam H 2O (hydrolyzed) [Michaelis complex] [acyl - enzyme]  Equation 1.1 !  the ß-lactam initially binds a PBP non-covalently to form (with dissociation constant Kd) the Michaelis complex. The PBP serine nucleophile then attacks the ß-lactam peptide bond to create the covalent acyl-enzyme complex (at the rate indicated by the constant k2). Nucleophilic attack by water would regenerate the enzyme; however, ß-lactam antibiotics are hydrolyzed extremely slowly by PBPs (reflected by low k3 values; Ghuysen et al., 1986, Frere et al., 1975) so in effect the acyl-PBP is now irreversibly inhibited (Fig 1.4b). PBP2a from Staphylococcus aureus is believed to resist such inactivation via a highly inefficient acylation step (its k2 is three orders of magnitude lower than ß-lactam sensitive PBPs; Lim and Strynadka, 2002). Meanwhile, PBP2a exhibits ß-lactam binding affinity (Kd) and deacylation efficiency (k3) values comparable to sensitive PBPs (Lu et al., 1999, GravesWoodward and Pratt, 1998). Thus, a ß-lactam with improved binding affinity (decreased Kd) would result in increased inhibition resulting from greater overall acylation efficiency (k2/Kd); in other words, the low probability that the Michaelis complex will yield an acylation event would be compensated for by increasing the frequency with which the Michaelis complex forms. This is thought to be the mechanism for increased acylation rates 51  observed for the cephalosporins nitrocefin and compound 1 (Lim and Strynadka, 2002, Lu et al., 1999). The narrow, extended, largely hydrophobic active site groove presumably places an additional restriction on binding by certain ß-lactams (note the bulky R1 group of methicillin in Fig. 1.1, for example). The remarkable ability of mecA to alone confer highlevel ß-lactam resistance represents an Achilles’ heel for its host organism: little evolutionary pressure remains to select for resistance by mutation of its endogenous PBPs. Thus drug development efforts against MRSA, for example, could succeed by striking a lone target (PBP2a). The extraordinary success of clinical trials for ceftobiprole, a next-generation ßlactam targeting PBP2a of MRSA and other multi-drug resistant bacteria, seems to reflect this (Jones, 2007). Recently, a whole-systems approach to the broad-spectrum ß-lactam resistance mechanism of MRSA has been explored in two independent computational modeling studies (Pienaar et al., 2008, Murphy et al., 2008) . The biological picture of broad-spectrum resistance involving PBP2a is complex and likely involve a number of other genes, for example hmrA and hmrB, which are able to confer very high levels of methicillin resistance (MIC 64-512 !g/ml) to MRSA while paradoxically rendering it susceptible to low-level exposure (#8 !g/ml) to the antibiotic. This has been termed called the “Eagle effect” and is discussed at length in Kondo et al., 2001. The clinical impact of PBP2a has been enormously detrimental. PBP2a is a highly effective single resistance determinant with apparently low genetic promiscuity. However, together these qualities appear to make it vulnerable to any antibiotic, a ß-lactam for instance, able to target it. This will be discussed in section 1.4.3, below.  52  1.4.2 S. pneumoniae PBP2x An alternative, in nearly every respect, to the resistant PBP in MRSA is exemplified by PBP2x from S. pneumoniae (strain 328). S. aureus PBP2a features a relatively consistent single variation on the TP theme (>90% amino acid sequence identity between MRSA strains; Ryffel et al., 1990) that uses global conformational changes to modulate acylation efficiency (k2, in Equation 1.1, above). In contrast, PBP2x is encoded by a highly heterogeneous group of mosaic genes, i.e. continuous blocks of DNA exchanged by homologous recombination, which confer ß-lactam resistance in the enzyme via reduction in binding affinity (increase in Kd for ß-lactams, in Equation 1.1). Interestingly, PBP2x may also be capable of employing the k2-based resistance strategy. A study of 89 strains from public and clinical databases uncovered a second group of PBP2x enzymes, 5259-PBP2x, that appears to share many of the same features of PBP2a (a closed active site, similar rearrangement of ß3 upon acylation) and thus perhaps a similar resistance mechanism (Pernot et al., 2004). 1.4.3 Overcoming Resistance: Ceftobiprole, a Next-Generation ß-Lactam As mentioned in 1.2.3, the isolation of the chemical cores of penicillin and cephalosporin C (6-APA and 7-ACA) permitted a great expansion of the storehouse of known antibiotics, by their substitution, respectively, at the lone variable R position or dual R1 and R2 groups (Fig. 1.1). Methicillin, for example, was put to use in the 1940s to overcome resistance by ß-lactamase-producing S. aureus strains. ß-lactamases are now known to be capable of hydrolyzing penicillins, cephalosporins, carbapenems, and monobactams (Galleni et al., 1995). Within two years of the introduction of methicillin, the first MRSA strains were reported. At present, MRSA exceeds 60% prevalence in some United States 53  hospitals (Doshi et al., 2009) and has become established as a community-acquired pathogen (Miller and Kaplan, 2009). Infection-control measures on hospitals wards have thus far been insufficient to contain the problem. Clearly new antimicrobial agents are required. Ceftobiprole is a clinically advanced (post-phase III), next-generation ß-lactam with anti-MRSA activity approved in Canada and Switzerland for complicated skin structure and soft-tissue infections and is presently awaiting approval by the United States Food and Drug Administration (Boucher et al., 2009). Its in vitro activity profile is impressive: it has been shown to be superior (with an MIC = 2 !g/ml) to linezolid and vancomycin in treating MRSA. Its activity is equivalent to ceftazidime and combined piperacillin-tazobactam in treating penicillin-resistant S. pneumoniae. Furthermore, it exhibits potential for empirical clinical use (i.e. prior to positive identification of the infectious organism) with a clinicallyrealistic MIC (of #4 !g/ml) against >90% of Enterobacter spp. tested and >50% of Acinetobacter spp. and P. aeruginosa tested (Jones, 2007). Activity against Acinetobacter spp. may prove to be of future clinical importance, since similar multi-drug-resistance risks are believed to be posed by these prevalent bacterial colonizers, which can become pathogenic to immune-compromised individuals (Giamarellou et al., 2008). Refer also to Bogdanovich et al., 2005 for extensive efficacy data (including MIC50, MIC90, and time-kill analysis data). In a set of double-blind, randomized clinical trials, ceftobiprole (administered alone, intravenously as the ester prodrug ceftobiprole medocaril) demonstrated bactericidal activity against MRSA, penicillin-resistant S. pneumoniae, and a range of other Gram-positive and Gram-negative species. In addition, ceftobiprole appears to be as safe and well tolerated as cephalosporins in general (Noel et al., 2008a, Noel et al., 2008b). Ceftobiprole is a cephalosporin with an R1 moiety consisting of an oxyimino 54  aminothiazolyl group (thought to be responsible for its stability against ß-lactamases) and a vinylpyrrolidinone group at the R2 position (believed to promote association with PBP2a and facilitate acylation; Livermore, 2006). See Fig. 1.11 for its structure and striking resemblance to the staphylococcal cell-wall stem peptides. Its PBP2a binding constant is indeed quite favourable (Kd = 0.87 !M) compared with ceftriaxone (Kd = 115 !M), imipenem (Kd > 0.5 mM), and methicillin (Kd > 0.5 mM; Hebeisen et al., 2001). Ceftobiprole also has strong affinity (IC50 = 1 !g/ml; corresponding to a Ki = 1.9 !M) for S. pneumoniae PBP2x, although it is ineffective against Enterococcus faecium PBP5 (Davies et al., 2006).  Figure 1.11: The chemical structure of ceftobiprole, highlighting its remarkable similarities to the PG stem peptide donor terminal DAla-D-Ala (red) and acceptor (blue); parts that overlap are shown in purple.  It is limited in its usefulness against metallo-ß-lactamase- and certain ESBLproducing pathogens, since it is generally hydrolyzed by these enzymes. Importantly, however, ceftobiprole is stable against the typically narrow-spectrum penicillinases produced by staphylococci, as well as TEM-1 derivatives, and appears relatively stable against 55  hydrolysis by AmpC ß-lactamases, likely in part by poorly inducing the production of these enzymes (Livermore, 2006). Ceftobiprole bears many hallmarks of a good anti-MRSA drug: it binds and inhibits the resistant target; binds and inhibits susceptible PBPs, ensuring bactericidal activity; it is stable against many ß-lactamases or induces their production poorly; it exhibits reasonable stability to human metabolic enzymes (e.g. cytochrome P450s; i.e. it has favourable pharmacokinetics); it penetrates well to target tissues; and finally it has an acceptably low level of human toxicity.  56  1.5  Objectives of the Present Thesis The flow of antibiotics from the drug-development pipelines of major pharmaceutical  companies is slowing to a trickle, particularly with respect to antibiotics of novel classes (Boucher et al., 2009); indeed, only a single antibacterial (telavancin) acts with a novel mechanism of action (it is a lipoglycopeptide cell wall synthesis inhibitor). Therefore, the research described in these pages was undertaken with a view to describing and understanding new means of overcoming ß-lactam resistance determinants: the ß-lactamases TEM-1 (E. coli) and OXA-10 (P. aeruginosa), both representing enzyme families able to cleave extended-spectrum ß-lactams; and the recalcitrant TP enzyme, PBP2a, able to confer broadspectrum ß-lactam resistance in MRSA, a notoriously difficult-to-treat pathogen that plagues hospitals (Doshi et al., 2009) and communities alike (Miller and Kaplan, 2009). Community strains that affect otherwise healthy individuals are worrisome in particular, as some aggressively produce Panton-Valentine leukocidin, causing rapidly destructive skin infections (Livermore, 2006). The experimental work described in the first half of this thesis (Chapters 2 and 3) deals with inhibitors of TEM-1 and OXA-10, Class A and Class D ß-lactamases, respectively, members of large families which feature many ESBL enzymes. The second half (Chapters 4 and 5) examines the next-generation cephalosporin, ceftobiprole: its mode of binding to its target, PBP2a, and characterization and analysis of resistance generated in vitro. Chapter 2 examines apparent bacterial countering of ß-lactamases by inhibitor proteins. The inhibitory mechanism is examined by comparison of the published atomicresolution structure of BLIP in complex with TEM-1 versus the novel crystal structure of BLIP-I and the BLIP-I•TEM-1 complex. The specificity of the interaction is examined by 57  contrast with an ultra-high-resolution (1.0 Å) structure determined for the non-inhibitory homologue BLP. Chapter 3 investigates a novel approach to ß-lactamase inhibition by a poorly hydrolysable, non-ß-lactam cyclobutanone, which attempts to geometrically mimic the conformation of penicillin thought to preferentially bind the ß-lactamase active site. The crystal structure of one such inhibitor in complex with OXA-10 is analyzed. Chapter 4 describes in detail the mode of ceftobiprole binding to PBP2a via the analysis of the crystal structure of their covalent complex. Critical aspects of the inhibitor interaction with the active site of PBP2a are identified in an effort to aid future drug development and critically evaluate current anti-MRSA drug development stratagems. Finally, while resistance to ceftobiprole was not observed in clinical trials (Jones, 2007) and only modest resistance was previously achieved in laboratory studies (Heller et al., 2004, Bogdanovich et al., 2005), Chapter 5 reports MRSA strains successfully selected for high-level ceftobiprole resistance. Several sets of mutations to PBP2a are described and their probable resistance mechanisms are analyzed by molecular modeling. It is hoped that the work described in this thesis will serve ongoing efforts to develop new antibiotics targeting broad-spectrum ß-lactamase-producing pathogens and methicillinresistant staphylococci and pneumonococci.  58  1.6  Postscript: A Brief Protein Crystallography Primer Despite the considerable flexibility (Mittermaier and Kay, 2006, Korzhnev and Kay,  2008) and the large number of degrees of conformational freedom of the backbone of even a small protein (100aa = ~200 # and $ angles with a wide range of permissible combinations; Ho et al., 2003), under certain conditions proteins may be induced to form crystals: ordered arrays of repeating units, relatable by symmetry operators (Rhodes, 2006). With relatively pure, stable protein in hand, a variety of parameters are screened, including protein and precipitant concentrations, ionic strength, and pH, in order to discover conditions suitable for crystal nucleation and growth. Dehydration techniques such as the hanging drop method are frequently employed to increase protein and precipitant concentrations in a controlled fashion until crystals grow to a suitable size for x-ray diffraction experiments, typically 0.05-0.5 mm in length. The goal of the diffraction experiment is to determine the solution of the function:  Equation 1.2 the graph of which is a three-dimensional representation of the electron density within the crystallographic unit cell, indicating the position of ordered atoms there. A model of the protein structure and any attending small molecules (e.g. ligands, buffer components, ions, ordered water molecules) may be built using this map. The principal challenge in crystallography (once protein in mg quantities is obtained in sufficient purity and crystal conditions are found which yield strongly diffracting, well-ordered crystals) is that, while amplitude F is experimentally determined, phase information ! is lost (refer to variables above). Thus the phase must be estimated indirectly by a host of possible techniques (for a 59  thorough and clear introduction see Taylor, 2003). These yield initial phase estimates only, permitting, in the best of circumstances, an electron density map that is interpretable, i.e. with connectivity consistent with a protein macromolecule. A protein model is built within this map. Then, using knowledge of organic chemistry (e.g. ideal bond lengths and angles), iterative rounds of phase-refinement (Murshudov et al., 1997) and model building (Emsley and Cowtan, 2004) ensue. One ultimately seeks to obtain a structural model that is reasonably consistent with the diffraction data, within experimental error.  60  1.7  References  ADACHI, H., OHTA, T. & MATSUZAWA, H. (1991) Site-directed mutants, at position 166, of RTEM-1 beta-lactamase that form a stable acyl-enzyme intermediate with penicillin. J Biol Chem, 266, 3186-91. AKOVA, M. (2008) Sulbactam-containing beta-lactamase inhibitor combinations. Clin Microbiol Infect, 14 Suppl 1, 185-8. ALBECK, S., UNGER, R. & SCHREIBER, G. (2000) Evaluation of direct and cooperative contributions towards the strength of buried hydrogen bonds and salt bridges. J Mol Biol, 298, 503-20. ALEXANDER, D. C. & JENSEN, S. E. (1998) Investigation of the Streptomyces clavuligerus cephamycin C gene cluster and its regulation by the CcaR protein. J Bacteriol, 180, 4068-79. ALLEN, H., MOE, L., RODBUMRER, J., GAARDER, A. & HANDELSMAN, J. (2009) Functional metagenomics reveals diverse beta-lactamases in a remote Alaskan soil. ISME J, 3, 243-51. ALTSCHUL, S. F., GISH, W., MILLER, W., MYERS, E. W. & LIPMAN, D. J. (1990) Basic local alignment search tool. J Mol Biol, 215, 403-10. AMBLER, R. P. (1980) The structure of beta-lactamases. Philos Trans R Soc Lond, B, Biol Sci, 289, 32131. BALTZ, R. H. (2008) Renaissance in antibacterial discovery from actinomycetes. Current Opinion in Pharmacology, 8, 557-63. BANERJEE, R., GRETES, M., WEBSTER, D., DERISI, J., & CHAMBERS, H.F. (2008) Characterization of a ceftobiprole-resistant, mecA-negative derivative of methicillin-resistant Staphylococcus aureus (MRSA). Presentation C1-3715, 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy and Infectious Diseases Society of America 46th Annual Meeting, Washington DC, Oct. 25-28, 2008. BARRETEAU, H., KOVAC, A., BONIFACE, A., SOVA, M., GOBEC, S. & BLANOT, D. (2008) Cytoplasmic steps of peptidoglycan biosynthesis. FEMS Microbiology Reviews, 32, 168-207. BATCHELOR, F. R., DOYLE, F. P., NAYLER, J. H. & ROLINSON, G. N. (1959) Synthesis of penicillin: 6-aminopenicillanic acid in penicillin fermentations. Nature, 183, 257-8. BHAVSAR, A. P. & BROWN, E. D. (2006) Cell wall assembly in Bacillus subtilis: how spirals and spaces challenge paradigms. Mol Microbiol, 60, 1077-90. BENNETT-LOVSEY, R. M., HERBERT, A. D., STERNBERG, M. J. & KELLEY, L. A. (2008) Exploring the extremes of sequence/structure space with ensemble fold recognition in the program Phyre. Proteins, 70, 611-25. BENNING, M. M., KUO, J. M., RAUSHEL, F. M. & HOLDEN, H. M. (1995) Three-dimensional structure of the binuclear metal center of phosphotriesterase. Biochemistry, 34, 7973-8. BERTSCHE, U., BREUKINK, E., KAST, T. & VOLLMER, W. (2005) In vitro murein peptidoglycan synthesis by dimers of the bifunctional transglycosylase-transpeptidase PBP1B from Escherichia coli. J Biol Chem, 280, 38096-101. BLUMBERG, P. M. & STROMINGER, J. L. (1972) Five penicillin-binding components occur in Bacillus subtilis membranes. J Biol Chem, 247, 8107-13. BOGDANOVICH, T., EDNIE, L. M., SHAPIRO, S. & APPELBAUM, P. C. (2005) Antistaphylococcal activity of ceftobiprole, a new broad-spectrum cephalosporin. Antimicrob Agents Chemother, 49, 4210-9.  61  BONECA, I. G., HUANG, Z. H., GAGE, D. A. & TOMASZ, A. (2000) Characterization of Staphylococcus aureus cell wall glycan strands, evidence for a new beta-N-acetylglucosaminidase activity. J Biol Chem, 275, 9910-8. BORN, P., BREUKINK, E. & VOLLMER, W. (2006) In vitro synthesis of cross-linked murein and its attachment to sacculi by PBP1A from Escherichia coli. J Biol Chem, 281, 26985-93. BOUCHER, H. W., TALBOT, G. H., BRADLEY, J. S., EDWARDS, J. E., GILBERT, D., RICE, L. B., SCHELD, M., SPELLBERG, B. & BARTLETT, J. (2009) Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis, 48, 1-12. BOX, S. J., HOOD, J. D. & SPEAR, S. R. (1979) Four further antibiotics related to olivanic acid produced by Streptomyces olivaceus: fermentation, isolation, characterisation and biosynthetic studies. J Antibiot, 32, 1239-47. BROTZU, G. (1948) Research on a New Antibiotic. Studies by the Institute of Hygiene of Cagliari. BROWN, A. G., BUTTERWORTH, D., COLE, M., HANSCOMB, G., HOOD, J. D., READING, C. & ROLINSON, G. N. (1976) Naturally-occurring beta-lactamase inhibitors with antibacterial activity. J Antibiot, 29, 668-9. BROWN, D. F. & REYNOLDS, P. E. (1980) Intrinsic resistance to beta-lactam antibiotics in Staphylococcus aureus. FEBS Lett, 122, 275-8. BROWN, S. & AMYES, S. G. (2005) The sequences of seven class D beta-lactamases isolated from carbapenem-resistant Acinetobacter baumannii from four continents. Clin Microbiol Infect, 11, 326-9. BROWN, S., YOUNG, H. K. & AMYES, S. G. (2005) Characterisation of OXA-51, a novel class D carbapenemase found in genetically unrelated clinical strains of Acinetobacter baumannii from Argentina. Clin Microbiol Infect, 11, 15-23. BROWN, V., GUERIN, P., LEGROS, D., PAQUET, C., PÉCOUL, B. & MOREN, A. (2008) Research in complex humanitarian emergencies: the Médecins Sans Frontières/Epicentre experience. PLoS Med, 5, e89. BRYSKIER, A. (2000) Perfecting the ring and extending the antibacterial spectrum: 'the multiple generations'. Clin Microbiol Infect, 6 Suppl 3, 13-21. BUMANN, D. (2008) Has nature already identified all useful antibacterial targets? Current Opinion in Microbiology, 11, 387-92. BUSH, K. (2008) Extended-spectrum beta-lactamases in North America, 1987-2006. Clin Microbiol Infect, 14 Suppl 1, 134-43. BUSH, K., JACOBY, G. A. & MEDEIROS, A. A. (1995) A functional classification scheme for betalactamases and its correlation with molecular structure. Antimicrob Agents Chemother, 39, 1211-33. CARBALLIDO-LOPEZ, R., FORMSTONE, A., LI, Y., EHRLICH, S. D., NOIROT, P. & ERRINGTON, J. (2006) Actin homolog MreBH governs cell morphogenesis by localization of the cell wall hydrolase LytE. Dev Cell, 11, 399-409. CHAIN, E. B. (1945) Nobel Lecture. Nobel Foundation, http://nobelprize.org/nobel_prizes/medicine/laureates/1945/chain-lecture.html, 1-34. CHAMBERS, H. F., SACHDEVA, M. J. & HACKBARTH, C. J. (1994) Kinetics of penicillin binding to penicillin-binding proteins of Staphylococcus aureus. Biochem J, 301 ( Pt 1), 139-44. CLAESSEN, D., EMMINS, R., HAMOEN, L. W., DANIEL, R. A., ERRINGTON, J. & EDWARDS, D. H. (2008) Control of the cell elongation-division cycle by shuttling of PBP1 protein  62  in Bacillus subtilis. Mol Microbiol, 68, 1029-46. CLSI (2007) Performance standards for antimicrobial susceptibility testing. CLSI approved standard M100-S17. Clinical and Laboratory Standards Institute, Wayne, PA. CONTRERAS-MARTEL, C., JOB, V., DI GUILMI, A. M., VERNET, T., DIDEBERG, O. & DESSEN, A. (2006) Crystal structure of penicillin-binding protein 1a (PBP1a) reveals a mutational hotspot implicated in beta-lactam resistance in Streptococcus pneumoniae. J Mol Biol, 355, 684-96. COOPER, R., WELLS, J. S. & SYKES, R. B. (1985) Novel potentiators of beta-lactam antibiotics. Isolation of SQ28,504 and SQ28,546 from Chromobacterium violaceum. J Antibiot, 38, 449-54. DANEL, F., HALL, L. M., GUR, D. & LIVERMORE, D. M. (1998) OXA-16, a further extendedspectrum variant of OXA-10 beta-lactamase, from two Pseudomonas aeruginosa isolates. Antimicrob Agents Chemother, 42, 3117-22. DANEL, F., PAETZEL, M., STRYNADKA, N. C. & PAGE, M. G. (2001) Effect of divalent metal cations on the dimerization of OXA-10 and -14 class D beta-lactamases from Pseudomonas aeruginosa. Biochemistry, 40, 9412-20. DASGUPTA, A., DATTA, P., KUNDU, M. & BASU, J. (2006) The serine/threonine kinase PknB of Mycobacterium tuberculosis phosphorylates PBPA, a penicillin-binding protein required for cell division. Microbiology, 152, 493-504. DAVIES, T. A., SHANG, W. & BUSH, K. (2006) Activities of ceftobiprole and other beta-lactams against Streptococcus pneumoniae clinical isolates from the United States with defined substitutions in penicillin-binding proteins PBP 1a, PBP 2b, and PBP 2x. Antimicrob Agents Chemother, 50, 2530-2. DE LENCASTRE, H., DE JONGE, B. L., MATTHEWS, P. R. & TOMASZ, A. (1994) Molecular aspects of methicillin resistance in Staphylococcus aureus. J Antimicrob Chemother, 33, 7-24. DEN BLAAUWEN, T., DE PEDRO, M. A., NGUYEN-DISTECHE, M. & AYALA, J. A. (2008) Morphogenesis of rod-shaped sacculi. FEMS Microbiol Rev, 32, 321-44. DEROUAUX, A., WOLF, B., FRAIPONT, C., BREUKINK, E., NGUYEN-DISTECHE, M. & TERRAK, M. (2008) The monofunctional glycosyltransferase of Escherichia coli localizes to the cell division site and interacts with penicillin-binding protein 3, FtsW, and FtsN. J Bacteriol, 190, 1831-4. DIVAKARUNI, A. V., LOO, R. R., XIE, Y., LOO, J. A. & GOBER, J. W. (2005) The cell-shape protein MreC interacts with extracytoplasmic proteins including cell wall assembly complexes in Caulobacter crescentus. Proc Natl Acad Sci U S A, 102, 18602-7. DMITRIEV, B. A., TOUKACH, F. V., HOLST, O., RIETSCHEL, E. T. & EHLERS, S. (2004) Tertiary structure of Staphylococcus aureus cell wall murein. J Bacteriol, 186, 7141-8. DORAN, J. L., LESKIW, B. K., AIPPERSBACH, S. & JENSEN, S. E. (1990) Isolation and characterization of a beta-lactamase-inhibitory protein from Streptomyces clavuligerus and cloning and analysis of the corresponding gene. J Bacteriol, 172, 4909-18. DOSHI, R. K., PATEL, G., MACKAY, R. & WALLACH, F. (2009) Healthcare-associated Infections: epidemiology, prevention, and therapy. Mt Sinai J Med, 76, 84-94. DRAMSI, S., MAGNET, S., DAVISON, S. & ARTHUR, M. (2008) Covalent attachment of proteins to peptidoglycan. FEMS Microbiology Reviews, 32, 307-20. EMSLEY, P. & COWTAN, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr, 60, 2126-32. ESSACK, S. Y. (2000) Treatment options for extended-spectrum beta-lactamase-producers. FEMS  63  Microbiol Lett, 190, 181-4. FAN, X., LIU, Y., SMITH, D., KONERMANN, L., SIU, K. W. & GOLEMI-KOTRA, D. (2007) Diversity of penicillin-binding proteins. Resistance factor FmtA of Staphylococcus aureus. J Biol Chem, 282, 35143-52. FINBERG, R. W., MOELLERING, R. C., TALLY, F. P., CRAIG, W. A., PANKEY, G. A., DELLINGER, E. P., WEST, M. A., JOSHI, M., LINDEN, P. K., ROLSTON, K. V., ROTSCHAFER, J. C. & RYBAK, M. J. (2004) The importance of bactericidal drugs: future directions in infectious disease. Clin Infect Dis, 39, 1314-20. FISHER, J. F., MEROUEH, S. O. & MOBASHERY, S. (2005) Bacterial resistance to beta-lactam antibiotics: compelling opportunism, compelling opportunity. Chem Rev, 105, 395-424. FLEMING, A. (1944) The discovery of penicillin. British Medical Bulletin. FLEMING, A. (1945) Nobel Lecture. Nobel Foundation, http://nobelprize.org/nobel_prizes/medicine/laureates/1945/fleming-lecture.html, 1-11. FLOREY, H. W. (1945) Nobel Lecture. Nobel Foundation, http://nobelprize.org/nobel_prizes/medicine/laureates/1945/florey-lecture.html, 1-12. FONZÉ, E., VERMEIRE, M., NGUYEN-DISTÈCHE, M., BRASSEUR, R. & CHARLIER, P. (1999) The crystal structure of a penicilloyl-serine transferase of intermediate penicillin sensitivity. The DD-transpeptidase of streptomyces K15. J Biol Chem, 274, 21853-60. FRANCESCHINI, N., BOSCHI, L., POLLINI, S., HERMAN, R., PERILLI, M., GALLENI, M., FRÈRE, J. M., AMICOSANTE, G. & ROSSOLINI, G. M. (2001) Characterization of OXA-29 from Legionella (Fluoribacter) gormanii: molecular class D beta-lactamase with unusual properties. Antimicrob Agents Chemother, 45, 3509-16. FRERE, J. M., GHUYSEN, J. M. & IWATSUBO, M. (1975) Kinetics of interaction between the exocellular DD-carboxypeptidase-transpeptidase from Streptomyces R61 and beta-lactam antibiotics. A choice of models. Eur J Biochem, 57, 343-51. FUDA, C., SUVOROV, M., SHI, Q., HESEK, D., LEE, M. & MOBASHERY, S. (2007) Shared functional attributes between the mecA gene product of Staphylococcus sciuri and penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus. Biochemistry, 46, 8050-7. GALLENI, M., LAMOTTE-BRASSEUR, J., RAQUET, X., DUBUS, A., MONNAIE, D., KNOX, J. R. & FRÈRE, J. M. (1995) The enigmatic catalytic mechanism of active-site serine beta-lactamases. Biochemical Pharmacology, 49, 1171-8. GAN, L., CHEN, S. & JENSEN, G. J. (2008) Molecular organization of Gram-negative peptidoglycan. Proc Natl Acad Sci USA, 105, 18953-7. GARAU, G., DI GUILMI, A. M. & HALL, B. G. (2005) Structure-based phylogeny of the metallobeta-lactamases. Antimicrob Agents Chemother, 49, 2778-84. GASTEIGER, E., GATTIKER, A., HOOGLAND, C., IVANYI, I., APPEL, R. D. & BAIROCH, A. (2003) ExPASy: The proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Research, 31, 3784-8. GEORGOPAPADAKOU, N. H. & LIU, F. Y. (1980) Binding of beta-lactam antibiotics to penicillin-binding proteins of Staphylococcus aureus and Streptococcus faecalis: relation to antibacterial activity. Antimicrob Agents Chemother, 18, 834-6. GEORGOPAPADAKOU, N. H., SMITH, S. A. & BONNER, D. P. (1982) Penicillin-binding proteins in a Staphylococcus aureus strain resistant to specific beta-lactam antibiotics. Antimicrob  64  Agents Chemother, 22, 172-5. GHUYSEN, J. M., FRERE, J. M., LEYH-BOUILLE, M., NGUYEN-DISTECHE, M. & COYETTE, J. (1986) Active-site-serine D-alanyl-D-alanine-cleaving-peptidase-catalysed acyl-transfer reactions. Procedures for studying the penicillin-binding proteins of bacterial plasma membranes. Biochem J, 235, 159-65. GIAMARELLOU, H., ANTONIADOU, A. & KANELLAKOPOULOU, K. (2008) Acinetobacter baumannii: a universal threat to public health? International Journal of Antimicrobial Agents, 32, 106-19. GIESBRECHT, P., KERSTEN, T., MAIDHOF, H. & WECKE, J. (1998) Staphylococcal cell wall: morphogenesis and fatal variations in the presence of penicillin. Microbiol Mol Biol Rev, 62, 1371-414. GLUSKER, J. P. (1994) Dorothy Crowfoot Hodgkin (1910-1994). Protein Sci, 3, 2465-9. GOFFIN, C. & GHUYSEN, J. M. (1998) Multimodular penicillin-binding proteins: an enigmatic family of orthologs and paralogs. Microbiol Mol Biol Rev, 62, 1079-93. GOLEMI, D., MAVEYRAUD, L., VAKULENKO, S., SAMAMA, J. P. & MOBASHERY, S. (2001) Critical involvement of a carbamylated lysine in catalytic function of class D beta-lactamases. Proc Natl Acad Sci U S A, 98, 14280-5. GOLEMI, D., MAVEYRAUD, L., VAKULENKO, S., TRANIER, S., ISHIWATA, A., KOTRA, L. P., SAMAMA, J. & MOBASHERY, S. (2000) The First Structural and Mechanistic Insights for Class D ß-Lactamases: Evidence for a Novel Catalytic Process for Turnover of ß-Lactam Antibiotics. Journal of the American Chemical Society, 122, 6132-6133. GORDON, E., MOUZ, N., DUÉE, E. & DIDEBERG, O. (2000) The crystal structure of the penicillin-binding protein 2x from Streptococcus pneumoniae and its acyl-enzyme form: implication in drug resistance. Journal of Molecular Biology, 299, 477-85. GRAVES-WOODWARD, K. & PRATT, R. F. (1998) Reaction of soluble penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus with beta-lactams and acyclic substrates: kinetics in homogeneous solution. Biochem J, 332 ( Pt 3), 755-61. GUILLAUME, G., VANHOVE, M., LAMOTTE-BRASSEUR, J., LEDENT, P., JAMIN, M., JORIS, B. & FRÈRE, J. M. (1997) Site-directed mutagenesis of glutamate 166 in two beta-lactamases. Kinetic and molecular modeling studies. J Biol Chem, 272, 5438-44. HACKBARTH, C. J., KOCAGOZ, T., KOCAGOZ, S. & CHAMBERS, H. F. (1995) Point mutations in Staphylococcus aureus PBP 2 gene affect penicillin-binding kinetics and are associated with resistance. Antimicrob Agents Chemother, 39, 103-6. HALL, L. M., LIVERMORE, D. M., GUR, D., AKOVA, M. & AKALIN, H. E. (1993) OXA-11, an extended-spectrum variant of OXA-10 (PSE-2) beta-lactamase from Pseudomonas aeruginosa. Antimicrob Agents Chemother, 37, 1637-44. HARDT, K., JORIS, B., LEPAGE, S., BRASSEUR, R., LAMPEN, J. O., FRERE, J. M., FINK, A. L. & GHUYSEN, J. M. (1997) The penicillin sensory transducer, BlaR, involved in the inducibility of beta-lactamase synthesis in Bacillus licheniformis is embedded in the plasma membrane via a fouralpha-helix bundle. Mol Microbiol, 23, 935-44. HARADA, S., ISHII, Y. & YAMAGUCHI, K. (2008) Extended-spectrum beta-Lactamases: Implications for the Clinical Laboratory and Therapy. The Korean journal of laboratory medicine, 28, 40112. HAREL, M., COHEN, M. & SCHREIBER, G. (2007) On the Dynamic Nature of the Transition State for Protein-Protein Association as Determined by Double-mutant Cycle Analysis and  65  Simulation. J Mol Biol. HAYHURST, E. J., KAILAS, L., HOBBS, J. K. & FOSTER, S. J. (2008) Cell wall peptidoglycan architecture in Bacillus subtilis. Proc Natl Acad Sci USA, 105, 14603-8. HEBEISEN, P., HEINZE-KRAUSS, I., ANGEHRN, P., HOHL, P., PAGE, M. G. & THEN, R. L. (2001) In vitro and in vivo properties of Ro 63-9141, a novel broad-spectrum cephalosporin with activity against methicillin-resistant staphylococci. Antimicrob Agents Chemother, 45, 825-36. HEINZ, U. & ADOLPH, H. W. (2004) Metallo-beta-lactamases: two binding sites for one catalytic metal ion? Cell Mol Life Sci, 61, 2827-39. HELFAND, M. S. & BONOMO, R. A. (2003) Beta-lactamases: a survey of protein diversity. Current drug targets Infectious disorders, 3, 9-23. HELFAND, M. S., TARACILA, M. A., TOTIR, M. A., BONOMO, R. A., BUYNAK, J. D., VAN DEN AKKER, F. & CAREY, P. R. (2007) Raman crystallographic studies of the intermediates formed by Ser130Gly SHV, a beta-lactamase that confers resistance to clinical inhibitors. Biochemistry, 46, 8689-99. HELLER, S., MARRER, E., PAGE, M. G. P., SHAPIRO, S. & THENOZ, L. (2004) Development of endogenous resistance by staphylococci to BAL9141 and comparators, abstract P675. Clin. Microbiol. Infect., 10, 163. HENDERSON, R. (1970) Structure of crystalline alpha-chymotrypsin. IV. The structure of indoleacryloyl-alpha-chyotrypsin and its relevance to the hydrolytic mechanism of the enzyme. Journal of Molecular Biology, 54, 341-54. HENZE, U. U. & BERGER-BÄCHI, B. (1995) Staphylococcus aureus penicillin-binding protein 4 and intrinsic beta-lactam resistance. Antimicrob Agents Chemother, 39, 2415-22. HERMANN, J. C., RIDDER, L., MULHOLLAND, A. J. & HOLTJE, H. D. (2003) Identification of Glu166 as the general base in the acylation reaction of class A beta-lactamases through QM/MM modeling. J Am Chem Soc, 125, 9590-1. HIRAMATSU, K. (1992) [Molecular genetics of MRSA]. Nippon Rinsho, 50, 938-44. HIRAMATSU, K., CUI, L., KURODA, M. & ITO, T. (2001) The emergence and evolution of methicillin-resistant Staphylococcus aureus. Trends Microbiol, 9, 486-93. HO, B. K., THOMAS, A. & BRASSEUR, R. (2003) Revisiting the Ramachandran plot: hard-sphere repulsion, electrostatics, and H-bonding in the alpha-helix. Protein Sci, 12, 2508-22. HODGKIN, D. C. (1949) The X-ray analysis of the structure of penicillin. Advancement of science, 6, 85-9. HOLTJE, J. V. (1998) Growth of the stress-bearing and shape-maintaining murein sacculus of Escherichia coli. Microbiol Mol Biol Rev, 62, 181-203. HUANG, W., ZHANG, Z. & PALZKILL, T. (2000) Design of potent beta-lactamase inhibitors by phage display of beta-lactamase inhibitory protein. J Biol Chem, 275, 14964-8. HUOVINEN, P., HUOVINEN, S. & JACOBY, G. A. (1988) Sequence of PSE-2 beta-lactamase. Antimicrob Agents Chemother, 32, 134-6. IKONOMIDIS, A., MICHAIL, G., VASDEKI, A., LABROU, M., KARAVASILIS, V., STATHOPOULOS, C., MANIATIS, A. N. & POURNARAS, S. (2008) In vitro and in vivo evaluations of oxacillin efficiency against mecA-positive oxacillin-susceptible Staphylococcus aureus. Antimicrob Agents Chemother, 52, 3905-8.  66  IMTIAZ, U., MANAVATHU, E. K., MOBASHERY, S. & LERNER, S. A. (1994) Reversal of clavulanate resistance conferred by a Ser-244 mutant of TEM-1 beta-lactamase as a result of a second mutation (Arg to Ser at position 164) that enhances activity against ceftazidime. Antimicrob Agents Chemother, 38, 1134-9. ISHINO, F., PARK, W., TOMIOKA, S., TAMAKI, S., TAKASE, I., KUNUGITA, K., MATSUZAWA, H., ASOH, S., OHTA, T., SPRATT, B. G. & ET AL. (1986) Peptidoglycan synthetic activities in membranes of Escherichia coli caused by overproduction of penicillin-binding protein 2 and rodA protein. J Biol Chem, 261, 7024-31. ITO, T. & HIRAMATSU, K. (1998) Acquisition of methicillin resistance and progression of multiantibiotic resistance in methicillin-resistant Staphylococcus aureus. Yonsei Med J, 39, 526-33. JABRI, E., CARR, M. B., HAUSINGER, R. P. & KARPLUS, P. A. (1995) The crystal structure of urease from Klebsiella aerogenes. Science, 268, 998-1004. JACOBY, G. A. (2006) Beta-lactamase nomenclature. Antimicrob Agents Chemother, 50, 1123-9. JACOBY, G. A. (2009) AmpC beta-lactamases. Clinical Microbiology Reviews, 22, 161-82, Table of Contents. JONES, R. N., BIEDENBACH, D. J., SADER, H. S., FRITSCHE, T. R., TOLEMAN, M. A. & WALSH, T. R. (2005) Emerging epidemic of metallo-beta-lactamase-mediated resistances. Diagn Microbiol Infect Dis, 51, 77-84. JONES, M. E. (2007) In-vitro profile of a new beta-lactam, ceftobiprole, with activity against methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect, 13 Suppl 2, 17-24. JONES, R. N., KIRBY, J. T. & RHOMBERG, P. R. (2008) Comparative activity of meropenem in US medical centers (2007): initiating the 2nd decade of MYSTIC program surveillance. Diagn Microbiol Infect Dis, 61, 203-13. JOUGHIN, B. A., GREEN, D. F. & TIDOR, B. (2005) Action-at-a-distance interactions enhance protein binding affinity. Protein Sci, 14, 1363-9. KANG, S. G., PARK, H. U., LEE, H. S., KIM, H. T. & LEE, K. J. (2000) New beta -lactamase inhibitory protein (BLIP-I) from Streptomyces exfoliatus SMF19 and its roles on the morphological differentiation. J Biol Chem, 275, 16851-6. KANNAPPAN, B. & GREADY, J. E. (2008) Redefinition of rubisco carboxylase reaction reveals origin of water for hydration and new roles for active-site residues. Journal of the American Chemical Society, 130, 15063-80. KARIMOVA, G., DAUTIN, N. & LADANT, D. (2005) Interaction network among Escherichia coli membrane proteins involved in cell division as revealed by bacterial two-hybrid analysis. J Bacteriol, 187, 2233-43. KATAYAMA, Y., ITO, T. & HIRAMATSU, K. (2000) A new class of genetic element, staphylococcus cassette chromosome mec, encodes methicillin resistance in Staphylococcus aureus. Antimicrob Agents Chemother, 44, 1549-55. KATAYAMA, Y., ROBINSON, D. A., ENRIGHT, M. C. & CHAMBERS, H. F. (2005) Genetic background affects stability of mecA in Staphylococcus aureus. J Clin Microbiol, 43, 2380-3. KATAYAMA, Y., ZHANG, H. Z. & CHAMBERS, H. F. (2003) Effect of disruption of Staphylococcus aureus PBP4 gene on resistance to beta-lactam antibiotics. Microb Drug Resist, 9, 32936. KATTAN, J. N., VILLEGAS, M. V. & QUINN, J. P. (2008) New developments in carbapenems.  67  Clin Microbiol Infect, 14, 1102-11. KEITH, K. E., OYSTON, P. C., CROSSETT, B., FAIRWEATHER, N. F., TITBALL, R. W., WALSH, T. R. & BROWN, K. A. (2005) Functional characterization of OXA-57, a class D betalactamase from Burkholderia pseudomallei. Antimicrob Agents Chemother, 49, 1639-41. KELLY, J. A., KNOX, J. R., MOEWS, P. C., HITE, G. J., BARTOLONE, J. B., ZHAO, H., JORIS, B., FRÈRE, J. M. & GHUYSEN, J. M. (1985) 2.8-A Structure of penicillin-sensitive D-alanyl carboxypeptidase-transpeptidase from Streptomyces R61 and complexes with beta-lactams. J Biol Chem, 260, 6449-58. KELLY, J. A. & KUZIN, A. P. (1995) The refined crystallographic structure of a DD-peptidase penicillin-target enzyme at 1.6 A resolution. Journal of Molecular Biology, 254, 223-36. KERFF, F., CHARLIER, P., COLOMBO, M. L., SAUVAGE, E., BRANS, A., FRERE, J. M., JORIS, B. & FONZE, E. (2003) Crystal structure of the sensor domain of the BlaR penicillin receptor from Bacillus licheniformis. Biochemistry, 42, 12835-43. KERN, T., HEDIGER, S., MULLER, P., GIUSTINI, C., JORIS, B., BOUGAULT, C., VOLLMER, W. & SIMORRE, J. P. (2008) Toward the characterization of peptidoglycan structure and proteinpeptidoglycan interactions by solid-state NMR spectroscopy. J Am Chem Soc, 130, 5618-9. KIM, M. K. & LEE, K. J. (1994) Characteristics of beta-lactamase-inhibiting proteins from Streptomyces exfoliatus SMF19. Appl Environ Microbiol, 60, 1029-32. KOCH, A. L. (1998) Orientation of the peptidoglycan chains in the sacculus of Escherichia coli. Res Microbiol, 149, 689-701. KONDO, N., KUWAHARA-ARAI, K., KURODA-MURAKAMI, H., TATEDA-SUZUKI, E. & HIRAMATSU, K. (2001) Eagle-type methicillin resistance: new phenotype of high methicillin resistance under mec regulator gene control. Antimicrob Agents Chemother, 45, 815-24. KORZHNEV, D. M. & KAY, L. E. (2008) Probing invisible, low-populated States of protein molecules by relaxation dispersion NMR spectroscopy: an application to protein folding. Acc Chem Res, 41, 442-51. KOZER, N., KUTTNER, Y. Y., HARAN, G. & SCHREIBER, G. (2007) Protein-protein association in polymer solutions: from dilute to semidilute to concentrated. Biophys J, 92, 2139-49. KOZER, N. & SCHREIBER, G. (2004) Effect of crowding on protein-protein association rates: fundamental differences between low and high mass crowding agents. J Mol Biol, 336, 763-74. KRUSE, T., BORK-JENSEN, J. & GERDES, K. (2005) The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex. Mol Microbiol, 55, 78-89. KUZIN, A. P., NUKAGA, M., NUKAGA, Y., HUJER, A. M., BONOMO, R. A. & KNOX, J. R. (1999) Structure of the SHV-1 beta-lactamase. Biochemistry, 38, 5720-7. LABISCHINSKI, H., GOODELL, E. W., GOODELL, A. & HOCHBERG, M. L. (1991) Direct proof of a "more-than-single-layered" peptidoglycan architecture of Escherichia coli W7: a neutron small-angle scattering study. J Bacteriol, 173, 751-6. LAFOND, M., COUTURE, F., VÉZINA, G. & LEVESQUE, R. C. (1989) Evolutionary perspectives on multiresistance beta-lactamase transposons. J Bacteriol, 171, 6423-9. LEDENT, P., RAQUET, X., JORIS, B., VAN BEEUMEN, J. & FRÈRE, J. M. (1993) A comparative study of class-D beta-lactamases. Biochem J, 292 ( Pt 2), 555-62. LESKI, T. A. & TOMASZ, A. (2005) Role of penicillin-binding protein 2 (PBP2) in the antibiotic  68  susceptibility and cell wall cross-linking of Staphylococcus aureus: evidence for the cooperative functioning of PBP2, PBP4, and PBP2A. J Bacteriol, 187, 1815-24. LEVY, S. B. (2001) Antibiotic resistance: consequences of inaction. Clin Infect Dis, 33 Suppl 3, S124-9. LIM, D. & STRYNADKA, N. C. (2002) Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol, 9, 870-6. LIRAS, P. & MARTÍN, J. F. (2006) Gene clusters for beta-lactam antibiotics and control of their expression: why have clusters evolved, and from where did they originate? Int Microbiol, 9, 9-19. LIVERMORE, D. (2006) Can beta-lactams be re-engineered to beat MRSA? Clin Microbiol Infect, 12 Suppl 2, 11-6. LIVERMORE, D. M. (2008) Defining an extended-spectrum beta-lactamase. Clin Microbiol Infect, 14 Suppl 1, 3-10. LIVERMORE, D. M. & WOODFORD, N. (2006) The beta-lactamase threat in Enterobacteriaceae, Pseudomonas and Acinetobacter. Trends Microbiol, 14, 413-20. LIVERMORE, D. M., HOPE, R., MUSHTAQ, S. & WARNER, M. (2008) Orthodox and unorthodox clavulanate combinations against extended-spectrum beta-lactamase producers. Clin Microbiol Infect, 14 Suppl 1, 189-93. LLINÁS, A., AHMED, N., CORDARO, M., LAWS, A. P., FRÈRE, J. M., DELMARCELLE, M., SILVAGGI, N. R., KELLY, J. A. & PAGE, M. I. (2005) Inactivation of bacterial DD-peptidase by beta-sultams. Biochemistry, 44, 7738-46. LODER, B., NEWTON, G. G. & ABRAHAM, E. P. (1961) The cephalosporin C nucleus (7aminocephalosporanic acid) and some of its derivatives. Biochem J, 79, 408-16. LOVERING, A. L., DE CASTRO, L., LIM, D. & STRYNADKA, N. C. (2006) Structural analysis of an "open" form of PBP1B from Streptococcus pneumoniae. Protein Sci, 15, 1701-9. LOVERING, A. L., DE CASTRO, L. H., LIM, D. & STRYNADKA, N. C. (2007) Structural insight into the transglycosylation step of bacterial cell-wall biosynthesis. Science, 315, 1402-5. LOVERING, A. L., GRETES, M. & STRYNADKA, N. C. (2008) Structural details of the glycosyltransferase step of peptidoglycan assembly. Curr Opin Struct Biol. LU, W. P., SUN, Y., BAUER, M. D., PAULE, S., KOENIGS, P. M. & KRAFT, W. G. (1999) Penicillin-binding protein 2a from methicillin-resistant Staphylococcus aureus: kinetic characterization of its interactions with beta-lactams using electrospray mass spectrometry. Biochemistry, 38, 6537-46. MACHEBOEUF, P., CONTRERAS-MARTEL, C., JOB, V., DIDEBERG, O. & DESSEN, A. (2006) Penicillin binding proteins: key players in bacterial cell cycle and drug resistance processes. FEMS Microbiology Reviews, 30, 673-91. MACHEBOEUF, P., DI GUILMI, A. M., JOB, V., VERNET, T., DIDEBERG, O. & DESSEN, A. (2005) Active site restructuring regulates ligand recognition in class A penicillin-binding proteins. Proc Natl Acad Sci U S A, 102, 577-82. MAINARDI, J. L., VILLET, R., BUGG, T. D., MAYER, C. & ARTHUR, M. (2008) Evolution of peptidoglycan biosynthesis under the selective pressure of antibiotics in Gram-positive bacteria. FEMS Microbiology Reviews, 32, 386-408. MAIQUES, E., UBEDA, C., CAMPOY, S., SALVADOR, N., LASA, I., NOVICK, R. P., BARBÉ, J. & PENADÉS, J. R. (2006) beta-lactam antibiotics induce the SOS response and horizontal transfer of virulence factors in Staphylococcus aureus. J Bacteriol, 188, 2726-9.  69  MARTIN, S. I. & KAYE, K. M. (2004) Beta-lactam antibiotics: newer formulations and newer agents. Infect Dis Clin North Am, 18, 603-19, ix. MARTÍNEZ-MARTÍNEZ, L. (2008) Extended-spectrum beta-lactamases and the permeability barrier. Clin Microbiol Infect, 14 Suppl 1, 82-9. MASSOVA, I. & MOBASHERY, S. (1998) Kinship and diversification of bacterial penicillin-binding proteins and beta-lactamases. Antimicrob Agents Chemother, 42, 1-17. MATAGNE, A., LAMOTTE-BRASSEUR, J. & FRÈRE, J. M. (1998) Catalytic properties of class A beta-lactamases: efficiency and diversity. Biochem J, 330 ( Pt 2), 581-98. MATTHEW, M. (1979) Plasmid-mediated beta-lactamases of Gram-negative bacteria: properties and distribution. J Antimicrob Chemother, 5, 349-58. MAVEYRAUD, L., GOLEMI, D., KOTRA, L. P., TRANIER, S., VAKULENKO, S., MOBASHERY, S. & SAMAMA, J. P. (2000) Insights into class D beta-lactamases are revealed by the crystal structure of the OXA10 enzyme from Pseudomonas aeruginosa. Structure, 8, 1289-98. MAVEYRAUD, L., GOLEMI-KOTRA, D., ISHIWATA, A., MEROUEH, O., MOBASHERY, S. & SAMAMA, J. P. (2002) High-resolution X-ray structure of an acyl-enzyme species for the class D OXA-10 beta-lactamase. J Am Chem Soc, 124, 2461-5. MAVEYRAUD, L., PRATT, R. F. & SAMAMA, J. P. (1998) Crystal structure of an acylation transition-state analog of the TEM-1 beta-lactamase. Mechanistic implications for class A betalactamases. Biochemistry, 37, 2622-8. MCDONOUGH, M. A., ANDERSON, J. W., SILVAGGI, N. R., PRATT, R. F., KNOX, J. R. & KELLY, J. A. (2002) Structures of two kinetic intermediates reveal species specificity of penicillinbinding proteins. Journal of Molecular Biology, 322, 111-22. MEMMI, G., FILIPE, S. R., PINHO, M. G., FU, Z. & CHEUNG, A. (2008) Staphylococcus aureus PBP4 is Essential for beta-lactam Resistance in Community-Acquired Methicillin Resistant Strains. Antimicrob Agents Chemother. MEROUEH, S. O., FISHER, J. F., SCHLEGEL, H. B. & MOBASHERY, S. (2005) Ab initio QM/MM study of class A beta-lactamase acylation: dual participation of Glu166 and Lys73 in a concerted base promotion of Ser70. J Am Chem Soc, 127, 15397-407. MILLER, L. G. & KAPLAN, S. L. (2009) Staphylococcus aureus: A Community Pathogen. Infect Dis Clin North Am, 23, 35-52. MINASOV, G., WANG, X. & SHOICHET, B. K. (2002) An ultrahigh resolution structure of TEM1 beta-lactamase suggests a role for Glu166 as the general base in acylation. Journal of the American Chemical Society, 124, 5333-40. MITTERMAIER, A. & KAY, L. E. (2006) New tools provide new insights in NMR studies of protein dynamics. Science, 312, 224-8. MOHAMMADI, T., KARCZMAREK, A., CROUVOISIER, M., BOUHSS, A., MENGINLECREULX, D. & DEN BLAAUWEN, T. (2007) The essential peptidoglycan glycosyltransferase MurG forms a complex with proteins involved in lateral envelope growth as well as with proteins involved in cell division in Escherichia coli. Mol Microbiol, 65, 1106-21. MORLOT, C., PERNOT, L., LE GOUELLEC, A., DI GUILMI, A. M., VERNET, T., DIDEBERG, O. & DESSEN, A. (2005) Crystal structure of a peptidoglycan synthesis regulatory factor (PBP3) from Streptococcus pneumoniae. J Biol Chem, 280, 15984-91. MULLER, P., EWERS, C., BERTSCHE, U., ANSTETT, M., KALLIS, T., BREUKINK, E.,  70  FRAIPONT, C., TERRAK, M., NGUYEN-DISTECHE, M. & VOLLMER, W. (2007) The essential cell division protein FtsN interacts with the murein (peptidoglycan) synthase PBP1B in Escherichia coli. J Biol Chem, 282, 36394-402. MULVEY, M. R., BRYCE, E., BOYD, D. A., OFNER-AGOSTINI, M., LAND, A. M., SIMOR, A. E. & PATON, S. (2005) Molecular characterization of cefoxitin-resistant Escherichia coli from Canadian hospitals. Antimicrob Agents Chemother, 49, 358-65. MURPHY, J. T., WALSHE, R. & DEVOCELLE, M. (2008) A computational model of antibioticresistance mechanisms in methicillin-resistant Staphylococcus aureus (MRSA). Journal of Theoretical Biology, 254, 284-93. MURSHUDOV, G. N., VAGIN, A. A. & DODSON, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr, 53, 240-55. NAAS, T. & NORDMANN, P. (1999) OXA-type beta-lactamases. Curr Pharm Des, 5, 865-79. NAAS, T., POIREL, L. & NORDMANN, P. (2008) Minor extended-spectrum beta-lactamases. Clin Microbiol Infect, 14 Suppl 1, 42-52. NEU, H. C. (1992) The crisis in antibiotic resistance. Science, 257, 1064-73. NICOLA, G., PEDDI, S., STEFANOVA, M., NICHOLAS, R. A., GUTHEIL, W. G. & DAVIES, C. (2005) Crystal structure of Escherichia coli penicillin-binding protein 5 bound to a tripeptide boronic acid inhibitor: a role for Ser-110 in deacylation. Biochemistry, 44, 8207-17. NOEL, G. J., BUSH, K., BAGCHI, P., IANUS, J. & STRAUSS, R. S. (2008a) A randomized, doubleblind trial comparing ceftobiprole medocaril with vancomycin plus ceftazidime for the treatment of patients with complicated skin and skin-structure infections. Clin Infect Dis, 46, 647-55. NOEL, G. J., STRAUSS, R. S., AMSLER, K., HEEP, M., PYPSTRA, R. & SOLOMKIN, J. S. (2008b) Results of a double-blind, randomized trial of ceftobiprole treatment of complicated skin and skin structure infections caused by gram-positive bacteria. Antimicrob Agents Chemother, 52, 37-44. NUKAGA, M., MAYAMA, K., HUJER, A. M., BONOMO, R. A. & KNOX, J. R. (2003) Ultrahigh resolution structure of a class A beta-lactamase: on the mechanism and specificity of the extendedspectrum SHV-2 enzyme. Journal of Molecular Biology, 328, 289-301. PADAYATTI, P. S., HELFAND, M. S., TOTIR, M. A., CAREY, M. P., CAREY, P. R., BONOMO, R. A. & VAN DEN AKKER, F. (2005) High resolution crystal structures of the trans-enamine intermediates formed by sulbactam and clavulanic acid and E166A SHV-1 ß-lactamase. J Biol Chem, 280, 34900-7. PADAYATTI, P. S., SHERI, A., TOTIR, M. A., HELFAND, M. S., CAREY, M. P., ANDERSON, V. E., CAREY, P. R., BETHEL, C. R., BONOMO, R. A., BUYNAK, J. D. & VAN DEN AKKER, F. (2006) Rational design of a beta-lactamase inhibitor achieved via stabilization of the trans-enamine intermediate: 1.28 A crystal structure of wt SHV-1 complex with a penam sulfone. Journal of the American Chemical Society, 128, 13235-42. PAETZEL, M., DANEL, F., DE CASTRO, L., MOSIMANN, S. C., PAGE, M. G. & STRYNADKA, N. C. (2000) Crystal structure of the class D beta-lactamase OXA-10. Nat Struct Biol, 7, 918-25. PAGE, M. G. (2000) Beta-Lactamase inhibitors. Drug Resist Updat, 3, 109-125. PAGE, M. G. (2008) Extended-spectrum beta-lactamases: structure and kinetic mechanism. Clin Microbiol Infect, 14 Suppl 1, 63-74. PALUMBI, S. R. (2001) Humans as the world's greatest evolutionary force. Science, 293, 1786-90.  71  PARK, H. U. & LEE, K. J. (1998) Cloning and heterologous expression of the gene for BLIP-II, a beta-lactamase-inhibitory protein from Streptomyces exfoliatus SMF19. Microbiology, 144 ( Pt 8), 21617. PARK, J. T. & UEHARA, T. (2008) How bacteria consume their own exoskeletons (turnover and recycling of cell wall peptidoglycan). Microbiol Mol Biol Rev, 72, 211-27, table of contents. PAYNE, D. J., CRAMP, R., WINSTANLEY, D. J. & KNOWLES, D. J. (1994) Comparative activities of clavulanic acid, sulbactam, and tazobactam against clinically important beta-lactamases. Antimicrob Agents Chemother, 38, 767-72. PEREZ-LLARENA, F. J., LIRAS, P., RODRIGUEZ-GARCIA, A. & MARTIN, J. F. (1997) A regulatory gene (ccaR) required for cephamycin and clavulanic acid production in Streptomyces clavuligerus: amplification results in overproduction of both beta-lactam compounds. J Bacteriol, 179, 2053-9. PERLSTEIN, D. L., ZHANG, Y., WANG, T. S., KAHNE, D. E. & WALKER, S. (2007) The direction of glycan chain elongation by peptidoglycan glycosyltransferases. J Am Chem Soc, 129, 12674-5. PERNOT, L., CHESNEL, L., LE GOUELLEC, A., CROIZÉ, J., VERNET, T., DIDEBERG, O. & DESSEN, A. (2004) A PBP2x from a clinical isolate of Streptococcus pneumoniae exhibits an alternative mechanism for reduction of susceptibility to beta-lactam antibiotics. J Biol Chem, 279, 16463-70. PERNOT, L., FRENOIS, F., RYBKINE, T., L'HERMITE, G., PETRELLA, S., DELETTRE, J., JARLIER, V., COLLATZ, E. & SOUGAKOFF, W. (2001) Crystal structures of the class D betalactamase OXA-13 in the native form and in complex with meropenem. J Mol Biol, 310, 859-74. PETROSINO, J., RUDGERS, G., GILBERT, H. & PALZKILL, T. (1999) Contributions of aspartate 49 and phenylalanine 142 residues of a tight binding inhibitory protein of beta-lactamases. J Biol Chem, 274, 2394-400. PIENAAR, E., WHITNEY, S., VILJOEN, H. & VAN RENSBURG, N. F. (2008) A model of the complex response of Staphylococcus aureus to methicillin. Journal of Theoretical Biology. POIREL, L., MAMMERI, H. & NORDMANN, P. (2004) TEM-121, a novel complex mutant of TEM-type beta-lactamase from Enterobacter aerogenes. Antimicrob Agents Chemother, 48, 4528-31. POWELL, A. J., TOMBERG, J., DEACON, A. M., NICHOLAS, R. A. & DAVIES, C. (2009) Crystal structures of penicillin-binding protein 2 from penicillin-susceptible and -resistant strains of Neisseria gonorrhoeae reveal an unexpectedly subtle mechanism for antibiotic resistance. J Biol Chem, 284, 1202-12. QUEENER, S. W. (1986) History and origins of beta-lactam antibiotics. IN QUEENER, S. F., WEBBER, J. A. & QUEENER, S. W. (Eds.) Clinical Pharmacology 4: Beta-Lactam Antibiotics For Clinical Use. New York, Marcel Dekker, Inc. REICHMANN, D., COHEN, M., ABRAMOVICH, R., DYM, O., LIM, D., STRYNADKA, N. C. & SCHREIBER, G. (2007) Binding hot spots in the TEM1-BLIP interface in light of its modular architecture. J Mol Biol, 365, 663-79. REICHMANN, D., PHILLIP, Y., CARMI, A. & SCHREIBER, G. (2008) On the contribution of water-mediated interactions to protein-complex stability. Biochemistry, 47, 1051-60. REYNOLDS, K. A., THOMSON, J. M., CORBETT, K. D., BETHEL, C. R., BERGER, J. M., KIRSCH, J. F., BONOMO, R. A. & HANDEL, T. M. (2006) Structural and Computational Characterization of the SHV-1 beta-Lactamase-beta-Lactamase Inhibitor Protein Interface. J. Biol.  72  Chem., 281, 26745-26753. RHAZI, N., CHARLIER, P., DEHARENG, D., ENGHER, D., VERMEIRE, M., FRÈRE, J. M., NGUYEN-DISTÈCHE, M. & FONZÉ, E. (2003) Catalytic mechanism of the Streptomyces K15 DD-transpeptidase/penicillin-binding protein probed by site-directed mutagenesis and structural analysis. Biochemistry, 42, 2895-906. RHODES, G. (2006) Crystallography Made Crystal Clear: A Guide for Users of Macromolecular Models!. 306. ROBINSON, D. A. & ENRIGHT, M. C. (2004) Multilocus sequence typing and the evolution of methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect, 10, 92-7. ROMEIS, T. & HOLTJE, J. V. (1994) Specific interaction of penicillin-binding proteins 3 and 7/8 with soluble lytic transglycosylase in Escherichia coli. J Biol Chem, 269, 21603-7. ROSSOLINI, G. M., D'ANDREA, M. M. & MUGNAIOLI, C. (2008) The spread of CTX-M-type extended-spectrum beta-lactamases. Clin Microbiol Infect, 14 Suppl 1, 33-41. RUDGERS, G. W., HUANG, W. & PALZKILL, T. (2001) Binding properties of a peptide derived from beta-lactamase inhibitory protein. Antimicrob Agents Chemother, 45, 3279-86. RUDGERS, G. W. & PALZKILL, T. (1999) Identification of residues in beta-lactamase critical for binding beta-lactamase inhibitory protein. J Biol Chem, 274, 6963-7 RUIZ, N. (2008) Bioinformatics identification of MurJ (MviN) as the peptidoglycan lipid II flippase in Escherichia coli. Proc Natl Acad Sci USA, 105, 15553-7. RYFFEL, C., TESCH, W., BIRCH-MACHIN, I., REYNOLDS, P. E., BARBERIS-MAINO, L., KAYSER, F. H. & BERGER-BÄCHI, B. (1990) Sequence comparison of mecA genes isolated from methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis. Gene, 94, 137-8. SALZBERG, L. I. & HELMANN, J. D. (2007) An antibiotic-inducible cell-wall associated protein that protects Bacillus subtilis from autolysis. J Bacteriol. SANTILLANA, E., BECEIRO, A., BOU, G. & ROMERO, A. (2007) Crystal structure of the carbapenemase OXA-24 reveals insights into the mechanism of carbapenem hydrolysis. Proc Natl Acad Sci USA, 104, 5354-9. SAUVAGE, E., HERMAN, R., PETRELLA, S., DUEZ, C., BOUILLENNE, F., FRÈRE, J. M. & CHARLIER, P. (2005) Crystal structure of the Actinomadura R39 DD-peptidase reveals new domains in penicillin-binding proteins. J Biol Chem, 280, 31249-56. SAUVAGE, E., KERFF, F., FONZÉ, E., HERMAN, R., SCHOOT, B., MARQUETTE, J. P., TABURET, Y., PREVOST, D., DUMAS, J., LEONARD, G., STEFANIC, P., COYETTE, J. & CHARLIER, P. (2002) The 2.4-A crystal structure of the penicillin-resistant penicillin-binding protein PBP5fm from Enterococcus faecium in complex with benzylpenicillin. Cell Mol Life Sci, 59, 1223-32. SCHROEDER, W. A., LOCKE, T. R. & JENSEN, S. E. (2002) Resistance to beta-lactamase inhibitor protein does not parallel resistance to clavulanic acid in TEM beta-lactamase mutants. Antimicrob Agents Chemother, 46, 3568-73. SELZER, T. & SCHREIBER, G. (2001) New insights into the mechanism of protein-protein association. Proteins, 45, 190-8. SEVERIN, A., VAZ PATO, M. V., SÁ FIGUEIREDO, A. M. & TOMASZ, A. (1995) Drastic changes in the peptidoglycan composition of penicillin resistant laboratory mutants of Streptococcus pneumoniae. FEMS Microbiol Lett, 130, 31-5.  73  SHEEHAN, J. C. & LOGAN, K. R. H. (1959) A general synthesis of the penicillins. J. Am. Chem. Soc. SHEEHAN, J. C., SHIBAHARA, S. & CHACKO, E. (1980) Simple beta-lactam compounds derived from 6-aminopenicillanic acid. J Med Chem, 23, 809-11. SILVAGGI, N. R., ANDERSON, J. W., BRINSMADE, S. R., PRATT, R. F. & KELLY, J. A. (2003) The crystal structure of phosphonate-inhibited D-Ala-D-Ala peptidase reveals an analogue of a tetrahedral transition state. Biochemistry, 42, 1199-208. SINGH, G. S. (2004) Beta-lactams in the new millennium. Part-I: monobactams and carbapenems. Mini reviews in medicinal chemistry, 4, 69-92. SINHA ROY, R., YANG, P., KODALI, S., XIONG, Y., KIM, R. M., GRIFFIN, P. R., ONISHI, H. R., KOHLER, J., SILVER, L. L. & CHAPMAN, K. (2001) Direct interaction of a vancomycin derivative with bacterial enzymes involved in cell wall biosynthesis. Chem Biol, 8, 1095-106. SLOVAK, P. M., PORTER, S. L. & ARMITAGE, J. P. (2006) Differential localization of Mre proteins with PBP2 in Rhodobacter sphaeroides. J Bacteriol, 188, 1691-700. SPRATT, B. G. & PARDEE, A. B. (1975) Penicillin-binding proteins and cell shape in E. coli. Nature, 254, 516-7. STAPLETON, P. D., WU, P. J., KING, A., SHANNON, K., FRENCH, G. & PHILLIPS, I. (1995) Incidence and mechanisms of resistance to the combination of amoxicillin and clavulanic acid in Escherichia coli. Antimicrob Agents Chemother, 39, 2478-83. STRYNADKA, N. C., ADACHI, H., JENSEN, S. E., JOHNS, K., SIELECKI, A., BETZEL, C., SUTOH, K. & JAMES, M. N. (1992) Molecular structure of the acyl-enzyme intermediate in betalactam hydrolysis at 1.7 A resolution. Nature, 359, 700-5. STRYNADKA, N. C., JENSEN, S. E., ALZARI, P. M. & JAMES, M. N. (1996a) A potent new mode of beta-lactamase inhibition revealed by the 1.7 A X-ray crystallographic structure of the TEM1-BLIP complex. Nat Struct Biol, 3, 290-7. STRYNADKA, N. C., JENSEN, S. E., JOHNS, K., BLANCHARD, H., PAGE, M., MATAGNE, A., FRERE, J. M. & JAMES, M. N. (1994) Structural and kinetic characterization of a beta-lactamaseinhibitor protein. Nature, 368, 657-60. STRYNADKA, N. C., MARTIN, R., JENSEN, S. E., GOLD, M. & JONES, J. B. (1996b) Structurebased design of a potent transition state analogue for TEM-1 beta-lactamase. Nat Struct Biol, 3, 68895. SUN, T., NUKAGA, M., MAYAMA, K., CRICHLOW, G. V., KUZIN, A. P. & KNOX, J. R. (2001) Crystallization and preliminary X-ray study of OXA-1, a class D beta- lactamase. Acta Crystallogr D Biol Crystallogr, 57, 1912-4. SUZUKI, E., HIRAMATSU, K. & YOKOTA, T. (1992) Survey of methicillin-resistant clinical strains of coagulase-negative staphylococci for mecA gene distribution. Antimicrob Agents Chemother, 36, 429-34. TAYLOR, G. (2003) The phase problem. Acta Crystallogr D Biol Crystallogr, 59, 1881-90. TERRAK, M., GHOSH, T. K., VAN HEIJENOORT, J., VAN BEEUMEN, J., LAMPILAS, M., ASZODI, J., AYALA, J. A., GHUYSEN, J. M. & NGUYEN-DISTECHE, M. (1999) The catalytic, glycosyl transferase and acyl transferase modules of the cell wall peptidoglycan-polymerizing penicillin-binding protein 1b of Escherichia coli. Mol Microbiol, 34, 350-64. THERRIEN, C. & LEVESQUE, R. C. (2000) Molecular basis of antibiotic resistance and betalactamase inhibition by mechanism-based inactivators: perspectives and future directions. FEMS  74  Microbiology Reviews, 24, 251-62. TIEN, M., BERLETT, B. S., LEVINE, R. L., CHOCK, P. B. & STADTMAN, E. R. (1999) Peroxynitrite-mediated modification of proteins at physiological carbon dioxide concentration: pH dependence of carbonyl formation, tyrosine nitration, and methionine oxidation. Proc Natl Acad Sci USA, 96, 7809-14. TIPPER, D. J. & STROMINGER, J. L. (1965) Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl Acad Sci USA, 54, 1133-41. TOMASZ, A., DRUGEON, H. B., DE LENCASTRE, H. M., JABES, D., MCDOUGALL, L. & BILLE, J. (1989) New mechanism for methicillin resistance in Staphylococcus aureus: clinical isolates that lack the PBP 2a gene and contain normal penicillin-binding proteins with modified penicillinbinding capacity. Antimicrob Agents Chemother, 33, 1869-74. TOTIR, M. A., PADAYATTI, P. S., HELFAND, M. S., CAREY, M. P., BONOMO, R. A., CAREY, P. R. & VAN DEN AKKER, F. (2006) Effect of the inhibitor-resistant M69V substitution on the structures and populations of trans-enamine beta-lactamase intermediates. Biochemistry, 45, 11895-904. VAN DEN ENT, F., LEAVER, M., BENDEZU, F., ERRINGTON, J., DE BOER, P. & LOWE, J. (2006) Dimeric structure of the cell shape protein MreC and its functional implications. Mol Microbiol, 62, 1631-42. VOLLMER, W. & BERTSCHE, U. (2008) Murein (peptidoglycan) structure, architecture and biosynthesis in Escherichia coli. Biochim Biophys Acta, 1778, 1714-34. VOLLMER, W., BLANOT, D. & DE PEDRO, M. A. (2008a) Peptidoglycan structure and architecture. FEMS Microbiol Rev, 32, 149-67. VOLLMER, W., JORIS, B., CHARLIER, P. & FOSTER, S. (2008b) Bacterial peptidoglycan (murein) hydrolases. FEMS Microbiology Reviews, 32, 259-86. VOLLMER, W., VON RECHENBERG, M. & HOLTJE, J. V. (1999) Demonstration of molecular interactions between the murein polymerase PBP1B, the lytic transglycosylase MltA, and the scaffolding protein MipA of Escherichia coli. J Biol Chem, 274, 6726-34. VON RECHENBERG, M., BLAKE, B. K., HO, Y. S., ZHEN, Y., CHEPANOSKE, C. L., RICHARDSON, B. E., XU, N. & KERY, V. (2005) Ampicillin/penicillin-binding protein interactions as a model drug-target system to optimize affinity pull-down and mass spectrometric strategies for target and pathway identification. Proteomics, 5, 1764-73. VON RECHENBERG, M., URSINUS, A. & HOLTJE, J. V. (1996) Affinity chromatography as a means to study multienzyme complexes involved in murein synthesis. Microb Drug Resist, 2, 155-7. WADA, A. & WATANABE, H. (1998) Penicillin-binding protein 1 of Staphylococcus aureus is essential for growth. J Bacteriol, 180, 2759-65. WALSH, C. (2000) Molecular mechanisms that confer antibacterial drug resistance. Nature, 406, 77581. WALSH, T. R. (2008) Clinically significant carbapenemases: an update. Curr Opin Infect Dis, 21, 36771. WANG, J., ZHANG, Z., PALZKILL, T. & CHOW, D. C. (2007a) Thermodynamic investigation of the role of contact residues of blip for binding to tem-1 beta -lactamase. J Biol Chem, 282, 1767617684. WANG, Z., NICHOLLS, S. J., RODRIGUEZ, E. R., KUMMU, O., HÖRKKÖ, S., BARNARD, J., REYNOLDS, W. F., TOPOL, E. J., DIDONATO, J. A. & HAZEN, S. L. (2007b) Protein  75  carbamylation links inflammation, smoking, uremia and atherogenesis. Nat Med, 13, 1176-84. WANG, J., PALZKILL, T. & CHOW, D. (2009) Structural insight into the kinetics and uCp of interactions between TEM-1 beta -lactamase and BLIP. J. Biol. Chem., 284, 595-609. WEIDENMAIER, C. & PESCHEL, A. (2008) Teichoic acids and related cell-wall glycopolymers in Gram-positive physiology and host interactions. Nat Rev Microbiol, 6, 276-87. WELDHAGEN, G. F. (2004) Integrons and beta-lactamases--a novel perspective on resistance. International Journal of Antimicrobial Agents, 23, 556-62. WELZEL, P. (2005) Syntheses around the transglycosylation step in peptidoglycan biosynthesis. Chem Rev, 105, 4610-60. WHICHARD, J. M., GAY, K., STEVENSON, J. E., JOYCE, K. J., COOPER, K. L., OMONDI, M., MEDALLA, F., JACOBY, G. A. & BARRETT, T. J. (2007) Human Salmonella and concurrent decreased susceptibility to quinolones and extended-spectrum cephalosporins. Emerging Infect Dis, 13, 1681-8. WHO (2007) World Health Organization Model List of Essential Medicines, 15th edition. 1-36. www.who.int/medicines/publications/essentialmedicines/en WILKE, M. S., HILLS, T. L., ZHANG, H. Z., CHAMBERS, H. F. & STRYNADKA, N. C. (2004) Crystal structures of the Apo and penicillin-acylated forms of the BlaR1 beta-lactam sensor of Staphylococcus aureus. J Biol Chem, 279, 47278-87. YIM, G., WANG, H. H. & DAVIES, J. (2006) The truth about antibiotics. Int J Med Microbiol, 296, 163-70. YUAN, Y., BARRETT, D., ZHANG, Y., KAHNE, D., SLIZ, P. & WALKER, S. (2007) Crystal structure of a peptidoglycan glycosyltransferase suggests a model for processive glycan chain synthesis. Proc Natl Acad Sci U S A, 104, 5348-53. ZAPUN, A., VERNET, T. & PINHO, M. G. (2008) The different shapes of cocci. FEMS Microbiology Reviews, 32, 345-60. ZHANG, H. Z., HACKBARTH, C. J., CHANSKY, K. M. & CHAMBERS, H. F. (2001) A proteolytic transmembrane signaling pathway and resistance to beta-lactams in staphylococci. Science, 291, 1962-5. ZHANG, Z. & PALZKILL, T. (2003) Determinants of binding affinity and specificity for the interaction of TEM-1 and SME-1 beta-lactamase with beta-lactamase inhibitory protein. J Biol Chem, 278, 45706-12. ZHANG, Z. & PALZKILL, T. (2004) Dissecting the protein-protein interface between betalactamase inhibitory protein and class A beta-lactamases. J Biol Chem, 279, 42860-6. ZHOU, X. Y., BORDON, F., SIROT, D., KITZIS, M. D. & GUTMANN, L. (1994) Emergence of clinical isolates of Escherichia coli producing TEM-1 derivatives or an OXA-1 beta-lactamase conferring resistance to beta-lactamase inhibitors. Antimicrob Agents Chemother, 38, 1085-9. ZIJDERVELD, C. A., AARSMAN, M. E. & NANNINGA, N. (1995) Differences between inner membrane and peptidoglycan-associated PBP1B dimers of Escherichia coli. J Bacteriol, 177, 1860-3. ZORZI, W., ZHOU, X. Y., DARDENNE, O., LAMOTTE, J., RAZE, D., PIERRE, J., GUTMANN, L. & COYETTE, J. (1996) Structure of the low-affinity penicillin-binding protein 5 PBP5fm in wild-type and highly penicillin-resistant strains of Enterococcus faecium. J Bacteriol, 178, 4948-57.  76  CHAPTER 2: INSIGHTS INTO POSITIVE AND NEGATIVE REQUIREMENTS FOR PROTEIN-PROTEIN INTERACTIONS BY CRYSTALLOGRAPHIC ANALYSIS OF THE ß-LACTAMASE INHIBITORY PROTEINS BLIP, BLIP-I AND BLP! 2.1  Introduction Protein-protein interactions frequently exhibit two contradictory characteristics:  stringent specificity and rapid rates of formation (Schlosshauer and Baker, 2004). The general principles underlying the formation of specific macromolecular complexes are applicable to fields ranging from protein engineering (Reichmann et al., 2008) to the evolution of complex molecular machines (Shi et al., 2005); from cell signaling (Tanoue and Nishida, 2003) to the rational design of novel therapeutic agents (Wells et al., 2007). A classic model for the study of protein-protein interactions is the enzyme-inhibitor pair TEM-1•BLIP, a clinically important Gram negative-borne Class A ß-lactamase and its well-studied macromolecular inhibitor. ß-lactams (penicillins, cephalosporins, etc.) are among the most widely prescribed antibiotics. Despite the large array of ß-lactam derivatives available, drug resistance remains widespread. This is due principally to the ß-lactamase enzymes, found in numerous bacterial species, including many pathogens (Livermore, 1995). While a handful of ß-lactamase inhibitors (clavulanate, sulbactam and tazobactam) are presently available for clinical use (Akova, 2008), these are by no means universally effective (Gupta et al., 2006). Therefore, in addition to helping elucidate general principles governing protein-protein interactions, further study of the BLIP system may provide insights into  !  A version of this chapter has been published. M Gretes, DC Lim, L de Castro, SE Jensen, SG Kang, KJ Lee, NCJ Strynadka. (2009) Insights into positive and negative requirements for protein-protein interactions by crystallographic analysis of the Beta-Lactamase Inhibitory Proteins BLIP, BLIP-I and BLP. J Mol Biol 389(2): 289–305.  77  novel means of ß-lactamase inhibition. Several BLIPs have been identified in species of the soil bacteria Streptomyces. The originally identified BLIP (from S. clavuligerus; Doran et al., 1990) has become a model for a broad range of structural, mutagenic, biophysical and computational studies (Strynadka et al., 1996, Huang et al., 1998, Albeck and Schreiber, 1999, Petrosino et al., 1999, Amstutz et al., 2002, Zhang and Palzkill, 2003, Joughin et al., 2005, Reichmann et al., 2005, Reynolds et al., 2006, Harel et al., 2007, Kozer et al., 2007, Wang et al., 2007). BLIP has two close probable homologues: BLIP-I (from S. exfoliatus) and BLP (S. clavuligerus). No other related proteins† are found in any Streptomyces genome that has been sequenced fully‡ or partially§. However, since only eight genomes of well over four thousand known Streptomyces species have to date been even partially sequenced, more BLIPs probably await discovery$. The structure of BLIP has been previously discussed in detail (Strynadka et al., 1994). In brief, it is a 165 aa, secreted protein composed of two apparently homologous domains, each featuring a helix-loop-helix motif packed against a four-stranded ß-sheet. The result is an extended eight-stranded ß-sheet with a solvent-exposed, slightly concave face that forms the interaction interface for a variety of ß-lactamases (Strynadka et al., 1996). This highaffinity (Ki ~0.5 nM) interface between BLIP and TEM-1 is comprised of a set of six mutually independent binding modules, named C1-C6 (Reichmann et al., 2005, Reichmann et al., 2007a). Many single and multiple mutants (of both enzyme and inhibitor) have been kinetically, thermodynamically and structurally characterized (Zhang and Palzkill, 2004,  † ‡ § $  one possible exception is a relatively weak match to a putative secreted esterase (NCBI Entrez Protein accession no. BAC68859 in S. avermitilis (e-value 5.5e-02, identity: 26/78aa). S. avermitilis MA-4680, S. coelicolor A3(2) M145, S. diversa, S. griseus griseus NBRC 13350 S. ambofaciens ATCC 23877, S. noursei ATCC 11455, S. peucetius ATCC 27952, S. scabies 87-22 www.genomesonline.org/BacteriaTree.html (Liolios et al., 2008)  78  Zhang and Palzkill, 2003, Reichmann et al., 2005, Reichmann et al., 2007a, Wang et al., 2007, Harel et al., 2007, Reynolds et al., 2006). In addition, several inhibitory peptides have been designed from key interacting residues of BLIP (Sun et al., 2005, Huang et al., 2000, Rudgers et al., 2001, Rudgers and Palzkill, 2001). It is consistently noted in the literature that the biological function of BLIP is to inhibit ß-lactamases in an escalating microbiological arms race. Yet this has not been proven. The idea that BLIP serves to inhibit TEM-1 in the wild is difficult to reconcile with the fact that TEM-1 is a gene product of the enteric mammalian gut bacterium Escherichia coli (Dougan et al., 2001, Hedges et al., 1977). However, evidence has emerged that numerous and diverse ß-lactamases also naturally occur in undisturbed soil microbial communities (Allen et al., 2009). Indeed, the aforementioned study identifies a class A ß-lactamase (EU408346) from an unculturable soil microbe that resembles E. coli TEM-1 with 43% aa sequence identity (E-value = 3 x 10-54; Altschul et al., 1990). Thus it appears that ß-lactamase inhibition remains a possible natural function of BLIP. The GC content of the EU408346 gene is 69%, consistent with that of the surrounding ~6.0 kbp genomic sequence (67%). For comparison, the TEM-1 gene from E. coli has GC content 49% (that of the E. coli K12 genome is 50%). This suggests, therefore, that the ß-lactamase gene is native to its host organism, although the total GC content and size of its genome is unknown. BLIP-I is a 157 aa protein isolated directly from the related bacterium S. exfoliatus. It has 37% aa sequence identity to BLIP and appears equivalent in terms of its secretion from the cell and ß-lactamase inhibition profile. There is also evidence for at least one common interaction interface residue: the mutation D49A in BLIP-I weakens its affinity for TEM-1 from Ki ~0.5 nM to 10 nM, closely mirroring values for BLIP and its respective D49A 79  mutant. Interestingly, although it is also able to strongly inhibit certain ß-lactamases (e.g. TEM-1 and bactopenase), BLIP-I differs from BLIP by playing a role in morphological differentiation, possibly by modulation of cell wall biosynthesis enzymes (Kang et al., 2000). BLP was discovered in S. clavuligerus by DNA sequence analysis of the cephamycinclavulanic acid biosynthesis gene cluster (Alexander and Jensen, 1998, Perez-Llarena et al., 1997). It is predicted to be 154 aa in length, processed from a 182 aa precursor polypeptide and secreted. Despite its apparent homology to BLIP (32% aa identity) and BLIP-I (42% aa identity), it has no detectable ß-lactamase inhibitory activity. The biological function of BLP is thus far also unclear; its gene knockout mutant (alone or in combination with a BLIP gene knockout) has yielded no apparent phenotypic differences. Unlike BLIP and BLIP-I, the BLP protein has not been previously purified nor studied directly. Despite their sequence similarity, BLIP, BLIP-I and BLP appear to differ functionally. To shed light on the reasons for this and to determine the precise structural features responsible for binding versus non-binding of TEM-1, we solved the crystal structures of BLIP-I alone and in complex with TEM-1 as well as BLP from two crystal forms. The interaction interface of BLIP is found to differ considerably from the corresponding interfaces of BLIP-I and BLP; our discussion highlights variations at key interaction hotspots and their likely consequences for TEM-1 binding. Unlike designed alanine mutants, natural variants of BLIP permit study of the effects on binding of extensive backbone rearrangements as well as numerous, simultaneous non-alanine mutations. The new structures presented also exhibit previously unreported multiple amino acid substitutions at individual binding hotspots. 80  2.2  Results and Discussion  2.2.1 Preliminary Investigation of BLP Function BLP is very likely to be secreted in its native context. Its signal peptide (and specific cleavage site; Fig. 2.1b) is predicted with a high degree of confidence by the SignalP server (Emanuelsson et al., 2007) and was able to direct the protein to the periplasm during recombinant E. coli expression. In addition, the molecular mass of BLP by SDS-PAGE and mass spectrometry suggests that the mature form is 154 aa in length. This is consistent with electron density observed in the crystal structures at the N-terminus. Previously, mutation of blp (the gene encoding BLP) has appeared to lead to no apparent physiological or morphological effects (Alexander and Jensen, 1998). Although blp appears to be efficiently transcribed (Perez-Llarena et al., 1997), null mutants in bli (the gene encoding BLIP) show no residual proteinaceous ß-lactamase inhibitory activity: thus BLP appears to lack the ability to inhibit ß-lactamases, or certainly at least the wide spectrum of these enzymes inhibited by BLIP. In vitro incubation with E. coli TEM-1, its extendedspectrum variant TEM-4, or Bacillus cereus penicillinase (Sigma P-6018) reveal no inhibition effects in nitrocefin assays. To ensure that the presence of an intact bli gene does not mask any effects of BLP mutation, we also prepared a double mutant deficient in bli and blp. No further phenotypic effects beyond those associated with the lone bli knockout were observed. Finally, because of phage display experiments suggesting that BLIP may bind PBPs (Huang et al., 1998), we attempted to determine if BLP binds another species' PBPs. Gel-shift assays however, did not indicate binding to serine ß-lactamase-like sensor domains of the Gram positive Staphylococcus aureus signaling proteins BlaR1 or MecR1.  81  Figure 2.1: Pair-wise amino acid sequence alignments generated from least-squares 3-D superposition of residues comparing BLIP with BLIP-I while in complex with TEM-1 (a) and comparing apo BLIP-I with BLP (b). Predicted secondary structure elements are shown and residues at the interaction interface (defined as those falling within 5 Å of TEM-1 in complex with it, or, for BLP, within 5 Å of TEM-1 as superposed on BLIPI) are indicated with colored stars. Residues sharing similar steric or chemical properties are boxed in blue and identical residues are also given red background. Signal peptides are aligned by sequence identity only and are numbered backward. 82  Figure 2.1 (cont’d): (c) shows the alignment of the signal peptides of BLIP and BLP. Note the greater identity between BLIP-I and BLP with respect to their signal peptides (66%) and mature protein sequences (42%; not depicted in the figure since (b) is not a direct sequence alignment) but the stark dissimilarity among their interface residues (19%). On the other hand, BLIP and BLIP-I share less mature-protein sequence identity (37%; not depicted in the figure since (a) is not a direct sequence alignment) and less signal peptide identity (44%) but a much greater relative corresponding similarity among interface residues (38%). Secondary structure elements (helices and strands) are indicated (labeled curlicues and arrows, respectively). 83  2.2.2 Structure Determination While the biological function for BLP is not yet clear, its lack of interaction with TEM-1 despite close sequence identity to BLIP and BLIP-I has interesting implications for the BLIP•TEM-1 protein-protein interaction model system. In order to determine the structural basis of interaction versus non-interaction with TEM-1 among the various BLIP family members, a host of crystal structures was solved by x-ray diffraction. The structure of BLIP-I was determined in the apo form (to 1.8 Å; PDB ID: 3GMV) and in complex with TEM-1 (to 2.1 Å; PDB ID: 3GMW). BLP was crystallized in two forms and its structure was determined to 1.7 Å (BLP SeMet derivative, phasing by single-wavelength anomalous diffraction, SAD; PDB ID: 3GMY) and to ultra high resolution ~1.0 Å (native BLP, phased using the partially-refined BLP SeMet structure as a search model for molecular replacement; PDB ID: 3GMX). In order to ensure optimal accuracy in the detailed comparison of all BLIP variants, the original data of the apo form of BLIP was refined again using current software (see Methods; PDB ID: 3GMU). Data collection and refinement statistics for all crystal structures are presented in Table 2.1.  84  Table 2.1: Crystallographic data collection and refinement statistics. Crystal Parametersa  C2 98.74 % 44.07 % 45.29 90, 104.58, 90  BLIP-I • TEM-1 complex P1 45.42 % 48.76 % 106.14 103.24, 91.29, 90.18  2.0 – 59.3  1.75 – 25.0  1.54056  BLIP (apo)  Space Group Cell Dimensions: a x b x c (Å)  C2 129.65 % 26.19 % 48.07  !, ß, " ( º )  90, 113.82, 90  BLIP-I (apo)  BLP (SeMet derivative) P212121  BLP (native), merged sets P21  101.17 % 67.20 % 41.20  35.77 % 99.93 % 41.64  90, 90, 90  90, 101.26, 90  2.1 – 24.3  1.7 – 50.0  0.91 – 29.63  1.0000  1.0000  0.9795  1.11588  16,546  146,984  250,024  231,602  1,131,540  8,609  19,263  46,703  30,474  114,297  1.9  7.6 (5.7)  5.4 (4.9)  7.6 (7.9)  9.9 (3.6)  84.6 (64.8)  91.0 (68.7)  94.7 (86.5)  96.5 (96.2)  86.1 (65.9)  I/!I  7.4 (1.8)  28.3 (11.6)  14.4 (6.5)  34.3 (10.2)  6.5 (2.3)  Rsymb  0.178  0.041 (0.094)  0.048 (0.111)  0.084 (0.230)  0.060 (0.265)  2.0 – 59.34  1.8 – 20.5  2.1 – 24.3  1.7 – 56.0  1.05 – 49.9  0.164 / 0.229  0.185 / 0.227  0.192 / 0.234  0.200 / 0.273  0.125 / 0.154  Resolution (Å) Wavelength (Å) Number of Reflections Number of Unique Reflections Average Redundancy Completeness (%)  Refinement Statisticsc Resolution Range (Å) Rwork / Rfree  rmsd bond 0.019 0.013 0.023 0.027 lengths (Å) rmsd bond 1.84 1.35 1.97 2.09 angles ( ° ) a Statistics for highest resolution shell are shown in parentheses b Rsym = &|(Ihkl)-<I>| / &(Ihkl) c 5% of reflections excluded from refinement and used to calculate Rfree  0.012 1.53  85  2.2.3 Sequence and Structure Comparisons The BLIP family fold (Fig. 2.2a) is unique: neither computational amino acid sequence-structure recognition (PHYRE; Bennett-Lovsey et al., 2008) nor 3-D structural matching software (DALI; Holm et al., 1992) yield similar folds with any reasonable degree of confidence. Some very low-probability hits were obtained using the protein structure comparison service SSM (Krissinel and Henrick, 2004): these include DNA-binding proteins, noted previously (Strynadka et al., 1996) with a similar topology (TATA-box binding protein) or a comparable large, concave ß-saddle binding surface (protein-binding transcriptional activator DCoH). All BLIP-I and BLP structures, as expected from the strong similarity of their primary sequences, share this fold, including internal pseudosymmetry consisting of two homologous domains each bearing a disulfide bridge.  86  Figure 2.2: Overall fold of BLIP (green), BLIP-I (blue) and BLP (magenta) in apo form (a) and BLIP and BLIP-I in complex with TEM-1 (b). In (a), BLIP family members are depicted as cartoons showing ß-strands (flat arrows), loops (thin noodles) and !-helices (thick cylinders) showing very strong similarity in number and position of secondary structure elements, with BLIP differing most, in particular at connecting loops. Nand C- termini are indicated. (b) depicts TEM-1 (chain a from the BLIP-I•TEM-1 complex) rendered as its solvent-excluded surface and, shown as ribbons, BLIP (green, chain b positioned by superposing its TEM-1 complex partner on TEM-1 from the BLIP-I•TEM-1 complex) and BLIP-I (blue, chain b from the BLIP-I•TEM-1 complex). Key TEM-1 activesite interacting residues of BLIP/BLIP-I (D49/D49; F142/L134,F135) are rendered as sticks (respectively, cyan/orange; brown/magenta). The molecules are rotated as indicated, with TEM-1 in (b) clipped for clarity. Note the ~4.5 Å shift by BLIP-I across the face of TEM-1 in the direction of the BLIP N-terminus and the similar insertion of residues into the TEM-1 active site cleft, despite loop differences. 87  This pseudosymmetry implies evolutionary origin from a tandemly-repeated domain. Indeed, BLIP, BLIP-I and BLP have internal sequence identity of 29%, 21% and 23%, respectively. Such self-symmetry is readily observed in the protein structure: each domain of BLIP, BLIP-I and BLP closely resembles its tandemly-repeated sister domain with an RMSD (averaged across all structures solved here) of ~0.80 Å (~52 C! pairs), ~1.1 Å (52 C! pairs) and ~1.0 Å (~50 C! pairs), respectively (Table 2.S1). The greater internal sequence and structural symmetry of BLIP may suggest it is evolutionarily nearer to the ancient gene duplication event that presumably gave rise to the BLIP scaffold. BLIP shares a greater proportion of identical amino acids with BLIP-I (37%) than with BLP (32%), while BLP possesses the greatest (42%) aa identity with BLIP-I. The differences in the degree of 3-D structural correspondence are far more pronounced. Iterative, secondary structure matching (Meng et al., 2006) revealed dramatic differences in the proportion of backbone C! atoms that could be superimposed (within 2.0 Å to give an overall RMSD <1.3 Å). Fewer than 65% of C! atom pairs correspond in the superposition of BLP on BLIP, ~75-80% in BLIP-I on BLIP and a remarkable ~91-94% in BLP on BLIPI. This strongly suggests a close evolutionary relationship between BLP and BLIP-I, rather unexpectedly as they are separated by both species and apparent function.  88  Table 2.S1: RMSD values among all observations of BLIP homologues. Internal superposition of pseudosymmetrical tandemlyrepeated domains is also listed. Total number of residues indicated in square brackets, number of superposable C-alpha atoms indicated in parentheses. Different molecules observed with the crystallographic asymmetric unit are indicated by .a, .b. Each molecule is numbered for ease of comparison. RMSD (CA atom pairs of BLIPs in complexes)  Internal Symmetry (of BLIPs only)  BLIP (apo)  BLPP21.a  BLPP21.b  0  1  2  BLIP (apo) [165] 0.71 (52) BLP-P21.a 1.10 1 0.93 (50) [154] (97) BLP-P21.b 1.17 2 1.00 (51) [154] (99) BLP-SeMet.a 1.11 3 1.00 (49) [154] (98) Blp-SeMet.b 1.14 4 1.06 (52) [154] (97) BLIP-1 (apo) 1.14 5 1.16 (52) [157] (122) BLIP•(TEM-1).b 0.60 6 0.87 (54) [165] (164) BLIP•(TEM-1).d 0.57 7 0.83 (52) [165] (164) BLIP-1•(TEM-1).b 1.17 8 1.10 (52) [157] (126) BLIP-1•(TEM-1).d 1.17 9 1.11 (52) [157] (126) vs. Different xtal condition Color key: Apo vs. (TEM-1)-complex  BLPSeMet .a 3  BLPSeMet .b 4  BLIP-1 (apo) 5  BLIP• (TEM-1) .b 6  BLIP• (TEM-1) .d 7  BLIP-1• (TEM-1) .b 8  0  0.48 (154) 0.50 0.46 (154) (154) 0.45 0.34 0.37 (152) (154) (154) 1.03 1.12 0.99 (145) (145) (138) 1.14 1.18 1.02 (95) (94) (87) 1.11 1.16 1.01 (93) (94) (87) 1.00 1.05 0.99 (145) (144) (141) 0.99 1.04 0.97 (145) (144) (141) vs. Other molecule in ASU BLP vs apo BLIP(-1)  1.08 (140) 1.10 1.18 (89) (122) 1.18 1.16 0.12 (96) (122) (165) 1.02 0.45 1.13 1.15 (141) (154) (117) (121) 1.01 0.47 1.14 1.11 (142) (155) (118) (118) vs. other active BLIP family member BLP vs (TEM-1)•BLIP(-1)  0.09 (154)  89  Comparison of signal peptide sequences supports this close evolutionary link. Signal peptides are short N-terminal sequences cleaved upon export of the nascent polypeptide chain from the cytoplasm and so do not become part of the protein final folded form. It follows that these are under lesser direct evolutionary selective pressure than residues forming an interaction interface. The frequency of random amino acid substitutions found at non-critical positions in the signal sequence could thus indicate evolutionary distance. The signal peptide of BLIP compares with that of BLIP-I (from a different species) and BLP (from the same species) with 44% and 40% sequence identity, respectively (Figs. 2.1a, c). BLIP-I and BLP, however, as with sequence and structural similarity, share substantially greater (66%) signal peptide sequence identity (Figs. 2.1b). It is, however, difficult to draw firm conclusions regarding the significance of such comparisons. Streptomyces species frequently share genetic information via horizontal gene transfer (Chater and Chandra, 2006) and, as noted above, so proportionally few related species have been studied there is no reasonable way to estimate the number of BLIP family members that exist in nature, let alone attempt to describe their phylogeny. 2.2.4 Surface Charge (APBS) and Long-Range Interactions Surface electrostatic charge generally affects the mutual affinity of interacting proteins by influencing the rate of formation of their initial encounter complex and their relative positioning within this complex (Albeck and Schreiber, 1999). However, since ionic strength does not greatly affect the rate of formation of the native BLIP•TEM-1 complex, it is thought that their association does not benefit much from long-range (>5 Å) electrostatic interactions, owing to their net negatively charged interaction surfaces (-2 e and -7 e, respectively; Albeck and Schreiber, 1999). However, their association rates can be 90  dramatically increased (up to 200-fold) by increasing the net positive charge of the BLIP interacting surface (Albeck and Schreiber, 1999, Selzer et al., 2000). BLIP-I would be expected to behave similarly as BLIP, bearing a comparable distribution of estimated negative charge at its interacting surface (Figs. 2.S1a, b). Although BLP appears to have relatively less net negative charge at this surface (Fig. 2.S1c), its apparent lack of net positive charge is consistent with its inability to bind TEM-1. We thus focus exclusively here on consideration of short-range interactions to explain the ability of each BLIP family member to associate with TEM-1.  91  Figure 2.S1: APBS modelling of charge potential of BLIP (a), BLIP-I (b) and BLP (c) at their interaction surface shows close correspondence between BLIP and BLIP-I, with a dramatic relative decrease in net negative charge observed in BLP (APBS surface is colored red, white and blue, corresponding to values of -5, 0 and +5 kT/e-, respectively, with shading by linear interpolation).  92  93  2.2.5 Overview of TEM-1 Interaction Interfaces The BLIP•TEM-1 interaction interface area is ~1266-1295 Å2, rather large among reported protein-protein interface areas (~700-1500 Å2; Lo Conte et al., 1999). The interface area of the BLIP-I•TEM-1 inhibition complex is somewhat smaller (~1177-1195 Å2), although interaction affinity in general appears to be relatively independent of interface size (Lo Conte et al., 1999). Indeed, BLIP and BLIP-I share an exceptionally high affinity for TEM-1–inhibition and binding constants Ki and Kd have reported values of ~0.1-0.6 nM (Rudgers and Palzkill, 1999, Petrosino et al., 1999, Strynadka et al., 1996) and ~0.05 nM (Kang et al., 2000), respectively. The BLIP and BLIP-I backbones are highly superposable in their respective apo forms (RMSD ~1.13 Å for 117 C! pairs) or in complex with TEM-1 (RMSD ~1.14 Å for 122 C! pairs). Upon binding TEM-1, BLIP-I undergoes a slight proximal rotation of its two homologous domains (RMSD ~0.45 Å for all C! pairs), comparable to the 5.3° increase in concavity of BLIP associated with its binding of the enzyme (RMSD ~0.59 Å for all C! pairs). It has been shown that the total number of contact atoms may be a good predictor of binding energy (Alsallaq and Zhou, 2007). This is supported by our data, which finds the number of potentially contacting, non-hydrogen atoms within 5Å of the interface of each complex (including sc and mc atoms) to be nearly identical: BLIP•TEM-1 has 625 (total for both complexes in the asymmetric unit cell) and BLIP-I•TEM-1 has 624. The free energy of binding for the respective complexes, calculated (as detailed below) from their published Ki values, is ~57 kJ/mol and ~59 kJ/mol. It thus appears that the two inhibitory BLIPs are similar with respect to sequence, structure and a modest conformational change upon binding, ultimately forming an 94  equivalent number of interatomic contacts. Given these points of comparison, it would not be unreasonable to expect a high degree of conservation in the specific protein-protein interactions at the interface with TEM-1. Yet while the number and type of interactions are largely comparable, the specific residues involved in these interactions frequently vary. Based on a structure-based sequence alignment of the TEM-1 in complex with BLIP and BLIP-I, there are 33 and 31 inhibitor residues, respectively, within 5 Å of the enzyme. Only 8 (~25%) are identical; a further 4 may be considered to have similar physicochemical properties (Fig. 2.1a). This high degree of variation in BLIP-I contact residues is accompanied by a ~4.5 Å translation of its backbone along the surface of TEM-1 in the direction of the inhibitor N-terminus (Fig. 2.2b). This yields the unexpectedly poor 3-D alignment of BLIP with BLIP-I while in complex with TEM-1 (Fig. 2.1a). The closely matching conventional alignment (of 37% identical sequences) could therefore, in the absence of experimental 3-D structural evidence, lead to a highly misleading BLIP-I•TEM-1 protein-protein interaction model. The position and number of ordered waters at the interface of each complex differs. This is in accordance with the recent suggestion (made by examining the BLIP•TEM-1 model interaction) that buried waters actually contribute little to the favorable energy of protein-protein interaction (Reichmann et al., 2008). To explain its lack of TEM-1 inhibition, BLP was superimposed on BLIP-I (its very close structural homologue) in its complex with TEM-1. BLP residues falling within 5 Å of TEM-1 were considered relevant to binding. There is a strong tendency toward residue variance at this binding surface: only 4/31 (13%) are identical (2 similar) compared with 42% overall identity (Fig. 2.1b) and 64% identical (16/25) residues buried in the protein core 95  (residue depth >1.5Å). Thus the BLIP fold may be viewed as a strongly conserved scaffold for generating diverse protein-protein interactions. 2.2.6 Examination of BLIP, BLIP-I and BLP at TEM-1 Interaction Hotspots Six energetically independent interaction “clusters” (C1-C6) have been identified at the BLIP•TEM-1 interface (Reichmann et al., 2007a; Fig. 2.3a). The change in the Gibbs' Free Energy of binding upon mutation to alanine [!!G(mut)] has been previously determined for a substantial number of such so-named 'hotspot' residues and these data are summarized in Table 2 (Wang et al., 2007, Reichmann et al., 2007a, Reichmann et al., 2005, Albeck et al., 2000, Petrosino et al., 1999, Albeck and Schreiber, 1999). Considerable differences are found among these cluster residues in the BLIP-I•TEM-1 complex. Specific interactions at these sites in both TEM-1 complexes (with BLIP and BLIP-I) are compared below. Also, those residues likely to interfere with BLP binding of TEM-1 are identified at these clusters by superposition of its structure on the far more closely related BLIP-I in complex with TEM-1.  96  Figure 2.3: View of the ß-lactamase-binding interfaces of BLIP (a) and BLIP-I (b) and corresponding interface of BLP (c). Alternate backbone conformations are indicated by overlaid translucent cartoons. Key interface residues are depicted as thick sticks (for (a) and (b), one of the molecules from each complex with TEM-1 was arbitrarily chosen) with alternate conformations shown in translucent wireframe (in (a) and (b) the largest deviations indicate the apo position). N- and C-termini are labeled. Residue non-carbon atoms are colored CPK; carbon atoms are colored by interaction cluster: C1 – cyan, C2 – brown, C3 – pink, C4 – purple, C5 – gold, C6 – grey. Both C3 and C4 cluster residues are largely absent from BLIP-I and BLP. The overall fold indicates strong correspondence in the number and position of secondary structure elements among all three proteins. Note the strong similarity of BLIP-I and BLP in all four extended loops, at least two of which differ substantially (in length or positioning) in BLIP.  97  98  2.2.6.1  C1 – favorable interactions by all BLIPs  In BLIP•TEM-1, C1 is centered around a loop bearing the key BLIP interacting residue D49 (Table 2.2, Fig. 2.3a), which mimics the carboxyl group common to nearly all ßlactams (Strynadka et al., 1996). It is positioned to form hydrogen bonds with TEM-1 residues S130, K234, S235 and R243 (Fig. 2.4a). BLIP-I bears an equivalent D49 residue (Fig. 2.3b) and several other residues in this interaction loop are conserved in sequence, if not in 3-D position (G48, Y50, Y53). While the hydrogen bonding network is not precisely conserved in the BLIP-I•TEM-1 interface, the number of potential TEM-1 hydrogen bonding partners is maintained (S70, S130, N132, K73) (Fig. 2.4a). Indeed, the reported Ki of the BLIP-I-D49A mutant is ~10 nM (Kang et al., 2000) which gives a calculated !!G(mut) ~13 kJ/mol, indicating a degree of interaction for this residue possibly greater than its equivalent in BLIP (!!G(mut) = 7.7 kJ/mol). Y50 in BLIP•TEM-1 points away from the enzyme binding pocket with a net detrimental effect on binding energy (its alanine mutant has a favorable !!G(mut) = -3.2 kJ/mol; Petrosino et al., 1999). Yet in BLIPI•TEM-1, Y50 points into the active site, forming a number of favorable interactions there, including hydrogen bonds with TEM-1-S70, -N132 and possibly -K73 (Fig. 2.4a).  99  Table 2.2: Key binding residues in BLIP, their loss of energy upon mutation !!G(mut) (Reichmann et al., 2007a) and their nearest corresponding BLIP-I and BLP residues. All values are expressed in kJ/mol. Module  BLIP aa  C1  D49A  C2  C2  K74A  F142A  nearest BLIP-I aa(s) D49  E70/ N109/ R154  Y71/ F135  nearest BLP aa Y47  L106  F132  !!G(mut)§ 7.7  16.6  10.1  TEM-1 contact aa(s)  !!G(loss)†  !!G(penalty)‡  S130 K234 R243 S130/S235/ R243 S130/K234/ S235/R243  1.5(a) 6.7(a) 4.6(a)  6.2 1 3.1  3.7(a)  4  7.1(a)  0.6  E104  6.6(b)  10  Y105 E104/Y105  13.9(b) 5.8(b)  2.7 10.8  E104  11.5(b)  -1.4  Y105 2.9(b) 7.2 E104/Y105 6.3(b) 3.8 C3 H148A none none 12.3 Q99 14.7(c) -2.4 N100 9.7(c) 2.6 C3 W150A none none 18.3 Q99 15.8(c) 2.5 N100 18.7(c) -0.4 C3 R160A none none 9.7 Q99 15.4(c) -5.7 N100 7.7(c) 2 C4 W112A none none 12.1 n/a n/a C4 W162A none none 10 V103 17.7(c) -7.7 E168 8.8(c) 1.2 C5 F36A I37 R37 14 P107 n/a M129 13.8(c) 0.2 C5 H41A L42 I39 14.4 P107 10.9(c) 3.5 C5 Y53A Y53 Q50 8.1 P107 9.9(c) -1.8 M129 15.4(c) -7.3 C6 S71A K125 R65 1.7 E110 21.0(c) -19.3 !!G(mut) >0 reflects a mutation having a net detrimental effect to binding; !!G(mut) <0 would indicate a net benefit to binding. !!G(loss) for the double mutant (BLIP and TEM-1) residues indicated is used to calculate !!G(penalty) for the unpaired TEM-1 residue resulting from the single mutant in BLIP. n/a, measurement not available. § -values from Reichmann et al., 2007a, Table 1 † -!!G(mut) for double/multiple mutant, values from: (a) Reichmann et al., 2005, Table 1; (b) Reichmann et al., 2007a, Table 3; (c) Reichmann et al., 2007a, Table 2. ‡ calculated for TEM-1 residue(s) indicated, left unpaired by mutation to alanine of BLIP contact residue.  100  Figure 2.4: Pairwise comparisons of BLIP vs. BLIP-I (a, c, e) and BLIP-I vs. BLP (b, d, f) at key BLIP•TEM-1 binding modules C1 (a, b), C2 (c, d) and C5 (e, f). In all panels TEM-1 and BLIP-I are chains a and b from the BLIP-I•TEM-1 complex. BLIP (green) is positioned relative to superposed TEM-1 (white) from the structure of its TEM-1 101  complex; TEM-1 varies little whether in its apo form [PDB: 1BTL] or when bound to either BLIP [PDB: 1JTG] or BLIP-I (RMSD <0.4 Å comparing >98% C! pairs in all cases). BLP (magenta) is superposed on BLIP-I as described in Section 2.3 (Materials and Methods). Certain secondary structure elements are rendered translucently in panels (a-d) and for clarity are omitted and rendered as noodles only in panels (c, d). TEM-1 residue E104 is shown in three alternate conformations in (c, d), with the lone conformation from the BLIP complex drawn in white, the two from the alternate conformations in BLIP-I in light blue. Note the close correspondence in the number and type of contacts made by BLIP and BLIP-I at all three clusters; there is some similarity between BLIP-I and BLP at C1, little at C2 and scarcely any at C5.  Most other interactions are likewise equivalently traded off between BLIP and BLIP-I (e.g. in BLIP, G48 interacts with TEM-1-Y105, while in BLIP-I it interacts with TEM-1V216). Additional hydrogen bonds are, however, made uniquely by this loop in BLIP-I, most notably BLIP-I-D49•TEM-1-A237 (potentially two sc-mc hydrogen bonds), BLIP-IA51•TEM-1-Y105 (mc-sc). Aromatic interactions involving Y50 in BLIP involve the TEM-1 residue M129 (a "-sulfur interaction; Ringer et al., 2007). BLIP-I-Y50 interacts with TEM-1Y105 ("-"; Sinnokrot and Sherrill, 2006) and TEM-1-K73 (cation-"; Dougherty, 1996). These interactions are depicted in Fig. 2.4a. Interestingly, the position of Y50 in the BLIPI•TEM-1 complex closely matches (with RMSD = 0.7 Å for all 12 atom pairs) that of Y50 in BLIP complexed with the D104K mutant SHV-1 enzyme (compare this with RMSD = 7.4 Å versus the Y50 position in the BLIP•TEM-1 complex; Reynolds et al., 2006). BLP-Y47 (Fig. 2.3c) would appear able to form several hydrogen bonds within the C1 cluster, specifically with TEM-1-S70, -K73 and -K234 (Fig. 2.4b). 2.2.6.2  C2 – favorable BLIP-I interactions; limited BLP interactions.  Interaction cluster C2 is dominated by the interaction of two BLIP residues: F142 and K74 (Table 2.2, Fig. 2.3a). F142 most notably makes hydrophobic contacts with TEM-1 102  residues Y105 and P167, including potential "-" stacking interactions with Y105 (Fig. 2.4c). K74 forms a key salt bridge with E104 (Reynolds et al., 2006), a possible cation-" interaction with Y105 as well as a hydrogen bond with F142 (sc-mc); guiding F142 into the TEM-1 active site has been postulated to be a major function of K74 (Wang et al., 2007). This two-residue BLIP cluster is largely replaced in BLIP-I by F135, E70 and Y71 (Fig. 2.3b). BLIP-I-F135 partly substitutes for BLIP-F142, hydrophobically interacting with TEM-1-Y105 and -P167 (Fig. 2.4c). Aromatic stacking ("–") interactions with TEM-1-Y105 involve BLIP-I-Y71. Several BLIP-I residues serve to satisfy the electronic requirements of E104; the salt bridge made by BLIP-K74 is replaced by E104 hydrogen bonding with several possible residues: BLIP-I-E70, -Y71 or -N109. Also, depending on the conformation of TEM-1-E104 (Fig. 2.4c), BLIP-I-R154 forms a salt bridge with either TEM-1-E104 or E168. In BLP the only interactions that would appear to be maintained involve BLP-F132 (taking the place of BLIP-F142 or BLIP-I-F135; Figs. 2.3c, 2.4d). No other interactions may form here: M67 takes the place of the primarily hydrogen bonding residue BLIP-I-E70; BLP-E68 replaces Y71, abolishing "-" stacking; BLIP-I-N109 is replaced by BLP-L106, which cannot favorably interact with charged residue E104. 2.2.6.3  C3 – no BLIP-I nor BLP interactions  Binding module C3 in BLIP comprises mainly residues H148, W150 and R160 (Table 2.2, Fig. 2.3a), making contacts primarily with the TEM-1 backbone in this region (most importantly, the BLIP-R160 guanidinium nitrogen hydrogen bonds with the TEM-1-N100 backbone amide oxygen). However, no BLIP-I residues make contact with TEM-1 (residues Q99, N100) here nor are any BLP residues (at least by structural alignment to BLIP-I) close 103  enough to make contact. It is interesting to note, however, that the most dramatic reported loss of BLIP•TEM-1 binding energy upon mutation at this module is thought to result from backbone rearrangements arising from the loss of the W150 side chain (Wang et al., 2007). 2.2.6.4  C4 – no favorable BLIP-I nor BLP interactions  Hydrophobic contacts BLIP-W112•TEM-1-L102 and BLIP-W162•TEM-1-V103 and BLIP-W162•TEM-1-P167 (binding module C4; Table 2.2, Fig. 2.3a) have no equivalents in BLIP-I (which has a positively-charged R154 at this position; shown as part of C2 in Fig. 2.3b) or in BLP (which features S151 here; see Fig. 2.3c). 2.2.6.5  C5 – favorable BLIP-I interactions; unfavorable BLP interactions  The binding hotspot cluster comprised of key BLIP residues F36, H41, Y53 (C5; Table 2, Fig. 2.3a) forms predominantly hydrophobic interactions (Wang et al., 2007) with TEM-1 residues V108, M129 and P107 (Fig. 2.4e). BLIP-I in fact increases the hydrophobicity of contact-forming residues in this key binding region by substituting V36, I37, L42, Y53 (Figs. 2.3b, 2.4e). As well, BLIP-I-Y53 adds an aromatic ("-") interaction with TEM-1-Y105 (only making hydrophobic contact to BLIP-Y53 since the distance and angle are inappropriate for "-bond stacking) and maintains the hydrophobic Y53-P107 interaction. Both BLIP-Y50 and BLIP-I-F44 are able to form an aromatic-sulfur interaction (not shown) with TEM-1-M129 as well. In contrast with the inhibitory BLIPs, BLP features several polar and charged residues in this region, including Q34, R37 and Q50 (Figs. 2.3c, 2.4f). It should be noted that hydrophobic residues BLP-I39 and -Y41 are closely matched with BLIP-I-L42 and -F44, respectively, and a distance of only ~2 Å separates the ring centres of BLP-H52 and BLIP-I104  Y53. However, the presence of polar and charged residues at this module would likely be highly destabilizing to any potential BLP•TEM-1 interaction. 2.2.6.6  C6 – favorable interactions by all BLIPs  Module C6 is most reliant upon the TEM-1 residue E110, which results in loss of 15.4 kJ/mol of interaction energy when mutated to alanine. Here, BLIP forms hydrogen bonds via the side chain hydroxyl groups of S71 and (in one molecule of the ASU) S113 (Fig. 2.3a). Loss of these hydroxyl groups does not substantially affect binding (Table 2) (Wang et al., 2007), however, and in any case BLIP-I and BLP both have positively-charged residues (K125 and R65, respectively; Figs. 2.3b, c) that could interact favorably with the electronegative TEM-1-E110. Specifically, the side-chain amine of K125 and guanidinium of R65 are each positioned to hydrogen bond with the E110 side-chain carboxyl group and conceivably aid in electrostatic attraction and positioning during initial complex formation. 2.2.7 Binding Energetics and Energetic Penalty From the structural analysis presented above, it appears that BLIP-I interactions at BLIP binding modules C1, C2, C5 and C6 are highly favorable, although interactions at C3 and C4 are absent. In contrast, when BLP is placed (by 3-D superposition) relative to TEM1 at a position equivalent to BLIP-I, it appears capable of interacting only at C1 and C6, with some possible partial interaction at C2. What can this tell us about the requirements for TEM-1 binding? Does this indicate that C3 and C4 are dispensable (since BLIP-I, without contacts here, still binds TEM-1 tightly) and that C2 and C5 are essential (since BLP lacks these) for complex formation? To begin to address these questions, we compared data from several published mutagenesis studies on the BLIP•TEM-1 complex. Reported values include the dissociation 105  constant Kd by surface plasmon resonance (SPR; Reichmann et al., 2007a, Reichmann et al., 2005), the association constant K and thermodynamic parameters by isothermal calorimetry (ITC; Wang et al., 2007) and the inhibition constant Ki measured by nitrocefin-monitored enzyme assay (Zhang and Palzkill, 2003, Zhang and Palzkill, 2004). Some studies measure binding affinity and some report inhibition data. In order to discuss these data interchangeably here, we need to establish that binding and inhibition are likely to represent the same phenomenon. The BLIP•TEM-1 association constant K (=1/Kd) measured by displacement ITC is 9.0 x 109 M-1 (Wang et al., 2007). This gives a calculated Ki of ~0.1 nM, within the reported range directly measured by enzyme inhibition assays. Comparing all available wild-type and mutant data, !G values calculated from inhibition studies (Ki) correlate reasonably well with those calculated from binding (ITC) studies (K) with a slope of 1.02, R2 = 0.777 (Fig. 2.5). This suggests that binding of TEM-1 by BLIP is indeed equivalent to inhibition. The relationship !G = RT•ln(Ki) = RT•ln(Kd) can be used to show that each !!G = +6 kJ/mol corresponds to an approximate order of magnitude loss of binding affinity and inhibition (i.e. a 10-fold increase in the value of Kd or Ki). Thus changes to BLIP (either mutations or differences among homologues) giving a combined !!G >40kJ/mol will reduce binding affinity for TEM-1 from the sub-nM to the mM range, abolishing its physiological relevance as an inhibitor. From the sub-nM values reported for the binding and inhibition constants, the net free energy change of the BLIP•TEM-1 interaction ranges from !G = -56.8 to !G = -59.0 kJ/mol.  106  Figure 2.5: Scatterplot of change in Gibbs Free Energy upon interaction of BLIP and TEM-1 calculated from enzyme inhibition [!G(Ki)] vs ITC binding studies [!G(K)] and corresponding linear regression (y = mx+b; m = 1.02, b = 0; R2 = 0.777). K values were converted to !G to account for the varying temperatures at which ITC experiments were conducted. The most dramatic outliers are those for which K values were measured at temperatures other than 25 ºC (6, 15 or 30 ºC), reflecting complex effects of temperature on the thermodynamics of binding; many outliers also suffer from the high measurement error intrinsic in displacement ITC (Wang et al., 2007).  107  Although 59.0 kJ/mol would thus appear to be the upper limit loss of binding energy, the sum of the reported mutagenesis values is !!G(mut) = 135 kJ/mol (Table 2.2). Indeed, the binding energy contributed by amino acid side chains within a cluster is often reportedly sub-additive, i.e. mutation of multiple intra-cluster residues results in a !!G(mut) up to 70% lower than that expected from the sum of the !!G values of single mutants (Reichmann et al., 2007a). For example, C2 mutants BLIP-K74A and TEM-1-E104A have !!G(mut) values of 16.6 and 6.5 kJ/mol, respectively, compared to the wild-type complex (Reichmann et al., 2007b). Mutating both residues at once yields a !!G(mut) of only 6.6 kJ/mol (Table 2.2). Thus it is clear that the total loss of a sc-sc interaction (by double alanine mutation) may, in some cases, be less detrimental to binding than the energetic penalty incurred by a lone unpaired interface residue (resulting from a single mutation); in the mutant BLIP-K74A only the positive charge at the interface is lost, placing the highly disfavored unpaired negative charge on TEM-1-E104 in an otherwise hydrophobic region of the interface (BLIPY143, -F142; TEM-1-Y105; Fig. 2.4c). We define here a conceptually useful quantity called “energetic penalty” for a mutation leading to a disfavored unpaired interaction: !!G(penalty) = !!G(mut) !!G(loss), where !!G(mut) is that observed for the single mutant and !!G(loss) is that observed for the multiple mutant (i.e. of the single-mutant residue and its key partner(s), representing the approximate value of the favorable interactions). For the mutation BLIPK74A described above !!G(penalty) = 10.0 kJ/mol (Table 2.2). Other examples of energetic penalty arising from mutations to the BLIP•TEM-1 system include (for residue pairs BLIP/TEM-1, respectively) the C1 hydrogen bonding pair D49/S130 (!!G(penalty) = 6.2 kJ/mol; Fig. 2.4a), the C2 hydrophobic contact pair F142/Y105 (7.2kJ/mol; Fig. 2.4c) 108  and the C5 pair H41/P107 (Fig. 2.4e); examples occur at every interaction cluster except C6 (Table 2.2). These support a general hypothesis that each measured unfavorable !!G(mut) for a single mutant in a protein-protein interaction in fact has two components: the loss in favorable binding energy, !!G(loss), and the introduction of disfavored unbalanced electrostatic charges or exposed hydrophobic surface area, !!G(penalty). Breaking !!G(mut) into its components permits a more nuanced discussion of the strong affinity of TEM-1 for BLIP and BLIP-I and its indifference toward BLP. !!G(penalty) at cluster C2 is particularly illustrative: unlike BLIP-I, BLP has no residues positioned to hydrogen bond or ion pair with the negative charge on TEM-1-E104. In addition, the charge on E104 is not only unpaired, it is faced with the hydrophobic residue BLP-L106 (Fig. 4d). Hydrophobic regions have a higher dielectric constant, resulting in higher potential energy (i.e. greater disfavorability) of charges placed there. Due to the greater hydrophobicity of leucine versus alanine, it is likely that BLP contact with TEM-1 would thus be highly disfavored here, likely in great excess of that corresponding to the !!G(penalty) for BLIP-K74A. There is a similar case in module C5, where !!G(penalty) = 3.5 kJ/mol in BLIPH41A, which removes the binding partner for TEM-1-P107 (Table 2.2). As with cluster C2, !!G(penalty) would likely be considerably increased by the introduction of unpaired polar and charged residues on one side of the interface. BLIP-I likely avoids such a binding penalty, featuring residues of modestly increased hydrophobicity in this region: namely BLIP-I-V36 in place of BLIP-S35 (Fig. 2.4e). BLP, on the other hand, presents several likely penalizing unpaired polar or charged residues Q34, R37 (placed near V216 in alignment with BLIP-I) and Q50 (near P107; Fig. 2.4f). 109  These examples of energetic penalty help to explain how a binding interface can be both high-affinity and strongly selective; binding incompatibility may be introduced with a relatively small set of high-penalty mutations, in particular where one member of a charge pair is replaced with a hydrophobic residue. Site-directed mutagenesis experiments would ideally test this hypothesis and its converse: artificial, non-alanine penalizing mutations introduced into BLIP/BLIP-I to abolish binding, and, conversely, non-alanine mutations introduced into BLP to relieve certain penalties described above and perhaps enable it to bind TEM-1. We have here suggested a framework for analyzing the probable effects of introducing non-alanine BLIP-I or BLP residues in place of a BLIP hotspot residue. More generally, this may permit more detailed interpretation and provide for further hypotheses to test in any protein-protein interaction system for which double mutant cycle data are available. 2.2.8 Concluding Summary BLIP and BLIP-I are homologous inhibitory proteins with similarly sized binding surfaces. Each protein has a sub-nM affinity for TEM-1 ß-lactamase, and forms an equivalent number of interactions with this enzyme. The specific interface formed by each, however, is dissimilar: there is a 4.5 Å lateral translation of BLIP-I relative to the BLIP position at the TEM-1 surface, structural dissimilarity between BLIP-I and BLIP (especially in contrast with the greater resemblance of BLIP-I to the non-TEM-1-binding BLP) as well as relatively low conservation (~25%) of residues making contact with TEM-1. Thus the BLIP fold represents a versatile framework upon which at least two distinct, high-affinity ßlactamase inhibitor interfaces may be built. The fold also permits strong specificity: BLP 110  bears even closer sequence and structural resemblance to BLIP-I than to BLIP yet shares half as many (~13%) identical potentially interacting residues with BLIP-I and exhibits no binding affinity for TEM-1. In contrast, a high proportion of residues buried in the protein core are conserved, maintaining the integrity of the BLIP fold itself. Interaction cluster comparisons and the concept of energetic penalty proposed here provide a framework for the understanding of the binding interface specificity among the BLIP family members and will hopefully prove useful in the study of other model proteinprotein interaction systems.  111  2.3  Materials and Methods  2.3.1 Molecular Biology 2.3.1.1  Construction of a recombinant plasmid for expressing bliA in E. coli  To produce functional BLIP-I in E. coli, an expression system was developed using pET-30a(+) vector (Novagen) that has an N-terminal His-tag and enterokinase cleavage site. The structural gene of bliA encoding mature BLIP-I was amplified by PCR from Streptomyces exfoliatus SMF19 using the primers BLIPEXP1 (forward strand): (5'-GTCGCGGGTACCGACGACGACGACAAGAATTCGGGCTTTTCGGCC-3') BLIPEXP2 (reverse strand): (5'-GGCGTCGGATCCTCAGGTCAGGCTGCGCTGGTAGCGGTACGTCAG-3'). The primers were designed to incorporate KpnI and BamHI sites, respectively. The KpnI/BamHI-digested PCR fragment was inserted between KpnI and BamHI sites of pET30a(+), creating pSMF1130. The positioning of bliA in pSMF1130 allowed the gene to be expressed by induction of the T7lac promoter with IPTG and the His6 tag facilitated the purification of BLIP-I using either a nickel- or cobalt-based affinity column. 2.3.1.2  Construction of a recombinant plasmid for expressing blp in E. coli  The blp gene (GenBank accession no. AF073897) was amplified from Streptomyces clavuligerus ATCC 27064 by PCR using the primers blp-fwd: (5'-GTTCCTTCATATGGTGAAGAAGACATGG-3') blp-rev: (5'-CTGCTATGGATCCCTTCAGGGCAGTTGG-3') with a plasmid construct carrying the ccaR-orf11-blp group of genes (Alexander and Jensen, 1998) as template. The amplified DNA fragment, flanked by NdeI and BamHI 112  restriction sites (shown in underline in bold font), was cloned into pET26b to give pET26bblp. The fidelity of the insert was verified by DNA sequence analysis. 2.3.1.3  Mutation of blp  To probe its physiological role, the blp gene, contained within a 3.7 kb EcoRI-BamHI fragment of S. clavuligerus ATCC 27064 genomic DNA cloned into pTZ18R, was mutated by insertion of a blunt ended apramycin resistance gene cassette (Alexander and Jensen, 1998) into the NotI site (also made blunt) near the 5' end of the gene. The resulting plasmid was converted into a Streptomyces-E. coli shuttle vector by fusion with the Streptomyces vector, pJOE829, at the HinDIII site and then transformed into S. clavuligerus whereupon the resident blp gene was replaced with the apr-disrupted version by homologous recombination to give blp::apr mutants. 2.3.1.4  Mutation of bli  A bli::tsr mutant was available from a previous study (Thai et al., 2001). 2.3.1.5  Use of protoplast fusion to prepare bli::tsr-blp::apr double mutants.  Protoplasts of both bli::tsr and blp::apr mutant strains of S. clavuligerus were prepared as described previously (Thai et al., 2001). Approximately 107 viable protoplasts of each mutant type were combined, sedimented by centrifugation at 3000 g for 7 min and washed once with 5 ml of P-buffer (Thai et al., 2001; specific composition noted in Zhang et al., 1997: 10.3% (w/v) sucrose, 0.025% (w/v) K2SO4, 0.202% (w/v) MgCl2, 0.005% (w/v) KH2PO4, 0.368% (w/v) CaCl2, 0.2 % (v/v) trace elements solution (from Okanishi and Gregory, 1970: per liter, 40 mg ZnCl2, 200 mg FeCl3•6H2O, 10 mg CuCl2•2H2O, 10 mg MnCl2•4H2O, 10 mg (NH4)6Mo7O24•4H2O, and 10 mg Na2B4O7•10H2O), 25 mM TES (N113  Tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid) buffer, pH 7.2). The sedimented protoplasts were resuspended gently in 0.8 ml of a 1:1 mixture of P-buffer and molten polyethylene glycol 1000 (cooled to 21 ºC after mixing). After 2 min at 21 ºC the protoplast mixture was diluted in P-buffer and aliquots were plated on R5B agar plates (Thai et al., 2001). After 40 h at 28 ºC, the regenerating protoplasts were overlaid with thiostrepton (thio) and apramycin (apr) at final concentrations of 5 µg/mL and 25 µg/mL respectively to counterselect the parental strains. After an additional eight days of incubation at 28ºC, developing fusants were patched onto MYM (Thai et al., 2001) plates and MYM+apr+thio plates. Almost all of the fusants grew on both types of plates. To eliminate heterokaryons and check for sporulation ability, the aprR-thioR fusants were plated onto ISP-4 (Difco) medium. No 'bald' colonies were noted and when a dilution series of the resulting spores was plated onto MYM+apr+thio, all aprR thioR progeny sporulated normally. The bli::tsr-blp::apr genotype of the double mutants was confirmed by Southern analysis. Genomic DNA from wild type S. clavuligerus, from the parental blp::apr and bli::tsr single mutants and from presumptive bli::tsr-blp::apr double mutants was digested with KpnI, separated by agarose gel electrophoresis and transferred to nylon membranes. Analysis with a radioactively labeled probe internal to the blp gene gave the expected band sizes of 3.7 kb for the wild type and bli mutant and 5.2 kb for the blp mutant and for four independent bliblp double mutants. Analysis with a probe internal to the bli gene gave the expected band sizes of 5.5 kb for the wild type and blp mutant and 6.5 kb for the bli mutant and the four bliblp double mutants, thus confirming the identity of the double mutants.  114  2.3.2 Protein Production and Purification 2.3.2.1  BLIP-I  Mature BLIP-I (NBCI Entrez Protein accession no. AAF74027) was recombinantly produced in Escherichia coli with a cleavable His6 tag at its N-terminus. Plasmid pSMF1130 was transformed into E. coli BL21(DE3)pLysS. Overnight cultures were grown with shaking in 40ml of Luria-Bertani (LB) medium at 37 ºC in the presence of 34.5 "g/ml chloramphenicol (chl) and 50 "g/ml kanamycin (kan). The 40ml overnight cultures were used to inoculate 2 L quantities of LB medium containing 34.5 "g/ml chl and 50 "g/ml kan. The cultures were then grown with shaking at 37 oC until A600 = 0.5-0.7. For induction, IPTG was added to each culture to a final concentration of 1 mM and the cultures were then allowed to grow an additional 3 h. Following the 3h induction, the cells were harvested and resuspended in 50 ml of ice-cold binding buffer (5 mM Tris-HCl, pH 8.0 and 500 mM NaCl). The cells were then sonicated to disrupt the cells and shear chromosomal DNA and insoluble material was pelleted by centrifugation. The insoluble material was resuspended in 50 ml of binding buffer containing 6 M urea and any residual insoluble material was removed by ultracentrifugation after 1 h incubation at 4 ºC. The supernatant was purified using a 1 ml HisTrap column (Pharmacia) under denaturing conditions according to the manufacturer's instructions. Binding buffer containing 10 mM imidazole was used to remove loosely bound protein from the column. His-tagged rBLIP-I was eluted using binding buffer containing 200 mM imidizole, pH 8.0. Purified His-tagged protein was cleaved by enterokinase (Novagen) to remove the His-tag and then rBLIP-I was purified via Mono Q chromatography.  115  2.3.2.2  BLP  BLP (NBCI Entrez Protein accession no. AAC32496) was recombinantly produced in E. coli BL21(DE3) using vector pET26b, without an affinity tag, fused to its native signal peptide for secretion into the periplasm. Following inoculation from an overnight starter culture, cells were grown at 37ºC, shaking at 200 RPM to OD600 ~0.5 and induced with 0.5 mM IPTG for 3 h at 37 ºC. The periplasmic fraction was obtained by osmotic shock: cells were centrifuged 20min at 4000g and resuspended in buffer R (30 mM TrisCl pH 8.5, 20% sucrose), 80mL/g wet cells; while stirring on ice, 0.5 M disodium EDTA was added dropwise to a final concentration of 1 mM and stirred ~10 min; cells were centrifuged 20 min at 8000 g and resuspended in ice cold 5 mM MgSO4 (80 ml/g wet cells) and stirred ~10 min; cells were centrifuged 20 min at 8000 g, the pellet was discarded and the supernatant (containing the periplasmic fraction) was concentrated in a stir cell concentrator (10K MWCO cellulose membrane; Amicon). 1 M TrisCl pH 8.5 and 5 M NaCl were added dropwise to 30 mM and 150 mM, respectively (buffer A). Periplasmic fraction was concentrated to 10 mg/ml (using the stir cell and centrifugal concentrators (10K MWCO, 15 ml and 0.6 ml versions; Millipore). The sample was purified by gel filtration fast protein liquid chromatography (FPLC) using an AKTA-FPLC equipped with a Superdex 75 10/30 column in buffer A, 0.5 ml/min @ 1.3 MPa (multiple, 500 "l injections). Molecular weight and purity (>95%) was confirmed by SDS-PAGE. 2.3.3 Enzyme Inhibition Kinetics and Electromobility Shift Assays Enzyme inhibition studies were performed for BLP using nitrocefin (Oxoid) and TEM-1, TEM-4, or Bacillus cereus penicillinase (Sigma P-6018). Each enzyme stock was diluted to a final concentration of 5 nM in buffer containing 50 mM sodium phosphate 116  buffer pH 7.0, 150 mM NaCl plus 1 mg/ml bovine serum albumin (B. cereus enzyme diluted in buffer from manufacturer containing zinc salts necessary for metallo-ß-lactamase activity). BLP stock solution was added to final concentrations of 5, 50, 500, or 5000 nM (100 "l final reaction volume). Following incubation for 5, 20, or 60 minutes at 21 ºC, the rate of nitrocefin hydrolysis was measured spectrophotometrically at 500 nm wavelength. No difference was discernable from BLP-free controls, so inhibition constants could not be calculated. To determine if BLP binds penicillin-binding-protein type sensor proteins, BLP was mixed in a 1:1 molar ratio (1000 ng each) with the serine ß-lactamase-like sensor domains of BlaR1 and MecR1 from Staphylococcus aureus +/- 1 mg/ml bovine serum albumin at 4 ºC (incubated for 18 h) or at 21 ºC (incubated 1 h) in 30 mM TrisCl pH 8.5 or 50 mM phosphate pH 7.0 (all buffers contain 150 mM NaCl). After incubation, BLP-containing samples were compared with BLP-only and BlaR1-/MecR1-only controls and observed for mobility shift on native PAGE gels (12% acrylamide) run at 100 V for 2 h (pH 6.8) or 5 h (pH 8.8). Duplicate gels were run in reverse polarity. 2.3.4 Crystallization, Data Processing and Structure Refinement The structure of BLIP unbound to TEM-1 was reported (Strynadka et al., 1994) prior to widespread use of the Protein Data Bank (Berman et al., 2000, Berman et al., 2003) and has not been deposited since. Original data for this apo form of BLIP was available in the NCJ Strynadka laboratory and was used in the re-refinement of the structure. Initial phase estimates were given by molecular replacement with Phaser (McCoy, 2007) using 1JTG chain b as a search model. All models (BLIP, BLIP-I and BLP) were built, adjusted and validated using Coot (Emsley and Cowtan, 2004) and refined using REFMAC (CCP4, 117  1994) and/or phenix.refine (Adams et al., 2004). ShelXL (Sheldrick, 2008) was used to validate refinement for the 1.0 Å-resolution BLP data. 2.3.4.1  BLIP-I  The hanging drop vapor batch method at 21 ºC was used to crystallize BLIP-I alone (20 mg/ml protein in 50 mM TrisCl pH 8.0, 0.05 M NaCl mixed 1:1 with 0.1 M TrisCl pH 8.5, 30% PEG 4000, 0.3 M lithium chloride) and in complex with TEM-1 (10 mg/ml of each protein in 50 mM TrisCl pH 8.0, 0.05 M NaCl mixed 1:1 with 0.2 M potassium phosphate pH 4.7, 20% PEG 3350). Data were collected at the Advanced Light Source synchrotron facility at Lawrence Berkeley National Laboratories on Beamline 8.2.1 using the Mar345 CCD detector. Phasing was performed by molecular replacement with Phaser (McCoy, 2007) using 1JTG chain b as a search model. Data collection strategies, unit cell refinement and data reduction was performed using the HKL2000 package (HKL Research). Models were built and refined as described above for BLIP. 2.3.4.2  BLP  Crystals of BLP were obtained in microbatch format at 21 ºC (drops under 0.6 ml Al's oil, covered in paraffin oil) at a protein concentration of 11 mg/ml in buffer A (0.03 M tris pH 8.5, 150 mM NaCl ) mixed 1:1 with 0.1 M TrisCl pH 8.5, 0.2 M sodium acetate trihydrate, 30% PEG 4000 (space group P212121). Crystals diffracting to very high resolution (space group P21) were grown as above mixed 1:1 instead with 0.1 M sodium cacodylate pH 6.5, 0.2 M magnesium acetate tetrahydrate, 20% PEG 8000. Molecular replacement was unsuccessful, so selenomethionine derivative crystals were grown via hanging drop using the P212121 condition (cryoprotection in mother liquor + 50% PEG 1000) described above and used for phase estimation by single-wavelength anomalous diffraction (SAD) followed by 118  solvent flattening with ShelX-C/D/E (Schneider and Sheldrick, 2002). Eight heavy atom sites were clearly identifiable (4 methionine residues in BLP, two molecules in the asymmetric unit) and, by inversion of the hand, yielded an interpretable map. Data for the P212121 selenomethionine derivative were collected at ALS Beamline 8.2.2 using the ADSC Q315 CCD detector. For the P21 crystal form, no cryoprotectant was used and two data sets (at 2# = 0º and 2# = 15º, respectively) were collected at the ALS Beamline 8.3.1 employing the ADSC Q210 CCD detector. The two P21 data sets were merged using HKL2000. For sub-Ångstrom data, it is generally recommended to employ refinement tools derived from those developed for small molecule crystallography, namely ShelXL (Sheldrick, 2008). This tool was not found to provide a distinct advantage (in terms of Rwork/Rfree or in map quality) versus REFMAC in the case of the BLP P21 data, so REFMAC alone was used for refinement and models were built as above. Hydrogens were included in both REFMAC and ShelXL refinements; occupancies were refined in ShelXL; unrestrained atomic refinement was attempted but led to unrealistic bond geometries. Anisotropic B factors were refined for both crystal forms. 2.3.5 Accession Numbers Coordinates and structure factors have been deposited in the Protein Data Bank with accession numbers 3GMU (BLIP), 3GMV (BLIP-I apo form), 3GMW (BLIP-I•TEM-1 complex), 3GMX (BLP native, 1 Å structure) and 3GMY (BLP Selenomethionine derivative).  119  2.3.6 Computational analysis Primary structure alignments and three-dimensional superposition was performed by the iterative, secondary structure-based matching algorithm MatchMaker (Meng et al., 2006) within UCSF Chimera (Pettersen et al., 2004). Protein-protein interaction surface area was determined by PISA (www.ebi.ac.uk/msd-srv/prot_int/pistart.html; Krissinel and Henrick, 2007) and is defined as one-half of the difference in the amount of solvent-accessible surface between the individual proteins and the complex. The Protein Interactions Calculator server (crick.mbu.iisc.ernet.in/~PIC; Tina et al., 2007) was used to ensure comprehensive contact analysis (including direct analysis of the previously-published BLIP•TEM-1 crystal structure 1JTG). Interacting residues were determined using UCSF Chimera (Pettersen et al., 2004), defined as those possessing one or more atoms within 5 Å of any TEM-1 atom. Buried residues were considered to be >1.5 Å residue depth from the protein surface, which was calculated by the DPX server (hydra.icgeb.trieste.it/dpx; Pintar et al., 2003). Electrostatic surfaces were calculated using PyMol (DeLano, 2002) and its corresponding Adaptive Poisson-Boltzmann Solver (APBS) plug-in (Lerner and Carlson, 2006). All figures were prepared using UCSF Chimera (Pettersen et al., 2004). The Gibbs Free Energy change upon binding was calculated according to the relationship !G = -RT•ln(K) = RT•ln(Ki). For all compiled/calculated thermodynamic data [e.g. calculations of !!G(penalty)], !!G(mut)=!!GKA(ma), the free energy change upon mutation, was calculated from the mass action coefficient of surface plasmon resonance measurements (Reichmann et al., 2007a). Although !!GKd is less prone to inaccuracies arising from protein concentration effects and instrument artifacts, !!GKA(ma) was used for 120  consistency (data for most single and multiple mutants were reported for this parameter) and also because it reflects rates of protein association (Reichmann et al., 2007a), giving a more comprehensive view of mutation effects.  121  2.4  Acknowledgements The authors thank Dr. Gunnar Olovsson for maintenance and operation of the  Centre for Blood Research Crystallography Hub at the University of British Columbia and Dr. Corie Ralston and staff at the Advanced Light Source at Lawrence Berkeley National Laboratories for assistance at the synchrotron beamline. For funding, NCJS thanks the Canadian Institutes of Health Research, Canadian Foundation for Innovation, Howard Hughes Medical Institute and Michael Smith Foundation for Health Research. SEJ thanks the Natural Sciences and Engineering Council of Canada for funding. KJL and SGK thank the KICOS (K20726000001-070100-00110) for financial support. For fellowships, MG thanks the Natural Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research. MG would also like in particular to thank the Strynadka laboratory members, especially Dr. Andrew Lovering, Dr. Trevor Moraes, Dr. Raz Zarivach, Dr. Hai-Zhong Zhang, Dr. Thomas Spreter, and Dr. Francesco Rao for technical advice and helpful discussions.  122  2.5  References  ADAMS, P. D., GOPAL, K., GROSSE-KUNSTLEVE, R. W., HUNG, L. W., IOERGER, T. R., MCCOY, A. J., MORIARTY, N. W., PAI, R. K., READ, R. J., ROMO, T. D., SACCHETTINI, J. C., SAUTER, N. K., STORONI, L. C. & TERWILLIGER, T. C. (2004) Recent developments in the PHENIX software for automated crystallographic structure determination. J Synchrotron Radiat, 11, 53-5. AKOVA, M. (2008) Sulbactam-containing &#x03B2;-lactamase inhibitor combinations. Clinical Microbiology and Infection, 14, 185-188. ALBECK, S. & SCHREIBER, G. (1999) Biophysical characterization of the interaction of the betalactamase TEM-1 with its protein inhibitor BLIP. Biochemistry, 38, 11-21. ALBECK, S., UNGER, R. & SCHREIBER, G. (2000) Evaluation of direct and cooperative contributions towards the strength of buried hydrogen bonds and salt bridges. J Mol Biol, 298, 503-20. ALEXANDER, D. C. & JENSEN, S. E. (1998) Investigation of the Streptomyces clavuligerus cephamycin C gene cluster and its regulation by the CcaR protein. J Bacteriol, 180, 4068-79. ALLEN, H., MOE, L., RODBUMRER, J., GAARDER, A. & HANDELSMAN, J. (2009) Functional metagenomics reveals diverse beta-lactamases in a remote Alaskan soil. ISME J, 3, 243-51. ALSALLAQ, R. & ZHOU, H. X. (2007) Energy landscape and transition state of protein-protein association. Biophys J, 92, 1486-502. ALTSCHUL, S. F., GISH, W., MILLER, W., MYERS, E. W. & LIPMAN, D. J. (1990) Basic local alignment search tool. J Mol Biol, 215, 403-10. AMSTUTZ, P., PELLETIER, J. N., GUGGISBERG, A., JERMUTUS, L., CESARO-TADIC, S., ZAHND, C. & PLUCKTHUN, A. (2002) In vitro selection for catalytic activity with ribosome display. J Am Chem Soc, 124, 9396-403. BENNETT-LOVSEY, R. M., HERBERT, A. D., STERNBERG, M. J. & KELLEY, L. A. (2008) Exploring the extremes of sequence/structure space with ensemble fold recognition in the program Phyre. Proteins, 70, 611-25. BERMAN, H., HENRICK, K. & NAKAMURA, H. (2003) Announcing the worldwide Protein Data Bank. Nat Struct Biol, 10, 980. BERMAN, H. M., WESTBROOK, J., FENG, Z., GILLILAND, G., BHAT, T. N., WEISSIG, H., SHINDYALOV, I. N. & BOURNE, P. E. (2000) The Protein Data Bank. Nucleic Acids Res, 28, 23542. CCP4 (1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr, 50, 760-3. CHATER, K. F. & CHANDRA, G. (2006) The evolution of development in Streptomyces analysed by genome comparisons. FEMS Microbiology Reviews, 30, 651-672. DELANO, W. L. (2002) The PyMOL Molecular Graphics System Palo Alto, CA, DeLano Scientific. DORAN, J. L., LESKIW, B. K., AIPPERSBACH, S. & JENSEN, S. E. (1990) Isolation and characterization of a beta-lactamase-inhibitory protein from Streptomyces clavuligerus and cloning and analysis of the corresponding gene. J Bacteriol, 172, 4909-18. DOUGAN, G., HAQUE, A., PICKARD, D., FRANKEL, G., O'GOARA, P. & WAIN, J. (2001) The Escherichia coli gene pool. Current Opinion in Microbiology, 4, 90-4.  123  DOUGHERTY, D. A. (1996) Cation-pi interactions in chemistry and biology: a new view of benzene, Phe, Tyr, and Trp. Science, 271, 163-8. EMANUELSSON, O., BRUNAK, S., VON HEIJNE, G. & NIELSEN, H. (2007) Locating proteins in the cell using TargetP, SignalP and related tools. Nat Protoc, 2, 953-71. EMSLEY, P. & COWTAN, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr, 60, 2126-32. GUPTA, V., DATTA, P., AGNIHOTRI, N. & CHANDER, J. (2006) Comparative in vitro activities of seven new beta-lactams, alone and in combination with beta-lactamase inhibitors, against clinical isolates resistant to third generation cephalosporins. Braz J Infect Dis, 10, 22-5. HAREL, M., COHEN, M. & SCHREIBER, G. (2007) On the Dynamic Nature of the Transition State for Protein-Protein Association as Determined by Double-mutant Cycle Analysis and Simulation. J Mol Biol. HEDGES, R. W., MATTHEW, M., SMITH, D. I., CRESSWELL, J. M. & JACOB, A. E. (1977) Properties of a transposon conferring resistance to penicillins and streptomycin. Gene, 1, 241-53. HOLM, L., OUZOUNIS, C., SANDER, C., TUPAREV, G. & VRIEND, G. (1992) A database of protein structure families with common folding motifs. Protein Sci, 1, 1691-8. HUANG, W., PETROSINO, J. & PALZKILL, T. (1998) Display of functional beta-lactamase inhibitory protein on the surface of M13 bacteriophage. Antimicrob Agents Chemother, 42, 2893-7. HUANG, W., ZHANG, Z. & PALZKILL, T. (2000) Design of potent beta-lactamase inhibitors by phage display of beta-lactamase inhibitory protein. J Biol Chem, 275, 14964-8. JOUGHIN, B. A., GREEN, D. F. & TIDOR, B. (2005) Action-at-a-distance interactions enhance protein binding affinity. Protein Sci, 14, 1363-9. KANG, S. G., PARK, H. U., LEE, H. S., KIM, H. T. & LEE, K. J. (2000) New beta -lactamase inhibitory protein (BLIP-I) from Streptomyces exfoliatus SMF19 and its roles on the morphological differentiation. J Biol Chem, 275, 16851-6. KOZER, N., KUTTNER, Y. Y., HARAN, G. & SCHREIBER, G. (2007) Protein-protein association in polymer solutions: from dilute to semidilute to concentrated. Biophys J, 92, 2139-49. KRISSINEL, E. & HENRICK, K. (2004) Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Crystallogr D Biol Crystallogr, 60, 2256-68. KRISSINEL, E. & HENRICK, K. (2007) Inference of macromolecular assemblies from crystalline state. J Mol Biol, 372, 774-97. LERNER, M. G. & CARLSON, H. A. (2006) APBS plugin for PyMOL. Ann Arbor, MI, University of Michigan LIOLIOS, K., MAVROMATIS, K., TAVERNARAKIS, N. & KYRPIDES, N. C. (2008) The Genomes On Line Database (GOLD) in 2007: status of genomic and metagenomic projects and their associated metadata. Nucleic Acids Res, 36, D475-9. LIVERMORE, D. M. (1995) beta-Lactamases in laboratory and clinical resistance. Clin Microbiol Rev, 8, 557-84. LO CONTE, L., CHOTHIA, C. & JANIN, J. (1999) The atomic structure of protein-protein recognition sites. J Mol Biol, 285, 2177-98. MCCOY, A. J. (2007) Solving structures of protein complexes by molecular replacement with Phaser. Acta Crystallogr D Biol Crystallogr, 63, 32-41.  124  MENG, E. C., PETTERSEN, E. F., COUCH, G. S., HUANG, C. C. & FERRIN, T. E. (2006) Tools for integrated sequence-structure analysis with UCSF Chimera. BMC Bioinformatics, 7, 339. OKANISHI, M. & GREGORY, K. F. (1970) Methods for the Determination of Deoxyribonucleic Acid Homologies in Streptomyces. J Bacteriol, 104, 1086-1094. PEREZ-LLARENA, F. J., LIRAS, P., RODRIGUEZ-GARCIA, A. & MARTIN, J. F. (1997) A regulatory gene (ccaR) required for cephamycin and clavulanic acid production in Streptomyces clavuligerus: amplification results in overproduction of both beta-lactam compounds. J Bacteriol, 179, 2053-9. PETROSINO, J., RUDGERS, G., GILBERT, H. & PALZKILL, T. (1999) Contributions of aspartate 49 and phenylalanine 142 residues of a tight binding inhibitory protein of beta-lactamases. J Biol Chem, 274, 2394-400. PETTERSEN, E. F., GODDARD, T. D., HUANG, C. C., COUCH, G. S., GREENBLATT, D. M., MENG, E. C. & FERRIN, T. E. (2004) UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem, 25, 1605-12. PINTAR, A., CARUGO, O. & PONGOR, S. (2003) DPX: for the analysis of the protein core. Bioinformatics, 19, 313-4. REICHMANN, D., RAHAT, O., ALBECK, S., MEGED, R., DYM, O. & SCHREIBER, G. (2005) The modular architecture of protein-protein binding interfaces. Proc Natl Acad Sci U S A, 102, 57-62. REICHMANN, D., COHEN, M., ABRAMOVICH, R., DYM, O., LIM, D., STRYNADKA, N. C. & SCHREIBER, G. (2007a) Binding hot spots in the TEM1-BLIP interface in light of its modular architecture. J Mol Biol, 365, 663-79. REICHMANN, D., RAHAT, O., COHEN, M., NEUVIRTH, H. & SCHREIBER, G. (2007b) The molecular architecture of protein-protein binding sites. Curr Opin Struct Biol, 17, 67-76. REICHMANN, D., PHILLIP, Y., CARMI, A. & SCHREIBER, G. (2008) On the contribution of water-mediated interactions to protein-complex stability. Biochemistry, 47, 1051-60. REYNOLDS, K. A., THOMSON, J. M., CORBETT, K. D., BETHEL, C. R., BERGER, J. M., KIRSCH, J. F., BONOMO, R. A. & HANDEL, T. M. (2006) Structural and Computational Characterization of the SHV-1 beta-Lactamase-beta-Lactamase Inhibitor Protein Interface. J. Biol. Chem., 281, 26745-26753. RINGER, A. L., SENENKO, A. & SHERRILL, C. D. (2007) Models of S/pi interactions in protein structures: comparison of the H2S benzene complex with PDB data. Protein Sci, 16, 2216-23. RUDGERS, G. W., HUANG, W. & PALZKILL, T. (2001) Binding properties of a peptide derived from beta-lactamase inhibitory protein. Antimicrob Agents Chemother, 45, 3279-86. RUDGERS, G. W. & PALZKILL, T. (1999) Identification of residues in beta-lactamase critical for binding beta-lactamase inhibitory protein. J Biol Chem, 274, 6963-71. RUDGERS, G. W. & PALZKILL, T. (2001) Protein minimization by random fragmentation and selection. Protein Eng, 14, 487-92. SCHLOSSHAUER, M. & BAKER, D. (2004) Realistic protein-protein association rates from a simple diffusional model neglecting long-range interactions, free energy barriers, and landscape ruggedness. Protein Sci, 13, 1660-9. SCHNEIDER, T. R. & SHELDRICK, G. M. (2002) Substructure solution with SHELXD. Acta Crystallogr D Biol Crystallogr, 58, 1772-9.  125  SELZER, T., ALBECK, S. & SCHREIBER, G. (2000) Rational design of faster associating and tighter binding protein complexes. Nat Struct Biol, 7, 537-41. SHELDRICK, G. M. (2008) A short history of SHELX. Acta Crystallogr A, 64, 112-22. SHI, T., BIBBY, T. S., JIANG, L., IRWIN, A. J. & FALKOWSKI, P. G. (2005) Protein interactions limit the rate of evolution of photosynthetic genes in cyanobacteria. Mol Biol Evol, 22, 2179-89. SINNOKROT, M. O. & SHERRILL, C. D. (2006) High-accuracy quantum mechanical studies of pipi interactions in benzene dimers. J Phys Chem A, 110, 10656-68. STRYNADKA, N. C., JENSEN, S. E., ALZARI, P. M. & JAMES, M. N. (1996) A potent new mode of beta-lactamase inhibition revealed by the 1.7 A X-ray crystallographic structure of the TEM-1BLIP complex. Nat Struct Biol, 3, 290-7. STRYNADKA, N. C., JENSEN, S. E., JOHNS, K., BLANCHARD, H., PAGE, M., MATAGNE, A., FRERE, J. M. & JAMES, M. N. (1994) Structural and kinetic characterization of a beta-lactamaseinhibitor protein. Nature, 368, 657-60. SUN, W., HU, Y., GONG, J., ZHU, C. & ZHU, B. (2005) Identification of beta-lactamase inhibitory peptide using yeast two-hybrid system. Biochemistry (Mosc), 70, 753-60. TANOUE, T. & NISHIDA, E. (2003) Molecular recognitions in the MAP kinase cascades. Cell Signal, 15, 455-62. THAI, W., PARADKAR, A. S. & JENSEN, S. E. (2001) Construction and analysis of b-lactamaseinhibitory protein (BLIP) non-producer mutants of Streptomyces clavuligerus. Microbiology, 147, 32535. TINA, K. G., BHADRA, R. & SRINIVASAN, N. (2007) PIC: Protein Interactions Calculator. Nucleic Acids Res, 35, W473-6. WANG, J., ZHANG, Z., PALZKILL, T. & CHOW, D. C. (2007) Thermodynamic investigation of the role of contact residues of blip for binding to tem-1 beta -lactamase. J Biol Chem. WELLS, T. J., TREE, J. J., ULETT, G. C. & SCHEMBRI, M. A. (2007) Autotransporter proteins: novel targets at the bacterial cell surface. FEMS Microbiol Lett, 274, 163-72. ZHANG, H., SHINKAWA, H., ISHIKAWA, J., KINASHI, H. & NIMI, O. (1997) Improvement of transformation system in Streptomyces using a modified regeneration medium. J Ferment Bioeng. ZHANG, Z. & PALZKILL, T. (2003) Determinants of binding affinity and specificity for the interaction of TEM-1 and SME-1 beta-lactamase with beta-lactamase inhibitory protein. J Biol Chem, 278, 45706-12. ZHANG, Z. & PALZKILL, T. (2004) Dissecting the protein-protein interface between betalactamase inhibitory protein and class A beta-lactamases. J Biol Chem, 279, 42860-6.  126  CHAPTER 3: CRYSTALLOGRAPHIC ANALYSIS OF THE OXA-10 ßLACTAMASE COVALENTLY MODIFIED AT ITS SERINE NUCLEOPHILE BY A NON-ß-LACTAM CYCLOBUTANONE INHIBITOR (JJ05-850) TO FORM A TETRAHEDRAL HEMIKETAL! 3.1  Introduction ß-lactam antibiotics are among the most successful and widely prescribed in history;  their efficacy and low toxicity arise from specific targeting of the biosynthesis enzymes of the bacterial cell wall (Tipper and Strominger, 1965, Demain and Elander, 1999). Their bactericidal effects and ubiquitous clinical use, however, together create massive evolutionary selective pressure that has led to presently widespread antibiotic resistance (Levy, 2001). The ß-lactamases are a large family of efficient hydrolytic enzymes that are rapidly disseminated among bacterial species (Weldhagen, 2004) and highly adaptable, via mutation, to the hydrolysis of large and varied sets of ß-lactam derivatives (Livermore, 2008, Bush, 2008). The majority of known ß-lactamases (structural classes A, C and D) employ a serine nucleophile and form a covalent intermediate during ß-lactam hydrolysis. Classes of ß-lactamase inhibitors available for clinical use include the penams (sulbactam, tazobactam) and the clavams (clavulanate); these are ß-lactams that acylate the serine nucleophile but are slowly hydrolyzed or irreversibly inactivate the enzyme (Therrien and Levesque, 2000). ß-lactamase resistance to such inhibitors, however, is widespread, in particular among the OXA family of Class D serine ß-lactamases (Totir et al., 2008, Page, 2000, Payne et al., 1994). Members of the OXA family are furthermore adept at hydrolyzing  !  A version of this chapter will be submitted for publication. M Gretes, JW Johnson, V Goodfellow, GI Dmitrienko, NCJ Strynadka. Crystallographic analysis of the OXA-10 ß-lactamase covalently modified at its serine nucleophile by a non-ß-lactam cyclobutanone inhibitor (JJ05-850) to form a tetrahedral hemiketal.  127  broad-spectrum, BLA-resistant ß-lactams (Brown and Amyes, 2005). Finally, no clinical inhibitors are yet available for the metallo-ß-lactamases (Class B), which function by activation of a water molecule for direct nucleophilic attack of the lactam ring without formation of a covalent intermediate (Walsh 2008). Thus a novel approach to ß-lactamase inhibition is needed. A related and relatively unexplored class is the non-lactam cyclobutanone analogues of ß-lactams. If covalently bound by the enzyme, these should be poorly hydrolyzed, since they lack the favourable leaving group represented by the lactam ring nitrogen atom. Partly as a consequence, however, ß-lactamase inhibition by members of this class has not been reported, with only one known to be a weak competitive inhibitor of R61 PBP (Ki = 1 mM). A method for increasing the affinity of such ß-lactam analogues for ß-lactamases was recently described (Johnson et al., 2008). JJ05-850 (1) is a cyclobutanone ß-lactam analogue bearing an !substituted methoxy group at its C3 position (Fig. 3.1). The 3!-OCH3 is believed to stabilize the exo-conformation (S pucker) thought to resemble that preferentially bound by the drug target, and thus indirectly increase the binding affinity of the molecule (IC50 of 61 "M with 30 minutes preincubation). The 3ß stereoisomer (1-3ß) is inactive.  Figure 3.1: Structures of cyclobutanone ß-lactam analogues: 1, used in this study; 1-3ß, its nonantibiotic stereoisomer at the C3 chiral centre.  In order to evaluate this hypothesis and explore the nature of BLA inhibition by 128  molecules of this class, we have determined the crystal structure of the Class D ß-lactamase OXA-10 covalently bound at its active site nucleophile to 1.  3.2  Results and Discussion 1 was used to covalently modify crystals of OXA-10 with considerable difficulty, as  exposure to inhibitor caused crystals to rapidly and severely degrade, likely via disruption of crystal contacts (see below). Nevertheless, a crystal structure of OXA-10 was obtained that is clearly covalently modified by a ligand at the Ser-67 nucleophile of two of four OXA-10 molecules in the asymmetric unit (ASU; Fig. 3.2). All molecules are superposable with RMSD 0.22-0.46 Å for #240 C! atoms. Ser-67 covalently linked to 1 via a hemiketal was modeled based on observed electron density without reference to published structures.  Figure 3.2: Electron-density (2Fo-Fc) maps of 1 (C atoms in orange or crimson) covalently bound to OXA-10 active site Ser-67 (C atoms in azure and lavender); views (a, 129  c) depict OXA-10 molecule A, (b, d) depict OXA-10 molecule B. Maps are contoured at 1.0$ and 0.5$ (dark and light colours, respectively). For comparison, electron density is also shown for the carbamylated Lys-70 (Kcx-70), a typical feature of the OXA ß-lactamases.  The surface area of the interface between the inhibitor and active site is 290 Å2 (molecule A) and 273 Å2 (molecule B). This is calculated by the LPC server† which, for context, estimates 362-371 Å2 as a theoretical maximum complementarity for 1 (Sobolev et al., 1999). Consistent with this computational result, numerous specific favorable interactions are observed in the inhibitor-enzyme complex (Fig. 3.3). Ideal hydrogen bonding‡ occurs between OXA-10 residues and inhibitor atoms as follows (listed as H donor•acceptor; distances given for molecules A / B; inhibitor atoms indicated in bold): Ser-115-O%H•C4’=O (3.1 Å / 2.9 Å), Thr-206-O%H•C4’=O (2.7 Å / 2.5 Å), Phe-208-NH•O6’ (2.7 Å / 2.5 Å), Arg250-N&2H-•O3’ (3.5 Å / 3.1 Å). A salt bridge is likely formed between C4’-O- and Arg-250-N&1H3+.  † ‡  http://bip.weizmann.ac.il/oca-bin/lpccsu according to parameters observed in a survey of small molecule interactions (Mills and Dean, 1996)  130  Figure 3.3: OXA-10 active site residues positioned to interact with 1 in ASU molecules A (a) and B (b). Potential hydrogen bonds are shown in cyan and magenta (those deviating from ideality by >0.4 Å or >20º are drawn transparently) and the possible halogen bond in green.  131  The following hydrogen bond donor•acceptor pairs deviate from their ideal angle by up to 20º: Thr-206-O%H•C4’-O- (2.9 Å / 2.7 Å) and Ser-67-NH•O6’ (2.9 Å). This latter interaction, along with Phe-208-NH forms the widely-conserved oxyanion hole (Hidayet et al., 1999). However, should O6’ be in the protonated state (which would deviate from convention and render hydrogen bonding with the Ser-67 backbone NH unfavorable) the backbone oxygen of Phe-208 may act as the hydrogen bond acceptor in the pair O6’•Phe208-O (2.8 Å / 2.9 Å). Two favorable sulfur-pi interactions (Ringer et al., 2007) are observed: S2•Phe-208 (4.3 Å / 3.9 Å), S2•Trp-102 (5.1 Å / 5.4 Å). This hydrophobic region of the active site pocket complements the hydrophobic edge of the inhibitor. Aside from the entropically favorable general hydrophobic interaction, several near-ideal VDW contacts are formed here. These include C5•Ser-115-Cß (3.5 Å), C4•Phe-208-Cß (3.4 Å / 3.5 Å), C3•Phe-208-Cß (3.7 Å / 3.5 Å), Clß•Phe-208-C$2 (3.4 Å / 4.1 Å); in molecule A only, Cl!•Leu-155-C$2 (4.0 Å). A diagram of interactions of 1 at the OXA-10 active site is given in Fig. 3.4.  Figure 3.4: Diagram of favorable interactions likely made with 1 at the OXA-10 active site (excluding VDW contacts and hydrophobic interactions): red lines represent hydrogen bonds; blue and green lines show a salt bridge and halogen bond, respectively; purple hashed lines indicate sulfur-" interactions. 132  A few apparent clashes are observed in the structure, where non-bonding atoms approach each other at distances less than the sum of their van der Waals (VDW) radii. The clash of Clß with the Phe-208 main chain N appears to be of particular concern: the protein backbone is less likely than a side chain to alter its conformation to accommodate inhibitor binding. Also, this N is critical for stabilizing the oxyanion formed in the course of nucleophilic attack. However, the apparent Clß clashes (2.5 Å and 2.2 Å in ASU molecules A and B, respectively, versus their VDW radii sum of 3.27 Å; Bondi, 1964) may be explained by halogen bonding. Although this is an underdeveloped area of study (e.g. halogen bonds have not been measured nor simulated ab initio for a dichlorine functional group) an aromatic halide-oxygen bond has been reported at distances as near as ~2.4 Å (Lu et al., 2007). Neither halogen bonding nor any other known interaction can resolve the clash between Cl! and Val-117-C% (2.8 Å and 2.5 Å in ASU molecules A and B, respectively; the sum of their VDW radii is 3.45 Å; Bondi, 1964). However, this clash as well as the relatively weak electron density observed for both Cl! and Clß can be explained by alternate conformations of 1, resulting in the Cl atoms’ occupation of a range of positions while its other atoms remain fixed within the crystal structure. This is supported by ab initio determination of two different energetically minimal conformational states of the methyl hemiketal (Fig. 3.5). Here, the chlorine atoms are each found to vary in position by an RMSD of ~1.4 Å. In particular, because of the relatively low energy of the transition state, rapid interconversion between these conformations, and thus occupation of intermediate states, is possible.  133  Figure 3.5: Alternative conformations of 1; (a) The methyl hemiketal formed by methanol in solution is used to model probable conformations of the OXA-10 Ser-67 hemiketal; (b) ab initio (RHF/6-31G(d)) optimized conformations of this methyl hemiketal, energy differences are given in kcal/mol; (c) least-squares superposition of the S-puckered (blue) and C4-puckered (gray) conformations of the methyl hemiketal of cyclobutanone 1. 1 is likely to inhibit the OXA family generally, since many of the interacting residues depicted in Fig. 3.4 are conserved. The oxyanion hole residues (Phe-208 and Ser-67) as well as the Thr-206 functionality are absolutely conserved (the variant OXA-9 still features the hydroxyl group in Ser-206). Arg-250 is conserved amongst groups 1 (OXA-10, -5, -7, -11, 13, -14, -16, -17, -19, -23) and 2 (OXA-2, -3, -15, -20, -21, LCR-1) enzymes. Apart from the novel nature of the stable hemiketal acylated state, the OXA-10•1 complex shares certain features with other reported OXA•inhibitor structures. The oxyanion hole appears to be universally conserved among the available acyl-enzyme structures of the OXA family. Also, both the OXA-13•imipenem and OXA-13•meropenem acyl-enzyme complexes (PDB IDs: 1H5X/1H8Y) feature probable salt bridge/hydrogen-bonding 134  interactions between both carboxylate atoms and Arg-250. However, of the three inhibitor-bound structures of OXA-10 available, the binding of all but one (moxalactam, PDB ID: 1K6R) differs remarkably from the OXA-13 inhibitor structures and the OXA-10 structure reported here. In addition, because of the pseudosymmetry of moxalactam with respect to its carboxylate groups (Fig. 3.S1), only one of its two OXA-10 binding modes resembles that of 1.  Figure 3.S1: Structure of moxalactam (Roberts, 1992).  OXA-10 in complex with penicillinic acid (1k54, Golemi et al., 2001) shares few binding features apart from similar occupation of the oxyanion hole, and the positioning of the heterocyclic sulfur atom. Otherwise the carboxylate and lactam nitrogen atom face bulk solvent, although the R1 oxygen atom interacts via a hydrogen bond with Kcx-70 at one of its carboxylate oxygen atoms. Likewise, the six-membered inhibitor ring of phenylboronate•OXA-10 complex (1k6s), lacking a carboxylate, closely matches the fivemembered ring in 1k54 and overlays poorly with the observed 1 orientation. Thus the binding orientation of 1 (especially the core ring atoms) appears to have much in common with that of the carbapenems imipenem and meropenem and, to some extent, the oxacephem moxalactam. The only other known cyclobutanone-containing protein crystal structure of possible 135  relevance is a cyclobutanone ß-lactam analogue bound, intact, to isopenicillin N synthase (PDB ID: 2jb4; Stewart et al., 2007), which is unrelated by sequence or structure to the ßlactamases. The authors credit the enhanced stability of the ring system to the cyclobutanone, and indeed identify the potential of these molecules as antibiotics or ßlactamase inhibitors. The structure presented here covalently bound to 1 nevertheless appears unique among those of OXA-10 with respect to its apparently stable hemiketal, tetrahedral state. Four-membered ring opening appears to be prevented by the nature of the C4 atom, a much poorer leaving group than the lactam nitrogen in the ß-lactam structures. Without the leaving of this group, the tetrahedral intermediate cannot collapse into the sp2-hypbridized acyl enzyme intermediate susceptible to nucleophilic attack by a hydrolytic water. In sum, covalent modification takes place despite the lack of a favorable leaving group and the C3-methoxy group itself is found to make few favorable active site contacts and in fact is relatively poorly ordered. In light of the lack of inhibitory power of the alternative C3-methoxy stereoisomer, our data supports the hypothesis that the primary function of the C3-methoxy substituent is to stabilize the inhibitor conformation favored in PBP active site binding. The lack of covalently bound 1 in OXA-10 molecules C and D is likely due to insertion of the loop bearing Lys-95 into the active site of the non-corresponding OXA-10 molecule of the other dimer of the ASU (A into D; B of the crystal symmetry mate into C). Lys-95 makes numerous favorable interactions with the opposing molecule active site (Fig. 3.6). These include hydrophobic interactions with Phe-208, Val-117, and Leu-155 as well as a near-ideal VDW contact (3.5 Å) between molecule B Lys-95-C' and molecule C Phe-208136  C$2. Hydrogen bonding may also take place between Lys-95-N( and Phe-208-O (3.1 / 3.3 Å) as well as to the tightly bound phosphate ion (the crystallization precipitant), which in turn hydrogen bonds to Ser-115-O%, Arg-250-N&1 and -N&2 as well as to a network of ordered water molecules in the active site.  Figure 3.6: Crystallographic contacts made by Lys-95 (a) of the symmetry-related molecule B (crimson) into the active site of molecule C (yellow); (b) of molecule A (azure) into the active site of molecule D (green) in the same ASU. Phosphate is shown in orange; position of the inhibitor is shown in transparent gray by superposition of an acylated monomer; oxygen atoms of ordered water molecules are shown as red spheres; and possible hydrogen bonds are shown in cyan and pink, respectively.  If the acylated chains are superimposed on the unacylated chains, it is clear that the interaction of Lys-95 here precludes substrate binding and acylation. In molecule C, Lys-95N( from molecule B would clash severely with Clß (1.0 Å) and C7 (1.5 Å). There are even more extensive clashes observed between the molecule A Lys-95 and the inhibitor (in its position in either A or B) superposed on the active site of molecule D: Lys95 would clash at 137  its N( with C7 (1.0Å), Clß (1.2-1.5 Å), and Cl! (1.6-1.9 Å), and most seriously at its C' with Clß (0.9-1.2 Å). Perhaps most importantly, Lys-95 stabilizes the well-ordered phosphate ion that takes the position otherwise occupied by the inhibitor carboxylate which would itself preclude binding. Together, Lys-95 and the oxyanion occlude much of the active site groove, including the Ser-67 nucleophile. The incompatibility of the favorable crystal contacts made by this residue insertion with inhibitor binding appears to be the reason for the failure to produce inhibitor-OXA-10 co-crystals (for 1 as well as several other related inhibitors). This also explains the difficulty in obtaining covalently modified crystal structures by inhibitor soaking: a great enough concentration of inhibitor must be soaked into the crystal to bind only the accessible active sites; if concentrations are too great or soaking time is too long, crystals begin to visibly degrade and diffraction power is lost. Of course, if inhibitor concentrations are too low, no hemiketal formation is observed. This problem is compounded by the strong occupation of the phosphate ion (or sulfate in alternative crystallization conditions) in the active site, observed in the same position in all related published structures and those solved in the course of unsuccessful soaking experiments in this study. Thus the covalently-bound inhibitor structure could only be obtained by long exposure (3 weeks) to modest inhibitor concentrations (see 3.4 Experimental Section).  138  3.3  Conclusions: Directions for Inhibitor Design Since 1 represents little more than a simple core structure of an inhibitor, there is  plenty of opportunity for structural modifications. To improve binding, more elaborate substituents should be placed at R1 and R2, perhaps similar to those featured in established OXA-10-binding drugs such as imipenem and meropenem, which make more extensive contacts with active site residues. In addition, the great many ß-lactam antibiotics cleaved by OXA-10 ought to offer a wealth of structural variants. In addition to increasing favorable active site interactions, possibly unfavorable interactions could be decreased e.g. by employing only single chlorine atom (in the stereo position of the halogen-bonding Clß atom) substituted at C6 mono-substituted chlorine. Finally, the potential of 1 and related cyclobutanones to inhibit Class B metallo-ß-lactamases should be explored functionally and structurally.  139  3.4  Experimental Section The synthesis of 1 was carried out as done previously (Johnson et al., 2008). OXA-10  was expressed and purified from a naturally producing isolate of Pseudomonas aeruginosa, as described in Danel et al., 1998. Crystals of OXA-10 were grown at 20 mg/ml protein in 1.4 M sodium/potassium phosphate pH 7.0. Fully-formed crystals were soaked for 21 days in this solution following the addition of 1 to a final concentration of 5 mM. Soaked crystals were plunge-cooled to 100K in liquid nitrogen after cryoprotection in inhibitor soaking solution containing glycerol at 12.5% (v/v). Diffraction data was collected using a rotating copper anode (Rigaku MSC Ltd.) and MAR2300 CCD detector (MAR Research). The data set was indexed, the unit cell refined, and reflections integrated using mosflm (Leslie, 2006) and scaled using scala in the CCP4 suite (CCP4, 1994). Since the space group and unit cell were closely comparable to 1e4d, initial phasing was achieved by rigid body refinement with CCP4-Refmac5 using this model (Hidayet et al., 1999). To avoid creating bias in the electron density map due to the ligand, the model was adjusted using Coot (Emsley and Cowtan, 2004) and refined to an R/Rfree = 0.215/0.248 prior to modeling of the inhibitor. Geometric restraints for 1 were generated using the PRODRG server (Schuttelkopf and van Aalten, 2004). Complete crystal data and refinement statistics are collected in Table 3.1. Least-squares superposition of protein molecules was by iterative, secondary structure matching (Meng et al., 2006) and figures (except Fig. 3.5) were prepared using UCSF Chimera (Pettersen et al., 2004). Ab initio calculations of energetically minimal conformations of 1 made use of GaussView (Dennington et al., 2003) and Gaussian-03 (Frisch et al. 2003). Structure optimizations and frequency calculations for the methyl hemiketal of 1 were done in the gas 140  phase and employed the restricted Hartree-Fock (RHF) method and the 6-31G(d) basis set. Determination of the energy differences between conformations involved an optimization and frequency calculation followed by addition of the zero-point energy correction for each conformation. The Sybyl 7.2 modeling package (SYBYL, 2007) was used for the leastsquares superposition of the S-puckered and C4-puckered conformations. PyMOL (DeLano, 2008) was used to generate graphics in Fig. 3.5.  Table 3.1: Data collection and refinement statistics. Crystal Parametersa Spacegroup P21 Cell Dimensions: a x b x c, Å 65.70 % 82.54 % 102.36 90.00, 94.63, 90.00 !, ß, ", º Resolution, Å 2.0 – 27.6 (2.00 – 2.11) Wavelength, Å 1.54179 No. Reflections 172,586 No. Unique Reflections 83,695 Average Redundancy 3.4 (3.4) Completeness, % 99.4 (99.4) 9.8 (2.2) I/!I b R sym 0.073 (0.350) Refinement Statisticsc Inhibitor Occupancy 0.60 Average B-Factor (Inhibitor), Å2 32.0 2 Average B-Factor (Solvent), Å 34.8 Average B-Factor (Protein), Å2 18.2 Rwork / Rfree 0.178 / 0.233 rmsd bond lengths, Å 0.019 rmsd bond angles, ° 1.77 a Statistics for highest resolution shell are given in parentheses b Rsym = &|(Ihkl)-<I>| / &(Ihkl) c 5% of reflections excluded from refinement and used to calculate Rfree  141  3.5  References  BONDI, A. (1964) van der Waals Volumes and Radii. J phys Chem, 68, 441-451. BROWN, S. & AMYES, S. G. (2005) The sequences of seven class D beta-lactamases isolated from carbapenem-resistant Acinetobacter baumannii from four continents. Clin Microbiol Infect, 11, 326-9. BUSH, K. (2008) Extended-spectrum beta-lactamases in North America, 1987-2006. Clin Microbiol Infect, 14 Suppl 1, 134-43. CCP4 (1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr, 50, 760-3. DEMAIN, A. L. & ELANDER, R. P. (1999) The beta-lactam antibiotics: past, present, and future. Antonie Van Leeuwenhoek, 75, 5-19. DANEL, F., HALL, L. M., GUR, D. & LIVERMORE, D. M. (1998) OXA-16, a further extendedspectrum variant of OXA-10 beta-lactamase, from two Pseudomonas aeruginosa isolates. Antimicrob Agents Chemother, 42, 3117-22. DELANO, W. L. (2008) PyMOL Molecular Graphics System; DeLano Scientific LLC: Palo Alto, CA. DENNINGTON, R., II; KEITH, T.; MILLAM, J.; EPPINNETT, K.; HOVELL, W. L.; GILLILAND, R. (2003) GaussView, Version 3.08; Semichem, Inc.: Shawnee Mission, KS. EMSLEY, P. & COWTAN, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr, 60, 2126-32. FRISCH, M. J.; TRUCKS, G. W.; SCHLEGEL, H. B.; SCUSERIA, G. E.; ROBB, M. A.; CHEESEMAN, J. R.; MONTGOMERY, J. A.; VREVEN, T.; KUDIN, K. N.; BURANT, J. C.; MILLAM, J. M.; IYENGAR, S. S.; TOMASI, J.; BARONE, V.; MENNUCCI, B.; COSSI, M.; SCALMANI, G.; REGA, N.; PETERSSON, G. A.; NAKATSUJI, H.; HADA, M.; EHARA, M.; TOYOTA, K.; FUKUDA, R.; HASEGAWA, J.; ISHIDA, M.; NAKAJIMA, T.; HONDA, Y.; KITAO, O.; NAKAI, H.; KLENE, M.; LI, X.; KNOX, J. E.; HRATCHIAN, H. P.; CROSS, J. B.; ADAMO, C.; JARAMILLO, J.; GOMPERTS, R.; STRATMANN, R. E.; YAZYEV, O.; AUSTIN, A. J.; CAMMI, R.; POMELLI, C.; OCHTERSKI, J. W.; AYALA, P. Y.; MOROKUMA, K; VOTH, G. A.; SALVADOR, P.; DANNENBERG, J. J.; ZAKRZEWSKI, V. G.; DAPPRICH, S.; DANIELS, A. D.; STRAIN, M. C.; FARKAS, Ö.; MALICK, D. K.; RABUCK, A. D.; RAGHAVACHARI, K.; FORESMAN, J. B.; ORTIZ, J. V.; CUI, Q.; BABOUL, A. G.; CLIFFORD, S.; CIOSLOWSKI, J.; STEFANOV, B. B.; LIU, G.; LIASHENKO, A.; PISKORZ, P.; KOMÁROMI, I.; MARTIN, R. L.; FOX, D. J.; KEITH, T.; AL-LAHAM, M. A.; PENG, C. Y.; NANAYAKKARA, A.; CHALLACOMBE, M.; GILL, P. M. W.; JOHNSON, B.; CHEN, W.; WONG, M. W.; GONZALEZ, C.; POPLE J. A. (2003) Gaussian-03, Revision B.04.; Gaussian Inc.: Pittsburg, PA. GOLEMI, D., MAVEYRAUD, L., VAKULENKO, S., SAMAMA, J. P. & MOBASHERY, S. (2001) Critical involvement of a carbamylated lysine in catalytic function of class D beta-lactamases. Proc Natl Acad Sci U S A, 98, 14280-5. HIDAYET, N. M., SHARAF, S. A., AREF, S. R., TAWFIK, T. A. & MOUBARAK, II (1999) Correlates of age at natural menopause: a community-based study in Alexandria. East Mediterr Health J, 5, 307-19. JOHNSON, J. W., EVANOFF, D. P., SAVARD, M. E., LANGE, G., RAMADHAR, T. R., ASSOUD, A., TAYLOR, N. J. & DMITRIENKO, G. I. (2008) Cyclobutanone mimics of penicillins: effects of substitution on conformation and hemiketal stability. J Org Chem, 73, 6970-82.  142  JONES, R. N., BIEDENBACH, D. J., SADER, H. S., FRITSCHE, T. R., TOLEMAN, M. A. & WALSH, T. R. (2005) Emerging epidemic of metallo-beta-lactamase-mediated resistances. Diagn Microbiol Infect Dis, 51, 77-84. LESLIE, A. G. (2006) The integration of macromolecular diffraction data. Acta Crystallogr D Biol Crystallogr, 62, 48-57. LEVY, S. B. (2001) Antibiotic resistance: consequences of inaction. Clin Infect Dis, 33 Suppl 3, S124-9. LIVERMORE, D. M. (2008) Defining an extended-spectrum beta-lactamase. Clin Microbiol Infect, 14 Suppl 1, 3-10. LU, Y. X., ZOU, J. W., WANG, Y. H., JIANG, Y. J. & YU, Q. S. (2007) Ab initio investigation of the complexes between bromobenzene and several electron donors: some insights into the magnitude and nature of halogen bonding interactions. The journal of physical chemistry A, 111, 10781-8. MENG, E. C., PETTERSEN, E. F., COUCH, G. S., HUANG, C. C. & FERRIN, T. E. (2006) Tools for integrated sequence-structure analysis with UCSF Chimera. BMC Bioinformatics, 7, 339. MILLS, J. E. & DEAN, P. M. (1996) Three-dimensional hydrogen-bond geometry and probability information from a crystal survey. J Comput Aided Mol Des, 10, 607-22. PAGE, M. G. (2000) Beta-Lactamase inhibitors. Drug Resist Updat, 3, 109-125. PAYNE, D. J., CRAMP, R., WINSTANLEY, D. J. & KNOWLES, D. J. (1994) Comparative activities of clavulanic acid, sulbactam, and tazobactam against clinically important beta-lactamases. Antimicrob Agents Chemother, 38, 767-72. PETTERSEN, E. F., GODDARD, T. D., HUANG, C. C., COUCH, G. S., GREENBLATT, D. M., MENG, E. C. & FERRIN, T. E. (2004) UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem, 25, 1605-12. RINGER, A. L., SENENKO, A. & SHERRILL, C. D. (2007) Models of S/pi interactions in protein structures: comparison of the H2S benzene complex with PDB data. Protein Sci, 16, 2216-23. ROBERTS, J. (1992) Cephalosporins. Kirk!Othmer Encyclopedia of Chemical Technology. SCHUTTELKOPF, A. W. & VAN AALTEN, D. M. (2004) PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr D Biol Crystallogr, 60, 1355-63. SOBOLEV, V., SOROKINE, A., PRILUSKY, J., ABOLA, E. E. & EDELMAN, M. (1999) Automated analysis of interatomic contacts in proteins. Bioinformatics, 15, 327-32. STEWART, A. C., CLIFTON, I. J., ADLINGTON, R. M., BALDWIN, J. E. & RUTLEDGE, P. J. (2007) A cyclobutanone analogue mimics penicillin in binding to isopenicillin N synthase. Chembiochem, 8, 2003-7. SYBYL Molecular Modeling Software; Tripos Inc. (2007): St. Louis, MO. THERRIEN, C. & LEVESQUE, R. C. (2000) Molecular basis of antibiotic resistance and betalactamase inhibition by mechanism-based inactivators: perspectives and future directions. FEMS Microbiology Reviews, 24, 251-62. TIPPER, D. J. & STROMINGER, J. L. (1965) Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl Acad Sci USA, 54, 1133-41. TOTIR, M. A., CHA, J., ISHIWATA, A., WANG, B., SHERI, A., ANDERSON, V. E., BUYNAK, J., MOBASHERY, S. & CAREY, P. R. (2008) Why clinically used tazobactam and sulbactam are poor inhibitors of OXA-10 beta-lactamase: Raman crystallographic evidence. Biochemistry, 47, 4094101.WALSH, T. R. (2005) The emergence and implications of metallo-beta-lactamases in Gram-  143  negative bacteria. Clin Microbiol Infect, 11 Suppl 6, 2-9. WELDHAGEN, G. F. (2004) Integrons and beta-lactamases--a novel perspective on resistance. International Journal of Antimicrobial Agents, 23, 556-62.  144  CHAPTER 4: STRUCTURAL BASIS FOR THE ACTIVITY OF CEFTOBIPROLE AGAINST METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS " 4.1  Introduction The spectacular success of the ß-lactam group of antibiotics in treating bacterial  infections has been blighted in recent times by the emergence of methicillin-resistant Staphylococcus aureus (MRSA) and other problematic “superbugs”. The medical revolution enabled by penicillin use in the 1940s was soon countered by strains of S. aureus able to produce a penicillinase (ß-lactamase). The introduction of methicillin in 1959 to overcome such ß-lactamase-producing strains was rapidly met with the emergence of the first MRSA strains in 1961. MRSA now poses a serious threat to global healthcare systems, with the reported prevalence of such strains in excess of 60% for some hospital settings in the United States (Doshi et al., 2009). Community acquired MRSA infections, affecting healthy individuals have also emerged (reviewed in Stryjewski and Chambers, 2008, Miller and Kaplan, 2009). Such trends are particularly worrisome as some degree of resistance has been reported for all of the therapeutics currently available to treat MRSA–from glycopeptides such as vancomycin (Walsh and Howe, 2002), to newer antibiotics such as linezolid (Tsiodras et al., 2001), daptomycin (Marty et al., 2006), and tigecycline (Bouchillon et al., 2009). The clear need to maintain effective treatments against MRSA and other “superbugs” may be met by a newer generation of ß-lactams (Fig. 4.1; Page, 2004, Guignard et al., 2005).  "  A version of this chapter will be submitted for publication. M Gretes, AL Lovering, F Danel, L De Castro, MGP Page, NCJ Strynadka. Structural basis for the activity of ceftobiprole against Methicillin-Resistant Staphylococcus aureus.  145  Of these, ceftobiprole (administered as the soluble prodrug ceftobiprole medocaril, and previously known as Ro 63-9141 or BAL9141) is the most clinically advanced, having successfully completed several phase III trials (Noel et al., 2008a, Noel et al., 2008b). To date it has been approved for clinical use in adult complicated skin and soft tissue infections by regulatory authorities in Canada and Switzerland, and is pending approval in Europe and the United States†, with USFDA approval pending only study site review (Boucher et al., 2009). Ceftobiprole is immensely promising as a therapeutic, but the precise details of the interactions with its target remain to be shown, and furthermore why it succeeds where other latter-generation ß-lactams fail.  †  www.basilea.com/template_loader.php?tplpage_id=19&_function=render&id=1 (last accessed 2009-0408)  146  Figure 4.1: Chemical structures of selected ß-lactams and PBP2a substrates. i) Comparison between effective and ineffective (e.g. IC50 > 40 mg/L; Lortholary et al., 2008) anti-MRSA ß-lactams. The generalized cephalosporin scaffold is provided at the top centre, with modifications to this occurring at the R1 and R2 positions. The figure illustrates that anti-MRSA efficacy is greatly affected by modification at the R2 position. Inhibitor efficacy is indicated, where known, by MIC90 (minimal inhibitory concentration of the compound at which growth of 90% of MRSA was inhibited) and IC50 (concentration inhibiting 50% of [3H] penicillin-labeling affinity for PBP2a; Guignard et al., 2005, Chambers, 1995, Franciolli et al., 1991, Fantin et al., 1996, Tsuji et al., 2003, Ida et al., 2002, Vouillamoz et al., 2004, Sumita et al., 1995). ii) Proposed steric equivalency of ceftobiprole to the PBP2a* transpeptidation substrates (Gly5 crossbridge acceptor and stem peptide D-Ala D-Ala donor). With the loss of the terminal D-Ala, a completed transpeptidation reaction in S. aureus results in the formation of a Gly-Ala bond. The lipophilicity of the acceptor Gly5 crossbridge may explain why hydrophobic, extended groups are effective at the R2 position.  147  148  The mechanism of action of ß-lactams is well documented (Waxman and Strominger, 1983), with the ß-lactam nucleus representing a steric mimic of the terminal D-Ala D-Ala stem peptide of cell wall peptidoglycan. Penicillin binding proteins (PBPs) are able to crosslink this stem peptide in a transpeptidation reaction, but form a poorly hydrolysable acyl-enzyme intermediate with the ß-lactam mimic. Acylated PBPs are unable to perform their regular functions in cross-linking the cell wall peptidoglycan. As a result, affected cells die; the weakened bacterial cell wall is unable to cope with internal osmotic pressure. Opportunity for resistance against ß-lactams arises from either degradation of the lactam ring (by ß-lactamases, which share structural homology to PBPs but are able to hydrolyze the acyl-enzyme intermediate) or through use of a PBP with intrinsically low ß-lactam affinity, able to maintain cell-wall crosslinking in the presence of ß-lactams. This latter strategy is pursued either by mutation of an endogenous PBP, important for PBP2x of penicillinresistant Streptococcus pneumoniae (Pernot et al., 2004), or by acquisition of an exogenous resistant PBP, characteristic of MRSA (Hanssen et al., 2004). Indeed, MRSA appears to employ this sole resistance mechanism, through the acquisition of the mecA gene, which encodes the PBP2a protein (often referred to as PBP2’). A definitive source for this gene transfer has yet to be found, but homology studies appear to implicate the related S. sciuri species (Fuda et al., 2007, Zhou et al., 2008). The advantage gained by PBP2a acquisition arises from the ability of this enzyme to produce crosslinked peptidoglycan in the presence of otherwise inhibitory ß-lactam concentrations. Peptidoglycan is assembled from lipid II precursors via sequential glycosyltransferase (GT) and transpeptidation (TP) reactions. These may be performed collectively by arrays of monofunctional enzymes, or individually with bifunctional proteins, bearing a distinct 149  catalytic site for each reaction (Lovering et al., 2007, Wilke et al., 2005). S. aureus has a single known bifunctional enzyme, PBP2. When MRSA is challenged by ß-lactams, the resilient TP activity of PBP2a is able to compensate for the inhibited endogenous TP enzymes (PBP1, PBP3, PBP4; Georgopapadakou and Liu, 1980, Wada and Watanabe, 1998) and TP functionality of PBP2 (Pinho et al., 2001). The GT function of PBP2 remains essential, and together PBP2 and PBP2a form a peptidoglycan that differs only slightly from its regular structure, possessing fewer crosslinks but still able to provide mechanical support to the cell (Pinho et al., 2001). Insight into the ability of PBP2a to resist ß-lactam acylation has been provided by x-ray crystallographic studies of a solubilized form of PBP2a: residues 23-668, bearing the mutation Tyr-23-Met (here denoted PBP2a*). Structures have been determined of the apo PBP2a* enzyme and PBP2a* forced to react with ß-lactams via the use of high inhibitor concentrations and extended reaction times (Lim and Strynadka, 2002). It was observed that the enzyme featured an unusually narrow active site groove, in which the active site serine nucleophile and local secondary structure elements are required to undergo a significant conformational change in order to react with ß-lactams. This alteration of the usual PBP architecture incurs an energetic cost for ß-lactam reactivity, but retains the ability to react with the physiological stem peptide substrate. The narrow active site groove likely serves to restrict access to the serine nucleophile – a process known to contribute to resistance in S. pneumoniae PBP2x (Dessen et al., 2001, Chesnel et al., 2003, Pernot et al., 2004). It was postulated that any anti-MRSA ß-lactam (like ceftobiprole and other next-generation cephalosporins; several are depicted in Fig. 4.1) would owe its effectiveness to increased hydrophobic interaction with PBP2a. Such increased favorability (and thus frequency of 150  formation) of the non-covalent Michaelis complex should facilitate acylation by compensating for the energetic cost of rearrangement (Lim and Strynadka, 2002). It is the purpose of this study to test this hypothesis and detail the precise nature of interactions between PBP2a and ceftobiprole, elucidating the features most important for recognition and reactivity. We present the 2.9 Å X-ray structure of a PBP2a*-ceftobiprole complex, and postulate the basis for its susceptibility to the new generation of ß-lactams. Besides providing insight into the basis for improved efficacy of the newest generation of antibacterials over traditional variants, it is hoped that this study may be used to guide the design of even more effective anti-MRSA therapeutics.  151  4.2  Results and Discussion  4.2.1 Examination of Current Anti-MRSA Drug Design Strategy in Light of the PBP2a*-Ceftobiprole Complex In order to test existing hypotheses about anti-MRSA ß-lactam drug design, we obtained the first structure of PBP2a* in complex with the most clinically advanced “nextgeneration” ß-lactam, ceftobiprole. PBP2a* was crystallized as described previously (Lim and Strynadka, 2002), and the ß-lactam acyl-enzyme was generated via crystal soaking (cocrystallization experiments were unsuccessful). Consistent with expectations, clear electron density indicating an acyl-enzyme species is observed around the nucleophile Ser-403 (Fig. 4.2). Importantly, this observation indicates that ceftobiprole behaves largely like traditional, generally non-effective ß-lactams; it does not elicit its anti-MRSA activity through an entirely novel mechanism.  Figure 4.2: Ceftobiprole adduct and representative electron density for monomer A, with all maps calculated to 2.9 Å resolution. The adduct is represented in stick form, with atom types colored by CMYK convention (C, yellow; N, dark blue; O, red; S, green). i) Final model, 2Fo-Fc map, contoured at 1 sigma. The R1 group is shown in both of the proposed 152  conformations, with equivalency up to the C7[C2] atom. ii) Results of test refinement with ceftobiprole in the “up” orientation only, both 2Fo-Fc (blue, 1 sigma) and Fo-Fc (green, 3 sigma) maps shown. The Fo-Fc difference density is continuous with the proximity of the atomic model, suggesting either multiple conformations or chemical modification of the terminal heterocycle of R1. With no evidence for the latter, and precedent for multiple R1 orientations in other transpeptidase structures, the modeling of two conformations was preferred. iii) As in ii), but with ceftobiprole modeled in the “down” state only.  Despite the moderate 2.9Å resolution of the diffraction data, it was obvious that the monomer A acyl-enzyme could not be described using a single conformation of the inhibitor (Fig. 4.2), and so we have modeled the R1 substituent in two positions. With Ser403 anchoring the cephalosporin nucleus, we choose to label these two conformations as the “up” and “down” states (Fig. 4.2). These two conformations arise from rotation about the ceftobiprole C7[C1-C2] bond, with equivalency of all groups (including the R2 substituent, cephalosporin nucleus and R1 amide bond) prior to the C7[C2] atom (for nomenclature guidance refer to Roberts, 1992. The two conformations collectively form a plane rich in heteroatoms. The aminothiadiazolyl ring forms VDW interactions with the sidechains of Gln-521 and Ala601 in the “down” position, with its NH2 group at 2.6Å from the carbonyl of Gly520 but at a planar angle unsuitable for hydrogen bonding (Mills and Dean, 1996; Pettersen et al., 2004). The aminothiadiazolyl ring shifts to displace Glu-602 in the “up” position, placing the R1 sulfur atom closer to the hydrophobic Tyr-446 and vinylpyrrolidinone group of the R2 substituent, possibly allowing favorable sulfur-! interactions (Ringer et al., 2007). The NH2 group in the “up” position is directed toward the mobile Arg-612 and Gln-613 residues, but at too great a distance (>4.2 Å) to interact; hydrogen bonding to the proximate (2.9Å) carboxylate of Glu-602 is impermissible at the observed angle. 153  The oxime group of ceftobiprole interacts with the amide of Glu-602 in the “down” conformation, and with the side-chain of Glu-602 in the “up” conformation. The x-ray data do not allow for refinement of the conformational occupancies, but we estimate the down state to be of marginally greater occupancy in this study (occupancies were refined 0.6 “down” : 0.4 “up”). Conversely, it is possible to model the acyl-monomer B (the less reactive of the two PBP2a* monomers, as judged from limited studies with other ß-lactams; Lim and Strynadka, 2002) in a single, “down” conformation. The proposed multiple conformations observed for the R1 substituent have no precedent in previous acyl-enzyme complexes of PBP2a*. However, despite this study being the first transpeptidase complex with a ß-lactam possessing this particular R1 substituent, both “up” and “down” conformations have been observed separately in studies involving the similar R1 group of cefotaxime (albeit these alternative conformations are observed only in different enzymes; Contreras-Martel et al., 2006, Macheboeuf et al., 2005, Silvaggi et al., 2005). In general, it is interesting to note that R1 substituents tend towards the “up” position in low molecular weight transpeptidases, and the “down” position in the high molecular weight enzymes. The two monomers present a further difference in that the R2 substituent of ceftobiprole is clearly visible in monomer A, but is either missing or disordered in monomer B. The presence/absence of the R2 group in other transpeptidase acyl-enzyme structures is apparently not governed by either enzyme or drug-type; the R2 group of nitrocefin cannot be observed in the DD-peptidase, yet is visible in PBP1b. In contrast, in the PBP1b:cefotaxime complex, the R2 group is unobserved (Sauvage et al., 2005, Macheboeuf et al., 2005). This particular case may be explained, however, by chemical reactivity: the acetoxy group of the cefotaxime R2 is a good leaving group and will likely be eliminated (as 154  an acetic acid molecule) rapidly following acylation, leaving a methylene group in its place. In contrast the R2 side chain of nitrocefin (absent in the DD-peptidase structure), being a poorer leaving group, should be retained after ring opening. The importance of visualizing the interactions between the ß-lactam R2 substituent and PBP2a* cannot be overstated – compounds varying substantially only at R2, identical at the lactam nucleus and similar at the R1 substituent, exhibit markedly different efficacy against MRSA: compare the less effective cefpirome to the experimental compound S-3578 (Fig. 4.1; Tsuji et al., 2003). Note also, despite differences at R1, substantial equivalence between R2 groups results in closely comparable efficacies for ceftobiprole and LB11058 (Fig. 4.1; Vouillamoz et al., 2004). Despite the evident importance of the R2 group, there is no consensus pharmacore for effective compounds, although the surprising lack of efficacy of the larger nitrocefin and cefodizime compounds indicates that negative charge in this region appears to be disfavored. This is in accordance with the partially hydrophobic, partially electronegative character of this region of the active site groove (Fig. 4.3).  155  Figure 4.3: Comparison between PBP2a* ceftobiprole and nitrocefin adducts. The ceftobiprole complex (yellow) and nitrocefin complex (blue) are shown with selected sidechains and ß-lactam ligands in stick form, colored by atom type (Ceftobiprole C, yellow; Nitrocefin C, light blue; N, dark blue; O, red; S, green). An orthogonal view of the two acylenzyme adducts is provided in the lower left inset. The chirality of the ceftobiprole R2 group causes it to lie out of plane with the nitrocefin dinitrobenzene moiety, and shifts the relative position of the Met641 region. The preference of the “up” orientation for the R1 group in the nitrocefin-bound structure results in a different conformation of the Gln521 sidechain, which stacks with the aminothiadiazolyl group of ceftobiprole in the “down” orientation.  156  In monomer A of our PBP2a*-ceftobiprole complex, residues Met-641, Thr-600 and Tyr-446 sandwich the rings of the R2 substituent. This position is approximately equivalent to the R2 co-ordination observed for enzyme complexes with earlier generation cephalosporins (represented by the position of nitrocefin in Fig. 4.3), both for resistant (PBP2a; Lim and Strynadka, 2002) and non-resistant enzymes (PBP1b; Macheboeuf et al., 2005). The electron density is clear for the Tyr-446 sidechain and ceftobiprole R2 group, but enzyme acylation has caused an apparent disordering of the Met-641 region. The apparent loss of PBP2a "-helicity upon acylation with ceftobiprole (discussed below) seems likely to arise from induced change in this region, as residue Met-641 initiates an "-helix (AA 641658) in both the apoenzyme and previous ß-lactam complexes. This conformational change is most probably a consequence of the chiral nature of the R2 group of ceftobiprole, which causes the pyrrolidine ring to lie out of plane with the nitrocefin di-nitrobenzene ring, and towards the region otherwise occupied by Met-641. This suggests that the interaction between the Met-641 side chain and the pyrrolidine ring is less defined than the !-! stacking interactions between Tyr-446 and the 3 conjugated double bonds (extending from the sixmembered cephalosporin ring to the carboxy group) of the R2 group. It is unclear whether this change is of consequence to the ability of ceftobiprole to acylate PBP2a effectively, but it should be noted that this stereochemistry is not likely to be a defining feature of ceftobiprole action as completely planar R2 substituents also show good anti-MRSA activity (Fig. 4.1). On the other hand, the PBP2a mutant Ser-649-Ala (postulated to perturb the alpha helix that also bears Met-641) arises in both of the MRSA strains in which high-level ceftobiprole resistance (MIC 128mg/L) has been generated via a mutated PBP2a (Banerjee et al., 2008). It may be that this mutant confers resistance by compromising 157  the flexibility of the Met-641 and thus its ability to accommodate the ceftobiprole R2 group. 4.2.2 Conformational Change in PBP2a Upon Acylation Both structural and kinetic studies have previously noted the importance of the conformational differences between the PBP2a apo enzyme and acyl-enzyme states (Roychoudhury et al., 1994, Lim and Strynadka, 2002, Fuda et al., 2004, Fuda et al., 2005). The changes involve a twisting of the central ß-sheet, most notably at strand ß3, altering the position and orientation of Ser598, Gly599 and Thr600 to avoid steric clashes with the reacted ß-lactam (Lim and Strynadka, 2002). Upon acylation with ceftobiprole, these backbone (and sidechain Cß) atoms twist with an RMSD ~1.2Å, comparable to that observed in the nitrocefin-PBP2a* structure (RMSD ~1.1Å). The neighboring Ala601 and Glu602 residues reposition themselves in response to the R1 substituent (all-atom RMSD ~1.5Å, compared with RMSD ~1.2 Å in complex with nitrocefin), as do residues Arg612 and Gln613 (all-atom RMSD ~2.9 Å; for comparison, 625 C" pairs of PBP2a* upon acylation vary with RMSD ~0.5 Å). The conformational change also involves movement of the N-terminus of "2 (backbone-plus-Cß, residues 403 and 404 have a RMSD ~0.9 Å) such that Ser-403 becomes more exposed and able to react. Apart from such detailed reorganization events, the opposing side of the active site cleft responds to acylation largely via a rigid body movement (RMSD ~1.0 Å) of regions 421-475 (RMSD ~0.9 Å) and 501-526 (RMSD ~1.1Å). Without information on the Michaelis complex with an unreacted ß-lactam, it is difficult to identify roles for the R1 and R2 groups in directly facilitating these conformational changes, since the only the endpoint of the acylation reaction is observed. A comparable study with the analogous BlaR ß-lactam sensor indicates that, at least in some cases, it may not be possible to distinguish secondary structure changes that take place at the 158  early stages of the reaction (e.g. substrate-binding) from those due to the acylation event itself (Thumanu et al., 2006). The planarity and hydrophobicity of R2 seem essential in allowing the lactam extended access to the narrow active site groove; these properties mediate a close fit with Tyr-446, Met-641 and Thr-600. The clustering of anti-MRSA lactams around cephalosporins (and some carbapenems but not penams—i.e. bearing the penicillin nucleus) indicates the importance of this planar interaction. The hydrophobic nature of the R2 group was initially proposed to allow a non-specific mode of interaction with PBP2a (Lim and Strynadka, 2002), and we observe that no hydrogen bonding occurs involving the CO or NH groups of the ceftobiprole R2 moiety. This lack of specificity would presumably allow “sliding” of the drug, with its distance to the Ser nucleophile varying at different points in the reaction cycle. With the available evidence we propose that the rigid-body movement of the outermost part of the active site (regions 421-475 and 501-526) occurs as a natural “breathing” of the enzyme, and that ceftobiprole and other effective ß-lactams are able to bind to the narrow cleft in its relatively exposed state via hydrophobic interaction with the extended R2 group. The binding event is associated with rearrangement of the elements on the opposing side of the cleft, most notably at ß3 and "2. Since ß3 residue Thr-600 contacts R2 and also forms part of the oxyanion hole, it appears to play a pivotal role in linking acylation to substrate binding in the course of ß3 strand rearrangement. The role of the ß-lactam in eliciting change at "2 is less obvious, although movement of Ser-403 into the acylation position seems only possible upon removal of steric clashes with "2 residue Lys-406 and ß3 upon formation of the Michaelis complex. Ring opening of the lactam forces the R1 and R2 groups into close proximity, an interaction well tolerated in the “up” position of 159  ceftobiprole, but strikingly detrimental (by examination of MIC values) in bulkier compounds like methicillin (Fig. 4.1; Guignard et al., 2005). Importantly, this difference in steric accommodation of the R1 group likely accounts for the increased acylation rate of penicillin G compared to methicillin, and results in a translation of the acyl-enzyme adduct in the active site and also a difference in co-ordination to the Ser-462-Xxx-Asn conserved motif (Lim and Strynadka, 2002), i.e. a hydrogen bond forms between the penicillin R1 carboxy group and N! of Asn-464. This close approach of the lactam to both the Ser-403Xxx-Xxx-Lys and Ser-462-Xxx-Asn motifs during acylation may also play a role in the general preference of PBP2a for cephalosporins over penams. To probe the relevance of acylation-induced conformational change in solution, we performed CD spectroscopy on the native soluble PBP2a, with the first 20 amino acids deleted (nPBP2a*). We then measured CD spectra of nPBP2a* upon acylation with the ßlactams ceftobiprole, cephalothin, benzyl penicillin, and imipenem. Reaction with all ßlactams affected nPBP2a* spectra at wavelengths 198-206 nm and 215-230 nm, with acylation resulting in increased ellipticity in both regions (Fig. 4.4). The increases in ellipticity in the 215-230 nm region correspond to predicted decreases in "-helical content with a shift in favor of random coil and ß-sheet structure. However, the prediction algorithm K2d (Andrade et al., 1993, Merelo et al., 1994) strongly overpredicts the "-helical content of nPBP2a* (giving estimates of ~70% compared with ~35% observed in the crystal structure) so rather than specific values, percentage change in the secondary structure upon covalent reaction with the ß-lactam is discussed here. Benzyl penicillin produced the strongest effects in this region of the spectrum, while cephalothin, imipenem and ceftobiprole had similar effects to one another. 160  Figure 4.4: Change in UV-CD spectrum during acylation of nPBP2a* by !-lactams. (a) UV-CD spectrum of native protein (solid line) and of nPBP2a* after 60 min incubation with 10 µM ceftobiprole (#),100 µM benzyl penicillin ($),100 µM cephalothin (!) and 1 mM imipenem (%). The spectrum of the unreacted &-lactam has been deducted in each case. (b) Difference spectra created by deducting the native protein spectrum from that of each of the acyl-enzyme spectra.  161  The changes in spectrum at shorter wavelengths were more difficult to examine because of the greater noise (due to ß-lactam absorbance) in the region 180-200 nm. However, ceftobiprole consistently produced less change in the 198-206 nm band whilst imipenem consistently produced the largest change. Averaging 10 separate experiments revealed that benzyl penicillin produced a significantly greater effect than cephalothin (Table 4.1). The changes in spectrum were largely associated with initial binding: within 2 minutes of ß-lactam exposure, and with acylation less than halfway complete, more than 70% of the change in secondary structure had occurred (Table 4.1). Table 4.1. Changes in UV-CD spectrum upon acylation of nPBP2a*. Integrated change in ellipticity Predicted secondary(mdeg•nm) Compound structure content 198-206 nm 210-230 nm change (%) 2 min 20 min 2 min 20 min None 0 0 0 0 Ceftobiprole -8.3% 221 272 1009 1136 Cephalothin -7.1% 298 312 830 848 Benzyl penicillin -16.7% 305 336 1390 1520 Imipenem -8.3% 540 584 980 1024 The thermal stability of nPBP2a* and the various acyl-enzyme complexes were compared using differential scanning calorimetry. The native nPBP2a* has a thermal transition (Tm) at 45.8 °C (Fig. 4.5). Ceftobiprole was the only ß-lactam observed to induce an increase in the thermal stability of the complex formed with nPBP2a* (Table 4.2). Such a change upon ligand binding is usually associated with the formation of additional interactions between the protein and ligand, which act to stabilize the complex. The cephalosporins nitrocefin and cephalothin each caused only a slight (yet reproducible) shift to a slightly lower temperature while causing a decrease in the peak height and inducing a 162  slight broadening of the transition. Benzyl penicillin and piperacillin produced larger shifts towards a lower Tm, suggesting that the acyl-enzyme formed with these penicillins had lost some stabilizing interactions during complex formation that are not compensated for by additional interactions with the inhibitor. Finally, imipenem produced a profound decrease in the thermal stability of the protein Tm, lowering the Tm by over 9 °C.  Figure 4.5: Differential scanning calorimetry traces for nPBP2a* and acyl-enzyme complexes formed with ß-lactams. (a) blue, heat capacity change in the presence of 1 mM imipenem, (b) magenta, heat capacity change in the presence of 100 "M benzyl penicillin, (c) green, heat capacity change observed for native protein, (d) red, heat capacity change in the presence of 10 "M ceftobiprole. The ß-lactam was also present in the reference cell and traces in the absence of protein showed only a steady decline over the temperature range studied (not shown). 163  Table 4.2: Thermal stability determined for nPBP2a* upon acylation. Peak width at Total heat Tm !Tm Peak height Compound half height capacity (°C) (°C) (mcal/°C) (°C) (mcal) None 45.8 6.00 3.75 26.25 Ceftobiprole 47.5 +1.7 3.04 6.88 23.32 Cephalothin 45.3 -0.5 3.83 4.03 24.95 Nitrocefin 45.2 -0.6 3.95 3.83 23.12 Piperacillin 43.4 -2.4 4.54 3.80 20.95 Benzyl penicillin 41.3 -4.5 4.48 3.75 19.31 Imipenem 36.5 -9.3 3.63 4.17 18.70 The changes in thermal stability of the acyl-enzyme complexes may influence kinetic measurements: for example, at 37 °C (a temperature often chosen for investigation of these kinetics; Graves-Woodward and Pratt, 1998), the native protein is at the beginning of the unfolding transition and can be expected to be unstable under these conditions whereas the imipenem complex (Tm 36.5°C) would be at its thermal transition midpoint. Even at 25°C, one should expect significant denaturation to occur if the measurements extend over hours (Fuda et al., 2004). This may suggest two different modes of PBP2a inhibition: destabilization (e.g. by imipenem) and irreversible inhibition by stabilization of the Michaelis complex followed by acylation (e.g. by ceftobiprole). The acylation of nPBP2a* by ceftobiprole proceeds more rapidly than with any other ß-lactam. This results in an acyl-enzyme complex that exhibits relatively little perturbation of the CD spectrum in the region 198-205nm and increased thermal stability. The acyl-enzyme complexes formed with all other ß-lactams tested showed more perturbation of the CD spectrum in the region 198-205nm and decreased thermal stability. These qualitative differences are consistent with greater stabilization of nPBP2a* by the formation of extended contacts made with ceftobiprole (over and above those made with other ß-lactams) 164  at the active site. 4.2.3 Understanding Resistance from a Susceptible/Resistant Protein Comparison Complementing the observation of PBP2a* in complex with both effective and ineffective ß-lactams (Lim and Strynadka, 2002), the structural determination of the susceptible transpeptidase of S. aureus PBP2 (Lovering et al., 2007) now permits more direct analysis of the resistance phenomenon. Although PBP2a and susceptible transpeptidases bind different peptidoglycan precursors, both enzymes utilize the same Gly5 crossbridge peptide substrate. Therefore useful comparisons can be made between the transpeptidase functionality of each. The differences between the PBP2a* apoenzyme and acyl enzymes, along with CD spectra indicating a structural rearrangement upon acylation, support the hypothesis that serine nucleophile accessibility is the major factor in determining whether acylation will occur. Such an hypothesis requires ready availability of the serine nucleophile in the susceptible transpeptidase PBP2, and a structural overlay of the two enzymes indeed confirms this (RMSD of apo enzymes’ serine nucleophiles ~2.1Å; Fig. 4.6). Point mutations of the endogenous PBP2 lead, at most, to so-called borderline resistance (termed BORSA; Nadarajah et al., 2006), and not the high level ß-lactam resistance achieved through acquisition of the foreign PBP2a.  165  Figure 4.6: Structural overlay to demonstrate serine nucleophile availability hypothesis. The methicillin-susceptible endogenous PBP2 (blue) is superimposed upon the resistant PBP2a* apoenzyme (green) and PBP2a*•ceftobiprole (yellow) structures. The ceftobiprole adduct is depicted as transparent, space-filled spheres (of the atoms’ Van der Waals radii, colored by atom type) to aid visualization of the limited space available to ßlactams when entering the active site pocket. The serine nucleophile of PBP2 (Ser-398) is vastly more exposed than that of PBP2a*, hence is presumably more “available” for acylation in vivo. Ser-403 of unacylated PBP2a* sits deeper in the active site cleft, and the shift of this residue to the orientation observed for the acylated complex requires a rearrangement of strand ß3.  Nevertheless, several BORSA-related mutations change a residue in the susceptible PBP2 to one bearing greater resemblance to that found in the corresponding position in PBP2a. These mutations may indicate small, incremental contributions to the resistance capability of PBP2a. The Pro-458-Leu mutation of PBP2 overlays structurally with position 166  Phe-466 in PBP2a, suggesting that a hydrophobic residue at this position may play some role in resistance to traditional ß-lactams. Similarly, the Asp-606-[Ala/Val] mutation of PBP2 corresponds to the hydrophobic Ile-614 of PBP2a. Pro-458 is located at the periphery of the active site, on the N-terminus of "5 (standard transpeptidase fold numbering, Lim 2002). Asp-606 interacts with the loop preceding the nucleophile (located at the N-terminus of "2). As neither Asp-606 or Pro-458 project into the potential ß-lactam binding cleft, we postulate that both mutations have secondary structure packing effects that alter the exposure of the Ser nucleophile or disrupt other elements of the active site (Pro-458 is also in close proximity to the conserved Ser-Xxx-Asn motif). It is becoming increasingly recognized that mutations distal to an enzyme site may affect catalysis via dynamic coupling of whole protein motion (Saen-Oon et al., 2008). The packing of "5 appears to be particularly important in both enzymes, as a PBP2a Val-470-Glu mutation was observed when attempting to generate increased resistance via passage of MRSA in a medium containing an investigational carbapenem (Katayama et al., 2004) and ceftobiprole itself (Banerjee et al., 2008). It appears that global, mosaic differences between the two known resistant and susceptible PBPs in S. aureus, have complex effects on the positioning of the serine nucleophile, yielding large differences in acylation rates. The high-affinity binding of ceftobiprole to PBP2a appears to overcome the inaccessibility of the serine nucleophile while retaining strong activity against susceptible PBPs (Jones, 2007, Livermore, 2006).  167  4.3  Conclusion The design of ceftobiprole and other anti-MRSA ß-lactams to increase contacts  between the R2 group and PBP2a follows the consensus that the molecule mimics both the acceptor and donor substrate peptides (Fig. 4.1, based on analysis of the DDcaboxypeptidase/transpeptidase from Streptomyces; Lee et al., 2001). Modification of the lactam nucleus in drug design must balance such considerations alongside ß-lactamase susceptibility and drug solubility (many of the compounds with increased affinity for PBP2a suffer from a decrease in serum solubility due to their greater hydrophobicity; Guignard et al., 2005). In tandem with the observation that altering the size and planarity of the R2 substituent assists in binding PBP2a, the results of this study indicate that there may be some benefit in investigating other substituents at the R1 position. Traditionally, modifications at R1 affect the ability of the compound to combat the hydrolytic ß-lactamase enzymes (Hebeisen et al., 2001). If the proposed multiple conformations for the R1 group in monomer A mirror the binding events that occur physiologically, it would be intuitive to use these interactions to guide drug design, through the proven relationship between drug dissociation constant and PBP2a acylation rate (Roychoudhury et al., 1994, Fuda et al., 2004). While the emergence of ceftobiprole-resistant MRSA strains appears to be unlikely (Bogdanovich et al., 2005, Heller et al., 2004), generation of in vitro resistance to ceftobiprole (Banerjee et al., 2008) and other investigational next-generation cephalosporins (Katayama et al., 2004) demonstrate that the probability of this is certainly non-zero. Widespread clinical use of such next-generation anti-MRSA drugs, while of immediate benefit to patients, will only increase this likelihood in the long term. This underscores the 168  urgency of continuing to develop improved anti-MRSA agents and it is hoped that the detailed protein-drug interactions and insights into conformational changes upon binding presented herein will assist this effort.  169  4.4  Materials and Methods  4.4.1 Protein Expression, Purification and Crystallization S. aureus PBP2a* was prepared as described previously (Lim and Strynadka, 2002). Briefly, a construct expressing residues 23-668 (with the mutation Tyr-23-Met) produced soluble protein, lacking the N-terminal transmembrane sequence. PBP2a* was purified to near homogeneity through sequential Q-Sepharose, CM-Sepharose, hydroxyapatite and Sephacryl-100 chromatography steps. The protein was concentrated to 20 mg/mL in a final buffer of 5 mM NaHCO3 pH 8.0, 150 mM NaCl. Crystals of the orthorhombic form of PBP2a* were grown using 1 mL of protein crystal stock mixed with 1 mL of reservoir solution (100 mM Hepes pH 7.0, 0.88 M NaCl, 20% v/v PEG 500 MME and 16 mM CdCl2) in a typical sitting drop vapor-diffusion experiment. 4.4.2 Derivatization, Data Collection and Structure Determination Ceftobiprole (Basilea Pharmaceutica) was solubilized in DMSO (with 0.2% v/v TFA, agitated for 20 min at room temperature) to give a stock solution of 20 mM. This was diluted 1:200 into PBP2a* cryoprotection solution (100 mM Hepes pH 7.0, 1 M NaCl, 28% v/v PEG 550 MME and 16 mM CdCl2). This derivatization solution was added directly to the crystallization drop (in three 1 "L increments, over a time course of 35 minutes) and left overnight to promote acylation. Acylated PBP2a* crystals were plunge cooled to 100K and screened for diffraction quality. Data were collected at Beamline 8.3.1 of the Advanced Light Source (Berkeley, CA), using x-rays of 1.116 Å wavelength. Full statistics are reported in Table 4.3.  170  Table 4.3: Data collection and refinement statistics. Crystal Parametersa Spacegroup P212121 Cell Dimensions: a x b x c, Å 80.8 # 103.5 # 186.5 Resolution, Å 2.9 (3.06-2.9) Wavelength, Å 1.11588 No. Reflections 106,806 No. Unique Reflections 32,591 Average Redundancy 3.3 (2.8) Completeness, % 91.9 (63.5) 15.9 (3.3) I/!I b R sym 0.07 (0.248) Refinement Statisticsc 1254 observable amino acid residues, 2 ASU Contents ceftobiprole molecules, 87 H2O, 7 Cd2+, 4 ClRwork / Rfree 0.212 / 0.287 rmsd bond lengths, Å 0.018 rmsd bond angles, ° 1.78 a Statistics for highest resolution shell are given in parentheses b Rsym = $|(Ihkl)-<I>| / $(Ihkl) c 5% of reflections excluded from refinement and used to calculate Rfree  4.4.3 Data Processing and Refinement Data were processed using MOSFLM, scaled with SCALA, and data file manipulations performed using the CCP4 suite of programs (CCP4, 1994). Due to the isomorphous nature of the crystals with those previously reported (Lim and Strynadka, 2002), the co-ordinates of the PBP2a* nitrocefin acyl-enzyme (PDB code 1MWS) were used to generate initial phases. The possibility of model bias arising from the inclusion of the nitrocefin adduct in the initial phase estimates was negated through deletion of the nitrocefin adduct from the model and the usage of the Prime and Switch procedure in RESOLVE (Terwilliger, 2004) to provide maps for the modeling of the ceftobiprole moiety. Geometric restraints for ceftobiprole were generated using the PRODRG server at Dundee 171  (Schuttelkopf and van Aalten, 2004). For the initial refinement stages, CNS was used (Brunger et al., 1998), with final stages using TLS in REFMAC (Murshudov et al., 1997). Due to the moderate resolution of the data, refinement utilized NCS restraints for residues 27-402 and 404-668, with the acyl-serine 403 excluded. The final model has an Rwork/Rfree of 0.212/0.287 and is of good stereochemical quality (Table 4.3). Figures were generated using PyMol (DeLano, 2002). 4.4.4 Circular Dichroism Spectroscopy Circular dichroism measurements were performed on nPBP2a* (native soluble PBP2a with the first 20 amino acids deleted) at 25 °C with the reactants dissolved in 20 mM Tris, adjusted to pH 7.5 with H2SO4, 1 M NaCl. All solutions were thoroughly degassed and passed through a 0.22 "m filter before use. The enzyme concentration was 5 "M and the ßlactam antibiotic was generally in the range 10 to 100 "M (final concentrations). The spectra between 180 and 300 nm were recorded using a 2 mm optical path cell in a !*-180 spectrometer (Applied Photophysics, Surrey, U.K.) using adaptive sampling at 1 nm intervals counting down to an error of ±0.05 nm with a maximum sample size of 500,000 for each wavelength. 4.4.5 Differential Scanning Calorimetry The thermal stability of nPBP2a* under various conditions was investigated by differential scanning calorimetry using the VP-DSC microcalorimeter (Microcal Incorporated). Solutions of 10 "M nPBP2a* in 20 mM Tris, adjusted to pH 7.5 with H2SO4, 1M NaCl with and without ß-lactam were used as samples. All solutions were thoroughly degassed before use and the reference cell in each experiment was filled with an aliquot of 172  buffer against which the protein solution had previously been dialyzed overnight. All samples were scanned from 30 to 70 °C, at a rate of 1 °C per minute and data were baseline corrected, smoothed using a Savitsky-Golay 9 point smoothing algorithm and analyzed using the Origin Scientific plotting software supplied with the instrument.  173  4.5  Acknowledgements and Funding The authors thank D Lim for his valuable efforts in the early stages of the project.  NCJ Strynadka is a Howard Hughes Medical Institute International Research Scholar, and this work was also funded by a Canadian Institutes of Health Research Fellowship Award to AL Lovering, as well as Michael Smith Foundation for Health Research fellowships to AL Lovering and M Gretes. We are grateful for beam time and assistance at the Advanced Light Source (ALS).  174  4.6  References  ANDRADE, M. A., CHACÓN, P., MERELO, J. J. & MORÁN, F. (1993) Evaluation of secondary structure of proteins from UV circular dichroism spectra using an unsupervised learning neural network. Protein Eng, 6, 383-90. BANERJEE, R., GRETES, M., BASUINO, L., STRYNADKA, N. & CHAMBERS, H. F. (2008) In Vitro Selection and Characterization of Ceftobiprole-Resistant Methicillin-Resistant Staphylococcus aureus. Antimicrob Agents Chemother, 52, 2089-96. BOGDANOVICH, T., EDNIE, L. M., SHAPIRO, S. & APPELBAUM, P. C. (2005) Antistaphylococcal activity of ceftobiprole, a new broad-spectrum cephalosporin. Antimicrob Agents Chemother, 49, 4210-9. BOUCHER, H. W., TALBOT, G. H., BRADLEY, J. S., EDWARDS, J. E., GILBERT, D., RICE, L. B., SCHELD, M., SPELLBERG, B. & BARTLETT, J. (2009) Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis, 48, 1-12. BOUCHILLON, S. K., IREDELL, J. R., BARKHAM, T., LEE, K. & DOWZICKY, M. J. (2009) Comparative in vitro activity of tigecycline and other antimicrobials against Gram-negative and Gram-positive organisms collected from the Asia-Pacific Rim as part of the Tigecycline Evaluation and Surveillance Trial (TEST). International Journal of Antimicrobial Agents, 33, 130-6. BRUNGER, A. T., ADAMS, P. D., CLORE, G. M., DELANO, W. L., GROS, P., GROSSEKUNSTLEVE, R. W., JIANG, J. S., KUSZEWSKI, J., NILGES, M., PANNU, N. S., READ, R. J., RICE, L. M., SIMONSON, T. & WARREN, G. L. (1998) Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr, 54, 90521. CCP4 (1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr, 50, 760-3. CHAMBERS, H. F. (1995) In vitro and in vivo antistaphylococcal activities of L-695,256, a carbapenem with high affinity for the penicillin-binding protein PBP 2a. Antimicrob Agents Chemother, 39, 462-6. CHESNEL, L., PERNOT, L., LEMAIRE, D., CHAMPELOVIER, D., CROIZÉ, J., DIDEBERG, O., VERNET, T. & ZAPUN, A. (2003) The structural modifications induced by the M339F substitution in PBP2x from Streptococcus pneumoniae further decreases the susceptibility to betalactams of resistant strains. J Biol Chem, 278, 44448-56. CONTRERAS-MARTEL, C., JOB, V., DI GUILMI, A. M., VERNET, T., DIDEBERG, O. & DESSEN, A. (2006) Crystal structure of penicillin-binding protein 1a (PBP1a) reveals a mutational hotspot implicated in beta-lactam resistance in Streptococcus pneumoniae. J Mol Biol, 355, 684-96. DELANO, W. L. (2002) The PyMOL Molecular Graphics System Palo Alto, CA, DeLano Scientific. DESSEN, A., MOUZ, N., GORDON, E., HOPKINS, J. & DIDEBERG, O. (2001) Crystal structure of PBP2x from a highly penicillin-resistant Streptococcus pneumoniae clinical isolate: a mosaic framework containing 83 mutations. J Biol Chem, 276, 45106-12. DOSHI, R. K., PATEL, G., MACKAY, R. & WALLACH, F. (2009) Healthcare-associated Infections: epidemiology, prevention, and therapy. Mt Sinai J Med, 76, 84-94. FANTIN, B., PIERRE, J., CASTÉLA-PAPIN, N., SAINT-JULIEN, L., DRUGEON, H., FARINOTTI, R. & CARBON, C. (1996) Importance of penicillinase production for activity of penicillin alone or in combination with sulbactam in experimental endocarditis due to methicillin-  175  resistant Staphylococcus aureus. Antimicrob Agents Chemother, 40, 1219-24. FRANCIOLLI, M., BILLE, J., GLAUSER, M. P. & MOREILLON, P. (1991) Beta-lactam resistance mechanisms of methicillin-resistant Staphylococcus aureus. J Infect Dis, 163, 514-23. FUDA, C., HESEK, D., LEE, M., MORIO, K., NOWAK, T. & MOBASHERY, S. (2005) Activation for catalysis of penicillin-binding protein 2a from methicillin-resistant Staphylococcus aureus by bacterial cell wall. Journal of the American Chemical Society, 127, 2056-7. FUDA, C., SUVOROV, M., SHI, Q., HESEK, D., LEE, M. & MOBASHERY, S. (2007) Shared functional attributes between the mecA gene product of Staphylococcus sciuri and penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus. Biochemistry, 46, 8050-7. FUDA, C., SUVOROV, M., VAKULENKO, S. B. & MOBASHERY, S. (2004) The basis for resistance to beta-lactam antibiotics by penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus. J Biol Chem, 279, 40802-6. GEORGOPAPADAKOU, N. H. & LIU, F. Y. (1980) Binding of beta-lactam antibiotics to penicillin-binding proteins of Staphylococcus aureus and Streptococcus faecalis: relation to antibacterial activity. Antimicrob Agents Chemother, 18, 834-6. GRAVES-WOODWARD, K. & PRATT, R. F. (1998) Reaction of soluble penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus with beta-lactams and acyclic substrates: kinetics in homogeneous solution. Biochem J, 332 ( Pt 3), 755-61. GUIGNARD, B., ENTENZA, J. M. & MOREILLON, P. (2005) Beta-lactams against methicillinresistant Staphylococcus aureus. Current Opinion in Pharmacology, 5, 479-89. HANSSEN, A. M., KJELDSEN, G. & SOLLID, J. U. (2004) Local variants of Staphylococcal cassette chromosome mec in sporadic methicillin-resistant Staphylococcus aureus and methicillinresistant coagulase-negative Staphylococci: evidence of horizontal gene transfer? Antimicrob Agents Chemother, 48, 285-96. HEBEISEN, P., HEINZE-KRAUSS, I., ANGEHRN, P., HOHL, P., PAGE, M. G. & THEN, R. L. (2001) In vitro and in vivo properties of Ro 63-9141, a novel broad-spectrum cephalosporin with activity against methicillin-resistant staphylococci. Antimicrob Agents Chemother, 45, 825-36. HELLER, S., MARRER, E., PAGE, M. G. P., SHAPIRO, S. & THENOZ, L. (2004) Development of endogenous resistance by staphylococci to BAL9141 and comparators, abstract P675. Clin. Microbiol. Infect., 10, 163. IDA, T., TSUSHIMA, M., ISHII, T., ATSUMI, K. & TAMURA, A. (2002) CP6679, a new injectable cephalosporin with broad spectrum and potent activities against methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa. J Infect Chemother, 8, 138-44. JONES, M. E. (2007) In-vitro profile of a new beta-lactam, ceftobiprole, with activity against methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect, 13 Suppl 2, 17-24. KATAYAMA, Y., ZHANG, H. Z. & CHAMBERS, H. F. (2004) PBP 2a mutations producing veryhigh-level resistance to beta-lactams. Antimicrob Agents Chemother, 48, 453-9. LEE, W., MCDONOUGH, M. A., KOTRA, L., LI, Z. H., SILVAGGI, N. R., TAKEDA, Y., KELLY, J. A. & MOBASHERY, S. (2001) A 1.2-A snapshot of the final step of bacterial cell wall biosynthesis. Proc Natl Acad Sci USA, 98, 1427-31. LIM, D. & STRYNADKA, N. C. (2002) Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol, 9, 870-6. LIVERMORE, D. (2006) Can beta-lactams be re-engineered to beat MRSA? Clin Microbiol Infect, 12  176  Suppl 2, 11-6. LORTHOLARY, O., LEFORT, A., TOD, M., CHOMAT, A. M., DARRAS-JOLY, C., CORDONNIER, C. & ONCO-HÉMATOLOGIE, C. D. R. S. L. I. E. (2008) Pharmacodynamics and pharmacokinetics of antibacterial drugs in the management of febrile neutropenia. The Lancet infectious diseases, 8, 612-20. LOVERING, A. L., DE CASTRO, L. H., LIM, D. & STRYNADKA, N. C. (2007) Structural insight into the transglycosylation step of bacterial cell-wall biosynthesis. Science, 315, 1402-5. MACHEBOEUF, P., DI GUILMI, A. M., JOB, V., VERNET, T., DIDEBERG, O. & DESSEN, A. (2005) Active site restructuring regulates ligand recognition in class A penicillin-binding proteins. Proc Natl Acad Sci U S A, 102, 577-82. MARTY, F. M., YEH, W. W., WENNERSTEN, C. B., VENKATARAMAN, L., ALBANO, E., ALYEA, E. P., GOLD, H. S., BADEN, L. R. & PILLAI, S. K. (2006) Emergence of a clinical daptomycin-resistant Staphylococcus aureus isolate during treatment of methicillin-resistant Staphylococcus aureus bacteremia and osteomyelitis. J Clin Microbiol, 44, 595-7. MERELO, J. J., ANDRADE, M. A., PRIETO, A. & MORAN, F. (1994) Proteinotopic Feature Maps. Neurocomputing. MILLER, L. G. & KAPLAN, S. L. (2009) Staphylococcus aureus: A Community Pathogen. Infect Dis Clin North Am, 23, 35-52. MILLS, J. E. & DEAN, P. M. (1996) Three-dimensional hydrogen-bond geometry and probability information from a crystal survey. J Comput Aided Mol Des, 10, 607-22. MURSHUDOV, G. N., VAGIN, A. A. & DODSON, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr, 53, 240-55. NADARAJAH, J., LEE, M. J., LOUIE, L., JACOB, L., SIMOR, A. E., LOUIE, M. & MCGAVIN, M. J. (2006) Identification of different clonal complexes and diverse amino acid substitutions in penicillin-binding protein 2 (PBP2) associated with borderline oxacillin resistance in Canadian Staphylococcus aureus isolates. J Med Microbiol, 55, 1675-83. NOEL, G. J., BUSH, K., BAGCHI, P., IANUS, J. & STRAUSS, R. S. (2008a) A randomized, doubleblind trial comparing ceftobiprole medocaril with vancomycin plus ceftazidime for the treatment of patients with complicated skin and skin-structure infections. Clin Infect Dis, 46, 647-55. NOEL, G. J., STRAUSS, R. S., AMSLER, K., HEEP, M., PYPSTRA, R. & SOLOMKIN, J. S. (2008b) Results of a double-blind, randomized trial of ceftobiprole treatment of complicated skin and skin structure infections caused by gram-positive bacteria. Antimicrob Agents Chemother, 52, 37-44. PAGE, M. G. (2004) Cephalosporins in clinical development. Expert Opin. Investig. Drugs, 13, 973-85. PERNOT, L., CHESNEL, L., LE GOUELLEC, A., CROIZE, J., VERNET, T., DIDEBERG, O. & DESSEN, A. (2004) A PBP2x from a clinical isolate of Streptococcus pneumoniae exhibits an alternative mechanism for reduction of susceptibility to beta-lactam antibiotics. J Biol Chem, 279, 16463-70. PETTERSEN, E. F., GODDARD, T. D., HUANG, C. C., COUCH, G. S., GREENBLATT, D. M., MENG, E. C. & FERRIN, T. E. (2004) UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem, 25, 1605-12. PINHO, M. G., DE LENCASTRE, H. & TOMASZ, A. (2001) An acquired and a native penicillinbinding protein cooperate in building the cell wall of drug-resistant staphylococci. Proc Natl Acad Sci USA, 98, 10886-91.  177  RINGER, A. L., SENENKO, A. & SHERRILL, C. D. (2007) Models of S/pi interactions in protein structures: comparison of the H2S benzene complex with PDB data. Protein Sci, 16, 2216-23. ROBERTS, J. (1992) Cephalosporins. Kirk!Othmer Encyclopedia of Chemical Technology. ROYCHOUDHURY, S., DOTZLAF, J. E., GHAG, S. & YEH, W. K. (1994) Purification, properties, and kinetics of enzymatic acylation with beta-lactams of soluble penicillin-binding protein 2a. A major factor in methicillin-resistant Staphylococcus aureus. J Biol Chem, 269, 12067-73. SAEN-OON, S., GHANEM, M., SCHRAMM, V. L. & SCHWARTZ, S. D. (2008) Remote mutations and active site dynamics correlate with catalytic properties of purine nucleoside phosphorylase. Biophys J, 94, 4078-88. SAUVAGE, E., HERMAN, R., PETRELLA, S., DUEZ, C., BOUILLENNE, F., FRÈRE, J. M. & CHARLIER, P. (2005) Crystal structure of the Actinomadura R39 DD-peptidase reveals new domains in penicillin-binding proteins. J Biol Chem, 280, 31249-56. SCHUTTELKOPF, A. W. & VAN AALTEN, D. M. (2004) PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr D Biol Crystallogr, 60, 1355-63. SILVAGGI, N. R., JOSEPHINE, H. R., KUZIN, A. P., NAGARAJAN, R., PRATT, R. F. & KELLY, J. A. (2005) Crystal structures of complexes between the R61 DD-peptidase and peptidoglycan-mimetic beta-lactams: a non-covalent complex with a "perfect penicillin". Journal of Molecular Biology, 345, 521-33. STRYJEWSKI, M. E. & CHAMBERS, H. F. (2008) Skin and soft-tissue infections caused by community-acquired methicillin-resistant Staphylococcus aureus. Clin Infect Dis, 46 Suppl 5, S368-77. SUMITA, Y., FUKASAWA, M., MITSUHASHI, S. & INOUE, M. (1995) Binding affinities of betalactam antibodies for penicillin-binding protein 2' in methicillin-resistant staphylococcus aureus. J Antimicrob Chemother, 35, 473-81. TERWILLIGER, T. (2004) SOLVE and RESOLVE: automated structure solution, density modification and model building. Journal of synchrotron radiation, 11, 49-52. THUMANU, K., CHA, J., FISHER, J. F., PERRINS, R., MOBASHERY, S. & WHARTON, C. (2006) Discrete steps in sensing of beta-lactam antibiotics by the BlaR1 protein of the methicillinresistant Staphylococcus aureus bacterium. Proc Natl Acad Sci USA, 103, 10630-5. TSIODRAS, S., GOLD, H. S., SAKOULAS, G., ELIOPOULOS, G. M., WENNERSTEN, C., VENKATARAMAN, L., MOELLERING, R. C. & FERRARO, M. J. (2001) Linezolid resistance in a clinical isolate of Staphylococcus aureus. Lancet, 358, 207-8. TSUJI, M., TAKEMA, M., MIWA, H., SHIMADA, J. & KUWAHARA, S. (2003) In vivo antibacterial activity of S-3578, a new broad-spectrum cephalosporin: methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa experimental infection models. Antimicrob Agents Chemother, 47, 2507-12. VOUILLAMOZ, J., ENTENZA, J. M., HOHL, P. & MOREILLON, P. (2004) LB11058, a new cephalosporin with high penicillin-binding protein 2a affinity and activity in experimental endocarditis due to homogeneously methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother, 48, 4322-7. WADA, A. & WATANABE, H. (1998) Penicillin-binding protein 1 of Staphylococcus aureus is essential for growth. J Bacteriol, 180, 2759-65. WALSH, T. R. & HOWE, R. A. (2002) The prevalence and mechanisms of vancomycin resistance in Staphylococcus aureus. Annu Rev Microbiol, 56, 657-75.  178  WAXMAN, D. J. & STROMINGER, J. L. (1983) Penicillin-binding proteins and the mechanism of action of beta-lactam antibiotics. Annu Rev Biochem, 52, 825-69. WILKE, M. S., LOVERING, A. L. & STRYNADKA, N. C. (2005) Beta-lactam antibiotic resistance: a current structural perspective. Curr Opin Microbiol, 8, 525-33. ZHOU, Y., ANTIGNAC, A., WU, S. W. & TOMASZ, A. (2008) Penicillin-binding proteins and cell wall composition in beta-lactam-sensitive and -resistant strains of Staphylococcus sciuri. J Bacteriol, 190, 508-14.  179  CHAPTER 5: IN VITRO SELECTION AND CHARACTERIZATION OF CEFTOBIPROLE-RESISTANT METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS " 5.1  Introduction Methicillin-resistant Staphylococcus aureus (MRSA) is a major cause of nosocomial and  community-associated illness throughout the world. Infection with MRSA results in diverse clinical manifestations ranging from minor skin infections to life threatening bacteremia and pneumonia. The recent, alarming emergence of vancomycin-intermediate and resistant MRSA strains highlights the need for new antibiotics to treat this highly drug-resistant pathogen. MRSA is resistant to most #-lactam antibiotics because it has acquired the gene mecA, most likely by horizontal transfer from coagulase-negative staphylococcal species (Wu et al., 1996). MecA encodes the low affinity penicillin-binding-protein PBP2a, which unlike other penicillin-binding proteins (PBPs) remains active and allows for cell wall biosynthesis at otherwise lethal #-lactam concentrations. Ceftobiprole (BPR) is an investigational, broadspectrum cephalosporin that binds to PBPs, including PBP2a, with high affinity. It is active against MRSA as well as other gram-positive and gram-negative pathogens, and is stable against the staphylococcal #-lactamase. For most clinical isolates of MRSA and methicillinsensitive Staphylococcus aureus (MSSA), BPR minimal inhibitory concentrations (MICs) are <2 "g/ml, below the preliminary breakpoint for BPR of 4 "g/ml (Noel, 2007). In vitro, the MIC90 for most MRSA and MSSA strains has been reported to be <4 "g/ml (Appelbaum,  "  A version of this chapter has been published. R Banerjee, M Gretes, L Basuino, N Strynadka, HF Chambers. (2008) In vitro selection and characterization of ceftobiprole-resistant methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother, 52(6): 2089–2096.  180  2006). BPR is currently in Phase III clinical studies to treat complicated skin and soft tissue infections as well as pneumonia. It has favorable pharmacokinetics in vivo and has been granted fast track approval by the FDA (Lodise et al., 2007). Although staphylococci have a proven ability to develop resistance to most antibiotics in clinical use, the potential for MRSA to become resistant to BPR is thought to be low, based on in vitro studies and clinical data. In an attempt to generate BPR-resistant mutants in vitro, Heller et. al. passaged 3 different MRSA strains on BPR-containing agar. After 50 passages the strains developed a BPR MIC of only 32 "g/ml (Heller et al., 2004). Similarly, other investigators performed 50 passages of MRSA and MSSA strains in 1 ml broth cultures containing subinhibitory BPR and found that the maximal BPR MIC achieved was 8 "g/ml, only a fourfold increase over that of the parental strain (Bogdanovich et al., 2005). Investigators in our laboratory have previously generated an MRSA strain resistant to L-695,256, another investigational #-lactam that binds tightly to PBP2a (Katayama et al., 2004). This strain, COL52, contains 5 amino acid changes in PBP2a and has a nafcillin MIC >2000 "g/ml, more than eightfold greater than strains with wild type PBP2a. We hypothesized that an analogous set of mutations in mecA, the gene encoding PBP2a, could confer resistance to BPR. Unlike the studies described above, we passaged large volume broth cultures in BPR to increase the inocula and the potential for identification of a resistant mutant.  181  5.2  Results  5.2.1 Generation of BPR-Resistant Mutants To generate BPR-resistant MRSA, 300 ml broth cultures of three plasmid transformants of strain COLnex were serially passaged in subinhibitory and increasing BPR concentrations for a total of 28 days. COLnex(pYK21) was BPR resistant initially, with the MIC for the strain being 16 "g/ml, above the BPR breakpoint of 4 "g/ml. Within 9 days of passaging, strains COLnex(pAW8) and COLnex(pYK20) became resistant to BPR, with MICs for the strains reaching 8 "g/ml (Fig. 5.1, inset). For all strains, BPR MICs continued to increase in a stepwise fashion until approximately 3 weeks (Fig. 5.1). For the pAW8 (mecA-negative) transformant, MICs plateaued for several days at 64, 128, and then 256 "g/ml of BPR. For the pYK20 and pYK21 transformants, MICs plateaued at 32, 64, and 128 "g/ml. The pAW8 transformant achieved the highest level of resistance. Population analyses demonstrated homogeneous resistance in all strains and confirmed the MICs obtained using broth cultures (Fig. 5.2). MICs increased as follows: 256-fold (from 1 to 256 "g/ml) for COLnex(pAW8), 32-fold (from 4 to 128 "g/ml) for COLnex(pYK20), and 8fold (from 16 to 128 "g/ml) for COLnex(pYK21). All the BPR-resistant, passaged mutants had significantly reduced doubling times compared to those of their parent strains. This difference was most pronounced for the mecA-negative strains COLnex(pAW8) and COLnex(pAW8)D28, the pAW8 transformant obtained after 28 days of passaging (Fig. 5.3).  182  Figure 5.1: BPR resistance developed in multiple steps during the serial passage of strains COLnex(pAW8) ($), COLnex(pYK20) (%), and COLnex(pYK21) (!). Strains were passaged each day in fresh media, at the previous day’s concentration of BPR or double this value. The maximum BPR concentration tolerated each day is indicated on the y-axis. Strains growing at the indicated concentration were used for the next passage. The inset is a closeup view of results from the first 2 weeks of passaging. The dashed line represents the preliminary BPR breakpoint of 4 "g/ml.  Figure 5.2: Population analyses showing BPR susceptibilities of prepassage strains [COLnex(pAW8) ($), COLnex(pYK20) (%), and COLnex(pYK21) (#)] and strains passaged in BPR for 28 days [COLnex(pAW8)D28 (! ); COLnex(pYK20)D28 (!); and COLnex(pYK21)D28 (!). The y axis indicates the number of cells (expressed as the log10 number of CFU per milliliter) growing on BPR-containing agar. 183  Figure 5.3: Growth curves of strains before and after BPR passage. COLnex(pAW8) ($), COLnex(pAW8)D28 (! ), COLnex(pYK20) (%), COLnex(pYK20)D28 (!), COLnex(pYK21) (#), COLnex(pYK21)D28 (!). OD578, optical density at 578 nm.  5.2.2 BPR-Resistant Strains Developed PBP2a Mutations The sequence of mecA was determined in post-BPR-passage strains COLnex(pYK20)D28 and COLnex(pYK21)D28. In COLnex(pYK20)D28, six new mutations arose in mecA; five resulted in amino acid changes within the transpeptidase domain of PBP2a and one change occurred in the non-penicillin-binding domain. MecA in COLnex(pYK21)D28 had four new mutations; three within the transpeptidase domain, one within the non-penicillin-binding domain (Fig. 5.4). Two identical changes occurred in both passaged COLnex(pYK20)D28 and COLnex(pYK21)D28: E447K and S649A. Intriguingly, the highest level of BPR-resistance was observed in COLnex(pAW8)D28, a strain lacking mecA entirely. The absence of mecA was confirmed by PCR given this surprising result.  184  Figure 5.4: Schematic of PBP2a and amino acid substitutions in BPR-passaged strains containing plasmid-carried mecA, COLnex(pYK20) and COLnex(pYK21). The day of serial passage and the corresponding amino acid changes identified are shown. In the schematic, vertical black lines indicate three penicillin-binding motifs, the arrowhead denotes a transmembrane anchor, the speckled region denotes the non-penicillin-binding domain (nonPBD), and the diagonally striped region denotes the transpeptidase domain. Underlined amino acid substitutions arose independently in derivatives of both COLnex(pYK20) and COLnex(pYK21). D0, D13, D15, and D28, days 0, 13, 15, and 28.  In the mecA-containing strains, BPR resistance increased with the number of mutations in PBP2a. Comparison of PBP2a sequences and BPR MICs from strains passaged for roughly 2 weeks and 4 weeks revealed that acquisition of S649A was associated with doubling of the BPR MIC from 64 to 128 in COLnex(pYK21)D28. Similarly, acquisition of S649A, F467Y, and R589K resulted in a fourfold increase in BPR MIC (from 32 to 128 "g/ml) in COLnex(pYK20)D28 (Fig. 5.4). To determine if mutations in other penicillin-binding proteins (PBPs) also occurred in BPR-resistant strains, PBPs 1-4 were sequenced in all pre- and post-passaged strains. No changes were identified in PBP1 in any strain. There were conservative amino acid 185  substitutions in PBP2 in COLnex(pYK20)D28 (A172V) and in PBP3 in COLnex(pYK21)D28 (D562E). There were two amino acid substitutions within PBP4 in COLnex(pAW8)D28 (E183A and F241R). 5.2.3 PBP2a Mutations Confer BPR-Resistance in mecA-Containing Strains To determine if the mutations in PBP2a were sufficient to confer BPR-resistance, we exploited the fact that our strains contained plasmid-encoded mecA. When plasmids containing mutant mecA (generated by passaging in BPR) were transformed into COLnex, this BPR-sensitive strain was converted into a BPR-resistant strain. The BPR MIC in transformed strains was similar to that of serially passaged strains suggesting that the presence of mutant mecA alone was sufficient to confer full BPR resistance. There was no difference between resistance due to mecA from COLnex(pYK20)D28 or COLnex(pYK21)D28. As expected, transformation of plasmid pAW8D28 did not confer BPR-resistance as this plasmid does not contain the mecA insert (Fig. 5.5A).  Figure 5.5: (A) Population analyses of COLnex transformed with plasmids derived from BPR-passaged strains, COLnex(pAW8)T ($), COLnex(pYK20)T (%), and 186  COLnex(pYK21)T (!). The y axis indicates the number of cells (expressed as the log10 number of CFU per milliliter) growing on BPR-containing agar. (B) Population analyses of BPR-resistant strains cured of plasmid by passage for seven days in the absence of tetracycline, COLnex(pAW8)c ($), COLnex(pYK20)c (%), and COLnex(pYK21)c (!). The y axis indicates the number of cells (expressed as the log10 number of CFU per milliliter) growing on BPR-containing agar.  Conversely, loss of PBP2a converted BPR-resistant strains into BPR-sensitive strains. Post-passage BPR-resistant strains were cured of their plasmids by serially passaging them in the absence of tetracycline. Their susceptibility to BPR was assessed by population analysis. When cured of the mecA-containing plasmid, COLnex(pYK20)c and COLnex(pYK21)c became highly susceptible to BPR (Fig. 5.5B). Only COLnex(pAW8)c remained resistant to BPR after loss of this plasmid. 5.2.4 Molecular Modeling of PBP2a Mutants Molecular models of PBP2a mutants were generated based on the crystal structure of wild type PBP2a from S. aureus (Lim and Strynadka, 2002). Amino acid substitutions found in COLnex(pYK20)D28 and COLnex(pYK21)D28 fall into three main groups. Group 1 mutations introduce charged or polar groups to the interior of the protein (not directly at the active site but nearby, within the transpeptidase domain) and make contact with a helix ($2) that plays a key role in acylation of #-lactam substrates and inhibitors. An example of a Group 1 mutation is V470E, which introduces a negatively charged or polar glutamate into the otherwise hydrophobic protein interior adjacent to helix "2; likewise, the Group 1 mutations F467Y and I563T both introduce a polar hydroxyl (–OH) group into the hydrophobic interior of the protein near this helix (Fig. 5.6A). Perturbation of this helix may have implications for the rate of inhibition of PBP2a as detailed below. 187  Figure 5.6: Structural perspectives of BPR-resistant PBP2a mutant forms. (A and B) Molecular modeling of mutant forms of PBP2a bound to the cephalosporin nitrocefin showing nitrocefin in blue (cephalosporin core structure), orange (variable group R1), and green (group R2); the native protein backbone and residues in gray; residues subject to mutation in cyan (native residue) and purple (corresponding mutant residue); and noncarbon atoms in CPK. (A) Group 1 mutations (F467Y, V470E, and I563T) likely affect BPR binding by perturbing helix 2 (shown as a yellow ribbon) by introducing adjacent polar or charged groups into the hydrophobic interior of PBP2a. (B) Group 2 substitutions 188  probably lower BPR-binding affinity by directly affecting the PBP2a active site. These substitutions include Y446L, predicted to disrupt van der Waals contacts and pi bond stacking (!-!) interactions between PBP2a and conjugated double-bond systems of cephalosporin R2 groups; E447K, which may interact electrostatically with E460 to reposition Y446; S649A, predicted to destabilize the alpha helix bearing M641 that interacts with aromatic or hydrophobic R2 groups; and S643N, which may alter the polarity at the active-site entrance. (C) APBS modeling of PBP2a surface charge potential shows native residues E150 and E239 located within an extended swath of negative charge in the putative dimerization domain (the APBS surface is colored blue, white, and red, corresponding to values of +15, 0, and –15 kT/e, respectively, with shading by linear interpolation). Each of these residues is mutated, respectively, in pYK20 and pYK21, possibly contributing to BPR resistance by influencing protein-protein interactions.  Group 2 substitutions occur within or proximal to the ß-lactam binding site and so most likely affect BPR binding directly. Group 2 substitutions include: S649A, which, by deleting the –OH group of Serine 649, abolishes a hydrogen bond that ordinarily stabilizes the alpha helix terminating in M641—a residue shown to interact with the R2 group of the cephalosporin nitrocefin (Lim and Strynadka, 2002); Y446L, which removes the aromatic ring of Tyrosine 446 that interacts with the nitrocefin R2-group (Lim and Strynadka, 2002); and E447K which causes a charge reversal (negative to positive) adjacent to the active site groove (Figs. 5.6B; 5.S1). Subsequent electrostatic attraction between mutant residue Lysine 447 and native residue Glutamate 460 (on the protein surface, ~4Å away) may further distort the active site perhaps also altering the position of Tyrosine 446. Finally, S643N may also affect substrate binding via a polarity change at the active site entrance (Fig. 5.6B). Of note, both S649A and E447K arose independently in COLnex(pYK20)D28 and COLnex(pYK21)D28. Group 3 substitutions occur far from the active site in a domain not involved in penicillin binding and are thus likely to mediate resistance indirectly via interactions with other proteins. Group 3 substitutions include E150K and E239K, found in both pYK20 and 189  pYK21 mutants, respectively. Each introduces a positive charge within the same patch of negatively charged protein surface located in the solvent-accessible, N-terminal region of PBP2a (Fig 5.6C). For the Group 3 mutations, therefore, increased resistance may result from modified protein-protein interactions that rely on the electronegativity of this region of the enzyme (see 5.3 Discussion, below).  Figure 5.S1: Supplemental figure showing comparative structures of !-lactam antibiotics. The differences in PBP2a-binding affinity are due to the nature of the R1 and R2 variable groups (1, cephalosporin backbone). To illustrate this, the chemical structures of ceftobiprole (2) and nitrocefin (3) are shown. For comparison, the structures of the penams penicillin G (4) and methicillin (5) are also given.  190  The remaining mutations do not fit neatly into any group, have no explicable effect, and may merely be incidental: I397T and R589K are located on the protein surface, do not involve formal charge differences and furthermore did not arise independently of other mutations. 5.2.5 Resistance in mecA-negative COLnex(pAW8) Appears to be Mediated by Chromosomal Genes The mecA-negative strain, COLnex(pAW8)D28, displayed the highest level of resistance to BPR among all strains used. This resistance mechanism was not plasmidmediated as the pAW8 plasmid contained no insert, and removal or addition of the plasmid had no effect on the level of BPR resistance. Resistance was not due to #-lactamase production as the COLnex strain and its transformants are all #-lactamase negative (data not shown). BPR resistance in COLnex(pAW8)D28 was also stable over at least 7 days of passaging in the absence of BPR.  191  5.3  Discussion This is the first study to demonstrate that a strain of MRSA can develop high level  BPR resistance in vitro and that BPR-resistance can be mediated by mutations in PBP2a. Unlike other groups, we were able to generate strains that were highly resistant to BPR most likely because we used higher inocula during serial passage of cultures. Maximal resistance to BPR required multiple mutations in PBP2a that occurred in at least three steps and developed after prolonged passage in subinhibitory BPR concentrations. The clinical implications of BPR resistance generated in vitro are unclear. Although the inocula used in our study were similar to concentrations seen in endocarditis or soft tissue abscesses (108 to 109 CFU per ml or gram of tissue; König et al., 1998, LaPlante and Rybak, 2004), cultures were passaged for 9 days before resistant mutants were detected (Fig. 5.1). Several days of infection with such high organism burden is unlikely to occur in an individual patient. However, a prolonged period at high concentration may occur with a strain of MRSA that is transmitted over time among several patients. Furthermore, the ability to develop and tolerate mutations in mecA or other genes mediating BPR resistance may depend on the genetic background, an issue that our experiments did not address. COL, the only parent strain used in our experiments, is an early MRSA strain (Sabath et al., 1972) that is homogeneously resistant, not particularly virulent, and not representative of MRSA strains existing outside the laboratory. Moreover, the BPRresistant strains exhibited significantly impaired growth compared to the parent strain, indicating that they had a fitness cost that may impair survival outside the laboratory (Fig. 5.3). We have shown that in mecA-containing strains, BPR resistance results from changes 192  in PBP2a alone. Because we used strains in which mecA had been excised from the chromosome and expressed from a plasmid, we were able to easily assess the effect of the addition or loss of mutant PBP2a on BPR susceptibility. The introduction of mutant PBP2a into a BPR-sensitive strain conferred full resistance to BPR. Conversely, the loss of mutant PBP2a converted a BPR-resistant strain into a fully sensitive one. The contribution of chromosomal loci in these strains appeared to be minor, indicating that in mecA containing strains, PBP2a is the overwhelming target for BPR. Our molecular modeling data suggest that PBP2a mutations can lead to BPR resistance through three different mechanisms: the inhibition of acylation, the inhibition of substrate binding, and interference with protein-protein interactions. All three mechanisms arose independently in both mecA-containing strains. PBP2a resists ß-lactam inactivation primarily via a highly inefficient acylation step that results from the high energetic cost needed to rearrange strand ß3 and helix "2 (which bears the PBP2a serine 403 nucleophile; Lim and Strynadka, 2002). The polar group 1 mutations (V470E, F467Y, and I563T; Fig. 5.6A) likely further inhibit the acylation step by perturbing helix "2 and increasing the energetic penalty associated with the acylation of S403. Cephalosporins with activity against PBP2a, including BPR (Davies et al., 2007) and nitrocefin (Lu et al., 1999), have improved initial binding to the protein and overall increased acylation efficiency compared to those of other ß-lactam antibiotics. The group 2 mutations identified here (S649A, Y446L, E447K, and S643N) appear to interfere with substrate binding by altering the functional groups that bind to BPR (Fig. 5.6B). Key substitutions appear to be S649A and E447K, as they both arose independently in COLnex(pYK20) D28 and COLnex(pYK21)D28. 193  PBPs, including PBP2a, are thought to be part of a large multi-membered cell wallsynthesizing holoenzyme complex (Pinho and Errington, 2003, Cabeen and Jacobs-Wagner, 2007). The N-terminal domain region of PBP2a, where group 3 substitutions occurred, is thought to be involved in PBP dimerization (InterPro† domain IPR005311). In addition, the farthest N-terminal domain may have a role in cell wall turnover, as it bears homology to YoeB, a protein that enhances cell survival in the presence of cell wall-targeting antibiotics by mediating autolysin activity via protein-protein interactions (Salzberg and Helmann, 2007). As the group 3 substitutions E150K and E239K (Fig. 5.6C) introduced positive charges within an area of otherwise negatively charged surface, they may affect cooperative interactions with other S. aureus PBPs in the process of ß-lactam resistance (Leski and Tomasz, 2005). Future studies that will be useful for the analysis of active-site architecture and the design of ß-lactam antibiotics will include enzyme kinetics, molecular dynamics, and drug resistance studies of PBP2a mutant forms containing one or more of the amino acid substitutions described here. The most surprising result of our study is that the highest level of BPR resistance developed in the mecA-negative strain COLnex(pAW8)D28. BPR resistance in this strain is unlikely to be explained by mutations in the high-affinity PBPs; no changes in PBP1, PBP2, or PBP3 were detected, and two amino acid changes in PBP4 were found. However, PBP4 is not essential and its inactivation in most cases has minimal effects on methicillin resistance (Katayama et al., 2003a). Unidentified chromosomal loci have been implicated previously in oxacillin resistance  †  www.ebi.ac.uk/interpro (Mulder et al., 2007)  194  in clinical MRSA isolates lacking mecA (Tomasz et al., 1989). A number of chromosomal genes may affect ß-lactam resistance in mecA-negative strains. Transposon mutagenesis experiments have previously identified a number of auxiliary genes that influence methicillin resistance and have proven or proposed roles in cell wall synthesis (murE, femA, and femV), the regulation of the global stress response (the B gene), or metabolism (protein kinase and ABC transporter genes; de Lencastre and Tomasz, 1994). In addition, proteins involved in the recruitment of PBPs to the septum (the site of new cell wall synthesis) or autolysins that control cell wall turnover may affect susceptibility to drugs active against the cell wall (Pinho and Errington, 2003). Future experiments will focus on identifying the novel resistance determinant in COLnex(pAW8)D28.  195  5.4  Materials and Methods  5.4.1 Reagents Ceftobiprole (BPR) solution was prepared fresh daily at a concentration of 2 mg/ml and was provided by Johnson and Johnson Pharmaceutical Research and Development. Tetracycline, Cefazolin, Vancomycin, and Nafcillin were obtained from Sigma Chemical Co, St. Louis, MO. Tetracycline was used at 10 !g/ml. 5.4.2 Bacterial Strains All strains were grown in Trypticase soy agar, Trypticase soy broth, or blood agar (Remel) at 37 degrees with aeration. Strains used in this study are listed in Tables 5.1 and 5.2. COLnex is the parental strain in which chromosomal SCCmec had been precisely excised by introduction of plasmid pSR encoding two site-specific recombinase genes, ccrA and ccrB. (Katayama et al., 2003b). Plasmid pAW8 contains a tetracycline resistance marker and has been previously described (Katayama et al., 2003b). Plasmid pYK20 contains wild type mecA cloned into the BamHI site of pAW8; pYK21 contains mutant mecA derived from strain COL52 cloned into pAW8 (Katayama et al., 2004); the native mecA promoter sequence was used. Table 5.1: Parental strains and phenotypes relevant to this studya. Strain Description Phenotype Reference Homogeneously methicillinCOL Mcr Tcr (Tomasz et al., 1989) resistant strain (Katayama et al., COLn Tcs derivative of COL Mcr Tcs 2003b) Antibiotic-selected COLn Highly methicillin (Katayama et al., COL52 variant with mutant mecA resistant; Tcs 2004) SCCmec excision strain (Katayama et al., COLnex Mcs Tcs derived from COLn 2003b) a SCCmec, staphylococcal cassette chromosome mec; Mcr, methicillin resistant; Tcr, tetracycline resistant; Tcs, tetracycline sensitive; Mcs, methicillin sensitive  196  Table 5.2: Derivatives of COLnex with plasmid-carried mecA used in this studyb. Strain Description Phenotype Reference (Katayama et al., 2003b, COLnex(pAW8) mecA-negative strain Mcs Tcr Tomasz et al., 1989) Strain expressing plasmid(Katayama et COLnex(pYK20) Mcr Tcr carried wild-type mecA al., 2003b) Strain expressing plasmidHighly (Katayama et COLnex(pYK21) carried mutant mecA from methicillin al., 2003b) COL52 resistant; Tcr BPR resistant; COLnex(pAW8)D28 Post-BPR passage strain This study Tcr BPR resistant; COLnex(pYK20)D28 Post-BPR passage strain This study Mcr Tcr BPR resistant; COLnex(pYK21)D28 Post-BPR passage strain This study Mcr Tcr Post-BPR passage strain BPR resistant; COLnex(pAW8)c This study cured of plasmid Tcs Post-BPR passage strain BPR sensitive; COLnex(pYK20)c This study cured of plasmid Mcs Tcs Post-BPR passage strain BPR sensitive; COLnex(pYK21)c This study cured of plasmid Mcs Tcs COLnex transformed with post-BPR passage BPR sensitive; COLnex(pAW8)T This study plasmid from Tcr COLnex(pAW8)D28 COLnex transformed with post-BPR passage BPR resistant; COLnex(pYK20)T This study plasmid from Mcr Tcr COLnex(pYK20)D28 COLnex transformed with post-BPR passage BPR resistant; COLnex(pYK21)T This study plasmid from Mcr Tcr COLnex(pYK21)D28 b Mcr, methicillin resistant; Tcr, tetracycline resistant; Tcs, tetracycline sensitive; Mcs, methicillin sensitive.  197  5.4.3 Multipassage Selection in BPR COLnex (pAW8), COLnex (pYK20), and COLnex (pYK21) were serially passaged for 28 days in subinhibitory BPR. For each strain, 300 ml of TSB containing 10 !g/ml tetracycline and varying concentrations of BPR were inoculated at a 1:100 dilution with an overnight culture containing 109 cfu/ml. ml. For each strain, approximately 1012 to 1013 cells are estimated to have been exposed over the entire 28 days. BPR concentration was doubled with each passage, as tolerated. 5.4.4 Plasmid Curing Strains were cured of plasmids by daily passaging in 10 ml TSB lacking antibiotics for 7 days. Cultures were then streaked out on blood agar, and individual colonies were replica plated onto TSA plates with or without tetracycline. PCR of the mecA gene was performed on those colonies that did not grow in tetracycline to confirm loss of plasmid. Plasmids containing mecA were included as positive controls in each PCR experiment. Growth rates of plasmid-cured strains were not determined. 5.4.5 Spectrophotometric ß-Lactamase Assay Broth cultures were grown overnight with tetracycline in the absence or presence of the ß-lactamase-inducing antibiotic cefoxitin at a concentration of 0.5 !g/ml. Cells were pelleted by centrifugation, resuspended in 2 volumes of phosphate-buffered saline, and aliquoted into 0.9-ml fractions. After the addition of 0.1 ml of 1 mM cephaloridine, each tube was incubated at 37°C for 0, 30, or 60 min. Cells were then pelleted by centrifugation at 13,000 x g for 3 min, and the absorbance of the supernatants at 254 nm was read.  198  5.4.6 Population Analysis Population analysis was done by the agar plate method in which approximately 109 CFU was serially diluted and quantitatively inoculated onto a series of TSA plates containing increasing concentrations of antibiotic. Drug plates were used within 1 week. 5.4.7 Growth Curves Volumes of 100 ml of TSB with 10 "g of tetracycline/ml were inoculated at a 1:100 dilution with overnight broth cultures. Cultures were grown at 37°C with aeration, and the optical density at 578 nm was measured every hour for 6 consecutive hours by using a spectrophotometer. All growth experiments were done in duplicate. 5.4.8 DNA Manipulations Primers used in this study are as follows: for mecA:  K23 (5’-TCGTGTCAGATACATTTCGATTG-3’) and K27 (5’-GTTGTAGCAGGAACACAAATGAATAAC-3’;  for PBP1:  PBP1F (5’-AGCAACAACCACAAACTAAGC-3’) and PBP1R (5’-CCTCGTCTACCTTAAAATTCTC-3’;  for PBP2:  PBP2F (5’-TGCATATCAACAAAAAGGTATTG-3’) and PBP2R (5’-CTATTTAGATGTTTCAAAATGTATG-3’);  for PBP3:  PFP3F (5’-GTTTGTTTTCACGTGAACAGAA-3’) and PBP3R (5’-ATTTTGGAATGTAGTTAACTGGG-3’);  for PBP4:  PBP4F (5’-GACATGACTGGGAAGGTGAATT-3’) PBP4R (5’-TAACACCTTTAGCTACACACGT-3’).  Genomic and plasmid DNA was prepared using the Qiagen DNeasy Kit or the Qiagen Qiaprep Spin Miniprep Kit, respectively, using an initial cell lysis step with 199  lysostaphin (Sigma). PCR was performed using 1.25 units of Taq DNA polymerase (All Star Scientific) per 20 "l reaction and a PTC200 thermal cycler (MJ Research) using the following conditions: 95o for 10 minutes, 95o for 30 seconds, 45o for 30 seconds, 70o for 3 minutes for 30 cycles. PCR products were analyzed by gel electrophoresis and purified using the Roche High Pure PCR Product Purification kit and the manufacturer’s instructions. Purified PCR products were sequenced by the University of California, Berkeley sequencing facility. Transformations were performed an Eppendorf 2510 electroporator. 5.4.9 Molecular Modeling Mutants were produced in silico from Protein Data Bank structures 1VQQ (S. aureus apo PBP2a residues 27-668) and 1MWS (S. aureus PBP2a residues 27-668 covalently bound to the cephalosporin nitrocefin) using UCSF Chimera (Pettersen et al., 2004). Mutant structures were energy minimized with Chimera employing the AMBER1994 force field (Cornell et al., 1995) with ANTECHAMBER charge determination (Wang et al., 2006). Modeling was confirmed independently with the Swiss-PDB Viewer (Guex and Peitsch, 1997) energy minimization routine, which uses the GROMOS96 force field (van Gunsteren et al., 1996) with 400 steps of steepest descent. Electrostatic surfaces were calculated (on 1VQQ structures only) using PyMol (DeLano, 2002) and its corresponding Adaptive Poisson-Boltzmann Solver (APBS) plug-in (Lerner and Carlson, 2006). All figures were prepared using UCSF chimera.  200  5.5  Acknowledgments This work was supported by grants from Johnson and Johnson Pharmaceutical  Research and Development and the Canadian Institutes of Health Research, Howard Hughes Medical Institute and Michael Smith Foundation for Health Research (to NCJS). MG is supported by fellowships from the Natural Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research. MG also thanks Dr. Andrew Lovering for productive discussions.  201  5.6  References  APPELBAUM, P. C. (2006) MRSA--the tip of the iceberg. Clin Microbiol Infect, 12 Suppl 2, 3-10. BOGDANOVICH, T., EDNIE, L. M., SHAPIRO, S. & APPELBAUM, P. C. (2005) Antistaphylococcal activity of ceftobiprole, a new broad-spectrum cephalosporin. Antimicrob Agents Chemother, 49, 4210-9. CABEEN, M. T. & JACOBS-WAGNER, C. (2007) Skin and bones: the bacterial cytoskeleton, cell wall, and cell morphogenesis. J Cell Biol, 179, 381-7. CORNELL, W. D., CIEPLAK, P., BAYLY, C. I., GOULD, I. R., MERZ, K. M., FERGUSON, D. M., SPELLMEYER, D. C., FOX, T., CALDWELL, J. W. & KOLLMAN, P. A. (1995) A Second Generation Force Field for the Simulation of Proteins, Nucleic Acids, and Organic Molecules. J. Am. Chem. Soc., 117, 5179-5197. DAVIES, T. A., PAGE, M. G., SHANG, W., ANDREW, T., KANIA, M. & BUSH, K. (2007) Binding of Ceftobiprole and Comparators to the Penicillin-Binding Proteins of Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus and Streptococcus pneumoniae. Antimicrob Agents Chemother. DE LENCASTRE, H. & TOMASZ, A. (1994) Reassessment of the number of auxiliary genes essential for expression of high-level methicillin resistance in Staphylococcus aureus. Antimicrob Agents Chemother, 38, 2590-8. DELANO, W. L. (2002) The PyMOL Molecular Graphics System Palo Alto, CA, DeLano Scientific. GUEX, N. & PEITSCH, M. C. (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis, 18, 2714-23. HELLER, S., MARRER, E., PAGE, M. G. P., SHAPIRO, S. & THENOZ, L. (2004) Development of endogenous resistance by staphylococci to BAL9141 and comparators, abstract P675. Clin. Microbiol. Infect., 10, 163. KATAYAMA, Y., ZHANG, H. Z. & CHAMBERS, H. F. (2003a) Effect of disruption of Staphylococcus aureus PBP4 gene on resistance to beta-lactam antibiotics. Microb Drug Resist, 9, 32936. KATAYAMA, Y., ZHANG, H. Z. & CHAMBERS, H. F. (2004) PBP 2a mutations producing veryhigh-level resistance to beta-lactams. Antimicrob Agents Chemother, 48, 453-9. KATAYAMA, Y., ZHANG, H. Z., HONG, D. & CHAMBERS, H. F. (2003b) Jumping the barrier to beta-lactam resistance in Staphylococcus aureus. J Bacteriol, 185, 5465-72. KÖNIG, C., SIMMEN, H. P. & BLASER, J. (1998) Bacterial concentrations in pus and infected peritoneal fluid--implications for bactericidal activity of antibiotics. J Antimicrob Chemother, 42, 227-32. LAPLANTE, K. L. & RYBAK, M. J. (2004) Impact of high-inoculum Staphylococcus aureus on the activities of nafcillin, vancomycin, linezolid, and daptomycin, alone and in combination with gentamicin, in an in vitro pharmacodynamic model. Antimicrob Agents Chemother, 48, 4665-72. LERNER, M. G. & CARLSON, H. A. (2006) APBS plugin for PyMOL. Ann Arbor, MI, University of Michigan LESKI, T. A. & TOMASZ, A. (2005) Role of penicillin-binding protein 2 (PBP2) in the antibiotic susceptibility and cell wall cross-linking of Staphylococcus aureus: evidence for the cooperative functioning of PBP2, PBP4, and PBP2A. J Bacteriol, 187, 1815-24.  202  LIM, D. & STRYNADKA, N. C. (2002) Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol, 9, 870-6. LODISE, T. P., JR., PYPSTRA, R., KAHN, J. B., MURTHY, B. P., KIMKO, H. C., BUSH, K., NOEL, G. J. & DRUSANO, G. L. (2007) Probability of target attainment for ceftobiprole as derived from a population pharmacokinetic analysis of 150 subjects. Antimicrob Agents Chemother, 51, 2378-87. LU, W. P., SUN, Y., BAUER, M. D., PAULE, S., KOENIGS, P. M. & KRAFT, W. G. (1999) Penicillin-binding protein 2a from methicillin-resistant Staphylococcus aureus: kinetic characterization of its interactions with beta-lactams using electrospray mass spectrometry. Biochemistry, 38, 6537-46. MULDER, N. J., APWEILER, R., ATTWOOD, T. K., BAIROCH, A., BATEMAN, A., BINNS, D., BORK, P., BUILLARD, V., CERUTTI, L., COPLEY, R., COURCELLE, E., DAS, U., DAUGHERTY, L., DIBLEY, M., FINN, R., FLEISCHMANN, W., GOUGH, J., HAFT, D., HULO, N., HUNTER, S., KAHN, D., KANAPIN, A., KEJARIWAL, A., LABARGA, A., LANGENDIJK-GENEVAUX, P. S., LONSDALE, D., LOPEZ, R., LETUNIC, I., MADERA, M., MASLEN, J., MCANULLA, C., MCDOWALL, J., MISTRY, J., MITCHELL, A., NIKOLSKAYA, A. N., ORCHARD, S., ORENGO, C., PETRYSZAK, R., SELENGUT, J. D., SIGRIST, C. J. A., THOMAS, P. D., VALENTIN, F., WILSON, D., WU, C. H. & YEATS, C. (2007) New developments in the InterPro database. Nucl. Acids Res., 35, D224-228. NOEL, G. J. (2007) Clinical profile of ceftobiprole, a novel beta-lactam antibiotic. Clin Microbiol Infect, 13 Suppl 2, 25-9. PETTERSEN, E. F., GODDARD, T. D., HUANG, C. C., COUCH, G. S., GREENBLATT, D. M., MENG, E. C. & FERRIN, T. E. (2004) UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem, 25, 1605-12. PINHO, M. G. & ERRINGTON, J. (2003) Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol Microbiol, 50, 871-81. SABATH, L. D., WALLACE, S. J. & GERSTEIN, D. A. (1972) Suppression of intrinsic resistance to methicillin and other penicillins in Staphylococcus aureus. Antimicrob Agents Chemother, 2, 350-5. SALZBERG, L. I. & HELMANN, J. D. (2007) An antibiotic-inducible cell-wall associated protein that protects Bacillus subtilis from autolysis. J Bacteriol, 189, 4671-80 TOMASZ, A., DRUGEON, H. B., DE LENCASTRE, H. M., JABES, D., MCDOUGALL, L. & BILLE, J. (1989) New mechanism for methicillin resistance in Staphylococcus aureus: clinical isolates that lack the PBP 2a gene and contain normal penicillin-binding proteins with modified penicillinbinding capacity. Antimicrob Agents Chemother, 33, 1869-74. VAN GUNSTEREN, W. F., BILLETER, S. R., EISING, A. A., HÜNENBERGER, P. H., KRÜGER, P., MARK, A. E., SCOTT, A. E. & TIRONI, I. G. (1996) Biomolecular Simulation: The GROMOS96 Manual and User Guide, Zürich, VDF Hochschulverlag AG an der ETH Zürich. WANG, J., WANG, W., KOLLMAN, P. A. & CASE, D. A. (2006) Automatic atom type and bond type perception in molecular mechanical calculations. J Mol Graph Model, 25, 247-60. WU, S., PISCITELLI, C., DE LENCASTRE, H. & TOMASZ, A. (1996) Tracking the evolutionary origin of the methicillin resistance gene: cloning and sequencing of a homologue of mecA from a methicillin susceptible strain of Staphylococcus sciuri. Microb Drug Resist, 2, 435-41.  203  CHAPTER 6: IMPLICATIONS, APPLICATIONS, DIRECTIONS, AND CONCLUSIONS 6.1  Major Contributions and Implications of Research The work described in the preceding chapters represents a substantial contribution to  knowledge in diverse fields of study: protein structure, protein-protein interactions, proteinligand interactions, inhibitor design, and antibiotic resistance. Major contributions are highlighted and discussed below in the context of established concepts and emerging research. 6.1.1 Form and Function of ß-Lactamase Inhibitory Protein Family Members The protein inhibitor BLIP has been studied in great structural and functional detail for nearly two decades. Its close relatives BLIP-I and BLP have, however, until now (Chapter 2) been minimally examined. BLIP-I forms a very high-affinity inhibitory complex with TEM-1, with a sub-nM dissociation constant (Kd) comparable to that of BLIP. It is closely related to BLIP by sequence (32% amino acid identity) and structure (RMSD 1.14 Å for 122 C" atoms). In complex with ß-lactamase TEM-1, BLIP-I exhibits similarity to the BLIP•TEM-1 complex with respect to surface area of interaction and the number and type of interactions formed. However, a nearly 5 Å lateral shift in binding position was observed, and as a result there are a number of differences at BLIP-I binding hotspots; indeed the conservation of residues observed to make contact with TEM-1 was substantially lower than that of residues in the core of the protein. This is in marked contrast with what would be expected by primary sequence alignment. Likewise, protein docking prediction, though promising, cannot yet reliably identify the specific protein-protein interaction interface (Ritchie, 2008). This highlights the importance of experimental structural data to confirm the 204  predictions of computational modeling, in particular for protein-protein interactions (Bordner and Gorin, 2008), especially where flexible loops may be key to binding, as they are in the BLIP•TEM-1 system. Also, the conclusions of functional annotation by homology (Li et al., 2008, Llewellyn and Eisenberg, 2008, Jensen et al., 2009) need to be cautiously considered in this context (i.e. qualitative differences in function are possible despite strong similarity). This is borne out in particular by the case of BLP, which appears to share a nearly identical backbone conformation with BLIP-I and a similar number of identical amino acids, yet does not function to bind TEM-1 or other tested ß-lactamases. In addition to providing information on specific BLIP family members, Chapter 2 also appears to confirm the importance of binding hotspots in protein-protein interactions (Reichmann et al., 2007), which eventually may find application in drug development (Wells and McClendon, 2007). Furthermore, Chapter 2 suggests a new conceptual tool (“energetic penalty”) for analysing the thermodynamics data from mutagenesis experiments in protein-protein interaction systems. This is a first attempt to employ a specific thermodynamics term to approximate the repulsive effects of an unpaired interacting amino acid residue at an interprotein interface. Energetic penalty allows a way to specifically describe one source of specificity within protein-protein interaction systems: binding can be abolished or attenuated via single mismatched mutations, especially involving electrostatic charges. Hopefully this will find application in ongoing studies in the field which employ BLIP as a model system, as BLIP-I and BLP are natural variants that may be viewed as the result of simultaneous, nonalanine mutagenesis incurring changes to the binding interface as well to the backbone conformation. These moreover may have some natural functional role. 205  The BLIP-TEM-1 model system remains an active area of research, with two structures of BLIP in complex with the Klebsiella ß-lactamase KPC in the Protein Data Bank, awaiting release upon publication [PDB IDs: 3e2l, 3e2k]. In addition, genome sequencing projects are ongoing for dozens of Streptomyces spp.† Some of these genomes may yield representatives of the BLIP family, and provide further fertile ground to examine their range of ß-lactamase inhibitory activity and the structural aspects of any discovered protein-protein complexes. 6.1.2 Cyclobutanone Analogues of ß-Lactams The structure of OXA-10 covalently modified by formation of a hemiketal with the cyclobutanone ß-lactam analogue JJ05-850 represents a successful, novel approach to ßlactamase inhibition (Chapter 3). This acyl-enzyme structure provides proof-of-principle for the use of such cyclobutanone analogues as a pharmacore for what may be considered a new subclass of the ß-lactam group of antibiotics. Furthermore, these findings appear to validate the use of chemical moieties for the sole purpose of stabilizing the conformation of the small molecule that is preferentially bound by an enzyme active site. According to transitionstate theory, catalysis is achieved by stabilization of the transition state (TS) i.e. the unfavourable potential energy (arising from stretched chemical bonding lengths and strained bonding angles) of the TS is counterbalanced by increased favourable interactions, necessarily of greater affinity than those made with either product or substrate. The closer an inhibitor resembles this transition state, therefore, the greater its affinity for the active site, and lesser the likelihood that a substrate could outcompete it for binding.  †  For example, see the Genomes OnLine Database v 2.0: www.genomesonline.org/BacteriaTree.html  206  The transition-state analogue (TSA) approach to drug development lately appears quite promising in light of extremely high-affinity inhibitors (with Ki in the nM to high pM range) generated via optimization to predicted TS bonds/angles by analysis of kinetic isotope effects (for a review of this approach see Schramm, 2007; for examples refer to Evans et al., 2007, Zhang et al., 2009). Affinity is critical since achieving and maintaining high concentrations of drugs in the clinic, i.e. in peripheral and deep tissues of infected patients, is a key challenge in drug development and delivery (Roberts and Lipman, 2009). Recently, the general conceptual approach to TS mimicry has been taken in the field of ß-lactamase inhibition to improve the affinity of a tazobactam derivative for the Class A enzyme SHV-1 (Padayatti et al., 2006). By modifying tazobactam with a negatively-charged carboxylate group, the trans-enamine intermediate (which is short-lived and thus likely to approach the structure of a high-energy transition state) appears to be stabilized via the carboxylate contacts with active site residues Lys-234, Ser-130, and the nucleophilic Ser-70. Adding conformational stabilization to the considerations used in drug design (as with inhibitors like JJ05-850) may expand the store of drug candidates for a number of target enzymes involved in various disease states. Inhibition of the OXA family of broad-spectrum ß-lactamases is an active research area. At the time of this writing, the coordinates of four OXA structures in complex with unknown classes of inhibitors are known and have been deposited into the Protein Data Bank for release pending manuscript publication. These include a penem-inhibited structure of OXA-1 (3g4p; Bethel et al., 2008) as well as undescribed inhibitors (“SA4-17”, “SA4-44”, “SA3-53”) in complex with the carbapenem-hydrolyzing OXA-24 (3fyz, 3fv7, and 3fzc, respectively). 207  6.1.3 Features of Ceftobiprole Binding to PBP2a Ceftobiprole (BPR) has been informally termed a “fifth generation” cephalosporin for its ability, unique among ß-lactams, to inhibit the MRSA ß-lactam-resistant transpeptidase PBP2a at clinically-relevant concentrations. BPR has been to designed to preferentially bind the narrow, extended, and predominantly hydrophobic PBP2a active site groove. Thus, through increased binding affinity (i.e. decreased Kd), it is able to overcome the increased energetic cost of acylation believed to underlie ß-lactam resistance. A detailed analysis of its binding interactions with its target, however, was not possible until the crystal structure presented in Chapter 4 was solved. As anticipated, a number of extended hydrophobic and !-! contacts are observed along the extended groove occupied by the BPR R2 group. The R1 group appears to contribute an unexpected degree of binding affinity by exhibiting alternative “up”/“down” conformations. The structural findings are consistent with the working hypothesis that substantial rearrangement of structural elements "2 and ß3 are required for acylation due to the otherwise inaccessible Ser-403 nucleophile (Fig. 4.6); BPR binding also engenders this rearrangement, with increased contacts presumably compensating for the energetic cost. This is supported by achieving acylation of the PBP2a crystal at inhibitor concentration of 100 "M, in contrast with concentrations of 20-40 mM required to acylate PBP2a with poorly-interacting inhibitors penicillin, methicillin, and nitrocefin (Lim and Strynadka, 2002). There are no other publicly available structures of PBP2a apart from these three, which were reported at the time its structure was first solved. Other drugs recommended for use against MRSA (vancomycin, linezolid, clindamycin, daptomycin, doxycycline, minocycline, Trimethoprim-sulphamethoxazole) do not target PBP2a, therefore structural 208  information on drug binding to PBP2a is unrelated to information regarding the binding of their targets (Peterson, 2008). The various shortcomings of these therapeutics (expense, toxicity, lack of bactericidal activity, contra-indication for patients with lung infections) are generally avoided by ß-lactams. Thus confirmation that specific inhibition of PBP2a is possible using a drug of this class raises the prospect of incremental improvement of a generally safe, effective group of cephalosporins to treat MRSA. 6.1.4 Mechanisms of Ceftobiprole Resistance Although no resistance to ceftobiprole was reported during clinical trials (Noel, 2007), Chapter 5 reports that resistance is indeed possible and may arise from multiple factors. Many PBP2a mutations in BPR-resistant strains are located in the transpeptidase (TP) domain and are predicted to alter BPR binding affinity. Three major types of mutation to PBP2a were thus identified through the reported modeling effort: (1) those affecting the energetic cost of acylation by perturbing helix "2, (2) those believed to affect substrate binding directly, and (3) those involving a positive-to-negative charge-reversal distal from the TP domain, which may function by influence the cell wall biosynthesis/degradation balance via modulated protein-protein interactions. Strategy (2) is particularly noteworthy, since PBP2a is believed to be intrinsically resistant to ß-lactams and clinically significant variants are little known; this is in contrast with the Streptococcus pneumoniae resistant TP, PBP2x, which exists as a wide variety of variants with decreased binding affinity, with no reported differences in acylation efficiency. The published structure of PBP2a covalently bound to nitrocefin (PDB ID: 1MWS) was used to predict these Group 2 mutation effects on the binding of BPR by computational modeling. This was a reasonable comparison in the absence, at that time, of direct observations of BPR 209  binding, since BPR and nitrocefin are both built on the characteristic cephalosporin scaffold, and share features common to all ß-lactams, including the atoms of the ß-lactam ring, the sulfur atom and the carboxyl group borne on the fused 5- or 6-membered ring, and the amide group proximal to the R1 variable group (Fig. 5.S1). These atoms are similarly oriented (with an RMSD between 14 atom pairs of only ~1.0 Å after superposition of PBP2a C$ atoms) in all available ß-lactam inhibitor-bound crystal structures (1MWS=nitrocefin-, 1MWT=penicillin G-, 1MWU=methicillin-bound). BPR and nitrocefin share additional structural features including a 5-membered sulfur-bearing heterocyclic ring in the R1 group and a system of conjugated double bonds extending into the R2 group. Group 2 mutation effects are postulated as follows. By abolishing a stabilizing hydrogen bond, the mutation S649A may distort the alpha helix containing Met-641, a residue that forms interactions with the R2 group of nitrocefin (Fig 5.6B). Any such disturbance to this key interacting residue would be expected to diminish binding of BPR, especially given the far greater hydrophobic character of its R2 group. S649A is likely a key resistance determinant, since it arose independently in both COLnex(pYK20)D28 and COLnex(pYK21)D28. The substitution Y446L appears to reduce BPR binding affinity by disrupting van der Waals contacts and abolishing highly favorable !-! interactions between the aromatic Tyrosine 446 and the extended conjugated double bond system of the nitrocefin (and by analogy that of the BPR) R2 group. Movement of native Y446 arising from electrostatic attractive forces between E447K and E460 could have similar effects. It is also worth noting the fairly conservative nature of the mutations to PBP2a (in particular of Group 1) that nevertheless appear to yield high-level BPR resistance. Penicillin resistance in Neisseria gonorrhoeae PBP2 appears to be based perhaps likewise on mutations that do not 210  substantially affect the enzyme structure. Rather, its variant residues are thought to perturb its thermal stability and flexibility, thereby providing resistance to penicillin while maintaining the ability to carry out catalysis on its natural substrate (Powell et al., 2008). Since mutations of all three types were observed in BPR-resistant strains (and confirmed by transformation of plasmid encoding resistant PBP2a into a BPR-susceptible strain), one may infer that this validates the hypotheses concerning the importance of all three of these corresponding resistance strategies: (1) modulation of energetic cost of acylation, (2) modulation of binding affinity, (3) interactions with other proteins, presumably members of the divisome or other CW remodeling enzymes such as autolysins (Pinho and Errington, 2003). Thus taking these potential resistance mechanisms into account in drug design are likewise important. Structure-based drug design efforts appear best able to address strategy (2) directly, improving drug affinity by increasing favourable active site contacts.  211  6.2  Applications and Recommendations for Related Research  6.2.1 ß-Lactamase Inhibitory Protein Family: Protein Engineering and Derived Inhibitors The best test of the hypotheses advanced in Chapter 2 to explain the non-TEM-1binding of BLP is to convert BLP to a TEM-1-binding protein via site-directed mutagenesis or domain-swapping to substitute its residues for those of BLIP-I. It would also be interesting to test the ability of BLIP-I to inhibit the range of ß-lactamases tested with BLIP (Table 1.3). These should include the SHV-1 D104K mutant (interesting because of the close conformational resemblance of BLIP-I-Y50 for the position of the otherwise distally located BLIP-Y50) as well as the clinically ascendant KPC Class A ß-lactamases. Alaninescanning mutagenesis should also be carried out on BLIP-I to test hypotheses concerning key binding residues identified from its structure. Double-mutant cycle mutagenesis at each hotspot thought to interact with TEM-1 or other ß-lactamases would gauge the general applicability of the energetic penalty concepts developed in the chapter. A number of small molecule inhibitors have been designed to mimic BLIP, in order to obtain a new inhibitor for clinical use. This effort could be expanded to include BLIP-I sequence information in the design of inhibitors. Promisingly, it has been shown that TEM1 inhibition by BLIP is unaffected by the clavulanic acid resistance mutation Arg-244-Ser and thus derived inhibitors may be used to overcome inhibitor-resistant ß-lactamases related to TEM-1. There are several good examples of proteinaceous inhibitors that have been reduced to their essential chemical components and recreated as potent analogous of small-molecule inhibitors (Leatherbarrow and Salacinski, 1991, Li et al., 1995, Yanofsky et al., 1996; for a recent, general discussion see Wells and McClendon, 2007). Cyclic peptide inhibitors based 212  on the BLIP domain 1 &-hairpin loop (Ala-46–Tyr-51) have, unfortunately, not thus far proven highly effective; their inhibition constants range in the hundreds of "M (Rudgers et al., 2001) and have not been improved over that of the random fragment peptide C30–D49 described previously (Rudgers and Palzkill, 2001). Another cyclic peptide derived from a region of BLIP optimized for maximum contact with TEM-1 (residues 41-50) gave 2-fold improved inhibition. This remains, however, four to six orders of magnitude poorer than inhibition of TEM-1 by various boronic acid (transition-state) inhibitors, modeled on crystal structures of penicillin-G in the TEM-1 active site (Strynadka et al., 1996b, Ness et al., 2000), and seven orders poorer than by BLIP itself. Clearly the inhibitory effectiveness of BLIP, in part due to whole-molecule charge-charge interactions (including many outside of the active site and perhaps removed from the enzyme-inhibitor interface) that guide docking of the BLIP macromolecule to its target &-lactamase (Selzer and Schreiber, 2001, Joughin et al., 2005), cannot be easily reproduced in a peptide removed from its protein context. The detrimental loss of affinity, however, is accompanied by a corresponding beneficial reduction in specificity. While inhibition constants remain poor for the A46-Y51 peptide (hundreds to thousands of "M), it is an improved inhibitor of SHV-1 in relative terms (owing to the poor, "M inhibition of SHV-1 by BLIP) and inhibits PC-1 and P99, which BLIP fails to inhibit at all. No further work has been published to date attempting to optimize this particular peptide inhibitor. Separately, another group has employed yeast-2-hybrid methods to screen a large library of semi-randomized peptides for TEM-1-binding partners (Sun et al., 2005). Using the polymerase chain reaction (PCR) to amplify a set of oligonucleotides containing random base pairs at selected positions, peptides of the form F-T-I-H-C-(x1)-(x2)-(x3)-A-A-G-D-Y213  Y-(x4)-(x5)-(x6)-(x7)-(x8)-(x9)-G-T-S-F were generated, where (xn) is any amino acid other than E, K, M, Q, or W; these amino acids appear to have been necessarily omitted because of the nature of the genetic code—i.e. to avoid the nonsense codons and to generate as evenly distributed a random sequence as possible. Thus, the ‘core’ of these semi-random peptides is formed by the A46–Y51 BLIP domain 1 loop, plus BLIP flanking residues IHC and TS. The authors do not elaborate on the reasons for choosing the remainder of the constant residues. With 9 positions of 15 possible amino acids, approximately 3 x 1010 (%159) peptides were theoretically generated. Because of experimental limitations the authors estimate they were only able to screen on the order of 107 (<0.1%) of these, although this is still a vast number. Remarkably, only two TEM-1-binding peptides were uncovered, only one of which had the ability to inhibit TEM-1 hydrolysis of ampicillin. The authors do not formally determine Ki but TEM-1 appears to be significantly inhibited at peptide concentrations above ~50"M. The discovery of this new BLIP-inspired inhibitory peptide is noteworthy because eight of its nine randomly-generated residues (all besides x5) were not previously predicted to contribute strongly to BLIP•TEM-1 binding (Petrosino et al., 1999, Rudgers and Palzkill, 1999, Strynadka et al., 1996a, Zhang and Palzkill, 2003, Zhang and Palzkill, 2004). Furthermore, in the inhibitory peptide position x5 is occupied by valine instead of tyrosine. The corresponding residue (Y53) in BLIP is critical for TEM-1 binding and is conserved in BLIP-I. Because of these successes with BLIP-derived inhibitors, there is a good chance that BLIP-I domain 1 loop-derived peptides will function as &-lactamase inhibitors in their own right. While no studies have reported the activity of BLIP-I-derived peptides, it would be of interest to determine the relative effectiveness of peptides derived from its partially divergent 214  domain 1 loop sequence to those of BLIP (E-S-G-D-Y-A versus A-A-G-D-Y-Y). Certain features of the BLIP-I•TEM-1 complex reported in Chapter 2 suggests they may have inhibitory activity: for example, the extensive interactions made by Y50 at the active site (Fig. 2.4a). 6.2.2 Further Development of ß-Lactam Analogues of Cyclobutanones, and Engineering of an OXA-10 Variant for Drug Lead Screening The successful inhibitor-soaking experiment reported in Chapter 3 informs hypotheses for derivatization of the lead compound JJ05-850 to improve its binding to OXA-10. The most obvious modification would be to remove the Cl! (Fig. 3.1) that appears to clash with Val-117-C". Contact improvements could be made to the C3 methoxy group, replacing it with another bulky !-substituent, to maintain the exo- conformational stabilization, while providing increased binding energy. This group could be replaced with a negatively charged carboxylate, for salt-bridge formation with the positively charged Arg-250 guanidinium, as observed with moxalactam in complex with OXA-10 (PDB ID: 1k6r). Compared with the putative hydrogen bond between Arg250 and the C3 methoxy oxygen atom, a salt bridge should provide a greater enthalpic benefit to compensate for the unfavourable entropic and enthalpic changes associated with desolvation of the drug and protein active site. With this in mind, increasing hydrophobic interactions at this moiety should provide even greater improvements to binding energetics. Either an aromatic group here or a conjugated double bond system would potentially allow highly favourable #-# interactions with Phe-208 or Trp-102, given an extended linker, possibly a #-sulfur interaction with Met-99. Refer to Figs. 3.3 and 3.4 for illustration of the potentially improved contacts above. The space of chemical derivatives should be explored by inhibition assays, 215  and may be limited by the feasibility of synthesis as well as solubility, especially for the aromatic/aliphatic substitutions suggested. While improved inhibitors of Class D ESBLs are of interest, clinical inhibitors of metallo-ß-lactamases (MBLs; Class B) are entirely unavailable at present. A major implication of the research in Chapter 3 is the potential use of related cyclobutanone compounds as inhibitors of MBLs. It is not feasible to target these enzymes by traditional mechanism-based inhibition, since no covalent intermediate is formed in the course of their hydrolytic mechanism, in which a water molecule is activated by one or more zinc ions for direct nucleophilic attack on the ß-lactam amide bond. The stability of the cyclobutanone may resist such direct hydrolysis or an MBL inhibitor could specifically be developed (e.g. a hydrate; conceptualized in Fig. 6.1). Efforts in the Strynadka lab to produce a co-crystal structure of the MBL IMP-1 with JJ05-850 are ongoing.  Figure 6.1: Proposed serine- and metalloß-lactamase inhibition by a cyclobutanone ßlactam analogue (Adapted from Johnson et al., 2008).  Of course the inhibitor (or variants) are expected to potentially bind PBPs as well as other classes of serine ß-lactamases (classes A and C). Thus the role of the cyclobutanone inhibitors like JJ05-850 as antibiotics in their own right should also be explored. Binding to susceptible PBPs could be tested by competition assays with radiolabeled substrates. If 216  binding is demonstrated, this would represent another potential advantage of the cyclobutanone ß-lactam mimics. An improved drug, active as both an antibiotic as well as a ß-lactamase inhibitor, could be given as a monotherapy. Alternatively, it could be given in conjunction with a common ß-lactam (as is the widely-prescribed clavulanate/amoxicillin combination; Ball, 2007), with potential synergistic effects. Finally, although difficulty in crystallizing inhibitor complexes arising from crystal deterioration has been attributed to a conformational change taking place upon binding (Maveyraud et al., 2002), my analysis suggests this is not entirely the case (Chapter 3). This could be addressed through site-directed mutagenesis to remove the active site interfering Lys-95 while improving crystal contacts through mutagenesis of residues in the surrounding loop. If successful, this OXA-10 inhibitor-cocrystallization variant would need to be shown to be comparable to wild-type OXA-10 in enzyme kinetics and inhibition experiments. Apart from explaining the origins of crystal degradation upon inhibitor binding and apparent lack of reactivity of the second active site of the OXA-10 physiological dimer, successful engineering of such a protein would greatly facilitate future studies of inhibitor binding and drug lead screening. 6.2.3 Further PBP2a Drug Development Based on Ceftobiprole Complex As with the cyclobutanone analogues, the PBP2a•BPR complex can guide the development of improved inhibitors. Presently, the design goal of BPR and other antiMRSA ß-lactams appears to be to increase contacts between the R2 group and PBP2a. This follows the consensus that the molecule mimics both the acceptor and donor substrate peptides (see Fig. 1.11, based on analysis of the DD-carboxypeptidase/TP from Streptomyces R61; Lee 2001). Chapter 4 validates this approach: altering the size and planarity of the R2 217  substituent indeed appears to assist in binding PBP2a. The results of this study indicate that there may be additional benefit in investigating alternative substituents at the R1 position. Traditionally, modifications at R1 have been made to improve stability against ß-lactamase hydrolysis (Hebeisen 2001). If the proposed multiple conformations for the R1 group in monomer A accurately reflect physiological binding, a double-headed R1 substituent that satisfies both binding conformations (Fig. 4.2) may have improved affinity for the PBP2a active site. Indeed, the branched R1 substituent of the anti-MRSA 2-oxaisocephem OPC20011 (Fig. 6.2; Guignard et al., 2005) may simultaneously bind to both “up” and “down” binding sites. It would be intuitive to use specific interactions here to guide drug design, through the proven relationship between drug dissociation constant and PBP2a acylation rate (Roychoudhury 1994, Fuda 2004). Modifications to R1 must of course be balanced against ß-lactamase susceptibility and drug solubility; many of the compounds with increased affinity for PBP2a suffer from a decrease in serum solubility due to their greater hydrophobicity (Guignard et al., 2005).  Figure 6.2: Chemical structure of investigational anti-MRSA cephalosporin OPC-20011 (Guignard et al., 2005).  The proposed equivalence of the R1 position to the donor peptide may mean that the multiple conformations observed in Chapter 4 (Fig. 4.2) result from the aminothiadiazolyl 218  ring of BPR attempting to satisfy binding to both the substrate Lys3 and D-iGln2 positions (“up” and “down” respectively, Figs. 4.1, 4.2). The cephalosporin nucleus and R2 substituent are proposed to mimic both the leaving D-Ala5 and incoming Gly5. This dual purpose is fulfilled more adequately in certain Gram-negative targets, due to the COOH group on the DAP acceptor peptide, which is missing in peptidoglycans that utilize a Lys or Gly5 acceptor substrate. This difference makes it more difficult to model the Gly5 acceptor over the R2 group, although the position of the carbonyl group of the vinylpyrrolidinone ring would suggest that BPR in fact mimics 3-4 glycine units (Fig. 4.1). The acceptor Gly5 is attached to the Lys-N' of a large, branched pentapeptide sugar substrate. Presumably, this region of the substrate binds near the N-terminal end of helix "11, and it would be interesting to observe whether the larger developmental anti-MRSA ß-lactams like RO4908463 and CS-023 (Guignard et al., 2005) exploit this extended binding pocket. Thus crystal structures of PBP2a acylated to these other experimental compounds should also be sought. The resolution of the PBP2a•BPR complex, as well as the occupancy of the inhibitor at the active site could perhaps be improved. It has been found that the activity of BPR and other cephalosporins against MRSA is improved at acidic pH (Lemaire et al., 2009). Since this improved activity may be related to acylation efficiency and PBP2a crystallizes in neutral pH (7.0), high-throughput robotics screening could be used to find a low-pH crystallization condition. If successful this could be used for co-crystallization (unsuccessful thus far with the present condition) or crystal soaking (which required extended soaking times to obtain the structure reported in Chapter 4). A higher-resolution structure would allow more detailed analysis of active site interactions, and perhaps better understanding of the alternate conformations of the R1 group. 219  6.2.4 Elucidation of Specific Mechanisms of PBP2a- and Non-PBP2a-Based Ceftobiprole Resistance The work in Chapter 5 was published prior to the solution of the PBP2a•BPR structure, presented here in Chapter 4. Now that the mode of binding of BPR to the enzyme is known, modeling of resistant mutants can now be done using the coordinates of the PBP2a-BPR complex (Fig. 6.3).  Figure 6.3: PBP2a acylated with BPR at Ser-403, modeled with BPR resistance-associated mutations.  The PBP2a variants examined here were only tested in vivo, and only in combination as they arose to confer resistance under selective pressure of passage through increasing concentrations of BPR. The individual mutations, therefore, should be made by site-directed mutagenesis to identify the major contributors to resistance. Relevant single and multiple mutants should be crystallized to test the results of in silico modeling. These crystals should be exposed to high concentrations of BPR (as done with methicillin in Lim and Strynadka, 220  2002) to determine if BPR-resistant PBP2a can be forced to bind these and whether binding orientation is affected, as may be expected with Group 2 mutants. Coimmunoprecipitation experiments coupled with mass spectrometry should also be employed using native and Group 3 mutant PBP2a (E150K/E239K) to attempt to identify affected binding partners. Although the transformation of plasmids bearing mutant mecA was sufficient to confer full BPR resistance to susceptible strains, conservative amino acid substitutions in fact occurred in PBP2 in COLnex(pYK20)D28 (A172V) and in PBP3 in COLnex(pYK21)D28 (D562E; Banerjee et al., 2008). There were two amino acid substitutions within PBP4 in COLnex(pAW8)D28 (E183A and F241R; their proximity to the active site is indicated in Fig. 6.4). The mutant PBP4 should also be characterized structurally, either via in silico modelling or crystallization, to determine likely effects on substrate binding. This would represent the first known endogenous PBP in S. aureus to develop resistance to ß-lactam inhibition.  221  Figure 6.4: Solvent-excluded surface rendering of S. aureus PBP4 (PDB ID: 1TVF) with mutations E183A (yellow) and F241R (pink) indicated. To emphasize the active site, the catalytic nucleophile Ser75 is shown in green, and a peptidomimetic boronic acid inhibitor placed in its position in complex with the close homologue PBP5 from E. coli.  A potential direct clinical application of this work would be a PCR-based screening tool for BPR-resistant MRSA. Detection of BPR resistance is important so that therapy can be discontinued in individuals with resistant infections and alternatives (listed in Shittu and Lin, 2006, Peterson, 2008) used instead. This would serve the interests of the patient by avoiding futile therapy and of public health by limiting proliferation of BPR-resistant MSRA. The most surprising result of our study is that the mecA-negative strain, COLnex (pAW8)D28, rapidly developed the highest level of BPR-resistance. The mechanism of BPR resistance in this strain appears to be mediated by one or more chromosomally-encoded genes. Unidentified chromosomal loci have been similarly implicated in oxacillin resistance 222  observed in some clinical MRSA isolates lacking mecA (Tomasz et al., 1989). Work is presently ongoing (Banerjee et al., 2008) to identify the novel resistance determinant(s) in non-mutant mecA COLnex(pAW8)D28.  223  6.3  References  BALL, P. (2007) The clinical development and launch of amoxicillin/clavulanate for the treatment of a range of community-acquired infections. International Journal of Antimicrobial Agents, 30 Suppl 2, S1137. BANERJEE, R., GRETES, M., WEBSTER, D., DERISI, J., & CHAMBERS, H.F. (2008) Characterization of a ceftobiprole-resistant, mecA-negative derivative of methicillin-resistant Staphylococcus aureus (MRSA). Presentation C1-3715, 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy and Infectious Diseases Society of America 46th Annual Meeting, Washington DC, Oct. 25-28, 2008. BETHEL, C. R., DISTLER, A. M., RUSZCZYCKY, M. W., CAREY, M. P., CAREY, P. R., HUJER, A. M., TARACILA, M., HELFAND, M. S., THOMSON, J. M., KALP, M., ANDERSON, V. E., LEONARD, D. A., HUJER, K. M., ABE, T., VENKATESAN, A. M., MANSOUR, T. S. & BONOMO, R. A. (2008) Inhibition of OXA-1 beta-lactamase by penems. Antimicrob Agents Chemother, 52, 3135-43. BORDNER, A. J. & GORIN, A. A. (2008) Comprehensive inventory of protein complexes in the Protein Data Bank from consistent classification of interfaces. BMC Bioinformatics, 9, 234. EVANS, G. B., FURNEAUX, R. H., KELLY, P. M., SCHRAMM, V. L. & TYLER, P. C. (2007) Transition state analogue inhibitors of N-ribosyltransferases: new drugs by targeting nucleoside processing enzymes. Nucleic Acids Symp Ser (Oxf), 63-4. GUIGNARD, B., ENTENZA, J. M. & MOREILLON, P. (2005) Beta-lactams against methicillinresistant Staphylococcus aureus. Current Opinion in Pharmacology, 5, 479-89. JENSEN, L. J., KUHN, M., STARK, M., CHAFFRON, S., CREEVEY, C., MULLER, J., DOERKS, T., JULIEN, P., ROTH, A., SIMONOVIC, M., BORK, P. & VON MERING, C. (2009) STRING 8-a global view on proteins and their functional interactions in 630 organisms. Nucleic Acids Research, 37, D412-6. JOHNSON, J. W., EVANOFF, D. P., SAVARD, M. E., LANGE, G., RAMADHAR, T. R., ASSOUD, A., TAYLOR, N. J. & DMITRIENKO, G. I. (2008) Cyclobutanone mimics of penicillins: effects of substitution on conformation and hemiketal stability. J Org Chem, 73, 6970-82. JOUGHIN, B. A., GREEN, D. F. & TIDOR, B. (2005) Action-at-a-distance interactions enhance protein binding affinity. Protein Sci, 14, 1363-9. LEATHERBARROW, R. J. & SALACINSKI, H. J. (1991) Design of a small peptide-based proteinase inhibitor by modeling the active-site region of barley chymotrypsin inhibitor 2. Biochemistry, 30, 10717-21. LEMAIRE, S., GLUPCZYNSKI, Y., DUVAL, V., JORIS, B., TULKENS, P. M. & VAN BAMBEKE, F. (2009) Activity of ceftobiprole and other cephalosporins against extracellular and intracellular (THP-1 macrophages, keratinocytes) forms of Methicillin-Sensitive (MSSA) and Methicillin-Resistant Staphylococcus aureus (MRSA). Antimicrob Agents Chemother. LI, B., TOM, J. Y., OARE, D., YEN, R., FAIRBROTHER, W. J., WELLS, J. A. & CUNNINGHAM, B. C. (1995) Minimization of a polypeptide hormone. Science, 270, 1657-60. LI, D., LIU, W., LIU, Z., WANG, J., LIU, Q., ZHU, Y. & HE, F. (2008) PRINCESS, a protein interaction confidence evaluation system with multiple data sources. Mol Cell Proteomics, 7, 1043-52. LIM, D. & STRYNADKA, N. C. (2002) Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol, 9, 870-6.  224  LLEWELLYN, R. & EISENBERG, D. S. (2008) Annotating proteins with generalized functional linkages. Proc Natl Acad Sci USA, 105, 17700-5. MAVEYRAUD, L., GOLEMI-KOTRA, D., ISHIWATA, A., MEROUEH, O., MOBASHERY, S. & SAMAMA, J. P. (2002) High-resolution X-ray structure of an acyl-enzyme species for the class D OXA-10 beta-lactamase. J Am Chem Soc, 124, 2461-5. NESS, S., MARTIN, R., KINDLER, A. M., PAETZEL, M., GOLD, M., JENSEN, S. E., JONES, J. B. & STRYNADKA, N. C. (2000) Structure-based design guides the improved efficacy of deacylation transition state analogue inhibitors of TEM-1 beta-Lactamase(,). Biochemistry, 39, 5312-21. NOEL, G. J. (2007) Clinical profile of ceftobiprole, a novel beta-lactam antibiotic. Clin Microbiol Infect, 13 Suppl 2, 25-9. PADAYATTI, P. S., SHERI, A., TOTIR, M. A., HELFAND, M. S., CAREY, M. P., ANDERSON, V. E., CAREY, P. R., BETHEL, C. R., BONOMO, R. A., BUYNAK, J. D. & VAN DEN AKKER, F. (2006) Rational design of a beta-lactamase inhibitor achieved via stabilization of the trans-enamine intermediate: 1.28 A crystal structure of wt SHV-1 complex with a penam sulfone. Journal of the American Chemical Society, 128, 13235-42. PETERSON, L. R. (2008) Currently available antimicrobial agents and their potential for use as monotherapy. Clin Microbiol Infect, 14 Suppl 6, 30-45. PETROSINO, J., RUDGERS, G., GILBERT, H. & PALZKILL, T. (1999) Contributions of aspartate 49 and phenylalanine 142 residues of a tight binding inhibitory protein of beta-lactamases. J Biol Chem, 274, 2394-400. PINHO, M. G. & ERRINGTON, J. (2003) Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol Microbiol, 50, 871-81. POWELL, A. J., TOMBERG, J., DEACON, A. M., NICHOLAS, R. A. & DAVIES, C. (2008) Crystal structures of penicillin-binding protein 2 from penicillin-susceptible and -resistant strains of N. gonorrhoeae reveal an unexpectedly subtle mechanism for antibiotic resistance. J Biol Chem. REICHMANN, D., RAHAT, O., COHEN, M., NEUVIRTH, H. & SCHREIBER, G. (2007) The molecular architecture of protein-protein binding sites. Curr Opin Struct Biol, 17, 67-76. RITCHIE, D. W. (2008) Recent progress and future directions in protein-protein docking. Curr Protein Pept Sci, 9, 1-15. ROBERTS, J. A. & LIPMAN, J. (2009) Pharmacokinetic issues for antibiotics in the critically ill patient. Crit Care Med, 37, 840-51; quiz 859. RUDGERS, G. W., HUANG, W. & PALZKILL, T. (2001) Binding properties of a peptide derived from beta-lactamase inhibitory protein. Antimicrob Agents Chemother, 45, 3279-86. RUDGERS, G. W. & PALZKILL, T. (1999) Identification of residues in beta-lactamase critical for binding beta-lactamase inhibitory protein. J Biol Chem, 274, 6963-71. RUDGERS, G. W. & PALZKILL, T. (2001) Protein minimization by random fragmentation and selection. Protein Eng, 14, 487-92. SCHRAMM, V. L. (2007) Enzymatic transition state theory and transition state analogue design. J Biol Chem, 282, 28297-300. SELZER, T. & SCHREIBER, G. (2001) New insights into the mechanism of protein-protein association. Proteins, 45, 190-8. SHITTU, A. O. & LIN, J. (2006) Newer antistaphylococcal agents: In-vitro studies and emerging  225  trends in Staphylococcus aureus resistance. Wounds. STRYNADKA, N. C., EISENSTEIN, M., KATCHALSKI-KATZIR, E., SHOICHET, B. K., KUNTZ, I. D., ABAGYAN, R., TOTROV, M., JANIN, J., CHERFILS, J., ZIMMERMAN, F., OLSON, A., DUNCAN, B., RAO, M., JACKSON, R., STERNBERG, M. & JAMES, M. N. (1996a) Molecular docking programs successfully predict the binding of a beta-lactamase inhibitory protein to TEM-1 beta-lactamase. Nat Struct Biol, 3, 233-9. STRYNADKA, N. C., MARTIN, R., JENSEN, S. E., GOLD, M. & JONES, J. B. (1996b) Structurebased design of a potent transition state analogue for TEM-1 beta-lactamase. Nat Struct Biol, 3, 68895. SUN, W., HU, Y., GONG, J., ZHU, C. & ZHU, B. (2005) Identification of beta-lactamase inhibitory peptide using yeast two-hybrid system. Biochemistry (Mosc), 70, 753-60. TOMASZ, A., DRUGEON, H. B., DE LENCASTRE, H. M., JABES, D., MCDOUGALL, L. & BILLE, J. (1989) New mechanism for methicillin resistance in Staphylococcus aureus: clinical isolates that lack the PBP 2a gene and contain normal penicillin-binding proteins with modified penicillinbinding capacity. Antimicrob Agents Chemother, 33, 1869-74. WELLS, J. A. & MCCLENDON, C. L. (2007) Reaching for high-hanging fruit in drug discovery at protein-protein interfaces. Nature, 450, 1001-9. YANOFSKY, S. D., BALDWIN, D. N., BUTLER, J. H., HOLDEN, F. R., JACOBS, J. W., BALASUBRAMANIAN, P., CHINN, J. P., CWIRLA, S. E., PETERS-BHATT, E., WHITEHORN, E. A., TATE, E. H., AKESON, A., BOWLIN, T. L., DOWER, W. J. & BARRETT, R. W. (1996) High affinity type I interleukin 1 receptor antagonists discovered by screening recombinant peptide libraries. Proc Natl Acad Sci U S A, 93, 7381-6. ZHANG, Y., LUO, M. & SCHRAMM, V. L. (2009) Transition states of Plasmodium falciparum and human orotate phosphoribosyltransferases. Journal of the American Chemical Society, 131, 4685-94. ZHANG, Z. & PALZKILL, T. (2003) Determinants of binding affinity and specificity for the interaction of TEM-1 and SME-1 beta-lactamase with beta-lactamase inhibitory protein. J Biol Chem, 278, 45706-12. ZHANG, Z. & PALZKILL, T. (2004) Dissecting the protein-protein interface between betalactamase inhibitory protein and class A beta-lactamases. J Biol Chem, 279, 42860-6.  226  APPENDIX I: PEER-REVIEWED PUBLICATIONS AND COPYRIGHT  Chapter 1 contains brief passages of excerpted text (with modifications as requested by the thesis examination committee), one figure, and one table reprinted from Current Opinion in Structural Biology, Vol. 18, Issue 5, authors Andrew L. Lovering, Michael Gretes and Natalie C.J. Strynadka, Structural details of the glycosyltransferase step of peptidoglycan assembly, Pages No. 534-543, Copyright (2008), with permission from Elsevier. Chapter 2 has been reprinted (with modifications as requested by the thesis examination committee) from Journal of Molecular Biology, Vol. 389, Issue 2, authors Michael Gretes, Daniel C. Lim, Liza de Castro, Susan E. Jensen, Sung Gyun Kang, Kye Joon Lee and Natalie C.J. Strynadka, Insights into Positive and Negative Requirements for Protein–Protein Interactions by Crystallographic Analysis of the &-Lactamase Inhibitory Proteins BLIP, BLIP-I, and BLP, Pages No. 289-305, Copyright (2009), with permission from Elsevier. Chapter 5 has been reprinted (with modifications as requested by the thesis examination committee) from Antimicrobial Agents and Chemotherapy, 2008, Volume 52, Issue 6, Pages No. 2089-2096, DOI:10.1128/AAC.01403-07 and reproduced/amended with permission from American Society for Microbiology.  227  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0228826/manifest

Comment

Related Items