Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Crystallographic studies of bacterial proteins involved in β-lactam resistance Lim, Daniel C. 2003

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

Item Metadata

Download

Media
831-ubc_2003-860078.pdf [ 23.28MB ]
Metadata
JSON: 831-1.0091377.json
JSON-LD: 831-1.0091377-ld.json
RDF/XML (Pretty): 831-1.0091377-rdf.xml
RDF/JSON: 831-1.0091377-rdf.json
Turtle: 831-1.0091377-turtle.txt
N-Triples: 831-1.0091377-rdf-ntriples.txt
Original Record: 831-1.0091377-source.json
Full Text
831-1.0091377-fulltext.txt
Citation
831-1.0091377.ris

Full Text

CRYSTALLOGRAPHIC STUDIES OF BACTERIAL PROTEINS INVOLVED IN P-LACTAM RESISTANCE by DANIEL C. LIM B.Sc, McMaster University, 1997 M.Sc, Queen's University, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Biochemistry and Molecular Biology; Faculty of Medicine We accept this thesis as conforming " to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Vancouver, Canada July 2003 © Daniel C. Lim, 2003 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission. Department of Biochemistry and Molecular Bi o l o g y The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date September 30, 2003 A B S T R A C T The P-lactams are an important class of antibiotics that target penicillin binding proteins (PBPs), bacterial enzymes that catalyze the final step of cell wall synthesis. The bacterial cell wall maintains cell shape and provides protection against osmotic lysis. PBPs catalyze the formation of peptide crosslinks in peptidoglycan, which are essential to the structural integrity ofthe cell wall. Due to the uniqueness ofthe peptidoglycan of bacterial cells (with no biochemical equivalent in humans), P-lactams have been highly successful in treating bacterial infections. Unfortunately bacteria have developed multiple resistance mechanisms that have substantially reduced the effectiveness of this class of antibiotics. Three main mechanisms of P-lactam resistance have been found in resistant bacteria: production of P-lactamase enzymes that hydrolyze and inactivate P-lactams, over-expression of multi-drug efflux pumps in Gram-negative bacteria that actively pump out P-lactams and other antibiotics, and the expression of resistant PBPs with unusually low affinities for P-lactams. The overall goal of this thesis is to use the technique of X-ray crystallography for the structural characterization of specific proteins involved in each of these three main mechanisms. PSE-4 is a class A P-lactamase produced by strains ofPseudomonas aeruginosa and is highly active for the penicillin derivative carbenicillin. The crystal structure of the wild type PSE-4 carbenicillinase has been determined to 1.95 A resolution by molecular replacement and represents the first structure of a carbenicillinase published to date. Most carbenicillinases are unique among class A P-lactamases in that residue 234 is an arginine (ABL standard numbering scheme), while in all other class A enzymes this residue is a lysine. Kinetic characterization of a R234K PSE-4 mutant reveals a 50 fold reduction in kcJKm and confirms the importance of Arg 234 for carbenicillinase ii activity. A comparison of the structure of the R234K mutant refined to 1.75 A resolution with the wild type structure shows that Arg 234 stabilizes an alternate conformation of the Ser 130 side chain, not seen in other class A P-lactamase structures. The molecular modelling studies presented here suggest that the position of a bound carbenicillin would be shifted relative to that of a bound benzylpenicillin in order to avoid a steric clash between the carbenicillin a-carboxylate group and the conserved side chain of Asn 170. The alternate conformation ofthe catalytic Ser 130 in wild type PSE-4 may be involved in accommodating this shift in the bound substrate position. MexR is a member of the MarR family of bacterial transcriptional regulators and is the repressor for the MexAB-OprM operon, which encodes a tripartite multidrug efflux system in Pseudomonas aeruginosa. Mutations in MexR result in increased resistance to multiple antibiotics due to over-expression of this efflux system. The crystal structure of MexR has been determined to 2.1 A resolution in the absence of effector. The four copies of the MexR dimer in the asymmetric unit are observed in multiple conformations. Analysis of these conformational states in the context of a model of the MexR-DNA complex proposed in this study suggests that an effector-induced conformational change may inhibit DNA binding by reducing the spacing of the DNA binding domains. The inhibited conformation is exhibited by one of the four MexR dimers, which contains an ordered C-terminal tail from a neighbouring monomer inserted between its DNA binding domains and which is proposed to resemble the MexR-effector complex. The results of this study indicate that MexR may differ from the other described member of this family, MarR, in the nature of its effector, mode of DNA binding and mechanism of regulation. The multiple antibiotic resistance of methicillin-resistant strains of Staphylococcus aureus (MRSA) has become a major clinical problem worldwide. The key determinant of the broad spectrum P-lactam resistance in MRSA strains is the penicillin binding protein 2a (PBP2a). Due to iii its low affinity for P-lactams, PBP2a provides transpeptidase activity to allow cell wall synthesis at P-lactam concentrations which inhibit the P-lactam sensitive PBPs normally produced by S. aureus. The crystal structure of a soluble derivative of PBP2a has been determined to 1.8 A resolution and provides the highest resolution structure for a high molecular mass PBP. Additionally, structures of the acyl-PBP complexes of PBP2a with nitrocefm, penicillin G and methicillin show for the first time P-lactam binding by a resistant PBP. An analysis of the PBP2a active site reveals the structural basis of its resistance and identifies features in P-lactams important for high affinity binding. iv A C K N O W L E D G E M E N T S I would like to thank my parents and brother for all the support I have received from them over the years. I am indebted to my supervisor Natalie Strynadka for her invaluable direction in my research and manuscript writing, for her faith in my abilities and for setting no limits on the resources available for my research. I would also like to thank my supervisory committee members Drs. Brett Finlay and Lawrence Mcintosh, who provided valuable guidance and a great deal of support for my work. I have benefited greatly from all the members ofthe Strynadka lab, both past and present. Special thanks go to Liza De Castro, who provided a tremendous amount of technical assistance and some truly delicious pastry. Tanya Hills has helped me a great deal with my cloning and organized our visit to Santa Claus. I would like to thank Dr. Steve Mosimann for many interesting and informative discussions. I must thank Richard Pfuetzner and Drs. Yu Luo and Gunnar Olovsson for their help. I would also like to thank Barbara Lelj Garolla di Bard, Shouming He, Manjeet Bains and Suzanne Perry for help with a multitude of things for my research. Important aspects ofthe work in my thesis were also made possible by contributions from our collaborators Drs. Roger Levesque, Francois Sanschagrin, Larry Blaszczak, Paul Skatrud and Keith Poole. I am also grateful to Paul Ellis and Mike Soltis at beam line 9-2 at the SSRL and to Joel Berendzen, Leon Flaks and Li-Wei Hung at beam line X8-C at the NSLS for their assistance with data collection at the synchrotrons. I am extremely grateful for the financial support from the Canadian Institutes of Health Research and from the Faculty of Medicine here at UBC for the Doctoral Research Award and for the Harry and Florence Dennison Fellowship in Medical Research. I am deeply honoured for having received the Zbarsky award, for which I would like to thank the Zbarsky family, and I would also like to thank all ofthe fellow graduate students who voted for me. v M A N U S C R I P T S A N D A U T H O R S H I P Chapters 3, 4 and 5 of this thesis describe work that were previously published. All manuscripts were written by myself and revised by my supervisor Dr. Natalie Strynadka. Chapter 3 contains portions from a manuscript published in Biochemistry [Lim, D., Sanschagrin, F., Passmore, L., De Castro, L., Levesque, R.C., Strynadka, N.C. Insights into the molecular basis for the carbenicillinase activity of PSE-4 beta-lactamase from crystallographic and kinetic studies. Biochemistry. 2001 40(2):395-402]. Both the wild type and mutant PSE-4 constructs were produced by Francois Sanschagrin in the laboratory of Dr. Roger Levesque. Crystallization conditions were initially determined by Liza De Castro and refined by myself. The structure was originally determined by molecular replacement by Lori Passmore. I determined suitable cryoprotection conditions for the crystals. Collection and processing of the high resolution data, structure refinement and analysis of both the wild type and mutant proteins were completed by myself. The wild type and mutant proteins used in these high resolution studies were purified by myself. The kinetic studies were carried out by Francois Sanschagrin. All the text and figures in the manuscript were prepared by myself with the exception ofthe methodology for the kinetic studies. Chapter 4 contains portions from a manuscript published in the Journal of Biological Chemistry [Lim, D., Poole, K. & Strynadka, N.C. Crystal structure of the MexR repressor ofthe mexRAB-oprM multidrug efflux operon of Pseudomonas aeruginosa. J . B i o l . Chem. 2002 277(32):29253-9.] The original tagged MexR construct was provided by Dr. Keith Poole, from which I created the untagged construct used for these studies. I determined the protocol for and carried out the purification of the protein. I established the crystallization and cryoprotection conditions. The data collection and processing, structure determination, refinement and analysis, the vi molecular modelling studies and electrophoretic mobility shift assays for DNA binding were completed by myself. All the text and figures in the manuscript were prepared by myself. Chapter 5 contains portions from a manuscript published in Nature Structural Biology [Lim, D. & Strynadka, N.C. Structural basis for the beta-lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nature Struct. Biol. 2002 (11):870-6.]. The original SauPBP2a* construct was provided by L. C. Blaszczak, and P. L. Skatrud (Infectious Disease Research, Eli Lilly and Company). A derivative plasmid used for the expression of this construct was created by Liza De Castro. I refined a refolding and purification protocol and determined crystallization conditions for this construct. I subsequently produced a more truncated derivative of this construct and established expression and purification protocols and crystallization conditions for this new construct. I prepared all the protein and crystals used in these studies. Data collection and processing, and structure determination, refinement and analysis were completed by myself. I prepared all the text and figures in the manuscript. vn L I S T O F A B B R E V I A T I O N S 6- APA 6-aminopenicillanic acid 7- ACA 7-aminocephalosporanic acid A 2 g 0 absorbance at 280 nm ABL class A P-lactmase BNL Brookhaven National Laboratory-CCD charge coupled device DTT dithiothreitol EDTA ethylene-diamine tetraacetic acid ESMS electrospray mass spectrometry HEPES 4-(2-hydroxyethyl)piperazine-l-ethanesulfonic acid HMM high molecular mass kDa kilodalton LB Luria Bertani medium LMM low molecular mass MAD multiwavelength anomalous diffraction MES 2-morpholinoethanesulfonic acid MIC minimum inhibitory concentration MOPS 2-morpholinopropanesulfonic acid MPD 2-methyl-2,4-pentanediol NCS non-crystallographic symmetry NMR nuclear magnetic resonance viii NSLS National Synchrotron Light Source OD600 optical density at 600 nm wavelength ORF open reading frame PAGE polyacrylamide gel electrophoresis PBP penicillin-binding protein PSE-4 Pseudomonas specific enzyme-4 PVDF polyvinylidene fluoride PEG polyethylene glycol rmsd root mean square deviation SauPBP2a penicillin-binding protein 2a from methicillin-resistant Staphylococcus aureus SDS sodium dodecylsulfate SeMet selenomethionine SSRL Stanford Synchrotron Research Laboratory TCEP tris-(carboxyethyl)phosphine hydrochloride UDP-GlcNAc uridine 5'-pyrophosphate-A^ -acetylglucosamine UDP-MurNAc uridine 5'-pyrophosphate-A^ -acetylmuramic acid UV ultraviolet v/v unit volume (mL) per unit volume (mL) w/v unit weight (g) per unit volume (mL) ix TABLE OF CONTENTS ABSTRACT ii ACKNOWLEDGEMENTS v MANUSCRIPTS AND AUTHORSHIP vi LIST OF ABBREVIATIONS viii TABLE OF CONTENTS x LIST OF TABLES xiii LIST OF FIGURES xiv Chapter 1: INTRODUCTION 1 1.1 Discovery of penicillin and the development of P-lactam antibiotics 1 1.2 Inhibition of bacterial cell wall crosslinking by P-lactams 3 1.3 Penicillin-binding proteins 8 1.4 P-lactam resistance mechanisms 16 1.4.1 P-lactamases IV 1.4.2 Multidrug efflux pumps 23 1.4.3 Resistant PBPs 26 1.5 Objectives of thesis 32 Chapter 2: MATERIALS AND GENERAL PROTOCOLS 34 2.1 Materials 34 2.2 DNA and protein sequence analysis 34 2.3 Over-expression of selenomethionine-substituted protein 35 2.4 Protein manipulation and storage 35 2.5 Protein mass analysis 36 2.6 Crystallization 36 2.7 Preparation of crystals for X-ray diffraction studies 37 2.8 Data collection and processing 37 2.9 Model building and refinement 39 2.10 Structure analysis 41 Chapter 3: PSE-4 42 3.1 Introduction 42 3.2 Materials 44 3.3 Methods 44 3.3.1 Protein over-expression and purification 44 3.3.2 Crystallization 46 3.4 Results 47 x 3.4.1 Data collection and processing 47 3.4.2 Structure determination and refinement 48 3.5 Discussion 51 3.5.1 Overall structure 51 3.5.2 Active site 52 3.5.3 Model of the acyl-enzyme intermediate 56 Chapter 4: MexR 59 4.1 Introduction 59 4.2 Materials 61 4.3 Methods 61 4.3.1 Creation of wild type untagged MexR construct 61 4.3.2 Protein over-expression and purification 62 4.3.3 Crystallization 65 4.4 Results 66 4.4.1 MexR DNA binding assays 66 4.4.2 Data collection 68 4.4.3 Structure determination and refinement 69 4.5 Discussion 71 4.5.1 Overall structure 71 4.5.2 Conformational flexibility ofthe MexR dimer 73 4.5.3 Molecular model of the MexR-DNA complex 75 4.5.4 Proposed allosteric mechanism of regulation of MexR 77 4.5.5 Contrast with MarR . 80 Chapter 5: PBP2a from MRSA 83 5.1 Introduction 83 5.2 Materials 86 5.3 Methods 86 5.3.1 Creation of a further truncated SauPBP2a construct and acylation-deficient mutant 86 5.3.2 Measurement of protein concentration 87 5.3.3 Over-expression, refolding and purification of SauPBP2aNA22+2 and SeMet-substituted SauPBP2a* 87 5.3.4 Over-expression and purification of soluble SauPBP2a* 90 5.3.5 Crystallization 91 5.3.6 Cryoprotection and P-lactam soaking 93 5.4 Results 94 5.4.1 Data collection 94 5.4.2 Structure determination and refinement 95 5.5 Discussion 97 5.5.1 Overall structure 97 5.5.2 Acylation with P-lactam 100 5.5.3 Structural basis for P-lactam resistance in SauPBP2a 103 5.5.4 Comparison with other resistant PBPs 105 5.5.5 Comparison of the P-lactam acyl-SauPBP2a* complexes 107 xi 5.5.6 Implications for transpeptidase activity 113 5.5.7 Effect of crystal packing on P-lactam binding 115 5.5.8 Insights for drug design 117 Chapter 6: CONCLUSIONS AND FUTURE DIRECTIONS 119 6.1 Summary and significance of results 119 6.2 Future studies 121 REFERENCES 126 Appendix - Publications arising from graduate work 157 xii LIST O F T A B L E S Table 3.4.1: PSE-4 X-ray data statistics 48 Table 3.4.2: PSE-4 structure refinement statistics 50 Table 3.5.1: Effect of R234K mutation on p-lactam hydrolysis by PSE-4 54 Table 4.4.1: MexR X-ray data statistics 69 Table 4.4.2: SeMet-substituted MexR structure refinement statistics 70 Table 5.4.1: SauPBP2a X-ray data statistics 94 Table 5.4.2: SauPBP2a structure refinement statistics 96 Table 5.5.1: Group occupancies and average 5-factors of P-lactam and Ser 403 atoms in SauPBP2a* 116 xiii LIST O F FIGURES Fig. 1.1.1: Structure of penicillin G 2 Fig. 1.2.1: Chemical structure of peptidoglycan 5 Fig. 1.2.2: Lipid II biosynthesis 6 Fig. 1.3.1 Comparison of D-Ala-D-Ala and penicillin G 10 Fig. 1.3.2: Crystal structures of PBPs 13 Fig. 1.3.3: Proposed acylation mechanism for R61 PBP 15 Fig. 1.4.1: Crystal structures of [3-lactamases 19 Fig. 1.4.2: Phylogenetic tree of penicillin-recognizing proteins 20 Fig. 1.4.3: Proposed catalytic mechanisms of class A P-lactamases 23 Fig. 1.4.4: Crystal structures of TolC and AcrB 25 Fig. 1.4.5: Overall reaction scheme for interaction of a PBP with a P-lactam 29 Fig. 1.4.6: The more open active site of Sp328 PBP2x* relative to that of R6 PBP2x* 31 Fig. 3.1.1: Structures of penicillin G and carbenicillin 42 Fig. 3.3.1: SDS-PAGE of PSE-4 fractions 46 Fig. 3.4.1: Representative electron density in PSE-4 active site 48 Fig. 3.5.1: Overall structure of PSE-4 51 Fig. 3.5.2: Comparison of the PSE-4 and TEM-1 structures 52 Fig. 3.5.3: Comparison of the active site region of the R234K PSE-4 mutant with the wild type PSE-4 and TEM-1 structures 55 Fig. 3.5.4: Model of the acyl-enzyme intermediate of PSE-4 with carbenicillin 56 Fig. 4.1.1: mexRAB-oprM operon 60 xiv Fig. 4.3.1: SDS-PAGE of SeMet-substituted MexR fractions 64 Fig. 4.3.2: Crystals of SeMet-substituted MexR 65 Fig. 4.4.1: Electrophoretic mobility shift assays of DNA binding by SeMet-substituted MexR 67 Fig. 4.5.1: Overall structure of the MexR dimer 72 Fig. 4.5.2: Overlap ofthe Ca traces ofthe eight monomers in the MexR asymmetric unit .... 73 Fig. 4.5.3: Overlap ofthe MexR dimers AB (blue) and CD (red) 73 Fig. 4.5.4: Proposed mechanism of regulation of MexR 76 Fig. 4.5.5: Binding of MexR to C-terminal tail from neighbouring molecule 79 Fig. 4.5.6: Comparison ofthe MexR and MarR dimers 80 Fig. 4.5.7: Comparison of the MarR salicylate binding sites with the corresponding regions in MexR 81 Fig. 5.3.1: SDS-PAGE of SauPBP2a* fractions 89 Fig. 5.3.2: Crystal of SauPBP2a* (NA22+2 construct) 91 Fig. 5.3.3: Crystals of native SauPBP2a* 93 Fig. 5.5.1: Structure of SauPBP2a* 98 Fig. 5.5.2: Interaction of SauPBP2a* with nitrocefin 102 Fig. 5.5.3: Overall reaction scheme for interaction of a PBP with a P-lactam 103 Fig. 5.5.4: Superposition ofthe active site region from the apo (yellow) and nitrocefin acyl-PBP (blue) structures 103 Fig. 5.5.5: Superposition ofthe active site region from the apo- (yellow) and acyl-PBP (purple) structures of R6 PBP2x 105 Fig. 5.5.6: Penicillin G acyl-PBP complex for SauPBP2a* 109 xv Fig. 5.5.7: Structure of the cephalosporin "compound 1" 109 Fig. 5.5.8: Widening of SauPBP2a* active site upon nitrocefin binding 110 Fig. 5.5.9: Methicillin acyl-PBP complex for SauPBP2a* 112 Fig. 5.5.10: Structures of novel cephalosporin compounds with increased activity against MRSA 118 xvi Chapter 1: INTRODUCTION 1.1 Discovery of penicillin and the development of p-lactam antibiotics In the summer of 1928, Sir Alexander Fleming happened upon his famous finding, that a contaminating mould (later identified as Penicillium notatum) resulted in the lysis of Staphylococcus aureus growing on a bacterial Petri plate (Fleming, 1928). Further work in his laboratory showed that the lysis was caused by a substance secreted by the mould into the culture medium, and which could be extracted from the culture broth filtrate using organic solvents. This substance was named penicillin and was highly potent against Gram-positive bacteria such as staphylococci, Streptococcus pyrogenes, and pneumococci. Furthermore, testing on animal tissues via direct application on wounds and human eyes and injection into mice and rabbits, showed that penicillin was nonirritating and nontoxic. However, experiments were not extended to injection into infected animals, and Fleming was discouraged by problems with the chemical purification of penicillin and with finding an effective way of concentrating it. The instability of the compound and difficulties in producing sufficient quantities were also problematic for other laboratories that subsequently attempted to purify penicillin. In 1939, the research group headed by H. W. Florey (the Oxford group) began work on developing penicillin as a chemotherapeutic agent. The Oxford group overcame many of the problems that had frustrated previous researchers and developed improved methods for isolating penicillin. They demonstrated successful use of penicillin to cure lethal streptococcal and staphylococcal infections in mice (Chain et al, 1940) and shortly after, carried out the first clinical trials in humans (Abrahams et al, 1941). The successful demonstration of the systemic chemotherapeutic activity of penicillin provided the stimulus for the development of techniques for 1 large scale production and purification of penicillin. Efforts from fermentation experts of the Northern Regional Research Laboratory (NRRL) of the U.S. Department of Agriculture and from U.S. pharmaceutical companies led to improvements in fermentation techniques and to identification of strains producing higher yields, making possible the mass production of penicillin (reviewed in Queener, 1986). Structural and chemical studies on penicillin provided a basis for the development of a wide range of derivative compounds. In 1943 the availability of crystalline penicillin allowed for the elemental analysis of penicillin, which showed that Fig. 1.1.1: Structure of penicillin G. The the penicillin produced in the U.S. (penicillin G) 6-aminopenicillanic acid (6-APA) portion is highlighted with thick lines, was structurally different from the penicillin Conventional numbering is shown for the bicyclic atoms. produced in England (penicillin F). The crystal structure of penicillin G determined in 1945 (Crowfoot et al, 1949) provided a framework for the available structural information and confirmed the unprecedented four-membered P-lactam ring structure (Fig. 1.1.1), a central feature of all p-lactam compounds. By 1949, six natural penicillins were known, which differed only in the nature of the side chain: 2-pentenyl (penicillin F), 3-pentenyl, j9-hydroxybenzyl (penicillin V), benzyl (penicillin G), w-amyl, and «-heptyl groups. Of these, penicillin G was the most potent and the only one developed for medicinal use. Precursor compounds for the side chain added to the culture medium were found to be incorporated into the penicillin produced. However, the diversity of compounds produced by this process of precursing was limited by the toxicity of precursor compounds. This problem was overcome with the discovery of 6-aminopenicillanic acid (6-APA), the penicillin nucleus (Fig. 1.1.1). 6-APA was initially 2 observed in Japan in the early 1950s, but was not exploited until 1956 when J. C. Sheehan synthesized 6-APA and converted it into the naturally occurring penicillin V by acylation of the nucleus with the side chain (Sheehan and Henery-Logan, 1959). This work in combination with the ability to produce large amounts of 6-APA by fermentation (Batchelor et al, 1959) made possible the synthesis of a large of number of penicillin derivative compounds: the semisynthetic penicillins (reviewed in Rolinson, 1998). The subsequent discovery of cephalosporin C produced by Cephalosporium acremonium, its derivatization via the 7-aminocephalosporanic acid (7-ACA) nucleus, and the discovery of carbapenems from Streptomyces, cephamycins from actinomycetes, and monobactams from Pseuodomonas, Gluconobacter, and Chromobacterium further expanded the range of P-lactam antibiotic compounds (reviewed in Queener, 1986). 1.2 Inhibition of bacterial cell wall crosslinking by P-lactams The bacterial cell wall is a mesh-like structure encompassing the cytoplasmic membrane. It is essential to the maintenance of cell shape and protects the cell from lysis due to osmotic stress. Studies on the effect of penicillin on bacterial cells indicated that P-lactam antibiotics target cell wall synthesis. Exposure to sub-lethal doses of penicillin results in abnormal morphology and imperfect fission (Gardner, 1940). Treatment with lethal doses of penicillin leads to bacterial lysis, which can be prevented in an isotonic medium. Penicillin-induced lysis is observed only in growing cells and requires the action of bacterial lytic enzymes that degrade the existing cell wall and are involved in normal cell wall remodelling during growth and division (reviewed inHoltje, 1998). Ultrastructural studies of S. aureus cells by electron microscopy (reviewed in Giesbrecht et al. ,1998) have provided further details on the lethal effects of penicillin on the cell wall. The degradation of the existing cell wall by lytic enzymes acting at points along the cell division plane, in conjunction with an 3 impairment of cell wall synthesis along the septal plane beneath these regions of lysis, result in the formation of pores in the cell wall. Ejection of cytoplasmic contents through these pores (due to the internal osmotic pressure) results in cell death. Further degradation of the cell wall (presumably by the continuing action of lytic enzymes) leads to cell disintegration and accounts for the observed lysis. The main structural component of bacterial cell walls is peptidoglycan, which consists of a carbohydrate polymer component (glycan strands of alternating residues of TV-acetylglucosamine and jV-acetylmuramic acid) and a peptide component (reviewed in Ghuysen, 1968). The peptide substituents are attached to the carboxylic group on the A^ -acetylmuramic acid residues and consist of L-Ala-y-D-Glu-L-R3-D-Ala (Fig. 1.2.1). In Gram-negative and some Gram-positive bacteria, L - R 3 is LL-diaminopimelic acid, hi Gram-positive bacteria such as S. aureus, L - R 3 is L-Lys, and the D-Glu residue is a-amidated. Crosslinking of the peptides occurs via the free amino group on L - R 3 of the acceptor strand to the penultimate D-Ala residue of the donor strand and may involve additional interbridge residues (pentaglycine in S. aureus). This results in a massive macromolecule (murein) that forms a mesh-like structure (sacculus) encompassing the entire cell on the extracellular face of the cytoplasmic membrane. 4 N-acetylglucosamine Fig. 1.2.1: Chemical structure of peptidoglycan. a) Portion of Gram-negative (E. coli) peptidoglycan. b) Portion of Gram-positive (S. aureus) peptidoglycan showing crosslink to peptide from neighbouring glycan strand. 5 Currently no direct method is available to study the architecture of murein in intact cells. Models of the three dimensional structure of murein have therefore been deduced from indirect evidence. Three models for the molecular configuration of murein have been discussed in the literature. The dominant model proposed that glycan strands are in an extended linear conformation oriented parallel to the cytoplasmic membrane and in rod-shaped bacteria, run perpendicular to the long axis of the cell (reviewed in Holtje, 1998). A related model also proposed that glycan strands are parallel to the cytoplasmic membrane but are otherwise randomly oriented (reviewed in Koch, 1998). In both models, the crosslinking of the glycan strands results in a two-dimensional layer. The relatively thin cell walls of Gram-negative bacteria such as Escherichia coli essentially consist of a monolayer, while the thicker Gram-positive cell walls are multilayered. An alternative scaffold model of murein structure proposed that the glycan strands are in fact oriented perpendicular to the membrane (Dmitriev et al, 1999; Dmitriev et al, 2003). In this model, the glycan strands adopt a helical conformation with four disaccharide units in each turn of the helix and with four peptide side chains extending outward in four directions. The peptide crosslinks are then formed perpendicular to the glycan strands and parallel to the membrane. Peptidoglycan biosynthesis begins with the assembly of the disaccharide-peptide repeating monomer in the cytoplasm (Fig. 1.2.2; reviewed in Holtje, 1998; van Fructose 6-phosphate GlmS. GlmM, GlmU t MurA MurB MurC MurD MurE UDP-GlcNAc J m phospho-enolpyruvate UDP-GlcNAc-enolpyruvate UDP-MurNAc J L-Ala UDP-MurNAc-L-Ala | • D-Glu UDP-MurNAc-L-Ala-D-Glu | • DL-diaminopimelic acid or L-Lys UDP-MurNAc-tripeptide MurF MraY MurG | • D-Ala-D-Ala UDP-MurNAc-pentapeptide J • undecaprenyl-phosphate lipid I | m UDP-GlcNac lipid II D-Ala + D-Ala Fig. 1.2.2: Lipid II biosynthesis. The enzymes involved are listed on the left and boxed. Heijenoort, 1998; vanHeijenoort, 2001). First, uridine 5'-pyrophosphate-A/-acetylglucosamine (UDP-GlcNAc) is formed in four successive steps from fructose 6-phosphate. The UDP-GlcNAc is then converted to uridine 5'-pyrophosphate-A^ -acetylmuramic acid (UDP-MurNAc), to which amino acid residues are successively added to yield aUDP-MurNAc-peptide intermediate. The MurNAc-peptide moiety of the UDP-MurNAc-peptide is then transferred to the membrane-bound acceptor undecaprenyl-phosphate to yield MurNAc(peptide)-pyrophosphoryl undecaprenol (lipid I). Following the addition of TV-acetylglucosamine, GlcNAc-MurNAc(peptide)-pyrophosphoryl-undecaprenol (lipid II) is produced. If present, amino acid residues for the interbridge peptide are then added. The disaccharide-peptide unit is translocated across the cytoplasmic membrane, where glycosyltransferases catalyze the elongation or formation of glycan strands by polymerization of the disaccharide units (reviewed in van Heij enoort, 2001). Details of the glycosyltransferase reaction are not well understood due to the lack of structural information on any of the bacterial cell wall synthetic glycosyltransferases. Formation of the peptide crosslinks between glycan strands during murein synthesis occurs via a transpeptidation reaction, in which the D-Ala-D-Ala peptide bond at the C-terminus of a donor peptide is broken (freeing the terminal D-Ala), and a new peptide bond is formed with an amino group on the acceptor peptide from a neighbouring glycan strand (Fig. 1.2.1b). The loss of a C-terminal D-Ala is therefore coupled to the formation of a peptide crosslink (Tipper and Strominger, 1965). In the absence of intracellular energy sources such as ATP hydrolysis, this new peptide bond at the cell exterior is formed at the energetic expense of the D-Ala-D-Ala peptide bond. It is this transpeptidation step that is specifically inhibited by penicillin and other B-lactam antibiotics. In the presence of penicillin, polymerization of the glycan strands can still proceed and results in the formation of uncrosslinked and poorly crosslinked soluble peptidoglycan, indicating that the 7 glycosyltransfer reaction and all preceding steps are not inhibited by penicillin (Terrak et al, 1999and reviewed in Ward, 1984). In all models of murein structure (two-dimensional layers or scaffold of crosslinked glycan strands perpendicular to the membrane), the peptide crosslinks are essential to the structural integrity of the sacculus. Without this crosslinking, the cell wall (and the entire bacterial cell in the case of lysis) is unable to remain intact. 1.3 Penicillin-binding proteins The transpeptidation reaction is catalyzed by bacterial enzymes known as penicillin binding proteins (PBPs). PBPs were so named because of their ability to form covalent complexes with penicillin, and were initially characterized using radioactively labelled P-lactams such as 14C-penicillin G (Blumberg and Strominger, 1972; Spratt and Pardee, 1975). Multiple PBPs have been detected in both Gram-positive and Gram-negative bacterial species using such labelling techniques and analysis of cell lysates by SDS-PAGE and autoradiography (Georgopapadakou and Liu, 1980). Both DD-transpeptidase and DD-carboxypeptidase activities have been detected in PBPs, the hydrolysis of an acyl-D-Ala-D-Ala bond (carboxypeptidase-catalyzed reaction) being analogous to a transpeptidation reaction with water as the acceptor. Additionally PBPs with DD-endopeptidase activity have also been characterized. Based on molecular models, it was proposed that penicillin is a structural analog of the acyl-D-alanyl-D-alanine portion of peptidoglycan stem peptides and therefore a competitive inhibitor of PBPs (Fig. 1.3.1a,b; Crowfoot era/., 1949; Tipper and Strominger, 1965). Penicillin can be thought of as an acylated cyclic dipeptide of L-cysteine and D-valine. The highly reactive amide bond of the p-lactam ring is equivalent to the D-alanyl-D-alanine peptide bond, while the carboxylate substituent on C3 of penicillin is equivalent to the peptide C-terminal carboxylate. It was further proposed that 8 the transpeptidation reaction catalyzed by PBPs involves cleavage of the D-Ala-D- Ala peptide bond and concomitant formation of an acyl-D-alanyl-PBP covalent intermediate with release of the terminal D-Ala (Fig. 1.3.1c). Transpeptidation is then completed by transfer ofthe acyl-D-alanyl group to the acceptor peptide. Interaction with penicillin involves a similar acylation reaction in which the p-lactam ring is cleaved with concomitant formation of a covalent penicilloyl-enzyme intermediate (Fig. 1.3.1 d). However, due to the cyclic nature of the p-lactam ring, the leaving group remains tethered and sterically blocks nucleophilic attack on the penicilloyl-serine ester by an acceptor peptide or water (Ghuysen et al, 1986). Furthermore, deacylation via recyclization of the highly strained four-membered p-lactam ring is unfavourable. Due to the stability of the penicilloyl-PBP intermediate, the enzyme is effectively trapped in an inactive state. Peptide mapping studies for two Bacillus DD-carboxypeptidases confirmed the formation of a covalent ester linkage to the same active site serine side chain hydroxyl for both P-lactam and peptide substrate (Yocum et al, 1979). 9 Fig. 1.3.1 :Comparison of D-Ala-D-Ala and penicillin G. Molecular models of a) D-Ala-D-Ala and b) penicillin G. c) Transpeptidation with formation of a transient acyl-PBP intermediate, d) Interaction of PBP with penicillin G with formation of a stable acyl-PBP intermediate. PBPs are predominantly extracellular / periplasmic proteins anchored or associated with the cytoplasmic membrane (reviewed in Ghuysen and Dive, 1994). Based on amino acid sequence, PBPs have been divided into two broad groups, the low molecular mass (LMM) PBPs (-20 to 50 kDa), and the multi-modular high molecular mass (HMM) PBPs (-60 to 120 kDa), which are each further divided into three classes (reviewed in Goffm and Ghuysen, 1998; Massova and Mobashery, 1998). Among the LMM PBPs, the class A enzymes are DD-transpeptidases, the class B enzymes exhibit both DD-carboxypeptidase and DD-transpeptidase activities, and the class C enzymes are DD-endopeptidases. LMM PBPs may be involved in modulating the extent of peptidoglycan crosslinking, cell wall breakdown and cell shape. The class A and B HMM PBPs are membrane-10 bound proteins anchored via a transmembrane segment near the N-terminus. The class A HMM PBPs are bifimctional enzymes, containing a glycosyltransferase domain following the transmembrane anchor and a DD-transpeptidase domain in the C-terminal region (reviewed in Di Guilmi et al., 2002). The glycosyltransferase domain catalyzes the polymerization of peptidoglycan strands from lipid II precursors and shows sequence homology to monofunctional glycosyltransferases also involved in glycan polymerization. Class A HMM PBPs are believed to associate with other proteins in multi-enzyme complexes that coordinate breakdown and synthesis of murein during growth of the sacculus (reviewed in Holtje, 1998; Vollmer et al., 1999). The class B enzymes also contain a transpeptidase domain in the C-terminal region but do not exhibit glycosyltransferase activity (Adam et al, 1997 and reviewed in Goffin and Ghuysen, 1998). The non-penicillin-binding (nPB) domain following the transmembrane anchor of class B HMM PBPs shares no sequence identity with the glycosyltransferase domains of class A HMM PBPs. The nPB domain is believed, at least in some members, to mediate protein-protein interactions and targeting to the divisome, a proposed complex of cytoplasmic, membrane and periplasmic proteins that are involved in septum formation during cell division (Marrec-Fairley et al., 2000 and reviewed in Nanninga, 1991). The class C HMM PBPs are P-lactam sensor / signal transduction proteins (BlaR) in Gram-positive bacteria and contain a C-terminal extracellular penicillin-binding sensor domain, an intracellular zinc metalloprotease domain and four predicted transmembrane helices (Hardt et al., 1997). P-lactam-binding to the sensor domain is believed to cause a conformational change, which results in the activation of the metalloprotease domain. Proteolysis of downstream target(s) then results in the inhibition of DNA binding by the Blal repressor protein to derepress P-lactamase expression (Zhang et al, 2001; Golemi-Kotra et al, 2003). 11 Crystal structures of a number of PBPs have provided valuable insights into their mechanisms of catalysis and inhibition by P-lactams. The crystal structure of the soluble LMM class A R61 PBP from Streptomyces R61 (Fig. 1.3.2a) provided the first view of the active site serine PBP fold and revealed a two-domain structure consisting of a predominantly a-helical domain and a mixed a/p domain (Kelly et al, 1982; Kelly et al, 1985; Kelly and Kuzin, 1995). From structures of p-lactam acyl complexes, the active site cavity was mapped at the interdomain interface and is defined by residues from the three motifs conserved in all PBPs. The nucleophilic Ser 62 from motif I (S-x-x-K) is located at the helix a2 N-terminus. Tyr 159 of motif II (S/Y-x-N/C) is located on a loop connecting helices a4 and a5, which lines one side of the active site cavity, while strand P3 defines the opposite side of the cavity and contains the conserved basic residue His 298 of motif m (H/K-T/S-G). Crystal structures have also been obtained for the LMM class A K15 PBP from Streptomyces K15 (Fonze et al, 1999), the LMM class B PBP5 from E. coli (EcoPBP5; Davies et al, 2001) and the HMM class B PBP2x from Streptococcus pneumoniae R6 (R6 PBP2x; Pares et al, 1996; Gordon et al, 2000), which all contain a similar overall fold and active site structure in their transpeptidase domains as the R61 PBP (Fig. 1.3.2b-d). 12 Fig. 1.3.2: Crystal structures of PBPs. a) R61 PBP from Streptomyces R61, b) K15 PBP from Streptomyces K15, c) EcoPBP5 from Escherichia coli, and d) transpeptidase domain of R6 PBP2x from Streptococcus pneumoniae R6. The conserved motifs I, II and III are coloured red, magenta and green, respectively. Helix oc2 and strand B3 are highlighted in dark blue and yellow, respectively. Proposed acylation mechanisms for the active site serine PBPs have emerged from the available structural data and have drawn parallels from the catalytic mechanism of serine proteases, 13 which also employ a nucleophilic active site serine residue that forms a covalent intermediate with the peptide substrate during the hydrolysis reaction (reviewed in Henderson, 1970). Using R61 PBP as an example (1.3.3), the nucleophilic Ser 62 (motif I) attacks the P-lactam / peptide carbonyl carbon resulting in an anionic tetrahedral intermediate, which is stabilized by an oxyanion hole formed by the backbone nitrogens of Ser 62 and Thr 301 (immediately following motif Ul). This oxyanion hole is analogous to the equivalent feature in serine proteases. The charge relay network in serine proteases consists of a catalytic triad of the critical serine, histidine and aspartate residues, in which the nucleophilic serine is activated due to a lowered p^ Ta. Based on their distances from the Ser 62 Oy, either Lys 65 (motif I) or Tyr 159 (motif II) in R61 PBP have been proposed as a possible general base in the acylation step, by abstracting a proton from the Ser 62 Oy. The hydrogen-bonding distances observed in P-lactam acyl complexes of R61 PBP seem to favour Tyr 159 (Kelly et al, 1985), while other structures such as that of the K l 5 PBP are more consistent with the motif I lysine as the general base (Rhazi et al., 2003). A significant reduction in the pKa of either side chain would be required, and may be facilitated by the overall positive charge of the active site. In several serine proteases that lack an essential histidine residue, an essential lysine residue has also been proposed as the general base in activating the nucleophilic serine (reviewed in Paetzel and Dalbey, 1997). During collapse of the tetrahedral intermediate, the proton abstracted from the Ser 62 is back-donated to the leaving group nitrogen atom via Tyr 159 (motif II), which acts as a proton donor, and involves Lys 65 either as a proton shuttle or providing electrostatic activation. Such a role for Tyr 159 is consistent with the crystal structure of a Henri-Michaelis (enzyme-substrate) complex with a glycyl-L-a-amino-E-pimelyl-D-Ala-D-Ala substrate obtained using R61 PBP inactivated by crosslinking of Lys 65 to Tyr 159 and His 108 (McDonough et al., 2002). For transpeptidation or carboxypeptide hydrolysis, deacylation is likely to be the reverse of acylation, with activation of the 14 acceptor peptide amino group or a water via proton abstraction by an anionic Tyr 159 (activated by Lys 65). The crystal structure of a tripeptide phosphonate inhibitor bound to R61 P B P as a deacylation transition state analog complex has been determined and supports such a role for Tyr 159 in deacylation (Silvaggi et al, 2003). The abstracted proton would then be passed to Ser 62 either directly or via Lys 65. Nucleophilic attack on the acyl-serine ester results in a second tetrahedral intermediate, the collapse of which results in acyl transfer to the acceptor peptide or water. Fig. 1.3.3: Proposed acylation mechanism for R61 PBP. The nucleophilic Ser 62 (motif I) is activated by Tyr 159 (motif II) and attacks the fi-lactam carbonyl carbon. A tetrahedral intermediate is formed and is stabilized by the backbone amides of Ser 62 and Thr 301 (oxyanion hole). During the collapse ofthe tetrahedral intermediate, Tyr 159 back-donates the proton abstracted from Ser 62 to the leaving group nitrogen. 15 The mechanism of hydrolysis of P-lactam acyl-PBP intermediates is less clear, and the lack of an obvious efficient deacylation machinery is consistent with the relatively poor deacylation rates of P-lactam acyl-PBP intermediates. In crystal structures of P-lactam acyl-complexes of R61 PBP, Tyr 159 is suitably positioned to activate a potential deacylating water (Kelly and Kuzin, 1995; McDonough et al, 2002). In the K15 PBP, both Lys 38 (motif I) and Ser 96 (motif H; analogous to Tyr 159 inR61 PBP) have been proposed as possible candidates for the catalytic base in deacylation (Rhazi et al, 2003). In contrast, the backbone carbonyls of His 151 in EcoPBP5 (Zorzi et al., 1996) and the corresponding residue Phe 450 in R6 PBP2x (Gordon et al, 2000) have been proposed to coordinate or assist the activation of a deacylating water. 1.4 P-lactam resistance mechanisms Despite the enormous success of antibiotics in controlling bacterial infections, incidence of resistance has been increasing at an alarming rate (reviewed in Neu, 1992; Levy, 2001). The relatively low cost of production, low toxicity, and the uniqueness (peptidoglycan being found only in eubacteria) and accessibility (extracellular location of the cell wall) of their targets have contributed to the success of the P-lactams as chemotherapeutic agents. However, the widespread use of p-lactam antibiotics has resulted in the selection and dissemination of resistant bacterial strains. The main mechanisms of p-lactam resistance involve the synthesis of hydrolytic enzymes that inactivate P-lactams, active efflux of the antibiotic from the periplasm of Gram-negative bacteria, and PBPs with reduced affinity for p-lactams (reviewed in Walsh, 2000). 16 1.4.1 P-lactamases As early as the 1940s, highly potent bacterial enzymes capable of inactivating penicillin were isolated from strains of both Gram-positive and Gram-negative bacteria resistant to penicillin (Abraham and Chain, 1940; Kirby, 1944). These enzymes catalyze the hydrolysis of the P-lactam ring of penicillin and other P-lactams, thereby making them unreactive towards PBPs. The production of P-lactamases has become the most common mechanism of p-lactam resistance (reviewed in Frere, 1995; Medeiros, 1997; Matagne et al, 1998; Philippon et al, 1998; Majiduddin et al., 2002). Based on primary sequence, p-lactamases can be grouped into four classes: A, B, C and D. The class A, C, and D enzymes are serine hydrolases, while the class B enzymes are metallo-enzymes that are generally regarded as distinct from the serine P-lactamases. The class C enzymes (-39 kDa) are primarily chromosomally encoded and are active against cephalosporins. Members of class D are penicillinases (-29 kDa) that are uniquely able to hydrolyze oxacillins. Clinically the most prevalent P-lactamases are the class A enzymes, which are mainly penicillinases, but point mutations in the TEM and SHV enzymes have increased the substrate spectrum of class A P-lactamases to include cephalosporins. The class A enzymes are -29 kDa proteins that are found in both Gram-negative and Gram-positive bacteria, can be either plasmid or chromosomally encoded and have also been found on transposons. For convenience, a standard consensus (ABL) numbering scheme has been adopted for the numbering of residues in class A P-lactamases (Ambler et al, 1991) and will be used hereafter. Based on the proposed similarities in the interaction of PBPs with P-lactams and acyl-D-Ala-D-Ala substrates, it was postulated that p-lactamases may have evolved from PBPs that gained the ability to rapidly hydrolyze the P-lactam acyl-enzyme intermediate (Tipper and Strominger, 1965). This was supported by initial comparisons between the overall folds of R61 PBP and class A P-17 lactamases horn Bacillus licheniformis 749/C (Kelly et al, 1986) and Bacillus cereus (Samraoui et al, 1986), which revealed a high degree of similarity in the overall arrangement of secondary structural elements. Subsequently, high resolution crystal structures of all four classes of B-lactamases (Fig. 1.4.1) have been determined and revealed that the serine B-lactamases share a common overall fold similar to that of the known PBP structures (Herzberg and Moult, 1987; Oefner et al, 1990; Carfi et al, 1995; Maveyraud et al, 2000; Paetzel et al, 2000). This conservation of overall structure exists despite a lack of sequence identity among the four classes of B-lactamases and the PBPs beyond the three active site motifs. Additionally there is an overall conservation of the spatial disposition of key structural features in the active sites including the nucleophilic serine and lysine residues from motif I, the hydroxylated residue from motif II, the conserved basic residue in motif HI, and the oxyanion hole. The observed structural similarities strongly support the evolution of the serine (3-lactamases from PBPs, and that these enzymes collectively constitute a superfamily of serine penicillin-recognizing proteins. 18 rJeacylating water Fig. 1.4.1: Crystal structures of (3-lactamases. a) class A p-lactamase from Staphylococcus aureus PC1. The omega loop and the Glu 166 residue conserved in all class A p-lactamases is highlighted in cyan. The deacylating water is shown as a red sphere, b) class B metallo-p-lactamase from Bacillus cereus. The active site zinc cations are shown as magenta spheres, c) class C p-lactamase from Citrobacter freundii. d) Oxa-10 class D p-lactamase from Pseudomonas aeruginosa. The colour scheme for secondary structural elements is the same as that for Fig. 1.3.2. The conserved motifs I, II and III are coloured red, magenta and green, respectively. 19 primordial PBP class D p-lactamase HMM class C PBPs LMM class A PBPs HMM class A PBPs HMM class B PBPs class A p-lactamase LMM class C PBPs class C p-lactamase LMM class B PBPs Fig. 1.4.2: Phylogenetic tree of penicillin-recognizing proteins. This is a graphical summary of multiple sequence alignment and cluster analyses of PBPs and p-lactamases. The evolutionary origin of the serine p-lactamases from PBPs is strongly supported by similarities in their overall structures. This figure was modified from Massova and Mobashery, 1998. In all three classes of serine P-lactamases, hydrolysis of the P-lactam substrate proceeds through an initial acylation reaction involving the nucleophilic serine (motif I) followed by a hydrolysis step that results in deacylation of the enzyme and release of the inactivated product. This overall reaction scheme is similar to that for the interaction of PBPs with P-lactams but differs in that the hydrolysis rates of acyl-enzyme intermediates of P-lactamases are rapid leading to high turnover rates. However despite similarities in their overall reaction schemes and conservation of active site motifs, the catalytic mechanisms for the three classes of serine P-lactamases appear to be quite 20 different, and details of these mechanisms have not been fully resolved. In the class C p-lactamase from Citrobacter freundii, Tyr 150 (motif II) is structurally equivalent to Tyr 159 in R61 PBP and is the proposed general base for both acylation (activation of Ser 64 of motif I) and subsequent deacylation (activation of deacylating water) with Lys 67 (motif II) and Lys 315 (motif III) stabilizing the Tyr 150 phenolate (Oefner et al, 1990; Lobkovsky et al, 1994; Dubus et al, 1996). However, studies using non-B-lactam compounds as mechanistic probes indicated the involvement of the substrate ring amine on the acyl-enzyme intermediate in activation of the deacylating water (Bulychev et al, 1997). The class D P-lactamases are the most recently characterized structurally, and consequently their catalytic mechanisms are the least understood. However the demonstration that carbamylation of Lys 70 (motif I) is critical for the activity of OXA-10, suggested that details of the catalytic mechanism of the class D P-lactamases may differ from those of other serine P-lactamases and PBPs (Golemi et al, 2001). The catalytic mechanism of class A P-lactamases has been studied most intensely (Galleni et al, 1995; Matagne and Frere, 1995; Matagne et al, 1998). A unique feature of the class A enzymes is Glu 166, a conserved residue in all class A P-lactamases on the omega loop (Fig. 1.4.1 a) which lines one side of the active site cavity (Ambler et al, 1991). It is generally accepted that deacylation involves activation of a highly coordinated and well-positioned nucleophilic water by Glu 166. The proton abstracted by Glu 166 may then be passed to Ser 70 (motif I) via Lys 73 (motif I) for collapse of the tetrahedral intermediate. This mechanism of deacylation is consistent with the structure of a boronate deacylation transition state analog in complex with the TEM-1 class A P-lactamase (Strynadka et al, 1996). In contrast, studies of the acylation mechanism have generated much controversy with two prominent mechanisms that propose either Lys 73 or Glu 166 as the general base. In one mechanism (Fig. 1.4.3a), Ser 70 is activated by a deprotonated Lys 73 with a 21 significantly lowered pA a^. Lys 73 shuttles the proton to Ser 130 (motif II), which protonates the leaving group nitrogen. Lys 234 (motif III) likely provides electrostatic assistance. This mechanism is consistent with the E166N mutant of TEM-1, which forms long-lived acyl-enzyme intermediates with penicillin (Adachi et al, 1991) and with detailed crystallographic studies ofthe penicilloyl-enzyme structure of this mutant (Strynadka et al., 1992). The ability of the E166N mutant to undergo acylation but not deacylation suggested an asymmetric mechanism, in which Glu 166 provided base catalysis in the deacylation but not the acylation step. In an alternative acylation mechanism (Fig. 1.4.1b), Glu 166 is the general base that activates Ser 70 via the deacylating water, which is also hydrogen-bonded to the Ser 70 Oy in the ground state (Herzberg and Moult, 1987; Lamotte-Brasseur et al, 1991). Reexamination of Glu 166 mutants have shown that mutations at this residue also impair acylation rates and indicated a role for this residue in both acylation and deacylation (Guillaume et al, 1997). More recently, a 0.91 A resolution structure of the SHV-1 class A (3-lactamase in the apo state shows the Ser 70 hydroxyl proton hydrogen-bonded to the water linking Ser 70 to Glu 166 (Nukaga et al, 2003), while a 0.85 A resolution structure of TEM-1 in complex with a phosphonate acylation transition state analog shows Glu 166 in a protonated state (Minasov et al, 2002). These observations argue for Glu 166 being the general base in acylation. The appreciable residual acylation rate of Glu 166 mutants maybe due to Lys 73 taking over the role of general base due to a lowered pA a^ resulting from the loss of the negatively charged Glu 166 carboxylate (Matagne et al, 1998). 22 a) Lys73 Ser70' O Glu166 0 Asn170 enzyme-substrate complex b) L v s 7 3 NH3+ ,-0 oj HOH O Glu166 O Asn170 enzyme-substrate complex Lys73 Ser130 I OH Lys73 Ser130 \ I 1 OH Lys73 HOH NH 3-/ Ser70 P S o Glu 166 tetrahedral intermediate 1 GIU166 NH 2 / Ser70 0~ S o N ' HO ° Lys73 OH acyl-enzyme intermediate Lys73 Ser130 \ I ] OH OH S o tetrahedral intermediate 2 Lys73 NH,-Ser70 y-H OH HOH 01 Glu166 ° tetrahedral intermediate 1 NH,-Ser70 O. ( O acyl-enzyme intermediate Ser70 OH s o 3D tetrahedral intermediate 2 Lys73 enzyme-product complex Lys73 Glu166 enzyme-product complex Fig. 1.4.3: Proposed catalytic mechanisms of class A P-lactamases. a) The proposed mechanism based on the stable acyl-enzyme intermediate of E166N mutant, in which an unprotonated Lys 73 is the general base for activation of the nucleophilic Ser 70 during acylation, and Glu 166 is the general base activating the deacylating water during deacylation. b) The proposed mechanism in which Glu166 is the general base in both acylation and deacylation. During acylation Glu 166 activates Ser 70 via the (deacylating) water molecule hydrogen-bonded to both residues. 1.4.2 M u l t i d r u g efflux pumps During his initial studies of the antibacterial properties of penicillin, Fleming had noted the greater sensitivity of Gram-positive bacteria to penicillin than Gram-negative species (Fleming, 1928). This intrinsic resistance of Gram-negative bacteria to antibiotics such as P-lactams results from a synergy of two factors, the outer membrane and multi-drug efflux. The outer membrane presents a permeation barrier to a number of antibiotics. Diffusion across the outer membrane is restricted by porin channels for hydrophilic molecules and by the low fluidity of the lipopolysaccharide leaflet for hydrophobic molecules. The second factor is the production of 23 tripartite multi-drug efflux systems that actively pump out a wide range of toxic and antibiotic agents both from the cytoplasm and from the periplasmic space (reviewed in Nikaido, 1996; Hancock, 1998; Nikaido, 1998; Nikaido, 2001). These efflux systems consist of an inner membrane drug-proton antiporter of the Resistance-Nodulation-cell Division (RND) family, an outer membrane channel component, and a periplasmic membrane fusion protein that is believed to somehow couple the inner and outer membranes for direct extrusion of antibiotics across both membranes. Crystal structures are available for two components ofthe AcrAB-TolC efflux system in is. coli (Fig. 1.4.4). The outer membrane component TolC was shown to be a trimer forming a large channel structure (>140 A long) spanning both the outer membrane and the periplasm (Koronakis et al, 1997). Four B-stands from each of the three monomers collectively form an outer membrane P-barrel porin-like domain, while oc-helices in the periplasmic region form a long a-helical barrel. The structure ofthe inner membrane component AcrB revealed a trimeric jellyfish-shaped structure consisting of a transmembrane region formed by helices (12 from each protomer) and large periplasmic regions containing putative sites of interaction with TolC (Murakami et al, 2002). The contribution of the AcrAB-TolC system to the intrinsic resistance of E. coli to P-lactams was demonstrated by the disruption of the acrAB genes, which resulted in significantly reduced MICs for a number of P-lactams (Mazzariol et al, 2000). It was proposed that upon entering the periplasm, amphiphilic compounds such as P-lactams partition partly into the outer leaflet ofthe cytoplasmic membrane (via their hydrophobic moieties) and are subsequently captured by the inner membrane component for active transport back to the external medium (Nikaido, 1996). Consistent with this proposal, the large central cavity (-5000 A 3 ) inside the AcrB trimer is accessible from the periplasm via 3 vestibules located at subunit interfaces. No structure is currently available for any of the periplasmic membrane fusion proteins. 24 outer membrane periplasm inner membrane Fig. 1.4.4: Crystal structures of TolC and AcrB. The TolC trimer (top) spans both the outer membrane and periplasmic space, while AcrB (bottom) spans the inner membrane and contains a large periplasmic region. The periplasmic region of AcrB is believed to interact with TolC for formation of a continuous channel from the cytoplasm to the cell exterior. The monomers in both proteins are coloured orange, blue and green. 25 Based on genomic sequence, a large number of RND-type multidrug efflux systems has been predicted in Pseudomonas aeruginosa, with 18 open reading frames (ORFs) encoding outer membrane proteins of the TolC family and 10 ORF pairs encoding members of the AcrAB family of efflux components (Stover et al, 2000). To date, five efflux systems ofthe RND-type have been characterized in P. aeruginosa:Mo\KB-0^xM (?OO\Q etai, 1993; Gotohef al, 1995),MexCD-OprJ (Poole et al, 1996a), MexEF-OprN (Kohler et al, 1997), MexXY-OprM (Mine et al, 1999), and MexJK-OprM (Schweizer, 2001; Chuanchuen et al, 2002). The MexAB-OprM system was the first of these systems described and shows the broadest substrate range (Poole et al, 1996b; Li et al, 1998). The mexAB-oprM operon is negatively regulated by the product of the mexR gene (Poole et al, 1996b; Srikumar et al, 2000). Mutations in mexR lead to hyper-expression of the mexAB-oprM operon, resulting in increased resistance to multiple antimicrobials including p-lactams, fluoroquinolones, tetracycline, macrolides, chloramphenicol, novobiocin, trimethoprim and sulphonamides (Srikumar et al, 1997; Srikumar et al, 1998; Saito et al, 1999; Srikumar et al, 2000). 1.4.3 Resistant PBPs Lacking an outer membrane, drug efflux cannot provide protection against P-lactams for Gram-positive bacteria, and P-lactamase-mediated resistance in Gram-positive cocci is found almost exclusively in staphylococci. However, staphylococcal P-lactamases are narrow-spectrum with relatively poor activity against semisynthetic anti-staphylococcal penicillins, cephalosporins, and carbapenems. P-lactamase production is not known to occur in streptococci, and enterococci also do not produce p-lactamases with the exception of a few strains that have acquired the blaZ P-lactamase gene from S. aureus. The dominant mechanism of P-lactam resistance in these Gram-26 positive bacteria is the expression of resistant PBPs that exhibit unusually low affinities for P-lactams (reviewed in Hakenbeck and Coyette, 1998; Chambers, 1999). In P-lactam-resistant strains of Streptococcus pneumoniae and the Gram-negative Neisseria gonorrhoeae, native chromosomally encoded PBPs are altered, resulting in increases in MICs for penicillin from -0.02 pg/mL in sensitive strains to 1-2 pg/mL in resistant strains. Highly resistant strains may show MICs of 4 to 8 pg/mL. Point mutations within the transpeptidase domains are often near the conserved motifs of HMM class B PBP genes in laboratory or clinical isolates and were associated with low levels of resistance. The level of resistance correlated with the extent of modification (number of mutations) and the number of PBPs modified (Chambers, 1999). An alternative strategy involves extensive modifications of PBP genes arising from introduction of exogenous DNA by transformation and homologous recombination with PBP genes from closely related species with intrinsic resistance (Spratt, 1988; Smithed al, 1991). Such recombination events have resulted in "mosaic" genes, in which blocks of the coding region have been swapped with those from a resistant homologue. Recombination allows a large number of mutations to be introduced more efficiently than via a series of single-step mutations. In enterococci and staphylococci, resistance is due to the expression of a low affinity PBP that in sensitive strains is not expressed or expressed in low amounts (enterococci), or absent (staphylococci). Relative to streptococci, enterococci are intrinsically more resistant to P-lactams with MICs of ~4 pg/mL for penicillin. This intrinsic resistance in enterococci is due to the presence of a native low affinity PBP (Hakenbeck and Coyette, 1998). In high level resistant strains, mutations result in increased expression or more commonly further reduction in the P-lactam affinity of this resistant PBP. Increases in MICs for ampicillin from 0.5 pg/mL in a more sensitive strain to 128 pg/mL in highly resistant strains have been observed (Ligozzi et al, 1996). The acquisition of 27 methicillin resistance in staphylococci is associated with the acquisition and expression of an extra-species low affinity PBP gene (mecA), which confers broad spectrum resistance to essentially all (3-lactams in clinical use (reviewed in Hiramatsu, 1995). Sequencing of the enterococcal low affinity PBP genes and the staphylococcal mecA gene identify them as a subgroup (BI) of the HMM class BPBPs (Matsuhashi et al, 1986; Matthews et al, 1987; elKharroubi etai, 1991;Piras etai, 1993; Signoretto et al, 1994; Ligozzi et al, 1996; Zorzi et al, 1996; Raze et al, 1998 and reviewed in Goffin and Ghuysen, 1998). The interaction of a PBP with a P-lactam inhibitor can be broken down into three steps: the rapid reversible formation of a non-covalent Michaelis complex, nucleophilic attack by the active site serine on the P-lactam ring (acylation) to form the covalent acyl-PBP intermediate, and hydrolysis of the acyl-PBP intermediate (deacylation) resulting in regeneration of active enzyme (Fig. 1.4.5; Frere et al, 1975; Ghuysen et al, 1986). Inhibition of PBPs by P-lactams is due to efficient formation (high k2) and slow hydrolysis (low k2) of the acyl-PBP intermediate. A comparison of the micro kinetic parameters of S. pneumoniae PBP2x from a sensitive (R6) and a highly resistant (CS109) strain (Lu et al, 2001) revealed that the reduction in overall affinity for penicillin G was largely due to a reduction in k2 (180 s"1 for R6 PBP2x* versus 0.56 s"1 for CS109 PBP2x*) with smaller contributions from changes in Kd (0.9 mM for R6 PBP2x* versus 4 mM for CS109 PBP2x*) and k3 (8 x 10"6 s"1 for penicilloyl-R6 PBP2x* versus 5.7 x 10"4 s"1 for penicilloyl-CS109 PBP2x*). Mutagenesis studies with R6 PBP2x have shown that replacement of Phe 450 with His, Glu or Asp can increase the deacylation rate by up to two orders of magnitude, apparently by mimicking the Glu 166 structural feature of class A P-lactamases (Chesnel et al, 2002). However this mutation has not been observed in clinical or laboratory isolates, and variants such as CS109 PBP2x utilize an alternative mechanism of deacylation enhancement, which remains to be 28 investigated. A closely related resistant variant (SP328 PBP2x; 96% amino acid sequence identity with CSI09 PBP2x for 666 aligned residues) shows deacylation rates comparable to those of a penicillin sensitive PBP2x (1 x 10"5 s"1 for R6 PBP2x and 3 x 10'5 s"1 for SP328 PBP2x; Di Guilmi et al, 2000). Michaelis K complex y acy l -PBP t, K d . Y ^ k2 . f Y . k3 ( P B P ) +(penici l l in) - * — • ( P B P j penicillin] • (PBP-penici l l in ] - j j -g- [ P B P ) + (penicilloate ) ! 2 V Fig. 1.4.5: Overall reaction scheme for interaction of a PBP with a 3-lactam. The p-lactam is represented by penicillin. K6 is the dissociation constant for the reversible formation of the non-covalent Michaelis complex. k2 is the first order rate constant for the formation of the acyl-PBP intermediate from the Michaelis complex. k3 is the first order rate constant for the hydrolysis of the acyl-PBP intermediate. The crystal structure for a soluble derivative of the resistant variant of PBP2x from S. pneumoniae Sp328 (Sp328 PBP2x*) has been determined (Dessen et al, 2001). Like the CS109 variant, Sp328 PBP2x is a "mosaic" resistant PBP containing 92 substitutions (82 in the soluble region) relative to the P-lactam-sensitive R6 PBP2x. The four molecules in the Sp328 PBP2x* asymmetric unit showed Ca rmsds of 0.5 to 0.6 A with the R6 PBP2x*structure (for 464 to 473 aligned pairs). A comparison of the Sp328 PBP2x structure with that of R6 PBP2x* (Pares et al, 1996; Gordon et al, 2000) revealed the disruption of a hydrogen-bonding network near the N-terminus of helix al (in the vicinity of the nucleophilic Ser 337) and a destabilization (disordering) of the loop regions consisting of residues 365 to 394 and 524 to 535 (Fig. 1.4.6), resulting in a more open active site and a displacement of the Ser 395 side chain (motif II). The hydrogen-bonding network near the N-terminus of helix al involves a buried structural water, and site-directed mutagenesis of the residues in contact with this water molecule were found to reduce the overall acylation efficiencies with P-lactams and with a thioester peptide substrate analogue (Mouz et al, 1998). The displacement of the catalytic Ser 395 side chain also likely plays a role in reducing the 29 acylation rate (k2) and p-lactam binding affinity (increased KA. The more open active site was proposed to facilitate binding of abnormal branched cell wall stem peptides present in the cell walls of resistant strains while reducing the number of contacts with and thus affinity for P-lactams. This proposal is consistent with observed differences in the cell wall stem peptide content of penicillin sensitive (>70% linear stem peptides) and resistant strains (>70% branched peptides with Ala-Ser or Ala-Ala dipeptides linked to the stem peptide Lys residues), and with the loss of penicillin resistance upon inactivation of the murMN genes, which are involved in the synthesis of branched cell wall peptides (Garcia-Bustos and Tomasz, 1990; Filipe and Tomasz, 2000). However since the structure of SP328 PBP2x* was determined in the apo state, details on the interactions between the active site and P-lactams are not clear. 30 a) b) Fig. 1.4.6: The more open active site of Sp328 PBP2x* relative to that of R6 PBP2x*. Stereo ribbon figures are shown for the transpeptidase domains of a) R6 PBP2x* and b) Sp328 PBP2x*. The orange regions in R6 PBP2x* (residues 365 to 394 and 524 to 535) define one side of the active site crevice and are disordered in Sp328 PBP2x*. The nucleophilic Ser 337 (motif I) side chain is shown in red ball-and-stick rendering, while the Ser 395 (motif II) side chain is shown in green ball-and-stick rendering. 31 1.5 Obj ectives of thesis The effectiveness of existing antibiotics is being diminished by the continuing emergence and dissemination of antibiotic resistance among pathogenic bacteria, and there is an urgent need for the development of novel antibiotic agents and of means to circumvent resistance mechanisms to extend the usefulness of existing antibiotics. Critical to the development of such agents will be a detailed understanding ofthe molecular components of resistance mechanisms. The overall aim of this thesis is the structural characterization of bacterial proteins involved in the three main modes of P-lactam resistance by X-ray crystallography. The first study is described in Chapter 3 and focussed on PSE-4, a representative member of a special group of class A P-lactamases (carbenicillinases) that are able to hydrolyze carbenicillin and other carboxy-substituted semisynthetic penicillins. The majority of carbenicillinases contain an arginine residue in place of the highly conserved Lys 234 (motif III) found in other class A P-lactamases. The importance of this unique feature of carbenicllinases was investigated through a structural analysis of the PSE-4 active site in the context of other class A P-lactamases of known structure. This work was published in Biochemistry (Lim et al, 2001). Chapter 4 describes the structure determination of the MexR repressor, which regulates the expression of the MexAB-OprM multi-drug efflux pump in P. aeruginosa. Mutations in the mexR gene result in over-expression of MexAB-OprM and increased MICs for p-lactams and a broad range of other antibiotics. The normal function of the MexAB-OprM efflux system is not known, and characterization of its regulator may provide hints as to the normal substrates of this efflux system, knowledge of which may be useful in the design of specific inhibitors of this system. A structural analysis of MexR was therefore undertaken to gain insights into its mode of DNA binding and how it may be regulated by effector binding. This work was published in the Journal ofBiological Chemistry (Lim etai, 2002). Chapter 5 describes the structure 32 determination ofthe ofthe low affinity PBP2a from methicillin-resistant S. aureus (MRSA). PBP2a is the key determinant of P-lactam resistance in MRSA, and an analysis of the active site both in the apo state and acylated with three different P-lactam compounds was conducted to determine features important for resistance and to provide insights for rational design of novel anti-MRSA agents. This work was published in Nature Structural Biology (Lim and Strynadka, 2002). 33 Chapter 2: MATERIALS AND GENERAL PROTOCOLS 2.1 Materials All reagents for DNA manipulation, protein purification and analysis, and crystal growth and manipulation were purchased from Fischer Scientific, Sigma Chemical Co., Amersham Biosciences, and Bio-Rad Laboratories unless otherwise specified. Pwo DNA polymerase, T4 DNA ligase, and DNA molecular size markers were purchased from Gibco BRL. The QIAquick Gel Extraction, QIAquick PCR Purification, and QIAprep Miniprep kits were purchased from QIAGEN. DNA oligonucleotides were purchased from Sigma-Genosys. Commercial screening kits, crystallization plates and crystal manipulation tools were purchased from Hampton Research. Nitrocefin was purchased from Oxoid Limited. All mutagenesis experiments utilized plasmid DNA propagated in Escherichia coli DH5a. All proteins were expressed from E. coli BL21 (A.DE3) strains using the Novagen pET® vector expression system. E. coli cells were transformed by electroporation using a MicroPulser Electroporator (Bio-Rad Laboratories). 2.2 DNA and protein sequence analysis DNA and protein sequencing were carried out by the staff at the Nucleic Acid Protein Service Unit (NAPS) at the University of British Columbia. DNA sequencing utilized the Applied Biosystems BigDye™ v3.1 Terminator Chemistry. The sequencing reactions were performed in an Eppendorf Mastercycler® gradient thermocycler using conditions detailed in the NAPS guidelines. The samples were then passed through SigmaSpin™ Post-Reaction Clean-Up Columns (Sigma 34 Chemical Co.) prior to submission to NAPS for drying and subsequent application on Applied Biosystems PRISM 377 automated sequencers. For N-terminal sequencing, proteins were separated by SDS-PAGE and electro-blotted on to a PVDF membrane prior to submission to NAPS. The Edman degradation reactions and analyses were performed by Suzanne Perry. 2.3 Over-expression of selenomethionine-substituted protein Seleonomethionine (SeMet)-substituted proteins were produced in the standard E. coli BL21 (A.DE3) expression strain. Typically, an overnight seed culture (grown in LB with antibiotic selection, 1 % culture volume) was washed and used to inoculate M9 medium (Sigma) supplemented with 2 mM MgS04, 0.1 mM CaCl2 and 4 g/L glycerol (Sambrook and Russel, 2001). Cultures were grown at 37 °C to an optical density at 600 nm (OD600) of-0.5, at which point 100 mg each of Thr, Lys and Phe, and 50 mg each of Leu, He and SeMet were added per litre of culture to suppress methionine biosynthesis. The culture was incubated for a further 15 minutes to allow remaining methionine to be depleted, before the addition of IPTG for induction (Doublie, 1997). The culture was shifted to 20 °C for overnight expression of soluble protein or maintained at 37 °C for expression of inclusion bodies. 2.4 Protein manipulation and storage All proteins were purified at 4 °C and concentrated by ultrafiltration using Ultrafree-15 centrifugal concentrators (Millipore Corporation) with molecular mass cutoffs below the molecular mass ofthe protein. Concentrated protein samples in 50 uL aliquots were rapidly frozen by plunging in liquid nitrogen and stored at -70 °C. Protein concentration in column eluate was monitored by 35 absorbance at 280 nm (A280). Protein purity was assessed by sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) with coomassie blue staining. 2.5 Protein mass analysis All mass spectrometry measurements were kindly performed by Dr. Shouming He (in the laboratory of Dr. Stephen Withers) on a PE-Sciex API 300 quadrupole mass spectrometer interfaced to a reverse phase column. Aqueous samples were submitted in quantities of 100 pg or greater. 2.6 Crystallization Crystallization trials were carried out using the standard hanging drop vapour diffusion method (McPherson, 1990) and 24-well crystallization plates (Hampton Research). Briefly, 1 uLof protein solution was pipetted on to a plastic cover slide and mixed with an equal volume of crystallization solution. The coverslide was then sealed over a well containing 0.5 mL of crystallization solution. Crystallization conditions are generally given as compositions ofthe well solutions. All solutions used for crystallization were filter sterilized with 0.2 um filters. Potential crystallization conditions were initially screened for using conventional sparse-matrix screens (Jancarik and Kim, 1991) with the commercial screening kits, Crystal Screens I & II and with a number of two dimensional grid screens. The screening kits contained pre-made solutions, while the solutions for the grid screens were mixed from stock solutions of precipitant, salts and buffers. The grid screens tested a range of conditions in which two parameters were varied, typically precipitant concentration and pH. 36 2.7 Preparation of crystals for X-ray diffraction studies For collection of X-ray diffraction data at room temperature, crystals were wet-mounted into quartz glass capillaries of 0.5 or 1.0 mm diameter. Typically, a crystal was drawn from its mother liquor or soaking solution into a glass capillary by pipetting. After the crystal was taken up, air was drawn in followed by a small volume of liquid. One end ofthe capillary was sealed with wax, and the crystal was then gently pushed into the air space with a strand of stretched glass capillary. After sealing the remaining end, the capillary was mounted on a goniometer with a piece of plasticine. An Oxford Cryosystems cryocooler was used to maintain crystals at 100 K. Crystals were transferred to a cryoprotectant solution and then mounted on cryo-loops. The loops were mounted on the goniometer with the cryo-stream momentarily blocked with apiece of plastic. Once the crystal was in place, the gas stream was quickly released. In preparation for synchrotron trips, crystals were transported in a Taylor-Wharton CP 100 dry dewar precooled during the previous night with liquid nitrogen. Crystals were first cryoprotected and transferred on cryo-loops into the cryo-stream. A diffraction image was then quickly collected to assess the diffraction quality ofthe crystal, and poorly diffracting crystals were discarded. Suitable crystals were transferred to liquid nitrogen using precooled cryo-tongs. While under liquid nitrogen, the crystals were transferred into vials, which were then stored on metal canes in the dewar. 2.8 Data collection and processing Diffraction data were collected using the oscillation method, in which a crystal was exposed to the X-ray beam and rotated about a spindle axis over a given oscillation range, while the resulting diffraction pattern was recorded on a detector. After a given exposure time, the detector was scanned and a digital image of the diffraction pattern was stored for analysis. A series of such images 37 comprised a complete data set. A typical oscillation range of 0.5 or 1.0° was used for most crystals. Exposure times were chosen to obtain sufficiently strong intensities and to make optimal usage of the dynamic range ofthe detector. Data collection at the home source utilized a Rigaku RU-200 rotating anode X-ray generator (50 kV, 100 mA) with Osmic focussing optics. Diffraction images were collected initially on a R-axis He image plate detector (Rigaku), which was subsequently replaced with a mar345 image plate detector (MAR Research). At synchrotron sources, data were collected on ADSC Quantum-4 CCD detectors. For multi-wavelength anomalous diffraction (MAD) data collection, an X-ray fluorescence scan was first measured for a selenomethionine (SeMet)-substituted crystal, from which the peak and inflection wavelengths were chosen to maximize f' and to minimize f, respectively (Hendrickson, 1991). MAD data were collected at wavelengths in the following order, peak, inflection and remote. Anomalous data were collected in single sweeps with no attempt to collect data in mirror or inverse beam geometry. Starting values for f and f' at the peak and inflection wavelengths were estimated from X-ray fluorescence spectra using CHOOCH (Evans and Pettifer, 2001), while for the remote wavelength, theoretical values (Sasaki, 1989) were used. The HKL package (Otwinowski, 1993) and the CCP4 software suite (Collaborative Computational Project, 1994) were used for data reduction and processing. Unit cell dimensions and initial crystal orientations were determined using the autoindexing feature of DENZO (Otwinowski, 1993). Optimal starting angles for data collection were determined using the program STRATEGY (Ravelli et al, 1997). Data reduction with DENZO involved reducing each diffraction image to a list of indexed reflections and their corresponding integrated spot intensities. The data were then combined together to produce a final set of reflections and intensities using SCALEPACK (Otwinowski, 1993), in which a scale factor and a 5-factor were applied to the intensities of each 38 frame to place all frames on a common scale. For anomalous data, Friedel mates were not merged. Intensities were converted to structure factor amplitudes and further processed using programs from the CCP4 software suite. The number of molecules in the unit cell was determined using the formula FM=(unit cell volume)/[w (molecular weight of protein)], where VM is likely to be in the range of 1.68 to 3.53 A3/dalton, based on measurements from 116 protein crystals, and where n is an integer (Matthews, 1968). The solvent content (Fsolv)was estimated using the formula F s o l v=l- l.23/VM, which assumes a partial specific volume of 0.74 cm3/g for most proteins. 2.9 Model building and refinement For electron density maps calculated from heavy atom-derived phases, atomic models of the protein structure were manually built using the Xfit program from the Xtalview software suite (McRee, 1999). In each case, an initial Ca trace was created using the Baton method, in which a distance constraint of 3.8 A between Ca atoms was imposed. A poly-alanine trace was then produced by substituting closest matching pentamers from a protein structure library for short regions of the Ca trace. Finally the side chains were automatically added using the known protein sequence. For structures determined by molecular replacement, the search model was used as the initial model and adjusted (with side chain substitutions if necessary) to fit the electron density map. Models for all proteins were refined to minimize the .K-factor, while conforming to geometric restraints derived from high resolution small molecule structures (Engh and Huber, 1991). The R-factor is defined as Rcryst = £|Fobs - Fcalc|/£|Fobs|, where Fobs and Fcalc are the observed and calculated structure factors respectively. All models were refined in CNS (Pannu and Read, 1996; Briinger et al., 1998). A small percentage (5 %) of the data was flagged for exclusion from refinement and for calculation of the free R-factor (i?free; Briinger, 1992). No low resolution or sigma cut-off was used, 39 and bulk solvent correction was applied to low resolution reflections (25 to 6 A ) . Initial models from manual building or from molecular replacement solutions were subjected to 20 steps of rigid body refinement to fine-tune the overall position and orientation of the molecule(s) in the asymmetric unit. Each model was then subjected to one cycle of refinement, which consisted of the following. The model was first minimized with 200 steps of positional refinement, in which the positions of the individual atoms in the model were refined. This was followed by torsion molecular dynamics refinement with simulated annealing using a temperature gradient (2500 to 300 K in the first round of refinement and 1000 to 300 K in subsequent rounds) and finally by 100 steps of positional refinement. Molecular dynamics at high temperatures allow the model to overcome torsional strain to refine to more energetically favourable conformations. This helps to prevent the model from being trapped in conformations of local minima. Following simulated annealing, restrained individual B-factor refinement was applied. In the final rounds of refinement, the weight for B-factor restraints was optimized to minimize i? f r e e. The water picking protocol implemented in CNS was used to position water molecules in the model using a cutoff of 4.0 G in the F0-Fc map. For structures of complexes, group occupancies of ligands were also refined. Following each cycle of refinement, sigmaA-weighted 2F0-FC and F0-Fc electron density map coefficients (Read, 1986) were calculated with CNS for map display in Xfit. Model refinement was completed by iterative rounds of manual building or adjustment in Xfit and refinement in CNS. Real-space /(-factors computed by CNS were used to identify improperly positioned water molecules that were in poor agreement with their local electron density. 40 2.10 Structure analysis Secondary structure assignments in the refined models were made with the program STRIDE (Frishman and Argos, 1995) with subsequent manual inspection and editing when necessary. Superposition of similar protein structures were made with the programs ALIGN (Cohen, 1997) and LSQKAB (Kabsch, 1976) from the CCP4 suite, and with the LSQFit protocol in Xfit (McRee, 1999). Structures were superimposed by minimizing the root mean square deviations (rmsd) between equivalent atoms, which were selected manually for Xfit and LSQKAB and automatically in ALIGN. Stereochemical parameters of models were calculated by PROCHECK (Laskowski et al., 1993) and CNS (Briinger et al., 1998). Ramachandran plots (Ramakrishnan and Ramachandran, 1965) were produced by PROCHECK. Figures of protein structures were created with MOLSCRTPT (Kraulis, 1991) and RASTER3D (Merritt and Murphy, 1994). Figures containing electron density maps were created with Xfit and RASTER3D or with O (Jones et al., 1991), MOLSCRTPT and RASTER3D. Figures containing molecular surfaces coloured by electrostatic or hydrophobic potentials were created with GRASP (Nicholls et al, 1993). 41 Chapter 3: PSE-4 Penicillin G 1 H N \ 2 / S " ' f a~ N-4 O Carbenicillin 1 H N' 4 O 3.1 Introduction Pseudomonas-specific enzyme (PSE)-4 is a class A P-lactamase (Boissinot and Levesque, 1990) produced by certain strains of Pseudomonas aeruginosa, a Gram-negative opportunistic pathogen that has become a significant problem in hospitals worldwide (reviewed in Bodey et al., 1983). PSE-4 and other carbenicillinases are classified as group 2c P-lactamases (reviewed in Bush et al., 1995) and show activity towards the penicillin derivative carbenicillin (Fig. 3.1.1; Acred et al, 1967) equal to or higher than that for aminopenicillins (Boissinot and Levesque, 1990; Bush et al, 1995). Carbenicillin was first Fig. 3.1.1: Structures of penicillin G introduced in 1967 and was the first p-lactam showing a n d carbenicillin. Carbenicillin contains an additional carboxylate clinically useful levels of activity against P. aeruginosa g r o u p a t t h e a position indicated by the arrow. Conventional numbering (Rolinson and Sutherland, 1967). It could be tolerated j s shown for the bicyclic atoms. in high doses and was able to safely and effectively treat P. aeruginosa infections, however resistance was soon reported (Lowbury et al., 1969). PSE-4 was first isolated from the carbenicillin-resistant strain Dalgleish in 1969 (Newsom, 1969; Newsom et al., 1970) and is the most commonly occurring P-lactamase in carbenicillin-resistant strains of P. 42 aeruginosa (Boissinot and Levesque, 1990). PSE-4 is closely related to a number of other carbenicillmases (Boissinot a/., 1989):PSE-1 (Huovinen and Jacoby, 1991), CARB-3 (Lachapelle et al, 1991), CARB-4 from P. aeruginosa and AER-1 from Aeromonas hydrophila (Sanschagrin et al., 1998), CARB-5 from Acinetobacter calcoaceticus var. anitratus (Choury et al., 2000), C ARB-6 Vibrio cholerae (Choury et al., 1999), N29 and GN79 from Proteus mirabilis (Sakurai et al., 1991; Ito and Hirano, 1997), and PSE-3 from Rhodopseudomonas capsulata sp.108 (Campbell et al, 1989). Residue 234 is arginine in all carbenicillinases sequenced to date with the exception of PSE-3 and AER-1, which like other class A B-lactamases have lysine at this residue. The ABL standard numbering scheme for class A B-lactamases will be used throughout (Ambler et al, 1991). Residue 234 is located within the binding pocket near the nucleophilic Ser 70. Previous molecular modelling studies have suggested that Arg 234 in PSE-4 may be involved in substrate recognition (Boissinot and Levesque, 1990). Mutagenesis studies in the class A enzyme TEM-1 have also implicated a role for the K234R substitution in the hydrolysis of a-carboxypenicillins (Lenfant et al, 1991; Lenfant et al, 1993) in addition to other residues conserved in carbenicillinases. To clarify the role of the conserved arginine in carbenicillinases and to identify other structural features important for carbenicillinase activity, crystallographic studies of the PSE-4 enzyme were initiated. The 1.95 A wild type structure and the structure of a R234K mutant refined to 1.75 A presented here provide the first structural data on a carbenicillinase. Kinetic and structural analyses of the wild type and R234K mutant enzymes reveal unique features in PSE-4 that are likely important for carbenicillin hydrolysis. 43 3.2 Materials Both the wild type and R234K mutant PSE-4 proteins were expressed from pET30a-based constructs created by Francois Sanschagrin then at the laboratory of Dr. Roger Levesque, who provided us with E. coli BL21 (A.DE3) strains harbouring these plasmids. 3.3 Methods 3.3.1 Protein over-expression and purification The cells were grown in Aharonowitz defined medium (Jensen et al, 1982) consisting of 1 g glycerol, 21 g MOPS, 0.6 g MgS04, and 2 g L-asparagine in 900 mL, supplemented with 4.5 g K2HP04 in 50 mL titrated to pH 7, 10 g Casamino acids (Sigma) in 50 mL, and 1 mL of Aharonowitz trace element solution containing 1 mg, FeS04"7H20, 1 mg MnCl2"H20, 1 mg ZnS04'7H20, and CaCl23H20. Ampicillin at a concentration of 128 pg/L was used for antibiotic selection. A 20 mL overnight culture was first pelleted by centrifugation and resuspended in 1 mL of fresh medium. The washed cells were then used to inoculate four 1 L cultures. The cultures were grown at 37 °C with vigorous shaking to an optical density (A600) of approximately 0.5. The cultures were then induced with 1 mM IPTG and grown for a further 4 hours at room temperature. Both mutant and wild type PSE-4 enzymes were purified using an osmotic shock protocol, followed by anion exchange chromatography and gel filtration. The cells were pelleted by centifugation at 5000x g for 10 minutes, resuspended in 80 mL of 20% sucrose in 33 mM Tris pH 8 and with 1 mM EDTA and incubated at room temperature for 10 minutes. The cells were then pelleted and resuspended in 40 mL of ice cold deionized H20 with vigorous shaking and vortexing. After incubation on ice for 10 minutes, the cells were centrifuged at 5000x g for 15 minutes. The 44 supernatant was then further centrifuged at 20,000x g for 30 minutes. Stock buffer solution (1M Tris pH 8) was added to the supernatant to a final concentration of 30 mM. All subsequent purification steps were carried out at 4°C. The sample was loaded at 2 mL/min on to a Q-sepharose FF column (1.5 x 10 cm) pre-equilibrated with 20 mM Tris pH 8. The column was then washed overnight with 800 mL of the same buffer at 1 mL/min. The protein was eluted with a linear gradient of 0 to 0.2 M NaCl at 1 mL/min over a volume of 100 mL. Peak fractions were analysed by SDS-PAGE (Fig. 3.3.1a). Fractions with the highest purity were pooled and concentrated to 0.5 mL by ultrafiltration. The sample was then loaded on to a sephacryl-100 column (2 x 65 cm) equilibrated with 150 mM NaCl in 50 mM Tris pH 8. The protein was eluted at 0.8 mL/min, and peak fractions with the highest purity as shown by SDS-PAGE were pooled (Fig. 3.3. lb), concentrated to 0.5 mL and filtered through a 0.1 (im filter. Final protein concentration was determined by measuring A 2 8 0 in 6 M guanidine hydrochloride and using an extinction coefficient of 1.031 cm"1 for a 1 mg/mL solution as estimated from the mature amino acid sequence (Expasy ProtParam web page, http://us.expasy.org/tools/protparam.html; Gill and von Hippel, 1989). Approximately 15 mg of enzyme were obtained from 4 L of culture. 45 Fig. 3.3.1: SDS-PAGE of PSE-4 fractions, a) Fractions off Q-sepharose FF column. The first lane is a sample of the crude osmotic shock fluid, b) Fractions off sephacryl-100 column. The numbers indicate the approximate molecular masses (kDa) of the molecular markers (lane 2). 3.3.2 Crystallization Conditions for the growth of small PSE-4 crystals (2 M ammonium sulfate) were initially determined by Liza De Castro by sparse matrix screening (Hampton Research). I refined these conditions using a pH versus divalent cation salt concentration grid screen to produce large single crystals diffracting to high resolution at the home source. Bipyramidal crystals of wild type PSE-4 measuring up to 0.6 mm in the longest dimension were grown using the hanging drop vapour diffusion method by mixing 3 pL of a 10 mg/mL protein solution with 1 pL of well solution consisting of 2 M ammonium sulfate, 50 mM MOPS pH 6.4 and 100 mM MgCl2. Crystals appeared 46 after 1 to 2 days and grew to maximum size after 1 to 2 weeks at room temperature. The R234K mutant crystals were grown in a similar manner but with 0.1 M sodium acetate pH 4.5 as the buffer. 3.4 Results 3.4.1 Data collection and processing Data were collected on the home source at 100 K with an R-axis Uc detector. Crystals were cryoprotected by stepwise transfer to a solution of 2 M ammonium sulfate and 35% (w/v) sucrose and mounted on the goniometer while blocking the cryostream. Based on systematic absences, the crystals belonged to either space group P4l2l2 or P432,2 with unit cell dimensions (wild type) of a = b = 95.2 A and c = 62.8 A , and one molecule per asymmetric unit with a VM of 2.45 A3/Da. Transient blockage of the cryostream during crystal mounting was critical to prevent modulation of the crystal packing into a perfect hemihedrally twinned P4, cell. Data statistics for wild type and R234K PSE-4 crystals are shown in Table 3.4.1. 47 Table 3.4.1: PSE-4 X-ray data statistics decayed wild type wild type R234K temperature (K) 295 100 100 unit cell axes: a, c (A) 95.4, 64.9 95.2, 62.8 95.0, 62.7 resolution (A)A 25-2.9 (3.0-2.9) 25-1.95 (2.02-1.95) 25-1.75 (1.81-1.75) observations 88,156 208,439 162,602 unique reflections 7,055 21,494 29,406 completeness (%)a 99.9 (100.0) 99.5 (96.5) 99.5 (98.6) average //a(7)a 22.3 (8.2) 33.0 (8.3) 25.4 (3.4) overall 5-factor (A2) 38.7 17.7 20.4 0.121 (0.342) 0.059 (0.219) 0.056 (0.359) a Values in parentheses correspond to the highest resolution shell. b R = 2|/av - L\l ZZ, where J a v is the average of all observations, 3.4.2 Structure determination and refinement The structure was initially determined by Lori Passmore (then an undergraduate student in the lab) by molecular replacement with the CCP4 program AMoRe (Navaza, 1994; Navaza, 2001) using a search Fig. 3.4.1: Representative electron density in PSE-4 active site. The SigmaA-weighted 2F0-FC electron model consisting of the TEM-1 p- density map is contoured at 1.5 a and to a resolution of 1.95 A. lactamase crystal structure Strynadka et al, 1992 with side chains modified to match the PSE-4 sequence. The correct space group was determined to be P412,2. Since the initial molecular replacement and structure refinement was performed using low resolution data of poor quality collected from two crystals at room temperature 48 (with clear signs of decay; Table 3.4.1), I collected a higher resolution and higher quality data set after producing larger crystals and establishing suitable cryoprotection conditions. I was then able to confirm the molecular replacement solution using an unmodified TEM-1 crystal structure (53.6% correlation coefficient and 45.1% R-factor after rigid body refinement) and complete the structure refinement. The refined model for wild type PSE-4 (with an R234A substitution and without water molecules) was used as an initial model for the R234K mutant. Following rigid body refinement, simulated annealing and individual 5-factor refinement, difference density for the Lys 234 side chain was clearly visible and unambiguously modelled. In both the wild type and R234K models, a sulfate group was positioned between Ser 130 and Arg 243 in the active site region to account for density that could not be modelled with any other known component of the crystallization conditions (Fig. 3.4.1). The occupancy of the sulfate group was refined using an initial 5-factor equal to the overall crystal average (Table 3.4.1), following which the 5-factors of the sulfate atoms were refined individually. Significant residual density remained around the sulfate group, which was not readily interpretable. The refinement statistics for both the wild type and R234K models are shown in Table 3.4.2. The rmsd between the R234K and wild type structures is 0.3 A for all protein atoms (2035 pairs). 49 Table 3.4.2: PSE-4 structure refinement statistics wild type R234K -^ cryst / -^ free no. of non-hydrogen atoms (protein / water / sulfate) rmsd from ideality bond lengths ( A ) bond angles (°) angle improper (°) average J5-factor ( A 2 ) main chain side chain water overall Luzzati coordinate error ( A ) resolution range (A) Ramachandran plot (% of residues in region) most favourable allowed generously allowed disallowed 0.167/0.212 0.177/0.209 2057/300/5 2050/258/5 0.0075 1.3 0.87 17.0 20.4 25.9 20.2 0.18 5.0-1.95 90.7 8.9 0.4 0.0 0.0097 1.4 0.97 18.0 21.2 31.1 20.9 0.18 5.0-1.75 92.0 7.2 0.8 0.0 50 3.5 Discussion 3.5.1 Overa l l structure The overall structure of PSE-4 is similar to those of other class A B-lactamases and consists of an a/p domain and a predominantly a-helical domain (Fig. 3.5.1). The oc/p domain contains a Fig. 3.5.1: Overall structure of PSE-4. a-helices are labelled from hi to hi 1, while p-strands are labelled from s1 to s5. The active site is indicated by a ball and stick representation of the Ser 70 side chain. central five-stranded antiparallel P-sheet with a-helices on both sides. The active site is located at the interface between the two domains. The Ca trace of PSE-4 shows rmsd values in the range of 1.3 to 2.7 A relative to those of other class A enzymes: the Streptomyces albus G (Dideberg et al, 51 1987), Enterobacter cloacae NMC-A (Swaren et al, 1998), Escherichia coli (strain TUH12191) TOHO-1 (Ibuka et al, 1999), Klebsiella pneumoniae SHV-1 (Kuzin et al, 1999), Staphylococcus aureus PCI (Herzberg, 1991), Bacillus licheniformis 749/C (Knox and Moews, 1991), and E. coli TEM-1 (Fig. 3.5.2a; Strynadka et al, 1992) P-lactamases. Fig. 3.5.2: Comparison of the PSE-4 and TEM-1 structures, a) Superposition of the Ca atoms of PSE-4 (thick blue trace) and TEM-1 (thin orange trace). The rmsd for 261 pairs of Ca atoms is 1.3 A. Residues 50 to 58, 226 to 228 and 250 to 256 show the largest deviations. Ser 70 indicates the location of the active site, b) Superposition of the active sites of wild type PSE-4 (thick rendering with yellow carbons) and TEM-1 (thin rendering with magenta carbons). 3.5.2 Active site The positions and orientations of the conserved catalytic residues in the wild type PSE-4 structure are similar to those of other class A p-lactamases and consist of Ser 70 (nucleophile), Lys 73 (potential general base for acylation; Strynadka et al, 1992; Swaren et al, 1995), Ser 130 (proton 52 transfer between the leaving group thiazolidine N 4 and Lys 73), Glu 166 (possible general base for acylation or deacylation; Lamotte-Brasseur et al, 1991; Strynadka et al, 1992), Asn 170 (coordination of deacylating water) and Arg 234 (depression of Ser 130 pKJ. A superposition of these residues with the corresponding residues in native TEM-1 shows a rmsd of 0.4 A for 47 pairs of main chain and side chain atoms (Figure 3.5.2b). The key differences between PSE-4 and the other class A penicillinases center around interactions from the unique Arg 234. The Arg 234 side chain is hydrogen-bonded to Wat 10 and the side chain hydroxyl of Ser 130. Ser 130 in the wild type PSE-4 structure is uniquely observed in two alternate conformations (Fig. 3.4.1,3.5.2b), which were each modelled with 50% occupancy. The 5-factors for the two conformers both refined to 14.9 A 2 . In all other class A P-lactamase structures, the conformation of the Ser 130 side chain is such that the Xi values are in the range of -120.5° to -163.5° (-153.4° in PSE-4). The alternate conformer ofthe Ser 130 side chain fa, = -69.7°) in PSE-4 is stabilized by hydrogen-bonding to the Nrd and Nr|2 atoms of Arg 234 with distances of 3.0 and 2.9 A , respectively. To investigate the importance of Arg 234 on carbenicillin hydrolysis by PSE-4, Km and ATCAT values for carbenicillin, ampicillin and penicillin G hydrolysis by wild type and R234K PSE-4 enzymes were measured by Francois Sanschagrin, then a graduate student at Dr. Roger Levesque's laboratory (Table 3.5.1; Lim et al, 2001). Relative to Lys 234 in other class A p-lactamases, Arg 234 in wild type PSE-4 is restricted in its ability to hydrogen-bond with substrate, as hydrogen-bonds are only favourable in the plane of the arginine guanidium group (Fig. 3.4.1,3.5.2b). It therefore seems unlikely that the reduction in carbenicillinase activity by the R234K mutation in PSE-4 results from loss of substrate interactions with Arg 234. This is consistent with the kinetic data, which shows that the R234K mutation has little effect on the Km for carbenicillin (Table 3.5.1). This also agrees with 53 the crystal structure of the TEM-l-penicillin G acyl-enzyme intermediate, which showed that Lys 234 makes only a minor electrostatic contribution to substrate binding (Strynadka et al, 1992). Table 3.5.1: Effect of R234K mutation on P-lactam hydrolysis by PSE-4 wild typea R234K" ^ca< CS"') K c J K m (xlO6 s'M1) ^ca , CS"') KjKm(xio6 s-VM-y relative to wild type Carbenicillin 68 ± 4 1200±120 17±2 129 ±62 57± 11 0.4 ± 0.2 / 0.02 Ampicillin 33 ± 3 1170 ±130 35 ±5 74 ± 11 840 ± 70 11 ± 2 / 0 . 3 Penicillin G 57 ±8 890 ± 40 16 ± 2 100 ±40 430 ± 70 4± 2/0.25 a Values for carbenicillin and ampicillin hydrolysis (wild type) were obtained from an earlier study (Savoiee^/., 2000). All other values were obtained by Francois Sanschagrin (Lim et al, 2001). The wild type and R234K mutant PSE-4 structures differ only in the local region of residue 234, with the remainder of the protein being essentially unchanged (Fig. 3.5.3a). Weaker density is seen for Wat 10 in the mutant structure (5-factor = 26.7 A 2 ) , which likely results from replacement ofthe Arg 234 Ns by the Lys 234 C5. The loss ofthe Arg 234 Nn2 creates a void in the R234K PSE-4 mutant that is filled by Wat 52 (5-factor =19.7 A 2 ) , which is not present in the wild type structure. Relative to Lys 234 in other class A structures, the conformation of Lys 234 in the R234K PSE-4 mutant is unusual (Fig. 3.5.3b) and positions the side chain NC, further away from the active site, so that it is hydrogen-bonded to the backbone oxygen of Thr 126 (2.9 A ) and to Wat 52 (2.7 A ) . It is not clear from the available structural data what factors stabilize the unusual conformation of Lys 234 in the PSE-4 R234K mutant, although it does not seem to result from the presence of Wat 52. Wat 52 is not hydrogen-bonded to any other atom and therefore does not provide any bridging interactions that would help to anchor the Lys 234 N<^. The conformation and hydrogen-bonding 54 pattern of Lys 234 in the R234K PSE-4 mutant seem possible in other class A structures but is apparently not favoured. Fig. 3.5.3: Comparison of the active site region of the R234K PSE-4 mutant with the wild type PSE-4 and TEM-1 structures, a) Superposition ofthe active site region near residue 234 ofthe wild type (thick rendering with yellow carbons) and R234K mutant (thin rendering with magenta carbons) PSE-4 structures. Density for Wat 10 is markedly reduced in the R234K mutant structure. Wat 52 is only seen in the R234K mutant, b) Superposition ofthe active site region near residue 234 ofthe R234K PSE-4 mutant (thick rendering with yellow carbons) and TEM-1 (thin rendering with magenta carbons). The position ofthe TEM-1 Lys 234 Nz atom is similar to those of other class A p-lactamases. Lys 234 in the R234K PSE-4 mutant may still act to lower the pKa of Ser 130, since the Lys 234 to Ser 130 Oy distance (3.0 A ) is still within the range observed in other class A structures (2.6 to 3.0 A ) . Interestingly, the side chain of Ser 130 in the R234K PSE-4 mutant is in a single conformation (Xi = -146.6°, 5-factor = 19.6 A 2 ) , the most significant structural difference resulting from the R234K substitution in PSE-4. 55 3.5.3 M o d e l of the acyl-enzyme intermediate To visualize the role of the Ser 130 alternate conformer in carbenicillin hydrolysis, the acyl-enzyme intermediate of PSE-4 with carbenicillin was modelled using the TEM-l-penicillin G complex structure (Strynadka et al, 1992). Fig. 3.5.4: Model of the acyl-enzyme intermediate of PSE-4 with carbenicillin. The carbenicilloyl moiety is shown with purple carbons. The a-carboxylate is in a steric clash with Asn 170. The penicillin G atoms were positioned into the active site of PSE-4 by superimposing the TEM-1 Ser 70 atoms on to those of PSE-4. A carboxylate group was then added to the Ca atom of penicillin G, using an ideal model of carbenicillin (Dobashi, 1998) as a guide. The resulting model (Fig. 3.5.4) shows the carbenicillin a-carboxylate in a steric clash with the Asn 170 N52 (1.7 A), a highly conserved residue in class A p-lactamases. The Asn 170 side chain forms a hydrogen-bonding network (observed in the vast majority of class A structures to date) with the side chain carboxylate of the proposed general base for deacylation, Glu 166, and the proposed deacylating water. Disruption of this hydrogen-bonding network is energetically unfavourable and would displace or eliminate the proposed deacylating water. Elimination of this water in an NI 70Q mutant of the PCI P-lactmase reduced the steady state rate of nitrocefm hydrolysis by 800 fold (Zawadzke et al., 1996). Thus the a-carboxylate of carbenicillin cannot be accommodated simply by movement of the Asn 170 side chain, and the position of carbenicillin in the PSE-4 substrate binding cavity is likely shifted relative to penicillin G in TEM-1. Given that Ser 130 is the only residue in a position to act as a proton donor to the substrate thiazolidine N 4 during the acylation step 56 (Strynadka et al., 1992), the alternate conformation of Ser 130 in wild type PSE-4 may accommodate the shifted position of the carbenicillin thiazolidine N4. Similar to both the wild type and R234K mutant PSE-4 enzymes (Table 3.5.1), TEM-1 exhibits comparable Km values for benzylpenicillin and carbenicillin (24 pM and 14 pM, respectively; Lenfant et al, 1993). The low rates of carbenicillin hydrolysis (relative to those for benzylpenicillin) for TEM-1 and the R234K PSE-4 mutant are largely due to low kcat values. In the case of TEM-1 the &cat for benzylpenicillin (1200 s"1) and carbenicillin (120 s"1) differed by an order of magnitude (Lenfant et al., 1993). That for these enzymes the rate of carbenicillin hydrolysis seems to be more dependent on turnover rate rather than substrate affinity is consistent with the conformation of Ser 130 being the key difference between the wild type and R234K mutant PSE-4 structures. Kinetic studies of a K234R TEM-1 mutant, which showed a 6 fold increase in kcat for carbenicillin with essentially no change in the &cat for benzypenicillin, confirmed the importance of Arg 234 in catalyzing carbenicillin hydrolysis (Lenfant et al, 1993). However increases in the Km values for carbenicillin (13 fold) and benzylpenicillin (10 fold) indicate that the K234R substitution in TEM-1 resulted in changes in the substrate binding pocket that impaired overall substrate binding (Lenfant et al, 1993). Differences between the TEM-1 and PSE-4 structures must account for the inability of TEM-1 to fully accommodate an arginine at residue 234. Indeed, these differences may even involve residues distant from the active site, as was found to be the case with trypsin, for which mutations outside of the active site and SI regions were necessary to introduce partial chymotrypsin activity (reviewed in Hedstrom, 1996). Given that certain structural features important for carbenicillinase activity may be very subtle, directed evolution (reviewed in Petrounia and Arnold, 2000; Tobin et al, 2000) may provide an efficient means to identify mutations in TEM-1 that would 57 confer high levels of carbenicilliase activity, particularly for positions distant from the active site and for which the contributions are less apparent from crystal structures. Such an approach was successful in isolating a TEM-1 triple mutant with a greater than 2000 fold increase in kcJKm for cefotaxime (Zaccolo and Gherardi, 1999). PSE-3 and AER-1 do not contain the K234R substitution and provide further examples of the importance of other factors for carbenicillinase activity. It will be interesting to see if alternate conformations of Ser 130 are also present in the K234R TEM-1 mutant, and perhaps in PSE-3 and AER-1, in which an alternate conformer of Ser 130 may be stabilized in a different way. 58 Chapter 4: MexR 4.1 Introduction The intrinsic resistance of Pseudomonas aeruginosa results from a synergy of low outer membrane permeability and chromosomally-encoded tripartite efflux systems (reviewed in Hancock, 1998; Nikaido, 1998; Nikaido, 2001; Poole, 2001). To date, five such efflux systems have been characterized in P. aeruginosa: MexAB-OprM (Fig. 4.1.1; Poole et al, 1993; Gotoh et al, 1995), MexCD-OprJ (Poole et al, 1996a), MexEF-OprN (Kohler et al, 1997), MexXY-OprM (Mine et al, 1999), and MexJK-OprM (Schweizer, 2001; Chuanchuen et al, 2002). These efflux systems consist of an inner membrane drug-proton antiporter of the Resistance-Nodulation-cell Division (RND) family (MexB, MexD, MexF, MexY and MexJ), an outer membrane channel component (OprM, OprJ and OprN), and a periplasmic membrane fusion protein (MFP - MexA, MexC, MexE, MexX and MexK). The MexAB-OprM system was the first multidrug efflux system described in P. aeruginosa and shows the broadest substrate range (Poole et al., 1996b; Li et al., 1998). The mexAB-oprM operon is negatively regulated by the product ofthe mexR gene (Poole et al., 1996b; Srikumar et al., 2000), which is located upstream of mexA and transcribed in the opposite direction. Mutations in mexR lead to hyper-expression of the mexAB-oprM operon, resulting in increased resistance to multiple antimicrobials including p-lactams, fluoroquinolones, tetracycline, macrolides, chloramphenicol, novobiocin, trimethoprim and sulphonamides (Srikumar et al, 1997; Srikumar et al., 1998; Saito et al., 1999; Srikumar et al., 2000). Signalling molecules utilized in quorum sensing have also been implicated as substrates for the efflux pumps in P. aeruginosa. Over-expression of the MexAB-OprM efflux system has been found to decrease intracellular accumulation of the P. 59 aeruginosa autoinducer 3-O-dodecanoyl-homoserine lactone (P AI-I) resulting in reduced expression of autoinducer regulated virulence factors (Evans et al, 1998; Pearson et al, 1999). outer membrane '/ Fig. 4.1.1: mexRAB-oprM operon. The mexAB-oprM genes encode a tripartite multi-drug efflux system consisting of outer membrane (OprM), inner membrane (MexB) and membrane fusion (MexA) components. The product ofthe mexR gene negatively regulates the expression of mexR, mexA, mexB and oprM. MexR is a member ofthe MarR family of bacterial transcriptional regulators (Sulavik et al, 1995). The marR gene is a component and negative regulator of the marRAB locus in Escherichia coli (Alekshun and Levy, 1997). Members ofthe MarR family regulate the expression of resistance to multiple antibiotics, organic solvents, detergents and oxidative stress agents, and of pathogenic factors (Miller and Sulavik, 1996; Alekshun and Levy, 1999b). By regulating the expression of the transcriptional activator MarA, MarR is able to regulate several MarA-regulated genes involved in antimicrobial resistance. The recently determined crystal structure of MarR in complex with the effector salicylate (Alekshun et al, 2001), revealed a dimeric structure with a winged helix DNA 60 binding motif (reviewed in Gajiwala and Burley, 2000) in each MarR monomer. Two molecules of the effector salicylate, which inhibits DNA binding by MarR (Alekshun and Levy, 1999a; Alekshun et al., 2001), were bound to each subunit and located on either side ofthe proposed recognition helix in the DNA binding domain (denoted SAL-A and SAL-B). To understand the specific regulation of MexAB-OprM expression in Pseudomonas and to gain further insights into the molecular regulation of the DNA binding activity of other members of the MarR family, the crystal structure of MexR was determined. Salicylate was found to have no effect on the DNA binding by MexR. The MexR-DNA complex was also modelled using the MexR operator sequence identified by previous footprinting studies (Evans et al., 2001). Based on the structural analysis of MexR and the model of the MexR-DNA complex, an allosteric mechanism of regulation of the DNA binding activity of MexR is proposed, that likely differs from that of MarR. 4.2 Materials A C-terminally His-tagged MexR construct (vector pKLEl) was obtained from our collaborators at the laboratory of Dr. Keith Poole (Evans et al, 2001). 4.3 Methods 4.3.1 Creation of wild type untagged MexR construct From the pKLEl plasmid, a PCR product containing only the wild type mexR open reading frame (encoding residues 1 to 147 with no tags or additional residues) was amplified using the p r i m e r s : 5'CAACTC C AT A T G A A C T A C C C C G T G A A T C C and 5'TGGTGGTGCTCGAGTCAAATATCCTCAAGCGGT (the Ndel and Xhol restriction sites are 61 underlined). The PCR product was cloned into the pET41a vector (Novagen) and used for the production of native and SeMet-substituted MexR protein. The construct was sequenced and confirmed to be in agreement with the published sequence (Poole et al, 1996b). 4.3.2 Protein over-expression and purification An overnight seed culture (grown in LB with 50 pg/mL of kanamycin) was used to inoculate (with 1:100 dilution) M9 medium supplemented with 2mM MgS04, O.lmM CaCl2 and 4 g/L glycerol (Sambrook and Russel, 2001). For the production of native protein, the medium was also supplemented with 2.5 g/L of Casamino acids. Cultures were grown at 37 °C to an OD600 of-0.5, at which point IPTG was added to a final concentration of 0.5 mM for the production of native protein. Cells were shifted to room temperature and MexR expression was allowed to proceed overnight. The same purification procedure was used for both native and SeMet-labelled proteins. Cells were pelleted by centrifugation at 5000 *g for 10 minutes and frozen at -80 °C. Frozen cell pellets were resuspended in a harvest buffer consisting of 5 mM sodium phosphate buffer (pH 7), 1 mM EDTA, 10% (v/v) glycerol and 14 mM B-mercaptoethanol. The cells were sonicated on ice (3 x 10 minutes, 30% pulse, 50% power), and insoluble material was removed by ultracentrifugation at 300,000 x g for 30 minutes. The supernatant was applied on to a SP-sepharose FF column (2.5 x 10 cm) equilibrated with harvest buffer. The protein was eluted with a linear gradient of 0 to 0.2 M NaCl (Fig. 4.3.1a). Unbound MexR protein in the initial flow-through was recovered and purified with a second pass over the SP-sepharose column. Peak fractions as assessed by SDS-PAGE were pooled, diluted 5 fold with harvest buffer and applied on to a heparin-agrose (Sigma) column (2.5 x 5 cm) equilibrated with harvest buffer (Fig. 4.3.1b). MexR protein was eluted with a linear 62 gradient of 0 to 0.5 M NaCl over 100 mL. Peak fractions were pooled, concentrated to ~0.5 mL and applied on to a Sephacryl-100 (Amersham Pharmacia Biotech) gel filtration column (2 x 65 cm) equilibrated with 10 mM Tris pH 8, 150 mM NaCl (Fig. 4.3.1c). Peak fractions were pooled and concentrated to 80 to 128 mg/mL as measured by the Bradford assay. The protein was assessed to be >90% pure by SDS-PAGE. Complete substitution of methionines in SeMet-labelled protein was confirmed by electrospray mass spectrometry (16,968 kDa observed and 16,964.6 kDa predicted for native MexR, and 17,200 kDa observed and 17,199.0 kDa predicted for SeMet-substituted MexR). Approximately 40 mg of protein were obtained per litre of culture. 63 a) 66.2 45 Fig. 4.3.1: SDS-PAGE of SeMet-substituted MexR fractions, a) Fractions off SP-sepharose FF column. The last lane is a sample of the pooled fractions from the preceding pass on the SP-sepharose FF column, b) Fractions off heparin-agarose column. The numbers indicate approximate molecular masses of the molecular markers on lane 1. c) Fractions off sephacryl-100 column. 64 4.3.3 Crystal l izat ion For crystallization, protein solutions were diluted to 8.8 mg/mL with water and with the addition of tris-(carboxyethyl)phosphine hydrochloride (TCEP) to a final concentration 10 mM. For native MexR, 1 pL of protein solution was mixed with an equal volume of reservoir solution consisting of 20 mM CaCl2,65 mM NaCl, 6 mM morpholinoethanesulfonic acid (MES; pH6), 4 mM sodium acetate (pH 5) and 10 mM dithiothreitol (DTT). For SeMet-substituted MexR, 2 uL of protein solution was mixed with an equal volume of reservoir solution consisting of 46 mM CaCl2, 105 mM NaCl, 6 mM MES (pH6), 4 mM sodium acetate (pH 5) and 10 mM DTT. Crystals of both native and SeMet-labelled MexR (Fig. 4.3.2) grew after equilibration over 1 mL reservoir solution for 12 to 24 hours at 20 °C reaching dimensions of up to -0.15 * 0.15 x 0.3 mm for native crystals and 0.2 x 0.2 x 0.5 mm for SeMet-substituted crystals. Fig. 4.3.2: Crystals of SeMet-substituted MexR. The longest dimension ofthe largest of these crystals is -0.5 mm. 65 4.4 Results 4.4.1 M e x R D N A binding assays An electrophoretic mobility shift assay (EMS A) was conducted to detect DNA binding by SeMet-substituted MexR to its operator DNA with and without prior treatment with Hinc II restriction endonuclease (Fig. 4.4.1a). All samples contained -250 ng of a 28 bp DNA oligonucleotide with the sequence 5'ATTTTAGTTAGACCTTATCAACCTTGTTT (the Hinc H site is in bold with the cleavage site indicated by "A", inverted repeats are underlined) corresponding to the MexR binding site LI identified by previous footprinting studies (Evans et al., 2001). The DNA used in lanes 3 and 4 were digested with Hinc II for 1 hour and 40 minutes at 37 °C. Prior to loading on to the gel, -13 pg of SeMet-substituted MexR protein were added to samples in lanes 2 and 4. The samples were loaded on to a 20% polyacrylamide gel buffered with 1X TBE and electrophoresed at 64 V until the xylene cyanol dye front reached the bottom of the gel (Sambrook and Russel, 2001). DNA fragments in the gel were then visualized with ethidium bromide staining. 66 a) 1 2 3 4 b) 1 2 3 4 « Fig. 4.4.1: Electrophoretic mobility shift assays of DNA binding by SeMet-substituted MexR. All gels were stained with ethidium bromide, a) DNA binding to DNA oligonucleotide corresponding to operator sequence with (lanes 3 and 4) and without (lanes 1 and 2) Hinc II digestion. SeMet-substituted MexR protein was added to lanes 2 and 4. Destruction of one of the inverted repeats in the MexR operator sequence by Hinc II digestion abrogates MexR binding, b) Electrophoretic mobility shift assays of DNA binding by SeMet-substituted MexR in the absence (lanes 1 and 2) and presence (lanes 3 and 4) of 5 mM sodium salicylate in the gel and running buffer. An EMS A (Fig. 4.4.1b) was also conducted to detect DNA binding by SeMet-substituted MexR in the absence (lanes 1 and 2) and presence of 5 mM sodium salicylate in the gel and running buffer. All samples contained -250 ng of a 28 bp DNA oligonucleotide corresponding to the MexR binding site II identified by previous footprinting studies (Evans et al, 2001), while -13 pg of SeMet-substituted MexR protein were added to samples in lanes 2 and 4. The electrophoresis conditions were the same as those detailed above. 67 4.4.2 Data collection MexR crystals were harvested in buffer containing 10 mM CaCl2, 6 mM MES pH6, 4 mM sodium acetate pH 5 and 10 mM DTT. Crystals were cryo-protected by gradual transfer in ten increments to 33% (v/v) PEG 400,20 mM CaCl2,128 mM NaCl, 6 mM MES (pH6), 4 mM sodium acetate (pH 5) and 10 mM DTT and were flash frozen in a nitrogen cold stream prior to data collection. The increase in salt concentration with the addition of PEG 400 was necessary to prevent shrinkage of the unit cell to a pseudo-tetragonal cell with pseudo-hemihedral twinning. Due to their superior diffraction over native crystals, data from SeMet-substituted crystals were used for both structure determination and refinement. Multi-wavelength anomalous diffraction (MAD) data were collected at beam line 9-2 at the Stanford Synchrotron Radiation Laboratory (SSRL), while a high resolution data set was collected at beam line X8-C at the National Synchrotron Light Source (NSLS). Based on systematic absences, the space group was determined to beP2,212I. Data statistics are shown in Table 4.4.1. 68 Table 4.4.1: M e x R X-ray data statistics peak remote high resolution wavelength ( A ) 0.979 0.932 1 cell axes ( A ) a 67.9 67.9 67.9 b 72.2 72.2 72.7 c 241.3 241.4 240.5 resolution (A)a 25.0-2.6 (2.69-2.6) 25.0-2.5 (2.59-2.5) 25-2.1 (2.18-2.1) observed reflections 192,756 164,169 352,292 unique reflections 69,848 78,058 63,205 overall completeness (%)" 99.1 (97.0) 99.0 (95.9) 90.3 (60.5) / / c (7 ) f l 22.5 (6.7) 21.3 (4.7) 30.7 (3.1) average 5-factor ( A 2 ) 49 48.1 34.5 3.3 (8.6) 2.8 (11.3) 4.5 (34.0) a Values in parentheses correspond to the highest resolution shell. b Rsym = E | / a v - I\l Eij, where 7 a v is the average of all observations, Iv 4.4.3 Structure determination and refinement The structure of M e x R was determined using M A D data collected from a single SeMet-substituted crystal. Nineteen out of 40 possible selenium sites were located using anomalous differences from the peak wavelength by the automated Patterson search routine in C N S (Briinger et al., 1998; Terwilliger and Berendzen, 1999). Five additional sites were located by log-likelihood gradient maps after refinement and phasing with the initial sites. A n overall figure of merit of 0.48 was obtained with C N S , which was improved to 0.59 after solvent flattening with R E S O L V E (Terwilliger, 2000). This resulted in a sufficiently interpretable electron density map to allow C a tracing of one of the four M e x R dimers in the asymmetric unit. This initial trace was manually 69 positioned into the density regions for the three remaining dimers, which were then used to derive non-crystallographic symmetry (NCS) operators. A significantly improved electron density map (figure of merit of 0.73) was obtained after NCS averaging and phase extension to 2.15 A resolution with the program DM (Collaborative Computational Project, 1994), from which a nearly complete model ofthe asymmetric unit was built. Due to the asymmetry of the MexR dimer, four-fold multi-domain averaging was done with the two monomers in each dimer treated as separate domains. Structure refinement statistics are shown in Table 4.4.2. Only one residue is in the disallowed region (Gin 90 in chain D), which is located in a region of poor electron density. Interhelical angles and distances were measured using the program PNTERHLX (Yap, 1998). Table 4.4.2: SeMet-substituted MexR structure refinement statistics *crys«/-Kfree 0.242/0.294 no. of non-H atoms (protein / water) 8,924 / 214 rmsd from ideality bond lengths ( A ) 0.0060 angles (°) 1.14 angle improper (°) 0.89 average 5-factor ( A 2 ) main chain 38.1 side chain 44.3 water 36.0 overall 42.1 Luzzati coordinate error ( A ) 0.3 resolution range ( A ) 5.0 - 2.1 Ramachandran plot (% of residues in region) most favourable 94.0 allowed 5.5 generously allowed 0.4 disallowed" 0.1 a Gin 90 (molecule D) is in a region of poor electron density. 70 4.5 Discussion 4.5.1 Overa l l structure The overall structure of M e x R (Fig. 4.5.1a) is predominantly a-helical and resembles that of M a r R (Alekshun et al, 2001). The M e x R dimer is triangular in shape and contains two winged helix D N A binding domains (residues 36 to 97 of each monomer), each connected via two long helices ( a l and a5) to a dimerization domain made up of the N - and C-terminal regions (residues 5 to 17 and 120 to 139) from the two monomers (Fig. 4.5.1a). Hydrophobic residues from a l , a5 and a6 are buried at the dimerization interface (Fig. 4.5.1b). The total buried surface area ranges from 4360 to 4930 A 2 for the four M e x R dimers in the asymmetric unit. A number of P. aeruginosa strains with increased resistance to the antibiotic nalidixic acid {nalB mutants), contain mutations in the mexR gene, which were found to disrupt dimerization, D N A binding or possibly protein stability (Adewoye et al, 2002). Based on the M e x R crystal structure, these mutations can be mapped to dimerization or D N A binding domains, or to the interdomain regions. The electrostatic interactions observed between the D N A binding domains of M a r R (Asp 67 of one monomer with A r g 73' from the other monomer and the reciprocal pair; Alekshun et al, 2001) are not present in M e x R . 71 Fig. 4.5.1: Overall structure ofthe MexR dimer. a) MexR dimer in ribbon representation. The secondary structure elements are labelled and coloured individually in the monomer on the right. The winged helix domain (coloured blue in the monomer on the left) consists of a 2 (H1) -p1 ( S 1 ) - a 3 ( H 2 ) - a 4 ( H 3 , recognition helix)-p2 ( S 2 ) - W 1 (wing)-p3 ( S 3 ) , where the terminology used by Gajiwala and Burley ( 2000 ) are given in brackets. The N- and C-terminii are labelled N and C, respectively, b) MexR dimer shown in similar orientation as in a) and with a molecular surface representation of one subunit highlighting the hydrophobic dimerization interface. Low, medium and high hydrophobic potentials are coloured grey, yellow and green, respectively. The second subunit is shown in ribbon respresentation (magenta). 72 4.5.2 Conformational flexibility of the M e x R dimer The molecular packing in the MexR crystal has provided eight independent observations of the MexR monomer structure. A significant degree of flexibility is seen in the MexR structure as evidenced by the large deviations between the eight copies of the MexR monomer in the asymmetric unit, with overall rmsd of 0.7 to 2.1 A relative to monomer A for 512 pairs of main chain atoms (Fig. 4.5.2). The regions encompassing helices al, a5 and a6 (residues 4 to 32 and 99 to 139) show main chain rmsds of 0.4 to 1.3 A relative to monomer A, with the largest differences seen between monomers A and B. Flexibility at loop regions (residues 33 to 34, 97 to 98 and 121 to 123) allows helix al to deviate by 17.0° and 4.9 A , a5 by 6.8° and 1.7 A and a6 by 12.1 ° and 8.2 A in helix orientation and midpoint position, respectively. In contrast to MarR, clear main chain density is observed for a5, including Ser 113, which corresponds to the poorly ordered region of Gly 116 in MarR (a proposed flexible hinge region; Alekshun et al, 2001). Fig. 4.5.2: Overlap of the Ca traces of the eight monomers in the MexR asymmetric unit. The monomers were superimposed using the main chain atoms of their winged helix domains. Chains A and B are highlighted with thick rendering in blue and red, respectively. Flexible loop regions allow for significant deviations in the orientations of helices a1, a5 and a6. 73 In addition to the conformational differences amongst individual monomers, there were significant variations in the relative disposition of the 2 monomers at the dimerization interface in each of the four independent dimers in the asymmetric unit. These positional variations in the MexR dimer gives rise to major differences in the disposition of the two DNA binding domains relative to each other (Fig. 4.5.3). The inter-helical interactions at the dimerization interface primarily consist of van der Waals contacts between hydrophobic side chains, which unlike hydrogen-bonds and salt bridges, do not require specific geometric arrangements for the interacting atoms and therefore could accommodate significant reorientation of the dimerization helices relative to each other. With the exception of the Wl region, the winged-helix domain moves as a relatively rigid and well ordered entity. There is little deviation between the eight copies in the asymmetric unit (rmsds of 0.3 to 0.6 A relative to chain A for main chain atoms in residues 33 to 84 and 93 to 98). Superposition of the four observed MexR dimers using main chain atoms from one DNA binding domain per dimer (residues 33 to 84 and 93 to 98) shows a maximum helix midpoint displacement of 7.7 A (between dimers AB and CD) for the recognition helix a4 in the other DNA binding domain. Fig. 4.5.3: Overlap of the MexR dimers AB (blue) and CD (red). The dimers were superimposed using the main chain atoms of the winged helix domains on the left. The recognition helices (a4) for chains B and C are highlighted with thicker rendering in cyan and orange, respectively. Flexibility in the dimerization domain allows significant movement of the DNA binding domains relative to each other. The Ca-Ca distances for Arg 73 and Arg 73' shown for the two dimers give an indication of the difference in the spacing of the DNA binding domains. 74 4.5.3 Molecular model of the MexR-DNA complex The MexR crystal structure and the MexR operator sequences identified by previous footprinting studies (Evans et al, 2001) provided geometric constraints to guide the modelling of the MexR-DNA complex. Both the MexR dimer and the operator DNA sequence contain a two-fold rotation axis relating the two MexR monomers and the inverted repeats, respectively, and these two axes are expected to coincide in the MexR-DNA complex. The inverted repeats are each five base pairs in length ( 5'GTTGA) and are spaced five base pairs apart. Assuming that the MexR operator DNA adopts a canonical linear B-DNA conformation with 10.5 bp per turn, the major grooves of the inverted repeats would be positioned on the same side of the DNA, to which MexR is expected to bind. Taking these geometric constraints into consideration, MexR was manually docked on to an ideal model of the operator DNA (5'AAATGTGGTTGATCCAGTCAACTATTTTG and its complement with inverted repeats underlined) produced with HyperChem (Hypercube, 1995). The MexR dimer was oriented such that the DNA binding domains faced the major grooves of the inverted repeats. Of the four MexR dimers in the asymmetric unit, dimer CD shows the largest spacing between the DNA binding domains with a Ca-Coc distance of 29.2 A between Arg 73 and Arg 73' (from helix ct4), which is close to the 34 A spacing between major grooves in linear B-DNA. Manual docking of dimer CD on to the operator DNA produced a model with a reasonable match of the spacing between the recognition helices (oc4 and cc4') with the spacing between the major grooves in the inverted repeats (Fig. 4.5.4a). Given the observed conformational adaptability of the dimerization interface, it is likely that in solution the spacing between the recognition helices may be further increased to provide an even closer match with the major groove spacing on the operator DNA. In our current model, the wings (Wl) are well positioned to make minor groove or phosphate 75 backbone contacts, presumably allowing for increased affinity in the interaction. Loss of one of the two inverted repeats on a 28 bp oligonucleotide corresponding to the MexR binding site II (Evans et al, 2001) by Hinc II digestion prevents binding by MexR (Fig. 4.4.1a) and is consistent with a single MexR dimer binding to both inverted repeats simultaneously. The predicted interactions of the recognition helices and wings with the major and minor grooves ofthe operator DNA for MexR would be similar to those observed in the CAP-DNA complex (Schultz et al, 1991; Parkinson et al, 1996). Fig. 4.5.4: Proposed mechanism of regulation of MexR. a) molecular modelling of MexR-DNA complex. The MexR dimer CD (ribbon representation) depicts the "open" or DNA-bound conformation. The DNA molecule is shown in stick representation, with the inverted repeats coloured green and highlighted by thick rendering. Residues on MexR which correspond to the MarR residues observed to be in contact with salicylate (Alekshun era/., 2001) are shown in stick rendering, b) insertion ofthe C-terminal tail (residues 140-147 shown with magenta carbons) from chain C in between the DNA binding domains results in a "closed" conformation depicted by dimer AB (ribbon representation), in which the reduced spacing between the DNA binding domains is incompatible with the spacing ofthe inverted repeats (green with thick rendering) of the operator. SigmaA-weighted 2F0-FC electron density is contoured at 1 a to 2.1 A resolution around the C-terminal tail. 76 4.5.4 Proposed allosteric mechanism of regulation of M e x R An analysis of the interactions of the MexR C-terminal tail region with a neighbouring MexR dimer in the crystal structure suggests that different conformations of the MexR dimer observed in the crystal structure may resemble each ofthe distinct DNA- and effector-bound conformations. The C-terminal tail immediately follows a6 and consists of residues 140 to 147. While the C-terminal tail is disordered in most ofthe MexR monomers, interpretable electron density is visible for the C-terminal tail of monomer C (average 5-factor of 38.7 A 2 for residues 140-147), which is inserted in between the DNA binding domains of dimer AB (Fig. 4.5.4c). Interestingly, dimer AB exhibits the shortest spacing between its DNA binding domains (Ca-Ca distance of 22.6 A between Arg 73 and Arg 73'). In contrast, dimer CD exhibits the largest spacing (Ca-Ca distance of 29.2 A between Arg 73 and Arg 73') between its DNA binding domains and shows no density for the C-terminal tail in the region between its DNA binding domains (Fig. 4.5.3). While dimer CD can be manually docked on to a linear model ofthe MexR operator DNA, the spacing ofthe recognition helices in dimer AB is incompatible with DNA binding as predicted by our model (Fig. 4.5.4a,b), so that the conformations shown by dimers CD and AB can be interpreted as open (able to bind DNA) and closed (unable to bind DNA) states, respectively. The observation of an ordered C-terminal tail in between the DNA binding domains in the closed conformation (but not in the open conformation) suggests that binding ofthe C-terminal tail or a ligand resembling this peptide can inhibit DNA binding by inducing the closed conformation. In the absence of an effector or an effector-like ligand, repulsion between positively charged side chains (Arg 21, His 41, Arg 63 and Arg 70) lining the crevice between the DNA binding domains likely help to maintain the MexR dimer in the DNA binding conformation (Fig. 4.5.5a,b). These charges are partially neutralized by Glu 145, Asp 146 and the C-terminal carboxylate on the C-77 terminal tail. Presumably the effector ligand will also have these acidic characteristics. Additionally Pro 143, Leu 144 and He 147 from the C-terminal tail make a number of van der Waals contacts with hydrophobic side chains from dimer AB. Peptide binding at a homodimeric interface was also observed in the crystal structure ofthe natriuretic peptide receptor dimer (human NPR-C) in complex with a 22-amino acid natriuretic peptide (CNP), in which peptide binding induced a 20 A closure of the membrane-proximal domains that interact with the peptide (He et al, 2001). Regulation of DNA binding activity by altering the spacing of the DNA binding domains in dimeric DNA binding proteins has also been observed in the tetracycline repressor (Orth et al, 2000) and the FadR repressor (van Aalten et al., 2001). However, the mechanism proposed here for MexR is unique in that each MexR homodimer binds one effector molecule in contrast to the situation of one dimer binding two effector molecules observed in these other repressor systems. 78 Fig. 4.5.5: Binding of MexR to C-terminal tail from neighbouring molecule, a) Close-up view of the interactions between the C-terminal tail of monomer C (magenta) with side chains on monomer A. b) Close-up view of the interactions between the C-terminal tail of monomer C (magenta) with the side chains on monomer B. Water molecules are shown as red spheres. An examination of the crystal packing of MexR predicts that inter-dimer interactions between the MexR C-terminal tail and DNA binding domains lead to the formation of MexR filaments, which is inconsistent with the high solubility of this protein. The interaction of the DNA binding domains with the C-terminal tail therefore likely occurs only under crystallization conditions, and the affinity is likely to be weak in solution, so that the C-terminal tail itself is unlikely to be the in vivo effector. However in the context of the MexR conformations observed in the crystal structure and the 79 proposed model of the MexR-DNA complex, the C-terminal tail has identified a potential effector binding site on MexR, and a conformation of the MexR dimer that is expected to be unable to bind DNA is apparently linked to interaction with the C-terminal tail. 4.5.5 Contrast with MarR An overall structure-based alignment of MexR (monomer D) with the MarR monomer gives an rmsd of 1.8 A for 116 Ca positions. This relatively large overall rmsd mostly results from conformational differences in the dimerization helices (al, a5 and a6), since a superposition of the DNA binding domains gives an rmsd of 1.2 A (for 256 main chain atoms). The multiple conformations observed in the crystal structure of MexR contrast with the situation in MarR, in which the dimer two-fold rotation axis coincides with a crystallographic two-fold axis, so that the MarR asymmetric unit contains only one monomer (Alekshun et al, 2001). The conformational flexibility seen in MexR is therefore not observed in MarR and may be accounted for in part by the salt bridge interactions, which tether the DNA binding domains of MarR, and which are not conserved in MexR. The single dimer conformation observed for MarR differs drastically from the observed MexR dimer conformations. A superposition of the MexR (CD) and MarR dimers using the main chain atoms from one DNA binding domain for each dimer shows that the second recognition helix (a4') of MarR is displaced 18.8 A relative to that of MexR (Fig. 4.5.6a). Fig. 4.5.6: Comparison of the MexR and MarR dimers. MexR (orange) and MarR (light blue) were superimposed using the main chain atoms of one DNA binding domain from each dimer. The recognition helices a4 and a4' are highlighted with thicker rendering and coloured red in MexR and dark blue in MarR. 80 Given the unusual "zigzag" arrangement of the recognition helices in MarR and the expected positions of the inverted repeats being on opposite faces of the MarR operator DNA (Martin et al, 1995), the mode of DNA binding by MarR is unclear and does not appear to resemble our proposed model of DNA binding for MexR. The salicylate binding sites in the MarR-salicylate complex were located on either side of the recognition helix (Fig. 4.5.7b; Alekshun et al, 2001). The SAL-A site is located in a crevice between the recognition helix (a4) and the |3-sheet (from which Wl loops out), while the SAL-B site is located on the opposite side of a4. In the absence of an apo MarR structure, the effect of salicylate binding to MarR on the dimer conformation (if any) is unclear, however, the observed positions of the SAL-A and SAL-B sites are distant from and seem unlikely to affect the conformation of the dimerization domain. Of the two salicylate binding sites observed in the MarR crystal structure, only for SAL-A are the structural characteristics similar for the corresponding region in MexR, while for SAL-B the binding of salicylate as observed in MarR would not be possible in MexR. However, sodium salicylate at a concentration of 5 mM is unable to dissociate Fig. 4.5.7: Comparison of the MarR salicylate binding sites with the corresponding regions in MexR. The putative salicylate binding sites (SAL-A and SAL-B) observed in the MarR crystal structure are indicated by the magenta salicylate molecules. The colouring scheme of the ribbon representation is as in Fig. 4.5.6. The MexR side residues are labelled in red, while the MarR residues are labelled in blue. 81 the MexR-DNA complex as assessed by an electrophoretic mobility shift assay (Fig. 4.4. lb), which directly indicates differences in effector specificity between MexR and MarR. The results of this study indicate that the effector binding and mode of regulation between MexR and MarR are very different despite clear similarities in sequence and structure. This raises the possibility that members of the MarR family of transcriptional regulators may be diverse in the types of effector molecules recognized and in the ways their DNA binding activity is regulated. The inability of salicylate to inhibit DNA binding by MexR and the interactions of the C-terminal tail with the DNA binding domains of the MexR dimer suggest that the effector for MexR may be a peptide signalling molecule or the C-terminus of a protein ligand. Further work is needed to identify the in vivo effector(s) for MexR, which will also shed light on the physiological function of the MexAB-OprM multidrug efflux system regulated by MexR. 82 Chapter 5: PBP2a from MRSA 5.1 Introduction Staphylococcus aureus is a major cause of hospital and community-acquired infections worldwide. Special strains known as methicillin-resistant Staphylococcus aureus (MRSA) are of particular concern due to their resistance to methicillin and all other clinically used P-lactam antibiotics (reviewed in Hiramatsu, 1995; Chambers, 1997; Hiramatsu et al., 2001). p-lactam resistance in S. aureus appeared with the introduction of penicillin in the 1940s, which was overcome by strains producing penicillinases, leading to the introduction in 1960 of methicillin, a semi-synthetic penicillin derivative resistant to hydrolysis by penicillinases (Rolinson et al, 1960). However, resistance developed quickly with the isolation of an MRSA strain in England just one year after the introduction of methicillin (Jevons, 1961). In the following decades MRSA strains have progressively spread to hospitals worldwide and have become a major health concern due to their high prevalence. MRSA strains account for over 60% of S. aureus clinical isolates in Japan, Singapore and Taiwan, over 50% in Italy and Portugal, and 34% in the United States (Diekema et al., 2001). More recently there has also been an increasing prevalence of methicillin resistance in S. aureus isolates from community-acquired infections, which hints at the establishment of MRSA strains as part of our normal flora by replacement of methicillin-susceptible strains (reviewed Chambers, 2001; Hiramatsu et al, 2002). Minimum inhibitory concentrations (MICs) of methicillin for susceptible staphylococci are in the range of 2.5 pg/mL or lower (Rolinson et al., 1960). Different levels of methicillin resistance resulting from multiple mechanisms have emerged in staphylococci. S. aureus strains that hyper-83 express penicillinase show resistance against methicillin (MIC of 4 pg/mL) that is borderline between susceptible and resistant strains (McDougal and Thornsberry, 1986). Borderline resistance is also conferred by alterations or over-expression of the native penicillin-binding proteins (PBPs). S. aureus normally produces four P-lactam-sensitive PBPs (Georgopapadakou and Liu, 1980; Ghuysen et al, 1986; Kuroda et al, 2001), consisting of 2 high molecular mass (HMM) class B PBPs (PBP 1 and 3), one HMM class A PBP (PBP 2), and one low molecular mass PBP (PBP 4). Decreased penicillin-binding affinities of PBPs 1 and 2, and overproduction of PBPs 2 and 4 have been reported in strains showing MICs of up to 16 pg/mL (Tomasz et al, 1989; Hiramatsu et al, 1992b; Suzuki et al., 1992; Chambers et al., 1994; Hackbarth et al., 1995; Henze and Berger-Bachi, 1995). Higher levels of methicillin resistance are shown by MRSA strains, which produce an additional 78 kDa HMM class B PBP designated PBP2a (PBP2' in Europe and Asia; Hartman and Tomasz, 1984; Utsui and Yokota, 1985; Reynolds and Fuller, 1986), a novel PBP distinct from any of the four native PBPs found in S. aureus (<21% sequence identity; Kuroda et al, 2001). PBP2a is encoded by the mecA gene (Matsuhashi et al, 1986; Matthews et al, 1987), which is highly conserved between MRSA strains (>90% sequence identity; Ryffel et al., 1990) and is located within a unique mobile genetic element, the staphylococcal cassette chromosome mec (SCCmec; Ito et al, 1999; Katayama et al, 2000). The SCCmec is a large DNA cassette (30 to 52 kb) integrated into the genome near the origin of replication and is believed to have been acquired via a horizontal transfer from an as yet unidentified staphylococcal species. The mec locus SCCmec also contains the mecR and mecl genes that encode regulatory elements that are functionally equivalent to the BlaR and Blal sensor-regulatory system that allow P-lactam-induced derepression of P-lactamase expression (Hiramatsu et al, 1992a; Hardt et al, 1997; Zhang et al, 2001). Due to the poor inducibility of the 84 MecR sensor by methicillin and other P-lactams, derepression in highly resistant strains is achieved via mutations or deletions in the mecA operator, meclox possibly other regulatory regions (reviewed in Hiramatsu et al, 2001). Based on its primary sequence, PBP2a belongs to subclass BI, and all characterized members of this group exhibit unusually low P-lactam affinities (reviewed in Goffm and Ghuysen, 1998). PBP2a is believed to act as a surrogate transpeptidase by remaining active at P-lactam concentrations that inactivate all ofthe native PBPs in S. aureus, thus sustaining cell wall synthesis at otherwise lethal P-lactam concentrations (reviewed in Hiramatsu, 1995; Chambers, 1997; Hiramatsu et al ,2001). MIC values for methicillin >2000 pg/mL for high level MRSA strains have been reported (Kondo et al, 2001). In addition to mecA, the SCCmec often contains additional resistance determinants, such that the SCCmec has been referred to as a "resistance island" (analogous to pathogenicity islands in Gram-negative bacteria; Ito et al, 1999 and reviewed in Hiramatsu et al, 2001). Consequently MRSA strains are often multidrug-resistant, with vancomycin being the drug of choice against MRSA infections. However an MRSA strain with low level vancomycin resistance was isolated in Japan in 1996 (MIC of 8 pg/mL; Hiramatsu et al, 1997), which was followed by similar strains isolated in other countries (reviewed in Hiramatsu et al, 2001; Walsh and Howe, 2002). More recently MRSA strains with high level vancomycin resistance (MIC > 32 pg/mL) were isolated in the United States (CDC, 2002b; CDC, 2002a). Both isolates contained the resistance element vanA, which was likely acquired from vancomycin-resistant enterococci. The acquisition of vancomycin resistance by MRSA underlines an urgent need for the development of novel antibiotics against MRSA. The crystal structure of PBP2a presented here reveals structural features responsible for its P-lactam resistance and provides important insights for the design of novel antibiotics against MRSA. 85 5.2 Materials S. aureus PBP2a (SauPBP2a) contains an N-terminal transmembrane anchor, which can be removed without affecting the B-lactam binding kinetics (Wu et al, 1992; Lu et al, 1999). Soluble derivatives of PBP2a were constructed by removal of the transmembrane segment and used for structure determination. A truncated construct of SauPBP2a derived from MRSA strain 27R (plasmid pEWS A3 7) was provided by our collaborators at Eli Lilly (Wu et al., 1992). This construct (designated SauPBP2aNA22+2) was created by removal of the N-terminal 22 residues with the addition of two residues to the N-terminus (Met and Val). The ORE for this construct was transferred into pET15b by Liza De Castro using Ncol and BamHI restriction sites. Methicillin was kindly provided by Manjeet Bains from Dr. Robert Hancock's laboratory. 5.3 Methods 5.3.1 Creation of a further truncated SauPBP2a construct and acylation-deficient mutant A construct (designated SauPBP2a*) was created by deletion of residues 2 to 23. A similar SauPBP2a construct from strain MI339 was shown to retain its penicillin-binding characteristics (Lu et al., 1999). The PCR product amplified from the SauPBP2aNA22+2-pET 15b construct using the primers: 5'GGGTTTGGTATATATTCCATGGCTTCAAAAGATAAAG (N-terminal primer with Ncol site underlined) and 5'GGCGTCGGATCCTTATTCATCTATATCGTATTTTTTATTAC (C-terminal primer with BamHI site underlined) was cloned into pET 15b. An acylation-deficient mutant of the SauPBP2a* construct was also created using a protocol based on the megaprimer method (Kammann et al, 1989; Sarkar and Sommer, 1990), in which the active site Ser 403 was mutated to Ala. In the first round of PCRs, an N-terminal fragment was amplified using the N-terminal primer 86 and 5 'CTTCACCAGGTGCAACTCAAAAA, and a C-terminal fragment was amplified using the C-terminal primer and 5'TTTTTGAGTTGC ACCTGGTGAAG. The two products overlapped in the region corresponding to residue 403, such that the S403A mutation was introduced by the primers. These two PCR products were isolated and used as template for a subsequent PCR using only the N-terminal and C-terminal primers to amplify the full length S403A mutant ORF product. Both constructs were sequenced and confirmed to be in agreement with published information on the gene and amino acid sequences (Wu et al, 1992; Sun et al, 1998). 5.3.2 Measurement of protein concentration Protein concentrations were determined by measuring A 2 8 0 in 6 M guanidine hydrochloride and using an extinction coefficient of 1.162 and 1.149 cm"1 for a 1 mg/mL solution of SauPBP2aNA22+2 or SauPBP2a*, respectively (Expasy ProtParam web page, http://us.expasy.org/tools/protparam.html; Gill and von Hippel, 1989). 5.3.3 Over-expression, refolding and purification of SauPBP2aNA22+2 and SeMet-substituted SauPBP2a* Native and SeMet-substituted SauPBP2aNA22+2 and SeMet-substituted SauPBP2a* proteins were expressed as inclusion bodies, refolded and purified using conditions based on previously established protocols (Wu et al, 1992; Lu et al, 1999). In each case, an overnight seed culture was grown in 20 mL LB (with 100 ug/L ampicillin) and used to inoculate 2 L of either LB (for native protein) or M9 medium (for SeMet-substituted proteins). Cells were grown to an OD600 of ~0.5 at 37 °C and induced with 1 mM IPTG. Following an induction period of 2.5 hours at 37 °C, the cells were harvested by centrifugation at 5000 xg for 10 minutes and freezing ofthe cell pellets at -70 °C. 87 For SeMet-substituted proteins, 10 mM P-mercaptoethanol was included in the buffers for the cell lysis, washing and refolding steps but not for gel filtration. The cell pellets were resuspended in 40 mL of 0.1 M Tris pH8 and 10 mM EDTA and lysed by sonication on ice (3x10 minutes, 30% pulse, 50% power). Inclusion bodies (and other insoluble material) were pelleted by ultracentrifugation at 300,000 xg for 30 minutes. The inclusion bodies were washed five times by resuspending in 20 mL of 25% sucrose, 1 % triton X-100 and 5 mM EDTA and centrifugation at 16,000 rpm in a Beckman JA-20 rotor for 20 minutes. The inclusion bodies were then washed once with 20 mL of 1 M urea and 10 mM Tris pH 7. The inclusion bodies were solubilized in 15 mL of 5 M guanidine-HCl, 50 mM Tris pH 8.0, 0.5 M NaCl and 0.01% thiodiglycol with gentle agitation at room temperature overnight. Insoluble material was removed by centrifugation at 16,000 rpm in a Beckman JA-20 rotor for 30 minutes. The denatured protein in the supernatant was refolded at 4 °C by addition of the sample (at a flow rate of 0.5 mL/min with a peristaltic pump) to a slowly stirring solution (1 L) of 50 mM NaHC03 pH 8, 0.5 M NaCl, 20% glycerol and 0.1 % thiodiglycol. The sample was left stirring for five hours, at which point 200 g of (NH4)2S04 were gradually added and dissolved. The sample was then applied at a flow rate of 1 to 2 mL/min to a 1.5 x 10 cm phenyl-sepharose column equilibrated with 1.5 M (NH4)2S04 and 10 mM NaHC03 pH 8. The protein was then eluted with a gradient of 1.5 M (NH4)2S04 to 0.1 M NaCl over 100 mL at 1 mL/min. Peak fractions were analysed by SDS-PAGE (Fig. 5.3.1a), pooled and concentrated to 0.5 mL. Particulate matter was removed by centrifugation at 14,000 rpm in an Eppendorf microfuge for 5 minutes. The sample was applied at 1 mL/min to a 2 x 65 cm sephacryl-100 gel filtration column equilibrated with 5 mM NaHC03 pH 8,150 mM NaCl and 0.1% thiodiglycol. Peak fractions were analysed by SDS-PAGE (Fig. 5.3.1b), pooled and concentrated to 0.5 mL or less. Loss of the N-terminal methionine and complete substitution of methionines in SeMet-substituted SauPBP2a* were confirmed by ESMS (-70% 88 73,823 and -30% 73,956 kDa observed and 73,932.6 kDa predicted for native SauPBP2aNA22+2, and 74,300 kDa observed and 74,480.6 kDa predicted for SeMet-substituted SauPBP2a*). Approximately 15 mg of protein were obtained per litre of culture, a) b) *»97 4 1 97.4 — •66.2 66 2 .. |f» 45 «g §31 31 w |21.5 —14.4 21.5 „ 14.4 c) d) 97.4-66.2 45 31-21.5-14.4"" e) 97.4 « 66.2 45 31 21.5 14.4— 97.4*,, 66.2 45 31H, 21.5 s14.4 Fig. 5.3.1: SDS-PAGE of SauPBP2a* fractions. Refolded SeMet-substituted SauPBP2a* fractions off: a) phenyl-sepharose column and b) sephacryl-100 column. Native SauPBP2a* fractions off c) Bio-Rex 70 carboxy column, d) hydroxylapatite column and e) sephacryl-100 column. The numbers indicate approximate molecular masses of the molecular markers (lane 1 in a) and lane 2 in the other gels). 89 5.3.4 Over-expression and purification of soluble SauPBP2a* A n overnight seed culture was grown in 10 m L L B (with 100 pg/L ampicillin), washed and used to inoculate 1 L of L B . Cultures were grown at 37 °C to an O D 6 0 0 o f ~0.5, heat shocked at 42 °C for 30 minutes and subsequently cooled on ice. I P T G was then added to a final concentration of 1 m M , and expression of SauPBP2a* was allowed to proceed overnight (-16 hours) at room temperature. The cells were harvested by centrifugation at 5000 xg for 10 minutes, resuspended in 40 m L of 50 m M K 2 H P 0 4 p H 8 and 1 m M E D T A and lysed by sonication (3 x 10 minutes, 30% pulse, 50% power). The lysate was ultracentrifuged at 300,000 xg for 33 minutes, and the supernatant was passed through a Q-sepharose F F column (2.5 x 2 cm) equilibrated with 50 m M K 2 H P 0 4 p H 8 at 1 mL/min. The majority of the SauPBP2a* protein was in the flow through (-50 mL) , which was dialysed against 500 m L of 50 m M K 2 H P 0 4 p H 7 overnight. The sample was then centrifuged at 16,000 rpm in a Beckman JA-20 rotor for 30 minutes and applied to a Bio-Rex 70 carboxy column (2.5 x 10 cm) equilibrated with 50 m M K 2 H P 0 4 p H 7. SauPBP2a* was eluted with a 0 to 1 M N a C l gradient over 100 m L at a flow rate of 1 mL/min . Peak fractions of the highest purity as assessed by S D S - P A G E (Fig. 5.3.1c) were pooled (-30 mL) and dialyzed against 500 m L of 50 m M K 2 H P 0 4 p H 7 overnight. Impurities that precipitated out were again removed by centrifugation at 16,000 rpm in a Beckman JA-20 rotor for 30 minutes. The supernatant was applied to a hydroxylapatite column (2.5 x 5 cm) and eluted with a 0 to 1.6 M KC1 gradient over 100 m L at a flow rate of 2 mL/min. Peak fractions were analysed by S D S - P A G E (Fig. 5.3.Id), pooled and concentrated to -0.5 m L . The sample was then centrifuged at 14,000 rpm in an Eppendorf microfuge and applied on to a sephacryl-100 gel filtration column (2 x 65 cm) equilibrated with 150 m M N a C l , 5 m M N a H C 0 3 p H 8. Peak fractions were analysed by S D S - P A G E (Fig. 5.3.le), pooled, concentrated to 80 to 160 mg/mL and rapidly frozen in liquid nitrogen in 50 u L aliquots. Loss of the 90 N-terminal methionine was confirmed by ESMS (73,552 kDa observed and 73683.3 kDa predicted). Approximately 40 mg of enzyme were obtained per litre of culture. 5.3.5 Crystallization Crystallization trials for SauPBP2a began with the SauPBP2aNA22+2 construct using protein solutions diluted to 15-30 mg/mL with 0.1 M NaCl. A large number of crystallization trials (consisting of over 3000 conditions) were screened, from which SauPBP2aNA22+2 was found to crystallize only in a C-centred Fig. 5.3.2: Crystal of SauPBP2a* (NA22+2 construct). The longest orthorhombic form (a = 108.9 A, b = 331.2 A, c dimension is -~1 mm. = 166.3 A). Initial crystals (Fig. 5.3.2) were grown in 28% PEG 400, 14 mM CdCl2, 0.65 M NaCl, 10 uM CuS04 and 0.1 M HEPES pH 7. Large single crystals grew overnight under these conditions but were of poor diffraction quality. Extensive additive screening, crystallization at 4 °C, crystallization underneath a 500 MHz NMR magnet, use of different cryoprotectants and room temperature capillary mounting failed to improve the diffraction quality of these crystals. Crystals were also grown (over 1-2 weeks) at -20 °C in 40% ethylene glycol, 6.4 mM CdCl2, 0.33 M NaCl and 2 uM CuS04, however no improvement in diffraction quality was achieved. Somewhat different crystallization conditions were subsequently found (56 mM CdCl2, 0.3 M NaCl and 10 mM HEPES pH 7). However despite their excellent external appearance, crystals grown under these conditions showed no improvement in diffraction quality. Crystallization in gels using ultra-low melting agarose (Sigma) was also attempted but failed to improve diffraction quality. 91 Due to the inability to obtain useable crystals of the SauPBP2aNA22+2 construct, crystallization of a more truncated construct (SauPBP2a*) was attempted. A key ingredient in the crystallization conditions ofthe SauPBP2aNA22+2 construct was CdCl2. Likewise SauPBP2a* was also crystallized with CdCl2 but required somewhat different conditions. SauPBP2a* crystals were grown at 18 °C using 0.5 pL of protein solution and 0.5 uL of well solution per drop. SeMet-substituted trigonal crystals were grown using 20 mg/mL protein solution (containing 0.1 M NaCl) and well solution consisting of 20% (v/v) PEG 550 MME, 0.95 M NaCl, 100 mM HEPES pH 7 and 16 mM CdCl2. Native trigonal crystals (Fig. 5.3.3a) were grown using 40 mg/mL protein solution (containing 0.1 M NaCl and 0.2 M ammonium acetate) and well solution consisting of 20% (v/v) PEG 550 MME, 0.96 M NaCl, 100 mM HEPES pH 7 and 24 mM CdCl2. Native primitive orthorhombic crystals (Fig. 5.3.3b) were grown using 20 mg/mL protein solution and well solution consisting of 20% (v/v) PEG 550 MME, 0.88 M NaCl, 100 mM HEPES; pH 7 and 16 mM CdCl2. Both the trigonal and primitive orthorhombic crystal forms of the SauPBP2a* construct showed superior diffraction to all of the crystals obtained for the SauPBP2aNA22+2 construct, and hereafter all crystals of SauPBP2a will refer to those of the SauPBP2a* construct. Primitive orthorhombic crystals of the S403A SauPBP2a* were grown under the same conditions as for the wild type protein. 92 Fig. 5.3.3: Crystals of native SauPBP2a*. a) Crystal of the trigonal form. The longest dimension is -0.6 mm. b) Crystals of the primitive orthorhombic form. The longest dimension of a single crystal was -0.5 mm. 5.3.6 Cryoprotection and P-lactam soaking Prior to data collection, trigonal crystals were cryoprotected by transfer in two steps to 22% (v/v) PEG 400, 0.7 M NaCl, 100 mM HEPES pH 7, 16 mM CdCl2, 10% (w/v) glucose and 0.1% (v/v) thiodiglycol, while primitive orthorhombic crystals were cryoprotected by transfer in two steps to 28% (v/v) PEG 550 MME, 1 M NaCl, 100 mM HEPES; pH 7, and 16 mM CdCl2. Nitrocefin (20 mM) was soaked into primitive orthorhombic crystals of SauPBP2a* in artificial mother liquor for -20 hours with subsequent cryoprotection. Penicillin G (20 mM) and methicillin (40 mM) were soaked into primitive orthorhombic crystals of SauPBP2a* in cryoprotectant solution for 19 to 21 hours. Primitive orthorhombic crystals of S403A SauPBP2a* were soaked in cryoprotectant solution containing nitrocefin (40 and 200 mM with some precipitation), penicillin G (100 and 400 mM) and methicillin (100 mM) for 0.5 to 3 hours, however in all cases no electron density could be seen for the (3-lactams in the active site region. Co-crystallization with P-lactams was unsuccessful for either the SauPBP2a* (no crystals) or S403A mutant (no P-lactam electron density in the active site). 93 5.4 Results 5.4.1 Data collection M A D data from a single SeMet-substituted trigonal crystal, low resolution data from a native trigonal crystal, higher resolution data from a SeMet-substituted trigonal crystal, and high resolution data from native apo and nitrocefin-soaked primitive orthorhombic crystals were collected at beam line X 8 - C at the National Synchrotron Light Source (NSLS) at the Brookhaven National Laboratory ( B N L ) . Data from the penicillin G and methicillin-soaked crystals were collected at the home source on a mar345 detector. A l l data were collected at 100 K . The data statistics are shown in Table 5.4.1. Table 5.4.1: SauPBP2a X-ray data statistics apo SeMet higher resolution SeMet apo native nitrocefin soaked native methicillin soaked native penicillin G soaked native space group P3.21 P3,21 P3,21 P2,2,2, P2,2,2, P2,2,2, P2,2,2, unit cell axes (A) a 140.8 141.1 140.9 80.9 80.6 80.7 80.5 b 140.8 141.1 140.9 100.6 100.8 103.2 103.3 c 146.4 146.7 146.9 186.2 187.2 186.5 186.8 wavelength (A) 0.979202 0.979413 0.932998 1 1 1 1 1.54 1.54 (peak) (inflection ) (remote) resolution range (A)a 30-2.7 30-3.0 30 - 3.0 25-2.45 25-3.0 25-1.8 25-2.0 25-2.6 25 - 2.45 (2.8 - 2.7) (3.1-3.0) (3.1-3.0) (2.5 - 2.45) (3.1 -3.0) (1.9-1.8) (2.1 -2.0) (2.7 - 2.6) (2.5 - 2.45) unique reflections 89,061 63,825 64,856 61,048 33,848 137,350 100,115 47,895 56,624 average multiplicity 5.5 3 4.5 3.7 6.7 3.3 5.2 2.9 3.7 completeness (%)a 100.0 98.0 99.6 98.0 99.4 97.5 96.6 98.1 97.4 (100.0) (99.9) (100.0) (95.5) (99.8) (97.7) (77.8) (99.7) (94.4) average I/G(J)' 26.8 23.4 30.4 21.7 28.1 23.4 16.8 12.7 18.9 (3.6) (4.7) (5.9) (2.9) (8.1) (2.7) (3.2) (2.0) (5.3) overall B-factor (A2) 71.9 75.3 74.3 62.5 77.4 24.8 24.9 69 49.5 p a,b sym 0.049 0.042 0.041 0.044 0.042 0.042 0.076 0.054 0.049 (0.392) (0.238) (0.233) (0.364) (0.149) (0.330) (0.312) (0.387) (0.207) a Values in parentheses correspond to the highest resolution shell. B RSYNL = Z| / a v - I\l E/j, where / a v is the average of all observations, 94 5.4.2 Structure determination and refinement MAD data were collected from a single SeMet trigonal crystal at peak, inflection and remote wavelengths. Using the peak anomalous data, the selenium sites were located with Shake-and-Bake (Miller et al., 1994; Weeks and Miller, 1999) and refined with SOLVE (Terwilliger and Berendzen, 1999). The resulting 28 of 32 possible sites were used for further refinement and phasing with SHARP (de La Fortelle and Bricogne, 1997) using the SeMet MAD and native data sets (to 3.0 A resolution), which produced figures of merit of 0.74 and 0.53 for centric and acentric reflections, respectively. Phase extension with RESOLVE (Terwilliger, 2000) to 2.7 A resolution produced a clearly interpretable electron density map (with an overall figure of merit of 0.74), from which an initial model was built using Xfit (McRee, 1999). The two molecules in the trigonal asymmetric unit were sufficiently different that NCS averaging was not helpful. An initial model for the primitive orthorhombic crystal form was obtained by molecular replacement with AMoRe (Navaza, 1994) using a monomer from the trigonal asymmetric unit as a search model. For each ofthe nitrocefin, penicillin G and methicillin acyl complexes, the corresponding acylated serine residue was created using HyperChem (Hypercube, 1995) to provide geometric restraints using XPL02D (Kleywegt, 1995) for refinement with CNS (Briinger et al. ,1998). The structure refinement statistics are shown in Table 5.4.2. The relatively high i?-factors for the refined structures are likely due to the high B-factors of the trigonal crystals, anisotropic diffraction of the primitive orthorhombic crystals and damage to the crystals due to P-lactam soaking, which adversely affected data quality. 95 Table 5.4.2: SauPBP2a structure refinement statistics higher resolution SeMeta apo native nitrocefin soaked native methicillin soaked native penicillin G soaked native space group P3,21 P2.2.2, P2.2.2, P212121 ^2,2,2, p work 0.269 0.240 0.235 0.242 0.234 0.323 0.277 0.274 0.302 0.296 No. of atoms: protein + P-lactam 10,082 10,133 10,192 10,040 10,189 water 7 482 394 67 100 ion 11 11 12 9 13 Average 6-factors (A 2): main chain 25.6 31.1 35.8 60.5 48.5 side chain 27.0 34.7 38.7 63.2 51.1 water 13.7 32.3 34.4 38.3 35.9 ion 19.8 31.4 41.6 46.7 44.2 P-lactam - - 43.8 64.9 50.2 overall 28.3 a 33.2 37.1 62.8 51.1 ' rmsd from ideality bond lengths (A) 0.0077 0.0095 0.0094 0.0048 0.0059 angles (°) 1.4 1.5 1.5 1.1 1.2 angle improper (°) 0.78 0.91 0.89 0.67 0.77 Luzzati coordinate error (A) 0.4 0.3 0.3 0.4 0.4 resolution range (A) 5.0-2.45 5.0-1.8 5.0-2.0 5.0-2.6 5.0 - 2.45 Ramachandran plot (% of residues in region): most favourable 83.9 91.5 89.9 83.7 86.3 allowed 14.3 7.9 9.6 15.5 12.9 generously allowed 1.4 0.6 0.4 0.6 0.5 disallowedb 0.4 0.0 0.2 0.2 0.3 Structure was refined against data sharpened with a negative 5-factor (-30 A 2 ) . Asn 305 is disordered in the apo structures and is in a strained conformation at a crystal contact in the p-lactamoyl complexes. Residues 266 (chain A), 246 (chain B) and 641 (chain B) in the trigonal asymmetric unit are in regions of poor electron density. 96 5.5 Discussion 5.5.1 Overa l l structure SauPBP2a* (Fig. 1) has overall dimensions of-130 x 60 x 58 A and consists of a bilobed non-penicillin-binding (nPB) domain and a C-terminal transpeptidase domain. The N-terminal lobe corresponds to an N-terminal extension unique to HMM PBPs of subclass Bl (reviewed in Goffin and Ghuysen, 1998). Using aDali search (Holm and Sander, 1993; Holm and Sander, 1995), this N-terminal extension was found to structurally resemble scytalone dehydratase (Lundqvist et al, 199'4; Chen et al, 1998), nuclear transport factor 2, P15 TAP fragment (Fribourg et al, 2001), and A5-3-ketosteroid isomerase (Wu et al, 1997) with rmsds of 2.3 A (101 Ca pairs), 2.8 A (100 Ca pairs), 2.8 A (97 Ca pairs), and 2.5 A (88 Ca pairs), respectively. However neither the hydrophobic ligand binding cavity nor the catalytic residues in these proteins are present in the SauPBP2a* N-terminal extension, so that the structural similarity with these proteins does not provide any insights into the possible function(s) of the N-terminal extension. Given that SauPBP2a is exogenous to S. aureus, it is unclear if the nPB domain actually serves a biological function in MRSA, although it is required for proper folding of the transpeptidase domain (Wu et al, 1994). Additionally, the length of the nPB domain (which positions the transpeptidase active site over 100 A from the expected C-terminus of the transmembrane anchor) suggests a possible structural role and potentially gives the transpeptidase active site substantial reach from the cell membrane. 97 Fig. 5.5.1: Structure of SauPBP2a*. The bilobed N-terminal nPB domain is coloured orange with the N-terminal lobe (N-terminal extension) coloured green. The transpeptidase domain is coloured blue with the position ofthe active site indicated by the red nitrocefin adduct (shown in stick rendering). The labelling scheme for secondary structural elements in the transpeptidase domain is based on that of R6 PBP2x* (Pares et al., 1996). The N- and C-terminii are labelled N and C, respectively. Shown to the left of the ribbon representation is a linear representation of the domain structure of SauPBP2a* with residue numbers shown in parentheses. 98 cytoplasmic face transmembrane aoehor (1=23) N-terminal cell membrane V V JI « oc3n|V2n 1 extension (27-138) a4n p12n x non-penicillin P^ >> binding /~\ V ^ ; domain /' a 6 n p5n I Mn4on (27-326) J t f * p 8 n f } t P 1 3 n (39n a7 transpeptidase domain (327-668) (310n | p11n 11n a10n 14n p6n Se . 99 The transpeptidase domain of SauPBP2a* shares a similar overall fold with other DD-transpeptidases and the serine P-lactamases (Paetzel et al, 2000 and reviewed in Massova and Mobashery, 1998). Structure-based alignments of the SauPBP2a* transpeptidase domain reveal low sequence identities (9.6 to 19.4%) and significant structural deviations with the corresponding domains from S. pneumoniae R6 PBP2x (R6 PBP2x; Pares et al, 1996; Gordon et al, 2000), Streptomyces R61 DD-carboxypeptidase (R61 PBP; Kelly et al, 1985), Streptomyces K15 DD-transpeptidase (K15 PBP; Fonze et al, 1999), and PBP5 from Escherichia coli (EcoPBP5; Davies et al, 2001) with rmsd values of 2.2 A (288 Ca pairs), 2.9 A (219 Ca pairs), 2.7 A (203 Ca pairs) and 3.0 A (198 Ca pairs), respectively. 5.5.2 Acyla t ion with P-lactam In both crystal forms, the asymmetric unit contained two SauPBP2a* molecules. Conformational differences between these four independent structures reveal significant flexibility in the nPB domain and in the regions surrounding the active site groove in the transpeptidase domain (rmsds for all main chain atoms relative to molecule A of the orthorhombic form of 0.9 to 1.4 A ) . Due to their higher resolution, the primitive orthorhombic crystals were used for a detailed analysis of the active site and its interaction with P-lactams. After soaking in P-lactam, molecule A in the orthorhombic asymmetric unit is observed in the acylated form (occupancies refined as 1.0) and allows for a structural analysis of P-lactam binding by the resistant SauPBP2a*. The active site of SauPBP2a* is located in an extended groove, with the nucleophilic Ser 403 at the N-terminus of helix a2 and with the backbone nitrogens of Ser 403 and Thr 600 forming the conserved oxyanion hole. Possible acylation mechanisms for HMM class B PBPs have been discussed in the context of analogous mechanisms proposed for class A P-lactamases (Strynadka et 100 al, 1992; Ishiguro and Imajo, 1996; Gordon et al, 2000), which suggest that in SauPBP2a*, both Lys 406 and the P-lactam carboxylate are possible candidates for the catalytic base that abstracts the Ser 403 Oy proton during nucleophilic attack. The high resolution structure of the nitrocefin acyl-SauPBP2a* complex shows that the Ser 462 Oy is hydrogen-bonded to the Lys 406 (2.6 A) and is also the most likely proton donor to the p-lactam leaving group nitrogen during acylation (Fig. 5.5.2a-d). The structure also shows that the hydrogen-bonding geometry between the Ser 403 Oy and either the Ser 462 Oy (158.7° from Ser 462 CP-Oy bond) or the p-lactam carboxylate (4.0 A) is unfavourable. Weak electron density is seen for a potential deacylating water (5-factor = 31.4 A2) well positioned for nucleophilic attack on the acyl-Ser 403 ester (2.9 A from the nitrocefoyl C7) and is coordinated by the backbone carbonyl of Tyr 519 (2.6 A). Tyr 519 is at the C-terminus of helix ot6, and the helix dipole of which may have a weak activating effect on the deacylating water. A similar deacylation mechanism was proposed for EcoPBP5 and R6 PBP2x (Gordon et al, 2000; Davies et al., 2001; Rhazi et al., 2003). The weak electron density of the putative deacylating water likely reflects a partial occupancy, which together with the lack of an efficient general base catalyst nearby are consistent with the slow deacylation of SauPBP2a* (Lu et al, 1999). 101 Fig. 5.5.2: Interaction of SauPBP2a* with nitrocefin. a) Structure of nitrocefin. Conventional numbering is shown for the bicyclic atoms, b) SigmaA-weighted 2F0-FC electron density (contoured at 1 a to 2.0 A resolution) for the nitrocefin-acylated Ser 403 as observed in the corresponding acyl-SauPBP2a* complex, c) Active site in the nitrocefin acyl-PBP structure. Nitrocefin is shown in stick rendering with magenta carbons, d) Active site in the native structure. Nitrocefin as it is observed in the nitrocefin acyl-PBP structure is shown in translucent stick rendering. 102 5.5.3 Structural basis for P-lactam resistance in SauPBP2a The interaction of a PBP with a P-lactam inhibitor begins with the rapid reversible formation of a non-covalent Michaelis complex (Fig. 5.5.3). This is followed by the nucleophilic attack bythe active site serine on the p-lactam ring to give a relatively stable covalent acyl-PBP intermediate (Frere et al, 1975; Ghuysen et al, 1986). A comparison of the micro kinetic parameters of SauPBP2a* with the P-lactam-sensitive R61 PBP and R6 PBP2x showed that the p-lactam resistance of SauPBP2a* is primarily due to extremely inefficient formation of the acyl-PBP intermediate (indicated by the first order rate constant for acylation k2 with penicillin G being three orders of magnitude lower for SauPBP2a*) and does not result from a poorer fit of the P-lactam in the active site (Kd for the formation of the Michaelis complex) nor from a more rapid breakdown (k3) of the acyl-PBP intermediate (Frere et al, 1975; Lu et al, 1999; Lu et al, 2001). Since typical Kd values for PBPs are relatively high (being in the millimolar range), the effectiveness of p-lactam inhibitors results from their ability to form stable covalent complexes with their targets. Therefore reducing the acylation rate (k2) is an effective strategy for achieving broad spectrum resistance. Michaelis is complex h acy l -PBP /, Kd. r — Y ^ k2 , * •> *3 ( P B P ) +(penici l l in) <—»• (PBPjpenic i l l in) • (PBP-penici l l in ) >• ( P B P ) + (penicilloate ' Fig. 5.5.3: Overall reaction scheme for interaction of a PBP with a (3-lactam. The p-lactam is represented by penicillin. Kd is the dissociation constant for the reversible formation of the non-covalent Michaelis complex. k2 is the first order rate constant for the formation of the acyl-PBP intermediate from the Michaelis complex. k3 is the first order rate constant for the hydrolysis of the acyl-PBP intermediate. 103 PBP complex structures of acylation rate is due to a distorted SauPBP2a* reveal that the poor The apo and nitrocefin acyl-helix oc2 N-terminus and at strand (33 novel conformational changes at the active site, which must undergo Fig. 5.5.4: Superposition ofthe active site region from the apo (yellow) and nitrocefin acyl-PBP (blue) structures. Nitrocefin (purple) is shown in thin stick rendering. Covalent binding of nitrocefin to SauPBP2a* requires conformational changes at strand p3 and at the N-terminus of helix oc2. for acylation to occur (Fig. 5.5.2c, d and 5.5.4). In the apo conformation of the helix cc2 N-terminus (residues 402 to 408), Ser 403 is in a poor position for nucleophilic attack. Upon acylation, the Ser 403 Ca, CP and Oy must move 1.1 A, 1.4 A and 1.8 A, respectively. Additionally, binding of nitrocefin requires a twisting of strand P3 (residues 594 to 603) due to a steric clash (1.6 A) between the nitrocefin carboxylate and the Gly 599 Ca in the apo structure. Superposition of the apo and nitrocefin acyl-SauPBP2a* structures also reveals that the twisted apo conformation of strand (33 is required to accommodate the apo conformation of the helix a2 N-terminus (Fig. 2e), in that a steric clash (2.5 A) can be seen between the Ser 598 backbone carbonyl in the acyl-PBP complex and the Ser 403 Cp in the apo structure. The kinetic parameters determined for the interaction of SauPBP2a* with p-lactams can then be rationalized on the basis that these conformational changes observed in SauPBP2a* impede the formation of the acyl-PBP intermediate (reduction of k2). Since values for Kd indicate that initial binding of P-lactams is not significantly impeded in SauPBP2a* (Lu et al, 1999), the position of nitrocefin in the initial Michaelis complex is proposed to differ from that observed in the acyl-PBP structure, so that strand P3 is not distorted during initial binding. The transition from the Michaelis complex to the acyl-PBP intermediate would then require an energetically costly rearrangement 104 involving the movement of the P-lactam into position for acylation. This in turn would require a twisting of strand P3 with a concomitant conformational change at the helix oc2 N-terminus to position Ser 403 for nucleophilic attack. In contrast to SauPBP2a, equivalent conformational differences between the apo and acyl-PBP structures are absent in the P-lactam-sensitive R61 PBP (Kelly f 1 * 5-5-5: Superposition of the active site region from the apo- (yellow) and acyl-PBP (purple) * / ioeo A D A D D D I structures of R6 PBP2x. et al, 1985) and R6 PBP2x (Fig 5.5.5; Gordon et al, 2000). This is consistent with a more rapid transition from the Michaelis complex to the acyl-PBP intermediate in these P-lactam-sensitive PBPs as reflected by their higher acylation rates (k2; Frere et al, 1975; Lu et al, 2001). 5.5.4 Compar ison with other resistant PBPs Naturally occurring mutations in the region of the helix a2 N-terminus in R6 PBP2x* significantly affect the acylation rate and support our proposal for the importance of the conformation of the helix ct2 N-terminus in properly positioning the Ser 337 side chain for nucleophilic attack. In particular, the T338A mutation is the most frequent mutation in resistant variants of R6 PBP2x and reduced the overall acylation rate 1.8 to 2.7 fold, while the reverse mutation in the resistant Sp328 PBP2x (a variant with 83 mutations from S. pneumoniae Sp328) increased the overall acylation rate 6 fold (Mouz et al, 1998). The crystal structure of Sp328 PBP2x* showed that the T338A mutation resulted in the disruption of a hydrogen-bonding network adjacent to the helix al N-terminus (Dessen et al, 2001). In light of the structural data for 105 SauPBP2a*, a reexamination of the structural differences between R6 PBP2x* and the resistant Sp328 PBP2x* suggests that conformational changes in the helix cc2 N-terminus and in strand S3 similar to those observed in SauPBP2a* may occur upon acylation of Sp328 PBP2x*, however confirmation of this will require structural information on a P-lactam acyl-PBP complex for Sp328 PBP2x* and a higher resolution structure of apo Sp328 PBP2x*. Shortly after the structure determination of SauPBP2a*, the crystal structure of a soluble derivative of Enterococcus faecium PBP5 (PBP5fm*) was published (Sauvage et al., 2002). Similar to SauPBP2a, PBP5fm is a HMM PBP of subclass Bl (Goffin and Ghuysen, 1998) and shares 35% sequence identity (for 603 aligned residues) with SauPBP2a (Zorzi et al, 1996). As with other members of subclass B1, PBP5fm is a resistant PBP with low P-lactam affinity and confers intrinsic resistance to Enterococci. Second order rate constants (k21K^) for interaction with penicillin G are similar for PBP5fm (15-24 M's1 for PBP5fm; Zorzi et al, 1996) and SauPBP2a* (16.5 NT's"1 for SauPBP2a*; Lu et al, 1999). The coordinates for PBP5fm* were not available for a direct structural comparison with SauPBP2a*, however the published ribbon figure of PBP5fm* shows an essentially identical overall fold to that of SauPBP2a*. Sauvage et al. (2002) primarily proposed that buried electrostatic interactions involving Arg 464 (Arg 445 in SauPBP2a), which is uniquely conserved in HMM PBPs of subclass B1, confer greater structural rigidity to the loop region lining one side of the active site groove and encompassing residues 451 to 465 (residues 432 to 446 in SauPBP2a), thereby impeding entry of p-lactams into the active site. Such a hypothesis predicts elevated Kd values for interaction with p-lactams, however since only second order rate constants (k21are available for PBP5fm*, their hypothesis cannot be substantiated with the available kinetic data. The structure of PBP5fm* was determined only in the penicillin G-acylated form (with no structure of the apo state available), and it is unclear at this time if there are conformational differences at helix 106 a2 and strand P3 as seen in SauPBP2a*. However given the similarities in sequence, structure and overall P-lactam affinity between SauPBP2a* and PBP5fm*, it is reasonable to expect that a reduced acylation rate (likely resulting from active site distortion as proposed for SauPBP2a*) is the key factor in the P-lactam resistance of PBP5fm. 5.5.5 Comparison of the P-lactam acyl-SauPBP2a* complexes The nitrocefin, penicillin G and methicillin acyl-PBP structures identify features responsible for the different affinities of SauPBP2a* for these P-lactam compounds. The R2 substituent of nitrocefin is wedged between Tyr 446 on one side of the active site groove and Thr 600 and Met 614 on the other side. The carboxylate group is hydrogen-bonded to Ser 598 (2.7 A) , the P-lactam leaving group nitrogen is hydrogen-bonded to Ser 462 (3.0 A), and N 9 on the R, substituent is hydrogen-bonded to the backbone carbonyl of Thr 600 (2.8 A ; Fig. 5.5.2a, c). Similar hydrogen-bonding interactions are seen in the penicillin G acyl-PBP structure (Fig. 5.5.6a-c), however the significant van der Waals contacts made by the nitrocefin R2 substituent are absent in the penicillin G acyl-PBP complex. The additional contacts made by the R2 substituent on cephalosporins such as nitrocefin (lacking in penicillins) are expected to increase binding affinity. This is consistent with the ~4 fold greater overall acylation rate (the second order rate constant k2IKA of nitrocefin over that of penicillin G (Graves-Woodward and Pratt, 1998). The cephalosporin compound 1 (Fig. 5.5.7) used by Lu et al, 1999 (Lu et al, 1999) contained an iodinated aromatic R2 substituent, which should resemble the nitrocefin R2 substituent (Fig. 5.5.2a) in hydrophobicity. A comparison of the micro kinetic parameters for compound 1 (Kd = 0.22 mM, k2 = 0.39 s"1) with those of penicillin G (Kd = 13.3 mM, k2 = 0.22 s"1) indicated that the greater overall acylation rate of nitrocefin relative to penicillin G largely resulted from a significantly reduced Kd. The van der Waals nature of the 107 contacts made by the nitrocefin and compound 1 R2 substituents (which do not require specific relative orientations of the interacting atoms) could also facilitate a repositioning of the P-lactam to accommodate the proposed rearrangements near the nucleophilic Ser 403 during the transition from the Michaelis complex to the acyl-PBP intermediate. Additionally, conformational flexibility of the region surrounding the SauPBP2a* active site is indicated by a significant widening of the active site groove upon nitrocefin binding (Fig. 5.5.8). This largely results from a Ca rmsd of 1.0 A for residues 424 to 478 and residues 510 to 524, which form one side of the active site groove opposite Ser 403. The flexibility of these loop regions is also expected to provide flexibility in the positioning of nitrocefin during initial binding and the subsequent transition to the acyl-PBP intermediate. A similar widening of the active site crevice upon p-lactam binding was also seen in R6 PBP2x* (Gordon et al, 2000). 108 Fig. 5.5.6: Penicillin G acyl-PBP complex for SauPBP2a*. a) Structure of penicillin G. Conventional numbering is shown for the bicyclic atoms, b) SigmaA-weighted 2F0-FC electron density (contoured at 1 a to 2.45 A resolution) for the penicillin G-acylated Ser 403 as observed in the corresponding acyl-SauPBP2a* complex, c) Active site in the penicillin G acyl-SauPBP2a* structure. Penicillin G is shown in stick rendering with magenta carbons. The nitrocefin-acylated Ser 403 as observed in the nitrocefin acyl-SauPBP* structure (Fig. 5.5.2c) is shown in thin green stick rendering. Fig. 5.5.7: Structure of the cephalosporin "compound 1". This structure was used by Lu et al., 1999 (Lu er al., 1999) and resembles nitrocefin (Fig. 5.5.2a) in hydrophobicity of substituents. 109 Fig. 5.5.8: Widening of SauPBP2a* active site upon nitrocefin binding. Residues 424 to 478 and 510 to 524 in the apo (blue) and acyl-enzyme (orange) states are shown as Ca traces. The nitrocefoyl group (red) is shown in stick rendering. Residues in strand B3 and at the N-terminus of helix a2 in the apo (blue) and acyl-enzyme (magenta) states are also shown as Ca traces. The poorer overall acylation efficiency of methicillin relative to penicillin G largely results from an extremely poor k2 value (0.0083 s"1), while the Kd value (16.9 mM) is comparable to that of penicillin G (LuetaL, 1999). The structure of the methicillin acyl-SauPBP2a* complex (Fig. 5.5.9a-c) shows that this poor acylation rate is a consequence of the position of the bound methicillin being translated along the active site groove relative to that observed in the nitrocefin and penicillin G acyl complexes. The acyl-Ser 403 CP, Oy, C 7 and C 9 atoms deviate by 0.4 A, 0.5 A, 0.5 A and 1.0 A, respectively between the methicillin and nitrocefin acyl complexes. This relative displacement of methicillin allows its bulky dimethoxy-phenyl (R,) substituent to be accommodated in the region occupied by the nitrocefin R, substituent. However this results in the distance between the Ser 462 Oy and the P-lactam leaving group nitrogen increasing from 3.0 A (nitrocefin acyl complex) to 3.7 A (methicillin acyl complex), retarding the proton transfer from Ser 462 to the P-lactam leaving 110 group nitrogen during acylation. The conformation of helix a2 is nearly identical in the nitrocefin and methicillin acyl-SauPBP2a* complexes (0.2 A main chain rmsd for residues 402 to 416), and the overall conformational changes in the helix al N-terminus and strand 63 resulting from acylation with methicillin are similar to the aforementioned changes resulting from acylation with nitrocefin. Thus the intrinsically slow acylation of SauPBP2a* works in conjunction with the restricted binding of methicillin within the narrow active site (Fig. 5.5. 9d) to provide enhanced resistance against bulky P-lactams such as methicillin. I l l Fig. 5.5.9: Methicillin acyl-PBP complex for SauPBP2a*. a) Structure of methicillin. Conventional numbering is shown for the bicyclic atoms, b) SigmaA-weighted 2F0-FC electron density (contoured at 1 a to 2.6 A resolution) for the methicillin-acylated Ser 403 as observed in the corresponding acyl-SauPBP2a* complex, c) Active site in the methicillin acyl-SauPBP2a* complex. Methicillin is shown in stick rendering with magenta carbons. The nitrocefin-acylated Ser 403 as observed in the nitrocefin acyl-SauPBP2a* structure (Fig. 5.5.2c) is shown in thin green stick rendering. The (3-lactam leaving group nitrogens of the methicilloyl and nitrocefoyl groups are indicated by "N 4 " and "N 5 ", respectively, d) GRASP (Nicholls et al., 1993) surface representation of the active site groove in the methicillin acyl-SauPBP2a* complex. Positively and negatively charged areas are coloured blue and red, respectively. Methicillin is shown in stick rendering with yellow carbons. 112 5.5.6 Implications for transpeptidase activity Biochemical and genetic studies have indicated a reduced transpeptidase activity of SauPBP2a relative to the native S. aureus PBPs. Analysis of the cell wall composition of an MRSA strain revealed a dramatic reduction in the degree of crosslinking of muropeptides isolated from the cell walls of MRSA grown in the presence of methicillin (de Jonge and Tomasz, 1993), as reflected by an increased relative abundance of muropeptide monomers (uncrosslinked) and dimers (single crosslink). However the absence of major changes in cell physiology morphology of MRSA (other than diffuse and abnormally wide septa) grown in the presence of methicillin hints at compensatory mechanisms that allow MRSA to cope with greatly reduced cell wall peptide crosslinking (de Jonge and Tomasz, 1993). Increases in doubling time were observed only at high concentrations of methicillin (>500 pg/mL). Several genetic factors involved in methicillin resistance in addition to mecA have been identified and likely fulfill this role. The glycosyltransferase domain ofthe class A PBP2 is essential for methicillin resistance. A mutant MRSA strain in which this domain is inactivated is viable in the absence of methicillin and produce cell walls containing shorter glycan strands containing fewer disaccharide units. However, the inability of this strain to grow in the presence of methicillin indicates a requirement for glycan strands of sufficient length to compensate for reduced peptide crosslinking (Pinho etai, 2001). Inactivation of lytH (which encodes a putative lytic enzyme with 30% sequence identity to the A^ -acetylmuramoyl-L-alanine amidase LytC of Bacillus subtilis) resulted in a dramatic increase in methicillin resistance (from 6.9 to 1600 pg/mL; Fujimura and Murakami, 1997). The inactivation of lytHmay help to restore a balance of breakdown and formation of cell wall crosslinks. Over-expression oihmrA (putative aminohydrolase) and hmrB (homologue of acyl carrier protein) also confers high level methicillin resistance, although the 113 biochemical basis for the increased resistance in this case is not well understood (Kondo et al., 2001). Since the acylation step is believed to be mechanistically the same in both P-lactam inhibition and transpeptidation, a reduced p-lactam acylation rate is expected to correlate with reduced transpeptidase activity. Indeed, the marked difference in solvent accessible surface area (Briinger et al., 1998) of the Ser 403 Oy in the apo (1.1 A 2 ) and acylated (11.7 A 2 ) conformations suggests that the reduced P-lactam acylation rate results from an overall reduction in the reactivity ofthe Ser 403 side chain. The solvent accessible surface area for the acylated conformation is within the range of values for the nucleophilic serine Oy of K15 PBP (6.5 A 2 ) , R61 PBP (8.9 A 2 ) , EcoPBP5 (26.1 A 2 ) , R6 PBP2x (13.2 A 2 ) , and Sp328 PBP2x (4.9, 7.5, 6.5 and 9.3 A 2 for molecules A, B, C and D, respectively). However suitable in vitro peptide substrates are not available to confirm a reduced transpeptidase activity of SauPBP2a (Graves-Woodward and Pratt, 1998). The apparently reduced transpeptidase activity of SauPBP2a in vivo may also result from the S. aureus cell wall stem peptides being sub-optimal substrates for the "foreign" SauPBP2a. It is possible that binding of an optimal peptide substrate to the SauPBP2a active site may induce or facilitate the conformational changes that increase the accessibility of the Ser 403 Oy. Since acylation is a key step in the inhibition of PBPs by P-lactams, our proposal that the SauPBP2a* active site effectively balances a retention of essential transpeptidase activity (via conservation of key catalytic residues) with a reduction of P-lactam affinity (via distortion of the active site to reduce catalytic efficiency) is likely applicable to Sp328 PBP2x and other resistant PBPs with reduced P-lactam acylation rates. Given the central role ofthe acylation step in inhibition by virtually all P-lactams (versus initial binding), reduction ofthe acylation rate is an effective means to achieve broad spectrum resistance. 114 5.5.7 Effect of crystal packing on P-lactam binding Despite numerous attempts at high concentrations, soaking of P-lactams into crystals of the acylation-deficient S403A SauPBP2a* mutant to obtain the crystal structure of a p-lactam Michaelis complex was unsuccessful. Since it was possible to obtain acyl-PBP complexes by soaking, access to the active site was not blocked by neighbouring molecules in the crystal lattice. Instead movement (Fig. 5.5.8) of the all alpha region relative to the alpha / beta region (required for entry of the P-lactam into the active site) was likely restricted by the overall crystal packing, which resulted in dramatically higher Kd values for P-lactam binding in the crystalline state. This again emphasizes the importance of the acylation step in the inhibition of PBPs by P-lactams. The effect of the crystal packing was more pronounced in molecule B of the primitive orthorhombic asymmetric unit, which showed no electron density for nitrocefin and significantly weaker electron density for penicillin G and methicillin (Table 5.5.1). Surprisingly, in molecule B of the primitive orthorhombic apo structure, the Ser 403 hydroxyl is pointing away from the active site and into the protein interior (x, =61.3° compared to %{ = -71.1 ° in molecule A) and results in complete burial of the Ser 403 Oy (0 A 2 solvent accessible surface). Since the affinity and reactivity of molecule A towards P-lactams is somewhat closer to levels observed in solution (in contrast to molecule B), molecule A of the primitive orthorhombic asymmetric unit was chosen for analysis of P-lactam binding. The basis for this distorted conformation of the Ser 403 side chain in molecule B is unclear, but is likely due to differences in the crystal environment between molecules A and B. Previously, this conformation of the nucleophilic serine side chain has only been observed in the K l 5 PBP (Fonze et al, 1999; Rhazi et al, 2003), however due to the relative disposition of Ser 35 to strand P3 in K l 5 PBP, the Ser 35 Oy remains relatively solvent accessible. An acyl-enzyme structure has not been determined for K l 5 PBP, and the conformation adopted by the Ser 35 side chain upon 115 acylation remains to be determined. The p-lactam affinity is further reduced in the trigonal crystal form of SauPBP2a*, in which acyl-PBP complexes for neither nitrocefin nor penicillin G could be obtained. Due to the poor resolution of the trigonal crystals, the basis for this further reduction in affinity is not entirely clear, however since the active sites in the asymmetric unit are again not blocked, restricted flexing ofthe active site cavity is again a likely explanation. Table 5.5.1: Group occupancies and average 5-factors of P-lactam and Ser 403 atoms in SauPBP2a* p-lactam Ser 403 group occupancy average 5-factor (A2) average 5-factor (A2) nitrocefin molecule A 1 43.9 28.1 molecule A 1 37.9 44.6 penicillin G molecule B a 0.97 45.2 55.3 molecule A 1 57.9 48.1 methicillin molecule B a 0.68 72.3 68.3 a For the penicillin and methicillin moieties in molecule B, the occupancies were refined with the 5-factors for the atoms initially set to the overall value for the crystal, following which individual 2?-factors were refined. Despite the effects of crystal packing on the active site structure, the structural basis for P-lactam resistance in SauPBP2a as proposed from the crystallographic studies presented here is consistent with previous kinetic studies of SauPBP2a* under solution conditions (Lu et al, 1999) and also with kinetic and structural studies of resistant variants of S. pneumoniae PBP2x (Mouz et al, 1998; Dessen et al, 2001; Lu et al, 2001). Additionally active site conformational differences between the apo and P-lactam-acylated states have not been observed in the structures of non-resistant PBPs, which were also obtained by crystallography (Kelly et al., 1985; Gordon et al., 2000). It would be interesting to have available a crystal structure of PBP5fm* from E.faecium in the apo 116 state. An active site conformational difference between the apo and B-lactam-acylated states of PBP5fm* would provide an independent confirmation of the observations for SauPBP2a*. The effect of crystal packing on the affinity / reactivity of SauPBP2a* towards P-lactams has not been observed for other PBPs and raises the possibility that the transpeptidase activity / P-lactam susceptibility of SauPBP2a (and possibly other related resistant PBPs) may be regulated by allosteric interactions with potential protein ligands that modulate the conformation of the active site. Allosteric regulation of catalytic activity has been proposed as a possible explanation for the increased reactivity of E. coli PBP3 towards the cephalosporin cephalexin in dividing cells relative to filamentous cells in which division was inhibited (Eberhardt et al., 2003). 5.5.8 Insights for drug design The ability of SauPBP2a to act as the sole DD-transpeptidase for cell wall synthesis under B-lactam selection has obviated the development of high level resistance in the native PBPs in MRSA. Thus, SauPBP2a provides a single target for the development of specific inhibitors against MRSA, which will be greatly facilitated by the structural information on the active site cavity presented here. Given that poor acylation efficiency is intrinsic to SauPBP2a, an important aspect for inhibitor design will be to improve binding affinity by increasing the number of non-covalent interactions between inhibitor and active site. In designing more effective P-lactam inhibitors, reduction of Kd is expected to play a large part in improving overall acylation efficiency (kj/KA. Consistent with this, the higher affinities of a number of newly developed cephalosporins for SauPBP2a correlate with their longer substituents (Fig. 5.5.10), which in light of the SauPBP2a* crystal structures, would provide better shape complementarity for the narrow active site groove and should allow for a larger number of non-covalent stabilizing interactions (Entenza et al., 2002; Fung-Tome et al., 2002). 117 These novel compounds exhibit MICs in the range of 2 to 8 pg/mL for methicillin-resistant staphylococcal strains as well as increased resistance against hydrolysis by P-lactamases. Alternatively, non-covalent non-P-lactam compounds that bind tightly to the active site without the need for acylation may also provide highly effective inhibitors. The pM IC50 values for SauPBP2a* shown by several dye compounds (Toney et al, 1998) and the high affinity (K{ = 26 uM) and selective binding of a competitive, noncovalent and non-P-lactam inhibitor of AmpC P-lactamase (Powers et al, 2002) maybe indicative of the potential of such inhibitors. Non-covalent inhibitors will not require the unfavourable conformational changes in SauPBP2a required for acylation and will also be insusceptible to P-lactamases. Fig. 5.5.10: Structures of novel cephalosporin compounds with increased activity against MRSA. a) BMS-247243. b) BAL9141. 118 Chapter 6: CONCLUSIONS AND FUTURE DIRECTIONS 6.1 Summary and significance of results The overall goal of this thesis was the structural characterization of specific bacterial protein molecules involved in the three major types of resistance mechanisms against the p-lactam class of antibiotics. The most common of these mechanisms is the production of P-lactamases. The development of numerous of P-lactam derivatives with broader spectrum and greater resistance to hydrolysis has been countered by the emergence of P-lactamase variants with novel substrate profiles (reviewed in Medeiros, 1997). This is well illustrated by the PSE-4 P-lactamase and other carbenicillinases which efficiently inactivate carbenicillin, a penicillin derivative with improved activity against Gram-negative pathogens such as Pseudomonas aeruginosa. The crystallographic and molecular modelling studies of PSE-4 (Chapter 3) reveals the structural importance of the K234R substitution (found in almost all carbenicillinases) in allowing an alternate mode of substrate binding to accommodate the more bulky carbenicillin and other related carboxy-substituted penicillins. The detailed structural information on the PSE-4 active site will be of use in the design of novel P-lactam compounds that are resistant to carbenicillinases. In Gram-negative bacteria, an alternative strategy to P-lactamase production is the over-expression of multi-drug efflux pumps that confer resistance against P-lactams and a broad range of other antibiotics. In the case of the MexAB-OprM efflux system in P. aeruginosa, mutations in its regulator MexR result in over-expression of the efflux system (due to derepression) and increased MICs to P-lactams and many other antibiotics. The structural analysis of MexR (Chapter 4) provides insights into a possible mode of regulation of the DNA binding of MexR involving a change in the 119 spacing of the recognition helices resulting from ligand binding between the DNA binding domains. The nature of this potential ligand binding site raises the possibility that the mexRAB-oprM operon may be regulated by a protein or peptide ligand(s). A greater understanding of the molecular mechanisms regulating the expression of MexAB-OprM will contribute to elucidating its normal physiological function and substrates, which currently are not known. Such information may eventually aid in the design of substrate-analog inhibitors. A major mechanism of P-lactam resistance in Gram-positive cocci is the expression of low affinity PBPs that arise either from modification or over-expression of endogenous PBP genes or by acquisition of extraspecies PBP genes. A prime example of resistant PBPs is SauPBP2a, the key determinant of broad spectrum P-lactam resistance in MRSA. The structure determination of SauPBP2a in the apo state and in acyl-enzyme complexes with three different P-lactam compounds reveals a distorted active site that must undergo conformational changes for acylation to occur (Chapter 5). The covalent interaction of p-lactams with their targets is a key step in their inhibition of PBPs. Consequently the reduced acylation rate in SauPBP2a results in lower overall affinities with essentially all P-lactams. The structures also reveal steric constraints ofthe narrow active site that hinder the binding of methicillin and other p-lactam compounds with bulky substituents designed to fit poorly into active sites of p-lactamases. SauPBP2a therefore exploits the very features of P-lactam compounds that had previously made them successful inhibitors. P-lactamases, the tripartite multi-drug efflux systems and possibly resistant PBPs like SauPBP2a can be thought of as proteins that have been recruited from other functions to provide P-lactam resistance. Structural, mechanistic and sequence analyses have provided strong evidence of the evolution of the serine P-lactamases from PBPs. The acquisition of an efficient deacylation mechanism and the ability to discriminate P-lactams from cell wall stem peptides were followed by 120 further evolution into numerous variants that collectively are able to efficiently bind to and hydrolyze an increasingly broader range of P-lactam substrates (Medeiros, 1997). An alternative and in some ways opposite strategy is exemplified by the evolution of resistant PBPs, which have retained their transpeptidase function and show poor reactivity towards P-lactams. The more restricted active site groove and reduced reactivity of the nucleophilic serine in SauPBP2a* contrasts with the alternate substrate binding modes and high catalytic efficiency of PSE-4. The effect of crystal packing on the active site structure of SauPBP2a* raises the possibility of allosteric regulation of its activity and P-lactam susceptibility. Allosteric regulation was also a proposed explanation for the observed cell division-stimulated increase in the acylation rate of Escherichia coli PBP3 by cephalexin (Eberhardt et al., 2003). If indeed PBP2a evolved from an allosterically regulated PBP, the acquisition ofmecA by Staphylococcus aureus and the use of this exogenous PBP as the sole DD-transpeptidase under methicillin selection may be viewed as the recruitment for general DD-transpeptidase function of a PBP that was activated at specific times or cellular locations in order to exploit the less active states of this PBP for P-lactam resistance. The MexAB-OprM efflux system provides a clear example of recruitment. The details of the normal mode of regulation of the mexRAB-oprM operon are not known. However the need for mutations in mexR and in other unidentified regulatory genes for increased multi-drug resistance indicate that the operon is not antibiotic-inducible and therefore did not originally evolve to confer multi-drug resistance (Srikumar et al, 2000). 6.2 Future studies Crystal structures of acyl-enzyme complexes of PSE-4 with penicillin G and carbenicillin will further aid efforts to design novel carbenicillinase-resistant P-lactams and provide confirmation of the alternate substrate binding mode suggested by the modelling studies. Such structural 121 information will help to define a minimum size of substituents that will hinder binding to the enzyme, which should greatly assist rational drug design efforts, given that any clinically useful anti-pseudomonal P-lactam must be non-toxic and must be able to penetrate the outer membrane. Crystallization of acyl-enzyme complexes of PSE-4 should be feasible using an E166N mutant as demonstrated by the penicilloyl-E 166N TEM-1 mutant structure (Strynadka et al., 1992). However, an analysis of the molecular packing in the wild type PSE-4 crystals shows blockage ofthe active site by neighbouring molecules. This was qualitatively confirmed by the lack of colour change of crystals of wild type PSE-4 crystals soaked in the chromogenic P-lactam nitrocefin (despite a colour change ofthe surrounding buffer due to trace concentrations of enzyme in solution). Crystals of acyl-enzyme intermediates of PSE-4 must therefore be obtained by co-crystallization ofthe E166N mutant in complex with P-lactam (using the crystallization conditions of the wild type enzyme) or by screening for additional conditions to obtain novel crystal forms with different packing arrangements that provide solvent access to the active site. Given the residual deacylation rate of E166N mutants (Guillaume et al, 1997) and the relatively long periods required for crystal growth, the latter approach is more likely to succeed. However, the deacylation rate ofthe E166N PSE-4 mutant remains to be determined and mutation of Glu 166 to other amino acid residues (such as glutamine or alanine) can also be attempted. To date, cocrystallization of the E166N PSE-4 mutant with P-lactam and the search for conditions that give an alternative crystal form have been unsuccessful. The availability of structural information on the active site of SauPBP2a* will allow structure-based design approaches (reviewed in Blundell, 1996) to help further the development of novel anti-MRSA P-lactam and potentially non-P-lactam compounds. Crystal structures of SauPBP2a* in complex with lead compounds (obtained by soaking or cocrystallization) can be used 122 to guide the introduction of modifications that optimize interactions with the active site and reduce dissociation constants (KA. Availability of high affinity P-lactam compounds with sufficiently low Kd values to overcome the effects of crystal packing may allow for the determination of a Henri-Michaelis complex structure of SauPBP2a*, which will provide further clarification on the details of the proposed active site conformational changes that retard acylation. To explore the possibility that protein ligands may allosterically regulate the activity and P-lactam susceptibility of SauPBP2a, a tagged construct of SauPBP2a can be immobilized on a column and potential interaction partners may be captured from MRSA cell lysate and identified by SDS-PAGE, ESMS and N-terminal sequencing. Proteins identified can be tested for their ability to interact with and alter the activity of SauPBP2a using P-lactam binding assays. Similar approaches were successful in "capturing" PBPs la, lb, lc, 2 and 3 from£. coli crude membrane extracts using immoblized MltA lytic transglycosylase (Vollmer et al., 1999). However since SauPBP2a is a protein of extraspecies origin in MRSA, its normal interaction partners may not be present in S. aureus. It may be more appropriate to look for potential allosteric regulators of other HMM subclass Bl PBPs that are endogenous to their hosts. One such target is the PBP2a homologue (-80% sequence identity to SauPBP2a) encoded by the mecAl gene, which is native to Staphylococcus sciuri (Couto et al, 1996; Wu et al, 1996). Mutations that cause over-expression of mecAl in S. sciuri resulted in increased resistance to methicillin, and expression oimecAl in a MRSA strain in which the mecA gene had been disrupted, rescued the methicillin resistance phenotype (Wu et al, 2001; Couto et al, 2003). It has been proposed that the mecAl gene may be the evolutionary precursor of mecA in MRSA, and thus S. sciuri may contain potential protein interaction partners of PBP2a. Other targets include the resistant PBPs native to enterococci. Identification of allosteric 123 regulators of PBPs of HMM subclass BI (if they exist), would shed light on the evolutionary origins of such highly resistant PBPs. Immobilization of a tagged MexR construct may also be used to find potential protein, peptide or possibly small molecule ligands from P. aeruginosa lysate. Growth phase regulation of MexAB-OprM expression has been observed with expression increasing in log to late log phase (Evans and Poole, 1999). However the molecular mechanisms involved in this regulation are not known, and it may be worthwhile to look for ligands that bind MexR in lysate from cells in log to late log phase. Lysate from cells in lag phase (when MexAB-OprM expression is minimal) may be used as a negative control. Lysate may also be initially screened for inhibitory activity against DNA binding by MexR to determine the involvement of MexR in the growth phase regulation of MexAB-OprM expression. Additionally, nalC mutant strains, which do not contain mutations in MexR but express MexAB-OprM at higher than normal levels, indicate the presence of additional regulatory elements (Srikumar et al., 2000), which maybe present in nalCcell lysate and possibly act on MexR. It is also possible that MexR may be regulated by post-translational modifications such as phosphorylation or proteolysis, and these may be detected by native/SDS PAGE or ESMS analysis of MexR treated with lysate. Alternatively small molecule libraries may be screened for binding to MexR and inhibition of DNA binding. Given the small size and high expression levels of MexR, NMR screening methods may be possible, allowing rapid detection of interactions between MexR and potential ligands by monitoring chemical shift changes of cross-peaks in a 15N-'H heteronuclear single quantum coherence spectrum in the presence of a compound mixture (Hajduk et al, 1999). Compounds identified by such an initial high throughput screen can subsequently be tested for their ability to inhibit DNA binding by MexR using electrophoretic mobility shift assays and more importantly for their ability to induce MexAB-OprM expression in vivo. The structures of inhibitory 124 compounds identified in this way may help to define a ligand profile for MexR and possibly provide further insights into the normal function of the MexAB-OprM system. Additionally, a crystal structure of MexR in complex with its operator DNA can be pursued to confirm the molecular modelling studies. In addition to screening crystallization conditions, the length and ends ofthe DNA fragment will also need to be varied (reviewed in Tan et al, 2000). Many protein-DNA complexes have been crystallized such that the DNA fragments pack end to end. The length of the DNA fragment will therefore need to be chosen such that an integral number of tandem repeats of the DNA fragment will give an integral number of turns (within a single unit cell). The choice of lengths must take into consideration potential bending ofthe DNA (when bound to MexR) and the choice of ends, which may be blunt or contain overhangs for Watson-Crick base pairing, or GGC or CGC Hoogstein triple strand pairing. 125 R E F E R E N C E S Abraham, E. P. and Chain, E. (1940). "An enzyme from bacteria able to destroy penicillin." Nature 146(3713): 837. Abrahams, E. P., Fletcher, C. M., Florey, H. W., Gardner, A. D., Heatley, N. G., et al. (1941). "Futher observations on penicillin." Lancet 2: 177. Acred, P., Brown, D. M., Knudsen, E. T., Rolinson, G. N. and Sutherland, R. (1967). "New semi-synthetic penicillin active against Pseudomonas pyocyanea." Nature 215(96): 25-30. Adachi, FL, Ohta, T. and 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(5): 3186-91. Adam, M., Fraipont, C, Rhazi, N., Nguyen-Disteche, M., Lakaye, B., et al. (1997). "The bimodular G57-V577 polypeptide chain of the class B penicillin-binding protein 3 of Escherichia coli catalyzes peptide bond formation from thiolesters and does not catalyze glycan chain polymerization from the lipid II intermediate." J Bacteriol 179(19): 6005-9. Adewoye, L., Sutherland, A., Srikumar, R. and Poole, K. (2002). "The mexR repressor of the mexAB-oprM multidrug efflux operon in Pseudomonas aeruginosa: characterization of mutations compromising activity." J Bacteriol 184(15): 4308-12. Alekshun, M. N. and Levy, S. B. (1997). "Regulation of chromosomally mediated multiple antibiotic resistance: the mar regulon." Antimicrob Agents Chemother 41(10): 2067-75. Alekshun, M. N. and Levy, S. B. (1999a). "Alteration of the repressor activity of MarR, the negative regulator of the Escherichia coli marRAB locus, by multiple chemicals in vitro." J Bacteriol 181(15): 4669-72. 126 Alekshun, M. N. and Levy, S. B. (1999b). "The mar regulon: multiple resistance to antibiotics and other toxic chemicals." Trends Microbiol 7(10): 410-3. Alekshun, M. N., Levy, S. B., Mealy, T. R., Seaton, B. A. and Head, J. F. (2001). "The crystal structure of MarR, a regulator of multiple antibiotic resistance, at 2.3 A resolution." Nat Struct Biol 8(8): 710-4. Ambler, R. P., Coulson, A. F., Frere, J. M., Ghuysen, J. M., Joris, B., et al. (1991). "A standard numbering scheme for the class A beta-lactamases." Biochem J 276 (Pt 1): 269-70. Batchelor, F. R., Doyle, F. P., Nayler, J. H. C. and Rolinson, G. N. (1959). "Synthesis of penicillin: 6-amino penicillanic acid in pencillin fermentations." Nature 183: 257-258. Blumberg, P. M. and Strominger, J. L. (1972). "Five penicillin-binding components occur in Bacillus subtilis membranes." J Biol Chem 247(24): 8107-13. Blundell, T. L. (1996). "Structure-based drug design." Nature 384(6604 Suppl): 23-6. Bodey, G. P., Bolivar, R., Fainstein, V. and Jadeja, L. (1983). "Infections caused by Pseudomonas aeruginosa." Rev Infect Pis 5(2): 279-313. Boissinot, M., Huot, A., Mercier, J. and Levesque, R. C. (1989). "Development of gene probes and evolutionary relationships ofthe PSE-4 bla gene to plasmid-mediated beta-lactamases of gram-negative bacteria." MolXeJIProbes 3(2): 179-88. Boissinot, M. and Levesque, R. C. (1990). "Nucleotide sequence of the PSE-4 carbenicillinase gene and correlations with the Staphylococcus aureus PCI beta-lactamase crystal structure." J Biol Chem 265(2): 1225-30. Brtinger, A. T. (1992). "The Free R Value: a novel statistical quantity for assessing the accuracy of crystal structures." Nature 355: 472-474. 127 Briinger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., etal. (1998). "Crystallography & NMR system: A new software suite for macromolecular structure determination." Acta Crvstallogr D Biol Crvstallogr 54 ( Pt 5): 905-21. Bulychev, A., Massova, I., Miyashita, K. and Mobashery, S. (1997). "Nuances of Mechanisms and Their Implications for Evolution of the Versatile Beta-Lactamase Activity: from Biosynthetic Enzymes to Drug Resistance Factors." J Am Chem Soc 119(33): 7619-7625. Bush, K., Jacoby, G. A. and Medeiros, A. A. (1995). "A functional classification scheme for beta-lactamases and its correlation with molecular structure." Antimicrob Agents Chemother 39(6): 1211-33. Campbell, J. I., Scahill, S., Gibson, T. and Ambler, R. P. (1989). "The phototrophic bacterium Rhodopseudomonas capsulata spl08 encodes an indigenous class A beta-lactamase." Biochem J 260(3): 803-12. Carfi, A., Pares, S., Duee, E., Galleni, M., Duez, C, et al. (1995). "The 3-D structure of a zinc metallo-beta-lactamase from Bacillus cereus reveals anew type of protein fold." Embo J 14(20): 4914-21. CDC (2002a). "Staphylococcus aureus resistant to vancomycin - United States, 2002." MMWR 51: 565-567. CDC (2002b). "Vancomycin-resistant Staphylococcus aureus - Pennsylvania, 2002." MMWR 51: 902. Chain, E., Florey, H. W., Gardner, A. D., Heatley, N. G., Jennings, M. A., etal. (1940). "Penicillin as a chemotherapeutic agent." Lancet 2: 226-228. Chambers, H. F. (1997). "Methicillin resistance in staphylococci: molecular and biochemical basis and clinical implications." Clin Microbiol Rev 10(4): 781-91. 128 Chambers, H. F. (1999). "Penicillin-binding protein-mediated resistance in pneumococci and staphylococci." J Infect Pis 179 Suppl 2: S353-9. Chambers, H. F. (2001). "The changing epidemiology of Staphylococcus aureus?" Emerg Infect Pis 7(2): 178-82. Chambers, H. F., Sachdeva, M. J. and Hackbarth, C. J. (1994). "Kinetics of penicillin binding to penicillin-binding proteins of Staphylococcus aureus." Biochem J 301 ( Pt 1): 139-44. Chen, J. M., Xu, S. L., Wawrzak, Z., Basarab, G. S. and Jordan, P. B. (1998). "Structure-based design of potent inhibitors of scytalone dehydratase: displacement of a water molecule from the active site." Biochemistry 37(51): 17735-44. Chesnel, L., Zapun, A., Mouz, N., Pideberg, O. and Vernet, T. (2002). "Increase ofthe deacylation rate of PBP2x from Streptococcus pneumoniae by single point mutations mimicking the class A beta-lactamases." Eur J Biochem 269(6): 1678-83. Choury,P., Aubert, G., Szajnert,M. F., Azibi,K., Pelpech,M., etai. (1999). "Characterizationand nucleotide sequence of CARB-6, a new carbenicillin-hydrolyzing beta-lactamase from Vibrio cholerae." Antimicrob Agents Chemother 43(2): 297-301. Choury, P., Szajnert, M. F., Joly-Guillou, M. L., Azibi, K., Pelpech, M., et al. (2000). "Nucleotide sequence ofthe bla(RTG-2) (CARB-5) gene and phytogeny of a new group of carbenicillinases." Antimicrob Agents Chemother 44(4): 1070-4. Chuanchuen, R., Narasaki, C. T. and Schweizer, H. P. (2002). "The MexJK efflux pump of Pseudomonas aeruginosa requires OprM for antibiotic efflux but not for efflux of triclosan." J Bacteriol 184(18): 5036-44. Cohen, G. E. (1997). "ALIGN: a program to superimpose protein coordinates, accounting for insertions and deletions." J Appl Crvstallogr 30: 1160-1161. 129 Collaborative Computational Project, n. (1994). "Collaborative Computational Project No. 4. The CCP4 suite programs for protein crystallography." Acta Crvstallogr D Biol Crvstallogr 50: 760-763. Couto, I., de Lencastre, H., Severina, E., Kloos, W., Webster, J. A., et al. (1996). "Ubiquitous presence of a mec A homo logue in natural isolates of Staphylococcus sciuri." Microb Drug Resist 2(4): 377-91. Couto, I., Wu, S. W., Tomasz, A. and de Lencastre, H. (2003). "Development of methicillin resistance in clinical isolates of Staphylococcus sciuri by transcriptional activation of the mecA homologue native to s." J Bacteriol 185(2): 645-53. Crowfoot, D., Bunn, C. W., Rogers-Low, B. W. and Turner-Jones, A. (1949). The X-ray crystallographic investigation of the structure of penicillin. The Chemistry of Penicillin. H. T. Clarke, Johnson, J. R. and Robinson, R. Princeton, Princeton University Press: 310-367. Davies, C, White, S. W. and Nicholas, R. A. (2001). "Crystal structure of a deacylation-defective mutant of penicillin-binding protein 5 at 2.3-A resolution." J Biol Chem 276(1): 616-23. de Jonge, B. L. and Tomasz, A. (1993). "Abnormal peptidoglycan produced in a methicillin-resistant strain of Staphylococcus aureus grown in the presence of methicillin: functional role for penicillin-binding protein 2A in cell wall synthesis." Antimicrob Agents Chemother 37(2): 342-6. de LaFortelle, E. andBricogne, G. (1997). Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methiods. New York, Academic Press. 130 Dessen, A., Mouz, N., Gordon, E., Hopkins, J. andDideberg, 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(48): 45106-12. Di Guilmi, A. M., Dessen, A., Dideberg, O. and Vernet, T. (2002). "Bifunctional penicillin-binding proteins: focus on the glycosyltransferase domain and its specific inhibitor moenomycin." Curr Pharm Biotechnol 3(2): 63-75. Di Guilmi, A. M., Mouz, N., Petillot, Y., Forest, E., Dideberg, O., et al. (2000). "Deacylation kinetics analysis of Streptococcus pneumoniae penicillin-binding protein 2x mutants resistant to beta-lactam antibiotics using electrospray ionization- mass spectrometry." Anal Biochem 284(2): 240-6. Dideberg, O., Charlier, P., Wery, J. P., Dehottay, P., Dusart, J., et al. (1987). "The crystal structure ofthe beta-lactamase of Streptomyces albus G at 0.3 nm resolution." Biochem J 245(3): 911-3. Diekema, D. J., Pfaller, M. A., Schmitz, F. J., Smayevsky, J., Bell, J., et al. (2001). "Survey of infections due to Staphylococcus species: frequency of occurrence and antimicrobial susceptibility of isolates collected in the United States, Canada, Latin America, Europe, and the Western Pacific region for the SENTRY Antimicrobial Surveillance Program, 1997-1999." Clin Infect Pis 32 Suppl 2: SI 14-32. Dmitriev, B. A., Ehlers, S. and Rietschel, E. T. (1999). "Layered murein revisited: a fundamentally new concept of bacterial cell wall structure, biogenesis and function." Med Microbiol Immunol (Berl) 187(3): 173-81. Dmitriev, B. A., Toukach, F. V., Schaper, K. J., Hoist, O., Rietschel, E. T., et al. (2003). "Tertiary structure of bacterial murein: the scaffold model." J Bacteriol 185(11): 3458-68. 131 Dobashi, A. (1998). 3D Pharamaceutical Plug-in Structure Database, Group of the Department of Structural Organic Chemistry. Doublie, S. (1997). "Preparation of selenomethionyl proteins for phase determination." Methods Enzvmol 276: 523-30. Dubus, A., Ledent, P., Lamotte-Brasseur, J. and Frere, J. M. (1996). "The roles of residues Tyrl50, Glu272, and His314 in class C beta-lactamases." Proteins 25(4): 473-85. Eberhardt, C, Kuerschner, L. and Weiss, D. S. (2003). "Probing the Catalytic Activity of a Cell Division-Specific Transpeptidase In Vivo with beta-Lactams." J Bacteriol 185(13): 3726-34. el Kharroubi, A., Jacques, P., Piras, G., Van Beeumen, J., Coyette, J., et al. (1991). "The Enterococcus hirae R40 penicillin-binding protein 5 and the methicillin-resistant Staphylococcus aureus penicillin-binding protein 2' are similar." Biochem J 280 ( Pt 2): 463-9. Engh, R. A. and Huber, R. (1991). "Accurate bond and angle parameters for X-ray protein structure refinement." Acta Crvstallogr A 47: 392-400. Entenza, J. M., Hohl, P., Heinze-Krauss, I., Glauser, M. P. and Moreillon, P. (2002). "BAL9141, a novel extended-spectrum cephalosporin active against methicillin-resistant Staphylococcus aureus in treatment of experimental endocarditis." Antimicrob Agents Chemother 46(1): 171-7. Evans, G. and Pettifer, R. F. (2001). "CHOOCH: a program for deriving anomalous-scattering factors from X-ray fluorescence spectra." J. Appl. Crvstallogr. 34: 82-86. Evans, K., Adewoye, L. and Poole, K. (2001). "MexR repressor of the mexAB-oprM multidrug efflux operon of Pseudomonas aeruginosa: identification of MexR binding sites in the mexA-mexR intergenic region." J Bacteriol 183(3): 807-12. 132 Evans, K., Passador, L., Srikumar, R., Tsang, E., Nezezon, J., et al. (1998). "Influence of the MexAB-OprM multidrug efflux system on quorum sensing in Pseudomonas aeruginosa." J Bacteriol 180(20): 5443-7. Evans, K. and Poole, K. (1999). "The MexA-MexB-OprM multidrug efflux system of Pseudomonas aeruginosa is growth-phase regulated." FEMS Microbiol Lett 173(1): 35-9. Filipe, S. R. and Tomasz, A. (2000). "Inhibition of the expression of penicillin resistance in Streptococcus pneumoniae by inactivation of cell wall muropeptide branching genes." Proc Natl Acad Sci U S A 97(9): 4891-6. Fleming, A. (1928). "On the antibacterial action of cultures of a penicillium, with special reference to their use in the isolation of B. influenzae." Br. J. Exp. Pathol. 10: 226-236. Fonze, E., Vermeire, M., Nguyen-Disteche, M., Brasseur, R. and 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(31): 21853-60. Frere, J. M. (1995). "Beta-lactamases and bacterial resistance to antibiotics." Moi Microbiol 16(3): 385-95. Frere, J. M., Ghuysen, J. M. and 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(2): 343-51. Fribourg, S., Braun, I. C, Izaurralde, E. and Conti, E. (2001). "Structural basis for the recognition of a nucleoporin FG repeat by the NTF2-like domain of the TAP/pl5 mRNA nuclear export factor." Moi Cell 8(3): 645-56. Frishman, D. and Argos, P. (1995). "Knowledge-based protein secondary structure assignment." Proteins 23(4): 566-79. 133 Fujimura, T. and Murakami, K. (1997). "Increase of methicillin resistance in Staphylococcus aureus caused by deletion of a gene whose product is homologous to lytic enzymes." J Bacteriol 179(20): 6294-301. Fung-Tome, J. C, Clark, J., Minassian, B., Pucci, M., Tsai, Y. PL, et al. (2002). "In vitro and in vivo activities of a novel cephalosporin, BMS-247243, against methicillin-resistant and -susceptible staphylococci." Antimicrob Agents Chemother 46(4): 971-6. Gajiwala,K. S. andBurley,S.K. (2000). "Winged helix proteins." Curr Opin Struct Biol 10(11:110-6. Galleni, M.,Lamotte-Brasseur, J.,Raquet, X.,Dubus, A.,Monnaie,D., etal. (1995). "Theenigmatic catalytic mechanism of active-site serine beta-lactamases." Biochem Pharmacol 49(9): 1171-8. Garcia-Bustos, J. and Tomasz, A. (1990). "A biological price of antibiotic resistance: major changes in the peptidoglycan structure of penicillin-resistant pneumococci." Proc Natl Acad Sci USA 87(14): 5415-9. Gardner, A. D. (1940). "Morphological effects of penicillin on bacteria." Nature 146: 837-838. Georgopapadakou, N. H. and Liu, F. Y. (1980). "Penicillin-binding proteins in bacteria." Antimicrob Agents Chemother 18(1): 148-57. Ghuysen, J. M. (1968). "Use of bacteriolytic enzymes in determination of wall structure and their role in cell metabolism." Bacteriol Rev 32(4): Suppl:425-64. Ghuysen, J. M. and Dive, G. (1994). Biochemistry of the penicilloyl serine transferases. Bacterial Cell Wall. J. M. Ghuysen and Hakenbeck, R. Amsterdam, Elsevier Science B. V.: 103-129. Ghuysen, J. M., Frere, J. M., Leyh-Bouille, M., Nguyen-Disteche, M. and Coyette, J. (1986). "Active-site-serine D-alanyl-D-alanine-cleaving-peptidase-catalysed acyl-transfer reactions. 134 Procedures for studying the penicillin-binding proteins of bacterial plasma membranes." Biochem J 235(1): 159-65. Giesbrecht, P., Kersten, T., Maidhof, H. and Wecke, J. (1998). "Staphylococcal cell wall: morphogenesis and fatal variations in the presence of penicillin." Microbiol Moi Biol Rev 62(4): 1371-414. Gill, S. C. and von Hippel, P. H. (1989). "Calculation of protein extinction coefficients from amino acid sequence data." Anal Biochem 182(2): 319-26. Goffin, C. and Ghuysen, J. M. (1998). "Multinodular penicillin-binding proteins: an enigmatic family of orthologs and paralogs." Microbiol Moi Biol Rev 62(4): 1079-93. Golemi, D., Maveyraud, L., Vakulenko, S., Samama, J. P. and 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(25): 14280-5. Golemi-Kotra, D., Cha, J. Y., Meroueh, S. O., Vakulenko, S. B. and Mobashery, S. (2003). "Resistance to beta -Lactam Antibiotics and Its Mediation by the Sensor Domain of the Transmembrane BlaR Signaling Pathway in Staphylococcus aureus." J Biol Chem 278(20): 18419-25. Gordon, E., Mouz, N., Duee, E. and Dideberg, O. (2000). "The crystal structure ofthe penicillin-binding protein 2x from Streptococcus pneumoniae and its acyl-enzyme form: implication in drug resistance." J Moi Biol 299(2): 477-85. Gotoh, N., Tsujimoto, H., Poole, K., Yamagishi, J. and Nishino, T. (1995). "The outer membrane protein OprM of Pseudomonas aeruginosa is encoded by oprK of the mexA-mexB-oprK multidrug resistance operon." Antimicrob Agents Chemother 39(11): 2567-9. 135 Graves-Woodward, K. and 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., et al. (1997). "Site-directed mutagenesis of glutamate 166 in two beta-lactamases. Kinetic and molecular modeling studies." J Biol Chem 272(9): 5438-44. Hackbarth, C. J., Kocagoz, T., Kocagoz, S. and 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(1): 103-6. Hajduk, P. J., Gerfm, T., Boehlen, J. M., Haberli, M., Marek, D., et al. (1999). "High-throughput nuclear magnetic resonance-based screening." J Med Chem 42(13): 2315-7. Hakenbeck, R. and Coyette, J. (1998). "Resistant penicillin-binding proteins." Cell Moi Life Sci 54(4): 332-40. Hancock, R. E. (1998). "Resistance mechanisms in Pseudomonas aeruginosa and other nonfermentative gram-negative bacteria." Clin Infect Pis 27 Suppl 1: S93-9. Hardt, K., Joris, B., Lepage, S., Brasseur, R., Lampen, J. O., et al. (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 four-alpha-helix bundle." Moi Microbiol 23(5): 935-44. Hartman,B. J. and Tomasz, A. (1984). "Low-affinity penicillin-binding protein associated with beta-lactam resistance in Staphylococcus aureus." J Bacteriol 158(2): 513-6. He, X., Chow, D., Martick, M. M. and Garcia, K. C. (2001). "Allosteric activation of a spring-loaded natriuretic peptide receptor dimer by hormone." Science 293(5535): 1657-62. 136 Hedstrom, L. (1996). "Trypsin: a case study in the structural determinants of enzyme specificity." Biol Chem 377(7-8): 465-70. Henderson, R. (1970). "Structure of crystalline alpha-chymotrypsin. IV. The structure of indoleacryloyl-alpha-chyotrypsin and its relevance to the hydro lytic mechanism of the enzyme." J Moi Biol 54(2): 341-54. Hendrickson, W. A. (1991).. "Determination of macromolecular structures from anomalous diffraction of synchrotron radiation." Science 254(5028): 51-8. Henze, U. U. and Berger-Bachi, B. (1995). "Staphylococcus aureus penicillin-binding protein 4 and intrinsic beta-lactam resistance." Antimicrob Agents Chemother 39(11): 2415-22. Herzberg, O. (1991). "Refined crystal structure of beta-lactamase from Staphylococcus aureus PCI at 2.0 A resolution." J Moi Biol 217(4): 701-19. Herzberg, O. and Moult, J. (1987). "Bacterial resistance to beta-lactam antibiotics: crystal structure of beta-lactamase from Staphylococcus aureus PCI at 2.5 A resolution." Science 236(4802): 694-701. Hiramatsu, K. (1995). "Molecular evolution of MRSA." Microbiol Immunol 39(8): 531-43. Hiramatsu, K., Aritaka, N., Hanaki, H., Kawasaki, S., Hosoda, Y., et al. (1997). "Dissemination in Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin." Lancet 350(9092): 1670-3. Hiramatsu, K., Asada, K., Suzuki, E., Okonogi, K. and Yokota, T. (1992a). "Molecular cloning and nucleotide sequence determination ofthe regulator region of mecA gene in methicillin-resistant Staphylococcus aureus (MRSA)." FEBS Lett 298(2-3): 133-6. Hiramatsu, K., Cui, L., Kuroda, M. and Ito, T. (2001). "The emergence and evolution of methicillin-resistant Staphylococcus aureus." Trends Microbiol 9(10): 486-93. 137 Hiramatsu, K . , Kihara, H . and Yokota, T. (1992b). "Analysis of borderline-resistant strains of methicillin-resistant Staphylococcus aureus using polymerase chain reaction." Microbiol Immunol 36(5): 445-53. Hiramatsu, K . , Okuma, K . , M a , X . X . , Yamamoto, M . , Hor i , S., et al. (2002). "New trends in Staphylococcus aureus infections: glycopeptide resistance in hospital and methicillin resistance in the community." Curr Opin Infect P i s 15(4): 407-13. Holm, L . and Sander, C. (1993). "Protein structure comparison by alignment of distance matrices." J M o i B i o l 233(1): 123-38. Holm, L . and Sander, C. (1995). "Pa l i : a network tool for protein structure comparison." Trends Biochem Sci 20(11): 478-80. Holtje, J. V . (1998). "Growth of the stress-bearing and shape-maintaining murein sacculus of Escherichia coli ." Microbiol M o i B i o l Rev 62(1): 181-203. Huovinen, P. and Jacoby, G . A . (1991). "Sequence of the PSE-1 beta-lactamase gene." Antimicrob Agents Chemother 35(11): 2428-30. Hypercube (1995). HyperChem. Gainesville, Hypercube Inc. Ibuka, A . , Taguchi, A . , Ishiguro, M . , Fushinobu, S., Ishii, Y . , et al. (1999). "Crystal structure of the E166A mutant of extended-spectrum beta-lactamase Toho-1 at 1.8 A resolution." J M o i B i o l 285(5): 2079-87. Ishiguro, M . and Imajo, S. (1996). "Modeling study on a hydrolytic mechanism of class A beta-lactamases." J M e d Chem 39(11): 2207-18. Ito, T., Katayama, Y . and Hiramatsu, K . (1999). "Cloning and nucleotide sequence determination of the entire mec P N A of pre-methicillin-resistant Staphylococcus aureus N315." Antimicrob Agents Chemother 43(6): 1449-58. 138 Ito, Y . and Hirano, T, (1997). "Carbenicillin-hydrolysing penicillinase mediated by a plasmid of Proteus mirabilis and its relationship to the PSE-type enzymes o f Pseudomonas aeruginosa." J A p p l Microbiol 83(2): 175-80. Jancarik, J. and K i m , S. H . (1991). "Sparse matrix sampling: A screening method for crystallization of proteins." J A p p l Crvstallogr 24: 409-411. Jensen, S. E. , Westlake, D . W . and Wolfe, S. (1982). "Cyclizationofdelta-(L-alpha-aminoadipyl)-L-cysteinyl-D-valine to penicillins by cell-free extracts of Streptomyces clavuligerus." J Antibiot (Tokyo) 35(4): 483-90. Jevons, M . P. (1961). ""Celbenin" resistant staphylococci." Br . Med. J. i: 124-125. Jones, T. A . , Zou, J . -Y. , Cowan, S. W . and Kjeldgaard, M . (1991). "Improved methods for building protein models in electron density maps and the location of errors in these models." Acta Crvstallogr A 47: 110-119. Kabsch, W . (1976). " A solution for the best rotation to relate two sets of vectors." Acta Crvstallogr A 32: 922-923. Kammann, M . , Laufs, J., Schell, J. and Gronenborn, B . (1989). "Rapid insertional mutagenesis of D N A by polymerase chain reaction (PCR)." Nucleic Acids Res 17(13): 5404. Katayama, Y . , Ito, T. and Hiramatsu, K . (2000). " A new class of genetic element, staphylococcus cassette chromosome mec, encodes methicillin resistance in Staphylococcus aureus." Antimicrob Agents Chemother 44(6): 1549-55. Kel ly , J. A . , Dideberg, O., Charlier, P., Wery, J. P., Libert, M . , et al. (1986). "On the origin of bacterial resistance to penicillin: comparison of a beta-lactamase and a penicillin target." Science 231(4744): 1429-31. 139 Kelly, J. A., Knox, J. R., Moews, P. C, Hite, G. J., Bartolone, J. B., et al. (1985). "2.8-A Structure of penicillin-sensitive D-alanyl carboxypeptidase-transpeptidase from Streptomyces R61 and complexes with beta-lactams." J Biol Chem 260(10): 6449-58. Kelly, J. A. and Kuzin, A. P. (1995). "The refined crystallographic structure of a DD-peptidase penicillin-target enzyme at 1.6 A resolution." J Moi Biol 254(2): 223-36. Kelly, J. A., Moews, P. C, Knox, J. R., Frere, J. M. and Ghuysen, J. M. (1982). "Penicillin target enzyme and the antibiotic binding site." Science 218(4571): 479-81. Kirby, W. M. M. (1944). "Extraction of a highly potent penicillin inactivator from penicillin resistant staphyolococci." Science 99(2579): 452-453. Kleywegt, G. J. (1995). "Dictionaries for Heteros." CCP4/ESF-EACBM Newsletter on Protein Crystallography 31: 45-50. Knox, J. R. and Moews, P. C. (1991). "Beta-lactamase of Bacillus licheniformis 749/C. Refinement at 2 A resolution and analysis of hydration." J Moi Biol 220(2): 435-55. Koch, A. L. (1998). "Orientation of the peptidoglycan chains in the sacculus of Escherichia coli." Res Microbiol 149(10): 689-701. Kohler, T., Michea-Hamzehpour, M., Henze, U., Gotoh, N., Curty, L. K., et al. (1997). "Characterization of MexE-MexF-OprN, a positively regulated multidrug efflux system of Pseudomonas aeruginosa." Moi Microbiol 23(2): 345-54. Kondo, N., Kuwahara-Arai, K., Kuroda-Murakami, H., Tateda-Suzuki, E. andHiramatsu, K. (2001). "Eagle-type methicillin resistance: new phenotype of high methicillin resistance under mec regulator gene control." Antimicrob Agents Chemother 45(3): 815-24. 140 Koronakis, V., Li, J., Koronakis, E. and Stauffer, K. (1997). "Structure of TolC, the outer membrane component of the bacterial type I efflux system, derived from two-dimensional crystals." Moi Microbiol 23(3): 617-26. Kraulis, P. J. (1991). "MOLSCRTPT: a program to produce both detailed and schematic plots of protein structures." J Appl Crvstallogr 24: 946-950. Kuroda, M., Ohta, T., Uchiyama, I., Baba, T., Yuzawa, H., et al. (2001). "Whole genome sequencing of meticillin-resistant Staphylococcus aureus." Lancet 357(9264): 1225-40. Kuzin, A. P., Nukaga, M., Nukaga, Y., Hujer, A. M., Bonomo, R. A., et al. (1999). "Structure of the SHV-1 beta-lactamase." Biochemistry 38(18): 5720-7. Lachapelle, J., Dufresne, J. and Levesque, R. C. (1991). "Characterization of the blaCARB-3 gene encoding the carbenicillinase-3 beta-lactamase of Pseudomonas aeruginosa." Gene 102(1): 7-12. Lamotte-Brasseur, J., Dive, G., Dideberg, O., Charlier, P., Frere, J. M., et al. (1991). "Mechanism of acyl transfer by the class A serine beta-lactamase of Streptomyces albus G." Biochem J 279 (Pt 1): 213-21. Laskowski, R. A., MacArthur, M. W., Moss, D. S. and Thornton, J. M. (1993). "PROCHECK: a program to check the stereochemical quality of protein structures." J Appl Crvstallogr 26: 283-291. Lenfant, F., Labia, R. and Masson, J. M. (1991). "Replacement of lysine 234 affects transition state stabilization in the active site of beta-lactamase TEM1." J Biol Chem 266(26): 17187-94. Lenfant, F., Petit, A., Labia, R., Maveyraud, L., Samama, J. P., et al. (1993). "Site-directed mutagenesis of beta-lactamase TEM-1. Investigating the potential role of specific residues on the activity of Pseudomonas-specific enzymes." Eur J Biochem 217(3): 939-46. 141 Levy, S. B. (2001). "Antibiotic resistance: consequences of inaction." Clin Infect Pis 33 Suppl 3: SI 24-9. Li, X. Z., Zhang, L., Srikumar, R. and Poole, K. (1998). "Beta-lactamase inhibitors are substrates for the multidrug efflux pumps of Pseudomonas aeruginosa." Antimicrob Agents Chemother 42(2): 399-403. Ligozzi, M., Pittaluga, F. and Fontana, R. (1996). "Modification of penicillin-binding protein 5 associated with high-level ampicillin resistance in Enterococcus faecium." Antimicrob Agents Chemother 40(2): 354-7. Lim, P., Poole, K. and Strynadka, N. C. (2002). "Crystal structure of the MexR repressor of the mexRAB-oprM multidrug efflux operon of Pseudomonas aeruginosa." J Biol Chem 277(32): 29253-9. Lim, P., Sanschagrin, F., Passmore, L., Pe Castro, L., Levesque, R. C, et al. (2001). "Insights into the molecular basis for the carbenicillinase activity of PSE-4 beta-lactamase from crystallographic and kinetic studies." Biochemistry 40(2): 395-402. Lim, P. and Strynadka, N. C. (2002). "Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus." Nat Struct Biol 9(11): 870-6. Lobkovsky, E., Billings, E. M., Moews, P. C, Rahil, J., Pratt, R. F., et al. (1994). "Crystallographic structure of a phosphonate derivative of the Enterobacter cloacae P99 cephalosporinase: mechanistic interpretation of a beta-lactamase transition-state analog." Biochemistry 33(22): 6762-72. Lowbury, E. J., Lilly, H. A., Kidson, A., Ayliffe, G. A. and Jones, R. J. (1969). "Sensitivity of Pseudomonas aeruginosa to antibiotics: emergence of strains highly resistant to carbenicillin." Lancet 2(7618): 448-52. 142 Lu, W. P., Kincaid, E., Sun, Y. and Bauer, M. D. (2001). "Kinetics of beta-lactam interactions with penicillin-susceptible and -resistant penicillin-binding protein 2x proteins from Streptococcus pneumoniae. Involvement of acylation and deacylation in beta-lactam resistance." J Biol Chem 276(34): 31494-501. Lu, W. P., Sun, Y.,Bauer,M. D., Paule, S.,Koenigs, P. M., etal. (1999). "Penicillin-bindingprotein 2a from methicillin-resistant Staphylococcus aureus: kinetic characterization of its interactions with beta-lactams using electrospray mass spectrometry." Biochemistry 38(20): 6537-46. Lundqvist, T., Rice, J., Hodge, C. N., Basarab, G. S., Pierce, J., et al. (1994). "Crystal structure of scytalone dehydratase—a disease determinant of the rice pathogen, Magnaporthe grisea." Structure 2(10): 937-44. Majiduddin, F. K., Materon, I. C. and Palzkill, T. G. (2002). "Molecular analysis of beta-lactamase structure and function." Int J Med Microbiol 292(2): 127-37. Marrec-Fairley, M., Piette, A., Gallet, X., Brasseur, R., Hara, H., et al. (2000). "Differential functionalities of amphiphilic peptide segments of the cell-septation penicillin-binding protein 3 of Escherichia coli." Moi Microbiol 37(5): 1019-31. Martin, R. G., Nyantakyi, P. S. and Rosner, J. L. (1995). "Regulation of the multiple antibiotic resistance (mar) regulon by marORA sequences in Escherichia coli." J Bacteriol 177(14): 4176-8. Massova, I. and Mobashery, S. (1998). "Kinship and diversification of bacterial penicillin-binding proteins and beta-lactamases." Antimicrob Agents Chemother 42(1): 1-17. Matagne, A. and Frere, J. M. (1995). "Contribution of mutant analysis to the understanding of enzyme catalysis: the case of class A beta-lactamases." Biochim Biophys Acta 1246(2): 109-27. 143 Matagne, A., Lamotte-Brasseur, J. and Frere, J. M. (1998). "Catalytic properties of class A beta-lactamases: efficiency and diversity." Biochem J 330 ( Pt 2): 581-98. Matsuhashi, M., Song, M. D., Ishino, F., Wachi, M., Doi, M., et al. (1986). "Molecular cloning of the gene of a penicillin-binding protein supposed to cause high resistance to beta-lactam antibiotics in Staphylococcus aureus." J Bacteriol 167(3): 975-80. Matthews, B. W. (1968). "Solvent content of protein crystals." J Moi Biol 33(2): 491-7. Matthews, P. R., Reed, K. C. and Stewart, P. R. (1987). "The cloning of chromosomal DNA associated with methicillin and other resistances in Staphylococcus aureus." J Gen Microbiol 133 (Pt7): 1919-29. Maveyraud, L., Golemi, D., Kotra, L. P., Tranier, S., Vakulenko, S., etai. (2000). "Insights into class D beta-lactamases are revealed by the crystal structure ofthe OXA10 enzyme from Pseudomonas aeruginosa." Structure Fold Des 8(12): 1289-98. Mazzariol, A., Cornaglia, G. and Nikaido, H. (2000). "Contributions of the AmpC beta-lactamase and the AcrAB multidrug efflux system in intrinsic resistance of Escherichia coli K-12 to beta-lactams." Antimicrob Agents Chemother 44(5): 1387-90. McDonough, M. A., Anderson, J. W., Silvaggi, N. R., Pratt, R. F., Knox, J. R., et al. (2002). "Structures of two kinetic intermediates reveal species specificity of penicillin-binding proteins." J Moi Biol 322(1): 111-22. McDougal, L. K. and Thornsberry, C. (1986). "The role of beta-lactamase in staphylococcal resistance to penicillinase-resistant penicillins and cephalosporins." J Clin Microbiol 23(5): 832-9. McPherson, A. (1990). "Current approaches to macromolecular crystallization." Eur J Biochem 189(1): 1-23. 144 McRee, D. E. (1999). "XtalView/Xfit—A versatile program for manipulating atomic coordinates and electron density." J Struct Biol 125(2-3): 156-65. Medeiros, A. A. (1997). "Evolution and dissemination of beta-lactamases accelerated by generations of beta-lactam antibiotics." Clin Infect Pis 24 Suppl 1: SI 9-45. Merritt, E. A. and Murphy, M. E. P. (1994). "Raster3D version 2.0. A program for photorealistic molecular graphics." Acta Crvstallogr D 50: 869-873. Miller, P. F. and Sulavik, M. C. (1996). "Overlaps and parallels in the regulation of intrinsic multiple-antibiotic resistance in Escherichia coli." Moi Microbiol 21(3): 441-8. Miller, R., Gallo, S. M., Khalak, H. G. and Weeks, C. M. (1994). "SnB: crystal structure determination via Shake-and-Bake." J Appl Cryst 27: 613-621. Minasov, G., Wang, X. and Shoichet, B. K. (2002). "An ultrahigh resolution structure of TEM-1 beta-lactamase suggests a role for Glu 166 as the general base in acylation." J Am Chem Soc 124(19): 5333-40. Mine, T., Morita, Y., Kataoka, A., Mizushima, T. and Tsuchiya, T. (1999). "Expression in Escherichia coli of a new multidrug efflux pump, MexXY, from Pseudomonas aeruginosa." Antimicrob Agents Chemother 43(2): 415-7. Mouz, N., Gordon, E., Di Guilmi, A. M., Petit, I., Petillot, Y., et al. (1998). "Identification of a structural determinant for resistance to beta-lactam antibiotics in Gram-positive bacteria." Proc Natl Acad Sci U S A 95(23): 13403-6. Murakami, S., Nakashima, R., Yamashita, E. and Yamaguchi, A. (2002). "Crystal structure of bacterial multidrug efflux transporter AcrB." Nature 419(6907): 587-93. Nanninga,N. (1991). "Cell division and peptidoglycan assembly in Escherichia coli." Moi Microbiol 5(4): 791-5. 145 Navaza, J. (1994). "AMoRe: an automated package for molecular replacement." Acta Crvstallogr A 50: 157-163. Navaza, J. (2001). "Implementation of molecular replacement in AMoRe." Acta Crvstallogr D Biol Crvstallogr 57fPt 10): 1367-72. Neu, H. C. (1992). "The crisis in antibiotic resistance." Science 257(5073): 1064-73. Newsom, S. W. (1969). "Carbenicillin-resistant pseudomonas." Lancet 2(7630): 1141. Newsom, S. W., Sykes, R. B. and Richmond, M. H. (1970). "Detection of a beta-lactamase markedly active against carbenicillin in a strain of Pseudomonas aeruginosa." J Bacteriol 101(3): 1079-80. Nicholls, A., Bharadwaj, R. and Honig, B. (1993). "GRASP: graphical representation and analysis of surface properties." Biophys J 64: A166-A166. Nikaido, H. (1996). "Multidrug efflux pumps of gram-negative bacteria." J Bacteriol 178(20): 5853-9. Nikaido, H. (1998). "Antibiotic resistance caused by gram-negative multidrug efflux pumps." Clin Infect Pis 27 Suppl 1: S32-41. Nikaido, H. (2001). "Preventing drug access to targets: cell surface permeability barriers and active efflux in bacteria." Semin Cell Dev Biol 12(3): 215-23. Nukaga, M., Mayama, K., Hujer, A. M., Bonomo, R. A. and Knox, J. R. (2003). "Ultrahigh resolution structure of a class A beta-lactamase: on the mechanism and specificity of the extended-spectrum SHV-2 enzyme." J Moi Biol 328(1): 289-301. Oefner, C, D'Arcy, A., Daly, J. J., Gubernator, K., Charnas, R. L., et al. (1990). "Refined crystal structure of beta-lactamase from Citrobacter freundii indicates a mechanism for beta-lactam hydrolysis." Nature 343(6255): 284-8. 146 Orth, P., Schnappinger, D., Hillen, W., Saenger, W. and Hinrichs, W. (2000). "Structural basis of gene regulation by the tetracycline inducible Tet repressor-operator system." Nat Struct Biol 7(3): 215-9. Otwinowski, Z. (1993). Oscillation data reduction program. Proceedings of the CCP4 Study Weekend: Data Collection and Processing. S. L., N., I. and S, B. Warrington, Daresbury Laboratory: 56-62. Paetzel, M. and Dalbey, R. E. (1997). "Catalytic hydroxyl/amine dyads within serine proteases." Trends Biochem Sci 22(1): 28-31. Paetzel, M., Danel, F., de Castro, L., Mosimann, S. C, Page, M. G., et al. (2000). "Crystal structure of the class D beta-lactamase OXA-10." Nat Struct Biol 7(10): 918-25. Pannu, N. S. and Read, R. J. (1996). "Improved structure refinement through maximum likelihood." Acta Crvstallogr A 52: 659-668. Pares, S., Mouz, N., Petillot, Y., Hakenbeck, R. and Dideberg, O. (1996). "X-ray structure of Streptococcus pneumoniae PBP2x, a primary penicillin target enzyme." Nat Struct Biol 3(3): 284-9. Parkinson, G., Wilson, C, Gunasekera, A., Ebright, Y. W., Ebright, R. E., et al. (1996). "Structure of the CAP-DNA complex at 2.5 angstroms resolution: a complete picture of the protein-DNA interface." J Moi Biol 260(3): 395-408. Pearson, J. P., Van Delden, C. and Iglewski, B. H. (1999). "Active efflux and diffusion are involved in transport of Pseudomonas aeruginosa cell-to-cell signals." J Bacteriol 181(4): 1203-10. Petrounia, I. P. and Arnold, F. H. (2000). "Designed evolution of enzymatic properties." Curr Opin Biotechnol 11(4): 325-30. 147 Philippon, A., Dusart, J., Joris, B. and Frere, J. M. (1998). "The diversity, structure and regulation of beta-lactamases." Cell Moi Life Sci 54(4): 341-6. Pinho, M. G., de Lencastre, H. and Tomasz, A. (2001). "An acquired and a native penicillin-binding protein cooperate in building the cell wall of drug-resistant staphylococci." Proc Natl Acad Sci USA 98(19): 10886-91. Piras, G., Raze, D., el Kharroubi, A., Hastir, D., Englebert, S., et al. (1993). "Cloning and sequencing of the low-affinity penicillin-binding protein 3r-encoding gene of Enterococcus hirae S185: modular design and structural organization of the protein." J Bacteriol 175(10): 2844-52. Poole, K. (2001). "Multidrug efflux pumps and antimicrobial resistance in Pseudomonas aeruginosa and related organisms." J Moi Microbiol Biotechnol 3(2): 255-64. Poole, K., Gotoh, N., Tsujimoto, PL, Zhao, Q., Wada, A., et al. (1996a). "Overexpression of the mexC-mexD-oprJ efflux operon in nfxB-type multidrug-resistant strains of Pseudomonas aeruginosa." Moi Microbiol 21(4): 713-24. Poole, K., Krebes, K., McNally, C. and Neshat, S. (1993). "Multiple antibiotic resistance in Pseudomonas aeruginosa: evidence for involvement of an efflux operon." J Bacteriol 175(22): 7363-72. Poole, K., Tetro, K., Zhao, Q., Neshat, S., Heinrichs, D. E., et al. (1996b). "Expression of the multidrug resistance operon mexA-mexB-oprM in Pseudomonas aeruginosa: mexR encodes a regulator of operon expression." Antimicrob Agents Chemother 40(9): 2021-8. Powers, R. A., Morandi, F. and Shoichet, B. K. (2002). "Structure-based discovery of a novel, noncovalent inhibitor of AmpC beta-lactamase." Structure (Camb) 10(7): 1013-23. 148 Queener, S. F. (1986). History and origins of beta-lactam antibiotics. Beta-Lactam Antibiotics For Clinical Use. S. F. Queener, Webber, J. A. and Queener, S. W. New York, Marcel Dekker, Inc. 3-15. Ramakrishnan, C. and Ramachandran, G. N. (1965). "Stereochemical criteria for polypeptide and protein chain conformations. U. Allowed conformations for a pair of peptide units." Biophys J 5(6): 909-33. Ravelli, R. B. G., Sweet, R. M., Skinner, J. M., Duisenberg, A. J. M. and Kroon, J. (1997). "STRATEGY: a program to optimize the starting spindle angle and scan range for X-ray data collection." J Appl Crvst 30: 551-554. Raze, D., Dardenne, O., Hallut, S., Martinez-Bueno, M., Coyette, J., et al. (1998). "The gene encoding the low-affinity penicillin-binding protein 3r in Enterococcus hirae S185R is borne on a plasmid carrying other antibiotic resistance determinants." Antimicrob Agents Chemother 42(3): 534-9. Read, R. J. (1986). "Improved Fourier coefficients for maps using phases from partial structures with errors." Acta Crvstallogr A 42: 140-149. Reynolds, P. E. and Fuller, C. (1986). "Methicillin-resistant strains of Staphylococcus aureus; presence of identical additional penicillin-binding protein in all strains examined." FEMS Microbiol Lett 33: 251-254. Rhazi,N., Charlier, P.,Dehareng,D., Engher, D., Vermeire, M., etai. (2003). "CatalyticMechanism ofthe Streptomyces KI 5 dd-Transpeptidase/Penicillin-Binding Protein Probed by Site-Directed Mutagenesis and Structural AnalysisQ." Biochemistry 42(10): 2895-906. Rolinson, G. N. (1998). "Forty years of beta-lactam research." J Antimicrob Chemother 41(6): 589-603. 149 Rolinson, G. N., Stevens, S., R., B. F., Wood, J. C. and Chain, E. B. (1960). "Bacteriological studies on a new penicillin." Lancet 2: 564. Rolinson, G. N. and Sutherland, R. (1967). "Carbenicillin, a new semisynthetic penicillin active against Pseudomonas aeruginosa." Antimicrobial Agents Chemother 7: 609-13. Ryffel, C, Tesch, W., Birch-Machin, I., Reynolds, P. E., Barberis-Maino, L., et al. (1990). "Sequence comparison of mecA genes isolated from methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis." Gene 94(1): 137-8. Saito, K., Yoneyama, H. and Nakae, T. (1999). "nalB-type mutations causing the overexpression of the MexAB-OprM efflux pump are located in the mexR gene of the Pseudomonas aeruginosa chromosome." FEMS Microbiol Lett 179(1): 67-72. Sakurai, Y., Tsukamoto, K. and Sawai, T. (1991). "Nucleotide sequence and characterization of a carbenicillin-hydrolyzingpenicillinase gene from Proteus mirabilis." J Bacteriol 173(21): 7038-41. Sambrook, J. and Russel, D. W. (2001). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, Cold Spring Harbor Laboratory Press. Samraoui, B., Sutton, B. J., Todd, R. J., Artymiuk, P. J., Waley, S. G., et al. (1986). "Tertiary structural similarity between a class A beta-lactamase and a penicillin-sensitive D-alanyl carboxypeptidase-transpeptidase." Nature 320(6060): 378-80. Sanschagrin, F., Bejaoui, N. and Levesque, R. C. (1998). "Structure of CARB-4 and AER-1 carbenicillin-hydrolyzing beta-lactamases." Antimicrob Agents Chemother 42(8): 1966-72. Sarkar, G. and Sommer, S. S. (1990). "The "megaprimer" method of site-directed mutagenesis." Biotechniques 8(4): 404-7. 150 Sasaki, S. (1989). KEK. Report 88-14. Tsukuba, Japan, National Laboratory for High Energy Physics. Sauvage, E., Kerff, F., Fonze, E., Herman, R., Schoot, B., et al. (2002). "The 2.4-A crystal structure of the penicillin-resistant penicillin-binding protein PBP5fm from Enterococcus faecium in complex with benzylpenicillin." Cell Moi Life Sci 59(7): 1223-32. Savoie, A., Sanschagrin, F., Palzkill, T., Voyer, N. and Levesque, R. C. (2000). "Structure-function analysis of alpha-helix H4 using PSE-4 as a model enzyme representative of class A beta-lactamases." Protein Eng 13(4): 267-74. Schultz, S. C, Shields, G. C. and Steitz, T. A. (1991). "Crystal structure of a CAP-DNA complex: the DNA is bent by 90 degrees." Science 253(5023): 1001-7. Schweizer, H. P. (2001). "Triclosan: a widely used biocide and its link to antibiotics." FEMS Microbiol Lett 202(1): 1-7. Sheehan, J. C. and Henery-Logan, K. R. (1959). "A general synthesis of the penicillins." J. Am. Chem. Soc. 81: 5838-5839. Signoretto, C, Boaretti, M. and Canepari, P. (1994). "Cloning, sequencing and expression in Escherichia coli of the low-affinity penicillin binding protein of Enterococcus faecalis." FEMS Microbiol Lett 123(1-2): 99-106. Silvaggi, N. R., Anderson, J. W., Brinsmade, S. R., Pratt, R. F. and 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(5): 1199-208. Smith, J. M., Dowson, C. G. and Spratt, B. G. (1991). "Localized sex in bacteria." Nature 349(6304): 29-31. 151 Spratt, B. G. (1988). "Hybrid penicillin-binding proteins in penicillin-resistant strains of Neisseria gonorrhoeae." Nature 332(6160): 173-6. Spratt, B. G. and Pardee, A. B. (1975). "Penicillin-binding proteins and cell shape in E. coli." Nature 254(5500): 516-7. Srikumar, R., Kon, T., Gotoh, N. and Poole, K. (1998). "Expression of Pseudomonas aeruginosa multidrug efflux pumps MexA-MexB-OprM and MexC-MexD-OprJ in a multidrug-sensitive Escherichia coli strain." Antimicrob Agents Chemother 42(1): 65-71. Srikumar, R., Li, X. Z. and Poole, K. (1997). "Inner membrane efflux components are responsible for beta-lactam specificity of multidrug efflux pumps in Pseudomonas aeruginosa." J Bacteriol 179(24): 7875-81. Srikumar, R., Paul, C. J. and Poole, K. (2000). "Influence of mutations in the mexR repressor gene on expression of the MexA-MexB-oprM multidrug efflux system of Pseudomonas aeruginosa." J Bacteriol 182(5): 1410-4. Stover, C. K., Pham, X. Q., Erwin, A. L., Mizoguchi, S. D., Warrener, P., et al. (2000). "Complete genome sequence of Pseudomonas aeruginosa PA01, an opportunistic pathogen." Nature 406(6799): 959-64. Strynadka, N. C, Adachi, H., Jensen, S. E., Johns, K., Sielecki, A., et al. (1992). "Molecular structure ofthe acyl-enzyme intermediate in beta-lactam hydrolysis at 1.7 A resolution." Nature 359(6397): 700-5. Strynadka, N. C, Martin, R., Jensen, S. E., Gold, M. and Jones, J. B. (1996). "Structure-based design of a potent transition state analogue for TEM-1 beta-lactamase." Nat Struct Biol 3(8): 688-95. 152 Sulavik, M. C, Gambino, L. F. and Miller, P. F. (1995). "The MarR repressor of the multiple antibiotic resistance (mar) operon in Escherichia coli: prototypic member of a family of bacterial regulatory proteins involved in sensing phenolic compounds." Moi Med 1(4): 436-46. Sun, Y., Bauer, M. D. and Lu, W. (1998). "Identification of the active site serine of penicillin-binding protein 2a from methicillin-resistant Staphylococcus aureus by electrospray mass spectrometry." J Mass Spectrom 33(10): 1009-16. Suzuki, E., Hiramatsu, K. and Yokota, T. (1992). "Survey of methicillin-resistant clinical strains of coagulase-negative staphylococci for mec A gene distribution." Antimicrob Agents Chemother 36(2): 429-34. Swaren, P., Maveyraud, L., Guillet, V., Masson, J. M., Mourey, L., et al. (1995). "Electrostatic analysis of TEM1 beta-lactamase: effect of substrate binding, steep potential gradients and consequences of site-directed mutations." Structure 3(6): 603-13. Swaren, P., Maveyraud, L., Raquet, X., Cabantous, S., Duez, C, etal. (1998). "X-ray analysis of the NMC-A beta-lactamase at 1.64-A resolution, a class A carbapenemase with broad substrate specificity." J Biol Chem 273(41): 26714-21. Tan, S., Hunziker, Y., Pellegrini, L. and Richmond, T. J. (2000). "Crystallization of the yeast MATalpha2/MCMl/DNA ternary complex: general methods and principles for protein/DNA cocrystallization." J Moi Biol 297(4): 947-59. Terrak, M., Ghosh, T. K., van Heijenoort, J., Van Beeumen, J., Lampilas, M., et al. (1999). "The catalytic, glycosyl transferase and acyl transferase modules of the cell wall peptidoglycan-polymerizing penicillin-binding protein lb of Escherichia coli." Moi Microbiol 34(2): 350-64. Terwilliger, T. C. (2000). "Maximum-likelihood density modification." Acta Crvstallogr D Biol Crvstallogr 56 ( Pt 8): 965-72. 153 Terwilliger, T. C. and Berendzen, J. (1999). "Automated MAD and MIR structure solution." Acta Crvstallogr D Biol Crvstallogr 55 (Pt 4): 849-61. Tipper, D. J. and 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(4): 1133-41. Tobin, M. B., Gustafsson, C. and Huisman, G. W. (2000). "Directed evolution: the 'rational basis for 'irrational' design." Curr Opin Struct Biol 10(4): 421-7. Tomasz, A., Drugeon, H. B., de Lencastre, H. M., Jabes, D., McDougall, L., et al. (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 penicillin-binding capacity." Antimicrob Agents Chemother 33(11): 1869-74. Toney, J. H., Hammond, G. G., Leiting, B., Pryor, K. D., Wu, J. K., et al. (1998). "Soluble penicillin-binding protein 2a: beta-lactam binding and inhibition by non-beta-lactams using a 96-well format." Anal Biochem 255(1): 113-9. Utsui, Y. and Yokota, T. (1985). "Role of an altered penicillin-binding protein in methicillin- and cephem-resistant Staphylococcus aureus." Antimicrob Agents Chemother 28(3): 397-403. van Aalten, D. M., DiRusso, C. C. and Knudsen, J. (2001). "The structural basis of acyl coenzyme A-dependent regulation ofthe transcription factor FadR." Embo J 20(8): 2041-50. vanHeijenoort, J. (1998). "Assembly of the monomer unit of bacterial peptidoglycan." Cell Moi Life Sd 54(4): 300-4. van Heijenoort, J. (2001). "Formation of the glycan chains in the synthesis of bacterial peptidoglycan." Glycobiology 11(3): 25R-36R. 154 Vollmer, W., von Rechenberg, M. and 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(10): 6726-34. Walsh, C. (2000). "Molecular mechanisms that confer antibacterial drug resistance." Nature 406(6797): 775-81. Walsh, T. R. and Howe, R. A. (2002). "The prevalence and mechanisms of vancomycin resistance in Staphylococcus aureus." Annu Rev Microbiol 56: 657-75. Ward, J. B. (1984). "Biosynthesis of peptidoglycan: points of attack by wall inhibitors." Pharmacol Ther25(3): 327-69. Weeks, C. M. and Miller, R. (1999). "The design and implementation of SnB v2.0." J Appl Crvst 32: 120-124. Wu, C. Y., Alborn, W. E., Jr., Flokowitsch, J. E., Hoskins, J., Unal, S., et al. (1994). "Site-directed mutagenesis of the mecA gene from a methicillin-resistant strain of Staphylococcus aureus." J Bacteriol 176(2): 443-9. Wu, C. Y., Hoskins, J., Blaszczak, L. C, Preston, D. A. and Skatrud, P. L. (1992). "Construction of a water-soluble form of penicillin-binding protein 2a from a methicillin-resistant Staphylococcus aureus isolate." Antimicrob Agents Chemother 36(3): 533-9. Wu, S., Piscitelli, C, de Lencastre, H. and 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(4): 435-41. Wu, S. W., de Lencastre, H. and Tomasz, A. (2001). "Recruitment of the mecA gene homologue of Staphylococcus sciuri into a resistance determinant and expression of the resistant phenotype in Staphylococcus aureus." J Bacteriol 183(8): 2417-24. 155 Wu, Z. R., Ebrahimian, S., Zawrotny, M. E., Thornburg, L. D., Perez-Alvarado, G. C., etai. (1997). "Solution structure of 3-oxo-delta5-steroid isomerase." Science 276(5311): 415-8. Yap, K. (1998). ESfTERHLX program. Toronto, University of Toronto. Yocum, R. R., Waxman, D. J., Rasmussen, J. R. and Strominger, J. L. (1979). "Mechanism of penicillin action: penicillin and substrate bind covalently to the same active site serine in two bacterial D-alanine carboxypeptidases." Proc Natl Acad Sci U S A 76(6): 2730-4. Zaccolo, M. and Gherardi, E. (1999). "The effect of high-frequency random mutagenesis on in vitro protein evolution: a study on TEM-1 beta-lactamase." J Moi Biol 285(2): 775-83. Zawadzke, L. E., Chen, C. C, Banerjee, S., Li, Z., Wasch, S., et al. (1996). "Elimination of the hydrolytic water molecule in a class A beta-lactamase mutant: crystal structure and kinetics." Biochemistry 35(51): 16475-82. Zhang, H. Z., Hackbarth, C. J., Chansky, K. M. and Chambers, H. F. (2001). "A proteolytic transmembrane signaling pathway and resistance to beta-lactams in staphylococci." Science 291(5510): 1962-5. Zorzi, W., Zhou, X. Y., Dardenne, O., Lamotte, J., Raze, D., et al. (1996). "Structure ofthe low-affinity penicillin-binding protein 5 PBP5fm in wild-type and highly penicillin-resistant strains of Enterococcus faecium." J Bacteriol 178(16): 4948-57. 156 Appendix - Publications arising from graduate work First author publications arising from Ph.D. studies 2002 Lim, D. & Strynadka, N.C. Structural basis for the beta-lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nature Struct. Biol. 2002, 9(11):870-876. 2002 Lim, D., Poole, K. & Strynadka, N.C. Crystal structure of the MexR repressor of the mexRAB-oprM multidrug efflux operon of Pseudomonas aeruginosa. J . Biol. Chem. 2002, 277(32):29253-29259. 2001 Lim, D., Park, H.U., De Castro, L., Kang, S.G., Lee, H.S., Jensen, S., Lee, K.J. & Strynadka, N.C. Crystal structure and kinetic analysis of beta-lactamase inhibitor protein-U in complex with TEM-1 beta-lactamase. Nature Struct. Biol. 2001, 8(10):848-852. 2001 Lim, D., Sanschagrin, F., Passmore, L., De Castro, L., Levesque, R.C. & Strynadka, N.C. Insights into the molecular basis for the carbenicillinase activity of PSE-4 beta-lactamase from crystallographic and kinetic studies. Biochemistry. 2001, 40(2):395-402. Additional publications during Ph.D. studies 2003 Moldoveanu, T., Hosfield, CM., Lim, D., Jia, Z. & Davies, P.L. Calpain silencing by a reversible intrinsic mechanism. Nature Struct. Biol. 2003, 10(5):371-378. 2002 Moldoveanu, T., Hosfield, CM., Lim, D., Elce, J.S., Jia, Z. & Davies, P.L. A Ca(2+) switch aligns the active site of calpain. Cell. 2002, 108(5):649-660. 157 2001 Luo, Y., Bertero, M.G., Frey, E.A., Pfuetzner, R.A., Wenk, M.R., Creagh, L., Marcus, S.L., Lim, D., Sicheri, F., Kay, C, Haynes, C, Finlay, B.B. & Strynadka, N.C. Structural and biochemical characterization of the type m secretion chaperones CesT and SigE. Nature Struct. B i o l . 2001, 8(12): 1031-1036. Publications arising from M.Sc. studies 2002 Lim, D. & Jia, Z. Heavy metal-mediated crystallization of Escherichia coli phytase and analysis of bridging interactions. Protein Pept. Lett . 2002, 9(4):359-365. 2000 Lim, D., Golovan, S., Forsberg, C.W. & Jia, Z. Crystal structures of Escherichia coli phytase and its complex with phytate. Nature Struct. B i o l . 2000, 7(2): 108-113. 158 

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.831.1-0091377/manifest

Comment

Related Items