Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Responses of Pseudomonas aeruginosa to sub-inhibitory antibiotics Brazas, Michelle Denise 2005

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

Item Metadata

Download

Media
831-ubc_2005-103809.pdf [ 17.52MB ]
Metadata
JSON: 831-1.0092345.json
JSON-LD: 831-1.0092345-ld.json
RDF/XML (Pretty): 831-1.0092345-rdf.xml
RDF/JSON: 831-1.0092345-rdf.json
Turtle: 831-1.0092345-turtle.txt
N-Triples: 831-1.0092345-rdf-ntriples.txt
Original Record: 831-1.0092345-source.json
Full Text
831-1.0092345-fulltext.txt
Citation
831-1.0092345.ris

Full Text

RESPONSES OF PSEUDOMONAS AERUGINOSA TO SUB-INHIBITORY ANTIBIOTICS by MICHELLE DENISE BRAZAS B.Sc. (Pharmacology), McMaster University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (MICROBIOLOGY AND IMMUNOLOGY) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2005 © Michelle Denise Brazas, 2005 ABSTRACT Emergence of resistance is never far behind the introduction of new antimicrobials. In response to antimicrobial challenge, bacteria have the capacity to alter their genomes in order to survive. Antibiotic environments play an important but poorly defined role in mediating such changes. The primary focus of this research was to further define the role of sub-inhibitory antimicrobials in the bacterial response to antimicrobial challenge, with particular attention to adaptive resistance responses. A greater understanding of this biological signature will help in evaluating dosing regimes, understanding mechanism of action and in designing new antimicrobials. The effect of sub-inhibitory ciprofloxacin was studied here, beginning with a global analysis using custom DNA microarray technology. Before conducting this analysis, a custom DNA microarray to 5,378 of the 5,570 open reading frames in the Pseudomonas aeruginosa genome was designed, constructed and validated, along with the accompanying set of RNA isolation, cDNA preparation and labelling, and microarray hybridization protocols. Analysis of the P. aeruginosa transcriptome following exposure to sub-inhibitory ciprofloxacin found expression changes in numerous genes. Prominent among these changes was up-regulation of the SOS DNA-repair response and the R2/F2 pyocin region. These changes were confirmed at both the transcription and protein level. Mutants in the R2/F2 pyocin region were found to be resistant to quinolones as well as other DNA damaging agents like mitomycin C, highlighting a role for this region in mediating susceptibility to DNA damage. I also found that sub-inhibitory ciprofloxacin, particularly 0.3pg/ml ciprofloxacin, induced adaptive resistance responses in P. aeruginosa. Similar responses were not observed in a R2/F2 pyocin mutant strain, although the extent of the role of the R2/F2 pyocin region in mediating adaptive resistance to ciprofloxacin awaits further characterization. Overall, this study indicated that sub-inhibitory concentrations of ciprofloxacin antimicrobial do play an important role in the development of resistance in P. aeruginosa, and highlight the clinical importance of better understanding bacterial-antimicrobial interactions. ii TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vii LIST OF FIGURES viii LIST OF ABBREVIATIONS xi ACKNOWLEDGEMENTS xiii INTRODUCTION 1 A. Cystic Fibrosis and Pseudomonas aeruginosa 1 a. Genetics of Cystic Fibrosis 1 b. Pathogenesis of Cystic Fibrosis 2 c. Infections in Cystic Fibrosis 5 d. Antibiotic Therapies and Clinical Outcomes 8 B. Resistance in Pseudomonas aeruginosa 10 C. Quinolone Antibiotics 13 a. Structure and Activity 13 b. Mechanism of Action 15 c. Quinolone Resistance Mechanisms in Pseudomonas aeruginosa 17 d. Effects of Sub-inhibitory Quinolone Concentrations on Pseudomonas aeruginosa.... 18 D. Pyocins in Pseudomonas aeruginosa 20 E. Microarray Analysis of Gene Expression 27 a. Microarray Platforms: Advantages and Disadvantages 28 b. Standards in Microarray Experimentation 29 c. Statistical Analysis of Microarray Data 33 d. Methods of Confirming Microarray Data 35 F. Rationale and Aims of this Study 36 MATERIALS AND METHODS 38 A. Bacterial Strains and Growth Conditions 38 B. Chemicals 38 C. Optimization and Development of Microarray Parameters 38 a. Genomic DNA Isolation 38 b. PCR amplification from Genomic DNA 40 c. Synthesis of 32P-labeled Probe 40 d. Preparation of Nylon Membrane DNA Macroarray 42 e. Hybridization and Image Analysis 43 f. Determination of Amplicon Size, Concentration and Volume 43 g. Determination of Hybridization Temperature and Solution 44 D. Microarray Construction 45 a. Primer Design 45 iii b. PCR Amplification, Purification and Amplicon Evaluation 45 c. Sequencing of Amplicons 46 d. Re-suspension and Plate Format Transfer 46 e. Synthesis of Quality Control Genes 46 f. Additional Microarray Features 47 g. Microarray Printing and Storage 49 E. Microarray Quality Assessment 49 a. Verification of Microarray Print Run 49 b. Cross Hybridization 50 F. Microarray Method Development 50 a. Determination of RNA Isolation Methods 50 b. Determination of Genomic DNA Treatment 52 c. Determination of Reverse Transcription Reaction : 52 d. Determination of Labeling Method 53 G. Experiments on Ciprofloxacin Treated Cultures 55 a. Determination of Minimum Inhibitory Concentrations 55 b. Growth Curve in Sub-inhibitory Ciprofloxacin 55 c. Time-Kill Assay 55 d. Light Microscopy 56 e. Microarray Experimentation 56 i. RNA Isolation and Evaluation 56 ii. Genomic DNA Treatment and RNA Evaluation 57 iii. Reverse Transcription Reaction 57 iv. cDNA Labeling and Purification 57 v. Microarray Slide Preparation and Sample Hybridization 58 vi. Post-hybridization Microarray Handling 58 vii. Scanning and Image Analysis 58 viii. Background Correction and Normalization 59 ix. Statistical Analysis 59 H. Confirmatory Experiments 59 a. Real-time PCR 59 b. Luminescence Assays 60 c. Twitching Assay 60 d. Transmission Electron Microscopy 61 e. Serial Selection of Genomic Loss of Pyocin/Phage Region 61 R E S U L T S 63 CHAPTER ONE: Design and Construction of a Pseudomonas Custom Microarray 63 A. Introduction 63 B . l . Optimization of Amplicon Size, Concentration and Volume 63 B. 2. Optimization of Hybridization Temperature and Hybridization Solution 65 C. 1. Evaluation of Amplicon Integrity 67 i. Agarose Gel Analysis of Amplicon Size and Uniqueness 67 ii. Sequencing Analysis 69 C.2. Evaluation of Amplicon Concentration 69 i. Agarose Gel Analysis 69 iv ii. Capillary Electrophoresis Analysis 71 C. 3. Evaluation of Print Run v » 71 D. Summary 75 CHAPTER TWO: Method Development for the Pseudomonas Custom Microarray 76 A. Introduction 76 B. Comparison of RNA Isolation Methods 76 C. Evaluation of Genomic DNase Treatments 77 D. Comparison of Reverse Transcription Reactions 80 E. Comparison of cDNA Labeling Methods 83 F. Summary 86 CHAPTER THREE: Quinolone Induction of Adaptive Resistance 87 A. Introduction 87 B. Determination of the Minimum Inhibitory Concentration to Ciprofloxacin 87 C. Growth Profile in Presence of Ciprofloxacin 88 D. Time-Kill Assays and Microscopy 88 D. 1. Observation of Adaptive Resistance to Sub-Inhibitory Ciprofloxacin 88 D. 2. Verification of Adaptive Resistance to Sub-Inhibitory Ciprofloxacin 94 E. Microarray Studies 96 E. 1. Expression Responses following Treatment with Sub-Inhibitory Ciprofloxacin 96 F. Microarray Confirmation Assays 103 G. Summary 103 CHAPTER FOUR: Involvement of Pyocin/Phage Expression in Quinolone Resistance 110 A. Introduction 110 B. Microarray Studies 110 B. l . Expression Responses of the R2/F2 Pyocin Region to Sub-inhibitory Ciprofloxacin 110 C. Microarray Confirmation Assays I l l C. l . Real-time PCR Analysis of Expression I l l C. 2. Luminescence Analysis of Expression 111 D. Analysis of Pyocin/Phage Expression 114 D. l . Luminescent Analysis of Pyocin/Phage Expression 114 i. Novobiocin Dose and Time Course 114 ii. Mitomycin Dose and Time Course 117 iii. Ceftazidime Dose and Time Course 117 E. Electron Microscope Analysis of Pyocin/Phage Release 122 F. Determination of the Minimum Inhibitory Concentrations for Pyocin/Phage Mutants 122 G. Time-Kill Assays for Pyocin/Phage Mutants 124 H. Serial Selection of Pyocin/Phage Loss with Sub-inhibitory Ciprofloxacin 127 I. Summary 129 DISCUSSION 133 A. Sub-inhibitory ciprofloxacin induces adaptive resistance in P. aeruginosa 133 B. Sub-inhibitory ciprofloxacin induces expression of R2/F2 pyocins 134 C. R2/F2 pyocin induction is related to DNA damage 137 D. R2/F2 pyocins have a role in adaptive resistance to sub-inhibitory ciprofloxacin 139 E. R2/F2 pyocin region is a fluoroquinolone susceptibility determinant 141 F. R2/F2 pyocins play a role in microbial diversity 142 G. Future directions 144 APPENDIX I: Sample data tracking of amplicon uniqueness, size and concentration for all open reading frames used in P. aeruginosa custom DNA microarray 167 APPENDIX II: Sample capillary electrophoresis data for all open reading frames used in P. aeruginosa custom DNA microarray 177 APPENDIX III: Genes up- and down-regulated in P. aeruginosa in response to various concentrations of ciprofloxacin 187 APPENDIX IV: Authors contributions to research and development 226 vi L I S T O F T A B L E S Table 1: Comparison of features of Pseudomonas aeruginosa pyocins 21 Table 2: R2/F2 pyocin-phage operon of Pseudomonas aeruginosa 26 Table 3: Bacterial strains used in this study 39 Table 4: Nucleotide sequences of primers used in this study 41 Table 5: Minimum inhibitory concentrations of various antimicrobials against various strains of P. aeruginosa 89 Table 6: Summary of custom DNA microarray findings and trends 97 Table 7: Expression changes in various genes induced by O.lx-, 0.3x- and lx-MIC ciprofloxacin 100 Table 8: Pyocin/phage operon and related genes induced by 0.3x- and lx-MIC ciprofloxacin.. 104 Table 9: Comparison of expression changes for various genes as analyzed by relative real-time PCR and custom P. aeruginosa microarray 112 Table 10: Minimum inhibitory concentrations of various antimicrobials against various strains of P. aeruginosa 125 vii LIST OF F I G U R E S Figure 1: Age specific prevalence of airway infections in patients with CF 6 Figure 2: Structure of quinolone core, nalidixic acid and ciprofloxacin 14 Figure 3: S-type pyocin organization 23 Figure 4: Schematic representation of R-type and F-type pyocins 24 Figure 5: Spotting pattern (version 3) for the P. aeruginosa custom DNA microarray 48 Figure 6: Effect of varying concentration and amplicon size on hybridization signal 64 Figure 7: Effect of volume and amplicon size on hybridization signal ..66 Figure 8: Effect of temperature and hybridization solution on hybridization signal 68 Figure 9: Sample agarose gel for evaluation of 96 well PCR and corresponding molecular ladder size range 70 Figure 10: Sample capillary electrophoresis digital agarose gel 72 Figure 11: SYBR Green analysis of upper left sub-grids of P. aeruginosa microarray 73 Figure 12: Terminal transferase Cy-3 label analysis of two sub-grids of the P. aeruginosa microarray 74 Figure 13: Comparison of RNA isolation methods 78 Figure 14: Comparison of RNA isolation methods with respect to rpoC transcript length 79 viii Figure 15: Comparison of various Ambion DNase-/ree treatment times for removal of genomic DNA contamination 81 Figure 16: Comparison of cDNA preparations from different initial amounts of total RNA 82 Figure 17: Comparison of direct and indirect cDNA labeling methods 84 Figure 18: Comparison of reverse transcriptase enzymes in the indirect labeling method 85 Figure 19: Growth curve of P. aeruginosa grown with various concentrations of ciprofloxacin.90 Figure 20: Survival ability of P. aeruginosa in 2x-MIC ciprofloxacin following growth in sub-inhibitory concentrations of ciprofloxacin 92 Figure 21: Survival ability of pretreated cultures of P. aeruginosa in 2x-MIC ciprofloxacin over a longer time frame 93 Figure 22: Gram stains of P. aeruginosa before, during and after the adaptive resistance assay.95 Figure 23: Relationship of genes with significant expression changes induced by O.lx-, 0.3x- and lx-MIC ciprofloxacin in P. aeruginosa 101 Figure 24: Analysis of twitch motility of P. aeruginosa cells cultured in the presence or absence of ciprofloxacin 106 Figure 25: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of ciprofloxacin 113 Figure 26: Growth curve of P. aeruginosa in various concentrations of novobiocin 115 ix Figure 27: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of novobiocin 116 Figure 28: Growth curve of P. aeruginosa in various concentrations of mitomycin 118 Figure 29: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of mitomycin 119 Figure 30: Growth curve of P. aeruginosa in various concentrations of ceftazidime 120 Figure 31: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of ceftazidime 121 Figure 32: Electron micrographs of supernatants from untreated and 0.3x-MIC ciprofloxacin treated P. aeruginosa strain H103 and PA0620::/wx cells 123 Figure 33: Survival ability of P. aeruginosa strain H103 and PA0620::/wx in 2x-MIC ciprofloxacin following growth in 0.3x-MIC ciprofloxacin 126. Figure 34: Growth curve of P. aeruginosa strain H103 and PA0620::/wx mutant 128 Figure 35: Potential relationships between ciprofloxacin and the development of resistance... 138 x L I S T O F A B B R E V I A T I O N S ABC ATP binding cassette asialoGM-1 asialoganglioside-1 ASL airway surface liquid ATCC American Type Culture Collection CCFF Canadian Cystic Fibrosis Foundation CF Cystic fibrosis CFF Cystic Fibrosis Foundation CFU colony forming units CFTR cystic fibrosis transmembrane conductance regulator ENaC epithelial Na + channel EtBr ethidium bromide GFP green fluorescent protein IL- interleukin-IROMP iron regulated outer membrane proteins LB Luria-Bertani LPS lipopolysaccharide MIC minimum inhibitory concentration M-MuLV Moloney murine leukemia virus OD optical density OM outer membrane ORF open reading frame PBS phosphate buffered saline PCR polymerase chain reaction QRDR quinolone resistance determining region RLU relative light unit RND resistance-nodule-cell division rRT-PCR relative real-time polymerase chain reaction RT reverse transcriptase RT-PCR reverse transcription polymerase chain reaction X I SCV small colony variant SDS sodium lauryl sulfate SSC Saline sodium citrate TEM transmission electron microscopy TNF-a tumor necrosis factor xii ACKNOWLEDGEMENTS First and foremost, I would like to thank Dr. Robert E. W. Hancock for his guidance and support throughout my project, for without his constant encouragement and confidence, I would never have taken on nor completed this body of work. He saw more in me than I believed possible, and for this I am grateful. I would also like to thank Dr. Robert Brunham, Dr. Rachel Fernandez, and Dr. Brett Finlay for their expertise and mentorship - I can only aspire. And to my friends and colleagues at Jack Bell Gene Array Facility, Xenon Genetics, Pathogenesis Co., and the University of Toronto Microarray Facility - many thanks for the knowledge and experiences. To my family - I consider this degree as much yours as it is mine, for it would not have been possible without your unwavering love and confidence in me. My immense gratitude extends to countless friends both near and far. I would not have survived this degree without the support and friendship of many people, whom I count myself lucky to have met and come to know well: S.F., M.O., M.S., M.K., K B . , C.R., S.M., K.T., J.B. and T.G. This research was funded by a Canadian Cystic Fibrosis Foundation scholarship and a National Science and Engineering Research Council scholarship granted to M.D.B., and by a Genome Canada research grant to R.E.W.H. xiii I N T R O D U C T I O N A. Cystic Fibrosis and Pseudomonas aeruginosa a. Genetics of Cystic Fibrosis Cystic fibrosis (CF) of the pancreas was first described in 1938 by Andersen (Andersen 1938). The syndrome was noted to occur in children who failed to thrive past two years of age. Autopsy studies found histopathological lesions of the pancreas, yet the established cause of death was found to be from an associated bronchopulmonary infection of Staphylococcus aureus. Production of a "thick, tenacious, greenish gray, purulent material" was also found within the airway lumen (Andersen 1938). This early work by Andersen describes the hallmarks of CF disease: pancreatic insufficiency and chronic airway infections with opportunistic bacterial pathogens, both of which are related to production of a viscous secretion. Since then, treatment has focused on combating these pathophysiological problems with great success. Where this disease once routinely killed patients during infancy, significant advances in nutritional and antimicrobial therapy have greatly improved life expectancy to a median survival age of 35.9 years as reported in 2003 (CCFF 2003). Moreover, advanced nutritional management has shifted the clinical manifestation of CF from intestinal blockade and malnutrition, to progressive loss of pulmonary function caused by chronic bacterial infection. Andersen in 1938 realized that CF was a disease of genetic origin (Andersen 1938). Cystic fibrosis is now known to be inherited as an autosomal recessive trait. The gene believed to be responsible for CF however, was only localized to the long arm of chromosome 7 in 1985 (Tsui, Buchwald et al. 1985; Wainwright, Scambler et al. 1985; White, Woodward et al. 1985), and only identified as the cystic fibrosis transmembrane conductance regulator (CFTR) gene in 1989 (Kerem, Rommens et al. 1989; Riordan, Rommens et al. 1989; Rommens, Iannuzzi et al. 1989). CFTR was found to be a member of the ATP binding cassette (ABC) family of proteins, and to function both as a chloride anion channel and channel regulator. Since its characterization, researchers have identified over 1200 mutations in CFTR responsible for CF (CCFF 2003a), the most common mutation being a deletion of phenylalanine at position 508 (AF508); 85% of all individuals with CF in Canada carry at least one copy of AF508 (CCFF 2003a). 1 Discovery of the CFTR gene brought hope of an imminent cure for cystic fibrosis. With one in every 2,500 children born in Canada being afflicted by CF (CCFF 2003a), this landmark finding held great potential for correcting the underlying defect in cellular ion transport. Although substantial progress toward this goal has been made, CF remains one of the most fatal genetic diseases fifteen years later, namely because the relationship between defects in CFTR and hypersusceptibility to bacterial infection has not been fully elucidated. b. Pathogenesis of Cystic Fibrosis A reasonable scientific explanation of why CF patients initially acquire and then fail to eliminate airway infections has plagued researchers for decades. Both immune function and architecture of the CF lung is essentially normal at birth, except for subtle abnormalities in mucus secretion. Whereas normal healthy airways maintain sterility despite a constant challenge from airborne viruses and bacteria, bacterial infection occurs at an early stage in CF airways (Abman, Ogle et al. 1991; Khan, Wagener et al. 1995). It is the pulmonary complications that result from the interplay between these bacterial infections and the inflammatory responses of the CF lung that account for 80% of CF patient morbidity and mortality (CFF 2003). Pathogenesis of CF in the lung likely begins with the airway epithelial layer. The upper respiratory tract is lined with ciliated epithelia, which are bathed on their apical surface by a thin liquid layer. This airway surface liquid (ASL) layer consists of a periciliary sol and a mucus gel, and is propelled upward out of the lung by the coordinated beating of the cilia (Proctor 1977; Lucas 1994). ASL is required for effective mucociliary clearance, which aided by cough, is the first line of defense against inhaled pathogens. ASL composition and volume reflect the salt and water absorptive and secretory functions of airway epithelia. Hence changes in the cellular transport ability of the epithelia are influenced both the composition and volume of the ASL, contributing to airway disease development. By means of altering the ASL composition and/or volume, mutations in CFTR contribute to CF disease progression. Defining how alterations in CFTR function result in ineffective mucociliary defense has led to two disparate hypotheses. In the high salt hypothesis, emphasis is placed on the function of CFTR as an anion channel (Smith, Karp et al. 1994; Smith, Travis et al. 1996; Zabner, Smith et al. 1998). Defective or absent CFTR results in reduced absorption of chloride ions across the apical membrane from the lumen, 2 hence a high salt concentration and consequently a high volume is observed in the ASL. High salt concentrations have been shown to interfere with the natural killing ability of several innate immune mechanisms like defensins (Goldman, Anderson et al. 1997), thereby facilitating bacterial colonization and airway disease development. In the low volume hypothesis, emphasis is placed on the function of CFTR as a channel regulator (Matsui, Grubb et al. 1998). CFTR negatively regulates amiloride-sensitive epithelial Na + channels (ENaC) located on the apical surface of epithelial cells (Egan, Flotte et al. 1992; Stutts, Rossier et al. 1997; Donaldson, Poligone et al. 2002), preventing sodium reabsorption from the lumen. Defective or absent CFTR, eliminates CFTR's inhibition of ENaC and results in increased Na + absorption. Chloride ions passively follow along other shunt pathways, leaving a reduced ion concentration in the lumen which drives increased water absorption. The net result of increased fluid absorption is depletion of the ASL volume and mucus dehydration, resulting in impaired mucociliary clearance and airway obstruction, both of which facilitate bacterial colonization and airway disease development. While mucus thickening and airway blockage may initiate CF pathogenesis, they do not explain the tendency of CF airways to become colonized by bacterial pathogens, especially chronic colonization by P. aeruginosa. As previously mentioned, it is the colonization of and subsequent inflammatory response to bacterial pathogens that is directly correlated with decline in CF lung function (CFF 2003). Infections are known to be initiated by bacterial adherence to host receptors, implying that increased adherence of inhaled organisms to airway epithelia may be critical for bacterial colonization. Studies of piliated strains of P. aeruginosa have shown greater adherence to CF epithelial cells compared to control cells (Prince 1992; Imundo, Barasch et al. 1995); adherence was reduced following introduction of wild type CFTR (Davies, Stern et al. 1997). Although clinical isolates of P. aeruginosa from chronically infected patients tend to be mucoid and non-piliated, piliated strains likely initiate the colonization process (Burns, Gibson et al. 2001). Regenerating respiratory epithelial cells from CF patients have also been shown to express increased levels of asialoganglioside-1 (asialoGM-1) glycolipid on their apical surface and consequently exhibit increased adherence of P. aeruginosa (Saiman and Prince 1993; de Bentzmann, Roger et al. 1996). Adherence was reversed by competitive inhibition with asialoGM-1 antibody (de Bentzmann, Roger et al. 1996), indicating direct bacterial interaction with asialoGM-1 glycolipids. The mechanism by which mutations in CFTR result in 3 undersialylation of these receptors is thought to be related to hyperacidification of the trans-Golgi network in CF epithelial cells (Poschet, Boucher et al. 2001). Again, while asialoGM-1 receptors may not interact with mucoid, non-piliated clinical isolates (Bryan, Kube et al. 1998), these adhesion sites likely play a role in initiating bacterial colonization. CFTR itself has also been shown to be a receptor for P. aeruginosa. Binding of P. aeruginosa lipopolysaccharide (LPS) core oligosaccharide to CFTR results in bacterial internalization and clearance. Cultured airway epithelial cells expressing the AF508 allele of CFTR were found to be defective in uptake of P. aeruginosa (Pier, Grout et al. 1996). Similarly, bacterial load in the lungs of a neonatal mouse model was increased by addition of exogenous oligosaccharide (Pier, Grout et al. 1996) and overexpression of CFTR in transgenic mice markedly accelerated bacterial clearance (Coleman, Mueschenborn et al. 2003), linking normal CFTR to host defense against bacterial colonization of the lung. Moreover, as for the studies on increased adherence to epithelial cells and asialoGM-1 receptors, only laboratory and nonmucoid clinical isolates of P. aeruginosa have been shown to bind to CFTR; mucoid strains of P. aeruginosa fail to bind CFTR and thus are not effectively cleared from the lung (Schroeder, Reiniger et al. 2001). Consistent with all of these findings, adaptation of P. aeruginosa within the CF airway is associated with modifications to the LPS structure (Ernst, Y i et al. 1999) and conversion to mucoidy (Deretic, Martin et al. 1993; Deretic, Schurr et al. 1994). Changes in bacterial adherence coupled with diminished bacterial clearance capabilities, likely predisposes CF patients to higher rates of airway colonization. Other innate immune responses to infection work in concert with mucociliary clearance and epithelial cell phagocytosis in healthy individuals to rid the lung of airborne pathogens. Airway epithelial cells secrete antimicrobial proteins and peptides into the ASL, including lysozyme and lactoferrin, as well as a- and p-defensins and cathelicidins, to protect the lung against colonization. While there is no evidence of defective antimicrobial peptide secretion in CF patients, altered salt concentrations of CF ASL have been shown to inactivate sensitive antimicrobial peptides, thereby permitting bacterial colonization (Smith, Travis et al. 1996; Goldman, Anderson et al. 1997). Furthermore, once infection with bacterial pathogens is initiated, CF airways exhibit an intense inflammatory response. Stimulated epithelial cells and resident macrophages release interleukin-8 (IL-8) (DiMango, Zar et al. 1995; Khan, Wagener et al. 1995), a potent chemoattractant for neutrophil infiltration into the CF lung. Other mediators including tumor necrosis factor-a (TNF-a) and IL-1 (Wilmott, Kassab et al. 1990), mediate 4 further neutrophil influx and provide additional stimulus for IL-8 production, so that neutrophil influx is sustained. Activated neutrophils are the primary effector cells in the pathogenesis of CF lung disease. Their release of massive amounts of elastase and other proteases, overwhelm intrinsic antiproteases (i.e. a-1 antitrypsin, secretory leukocyte protease inhibitor) (Birrer, McElvaney et al. 1994). Lungs of older CF patients are dominated by persistent neutrophil infiltration, and elevated levels of IL-8 and neutrophil elastase (Bonfield, Panuska et al. 1995). Indiscriminant destruction by neutrophil elastase damages airway epithelia, promoting further inflammatory responses in the airway. With respect to acquired immune responses, there is no evidence of any defects in CF patients that might explain the inability of CF patients to clear airway infections. Outside of the respiratory tract, neither infection rates or severity differ between CF and normal patients (Chmiel, Berger et al. 2002). Furthermore, CF patients mount a significant humoral response to P. aeruginosa antigens, to the extent that the early appearance of these antibodies may improve detection methods of P. aeruginosa infections in young children (Burns, Gibson et al. 2001; West, Zeng et al. 2002). Despite this early and sustained immune response to P. aeruginosa, CF patients are unable to clear P. aeruginosa from their airways. The dominant presence of neutrophils and proteases in the CF lung also plays a role in the ineffectiveness of the acquired immune response. Successful antibody mediated killing depends upon intact complement and Fc receptors on phagocytes. The CF lung however, is neutrophil rich and concentrated with proteases that work to cleave complement and Fc receptors, thereby reducing opsonophagocytosis (Chmiel, Berger et al. 2002). Coupled with unresolved and persistent inflammation, airways become progressively dilated and bronchiectatic, eventually crippling the ability of the lung to provide respiration. In summary then, failure of CF lungs to clear initial bacterial infections leads to chronic infection and thus persistent inflammation. This damages the lung architecture and leads to bronchiectasis, progressive airway obstruction and ultimately death of the CF patient. c. Infections in Cystic Fibrosis Data gathered from the Cystic Fibrosis Patient Registry highlights patterns in the overall prevalence and age-related distribution of pathogens in the CF population (Figure 1, (CFF 5 Resp i ra to ry Infect ions v s . A g e 100 -i 0 i — r-0to1 2t05 6to10 11to17 18to24 25to34 35to44 45+ Age Overall Percentage in 2003: P. aeruginosa 57.2% S. aureus 51.1% H. influenza 16.8% S. maltophilia 11.0% B. cepacia 3.1% —MRSA11.8% Figure 1: Age specific prevalence of airway infections in patients with CF. Organisms reported to the U.S. Cystic Fibrosis Patient Registry, 2003. Overall percentage of patients (all ages) who had at least one respiratory tract culture (sputum, bronchoscopy, oropharyngeal or nasal) performed in 2003 that was positive for the following organisms: Pseudomonas aeruginosa (red line), 57.2%; Staphylococcus aureus (green line), 51.1%; Haemophilus influenzae (dark blue line), 16.8%; Stenotrophomonas maltophilia (yellow line), 11.0%; Burkholderia cepacia (black line), 3.1%; and methicillin-resistant Staphylococcus aureus (purple line), 11.8%. Reprinted with permission from reference Cystic Fibrosis Foundation Patient Registry Annual Data Report 2003. 6 2003)). Initial cultures from young CF patients are most often positive for S. aureus and Haemophilus influenzae (Figure 1). As initially noted by Andersen in 1938, CF patients failed to thrive past two years of age, the cause of death being attributed to bronchopulmonary infections of S. aureus (Andersen 1938). Introduction of antimicrobial therapies directed against S. aureus has greatly improved patient lifespan (Thomassen, Demko et al. 1987), although recent studies have indicated prophylactic antistaphylococcal therapy in infants and young children with CF enhances colonization with P. aeruginosa (Ratjen, Comes et al. 2001; Stutman, Lieberman et al. 2002). S. aureus and H. influenzae infections are typically followed by P. aeruginosa infections. Of these infecting organisms, P. aeruginosa is the most prevalent and significant pulmonary pathogen. It is the cause of complicating chronic infections in 80% to 90%) of CF patients (Figure 1) (Hutchison and Govan 1999; CFF 2003). The mean age of P. aeruginosa infection is 15 months or 21 months, the difference depending on the method of detection (Burns, Gibson et al. 2001). Significant correlation however exists between acquisition of P. aeruginosa infection and progressive loss of lung function and mortality in CF patients (Nixon, Armstrong et al. 2001; Emerson, Rosenfeld et al. 2002). The source of P. aeruginosa has not been clearly established, although the wide variety of genotypes seen in isolates from CF patients suggests an environmental reservoir (Burns, Gibson et al. 2001). Although the strain initially acquired is typically an environmental isolate (Burns, Gibson et al. 2001), along the course of disease progression, P. aeruginosa alters many of its phenotypic characteristics. Alteration of these phenotypes appears to be selected within the CF airway (Mulvey, Lopez-Boado et al. 1998) and occurs more frequently with increasing length of infection. With 5,570 predicted open reading frames (ORFs), the genetic complexity of the P. aeruginosa genome affords this organism a tremendous ability to adapt to its environment (Stover, Pham et al. 2000). While initial isolates express a 'smooth' LPS, containing O-side chains and little to no extracellular mucoid polysaccharide (alginate) (Pier 1985), later isolates lose the O-side chain on LPS (Hancock, Mutharia et al. 1983) and acquire a distinctive LPS acylation pattern (Ernst, Yi et al. 1999). Late isolates also tend to be more resistant to antibiotics and frequently mucoid. Conversion to mucoidy is an important phenotypic change because establishment of chronic P. aeruginosa infections in the CF lung correlates with transition of the microbe to a mucoid phenotype (Henry, Mellis et al. 1992). Mucoid conversion is 7 chromosomally encoded (Deretic, Schurr et al. 1994) and hypothesized to be selected for by the CF lung (Mulvey, Lopez-Boado et al. 1998). Indeed, mucoid forms of P. aeruginosa account for 90% of P. aeruginosa isolates from CF patients (Pedersen, Hoiby et al. 1992). Other phenotypic changes include loss of flagella dependent motility (Luzar and Montie 1985; Luzar, Thomassen et al. 1985) and increased auxotrophy (Taylor, Hodson et al. 1992; Taylor, Hodson et al. 1993; Thomas, Ray et al. 2000). P. aeruginosa has also recently been shown to adopt a biofdm mode of growth in the CF lung (Singh, Schaefer et al. 2000; Drenkard and Ausubel 2002). Biofdm communities exhibit characteristically slow growth, and are inherently resistant to antibiotics and thus difficult to eradicate even from immunocompotent individuals (Costerton, Stewart et al. 1999), characteristics which parallel those of CF airway infections. More recently, a high frequency of hypermutability has been identified in P. aeruginosa isolates from CF patients (Oliver, Canton et al. 2000). The selective conditions found in the CF lung including a high degree of compartmentalization, a deteriorating lung environment, and continuous immune defense and antimicrobial challenges, are proposed to mediate this rate of evolution. Furthermore, CF isolates of P. aeruginosa have genomes larger than the laboratory strain, indicative of adaptation and thus acquisition of new genes from its environment (Spencer, Kas et al. 2003). In addition to P. aeruginosa infections, CF patients may also become infected with Burkholderia cepacia, Stenotrophomonas maltophilia, Achromobacter xylosoxidans and fungi including Aspergillus and nontuberculous mycobacteria (Figure 1). Of these organisms, B. cepacia is the most serious, being associated with a more dramatic decline in lung function and increased mortality (Isles, Maclusky et al. 1984; Thomassen, Demko et al. 1985). d. Antibiotic Therapies and Clinical Outcomes The hallmark in CF research and therapy has been the introduction of antibiotics into the CF treatment strategy. Progress in patient management and life expectancy has come predominantly from antibiotic control of pulmonary infections, in particular infections caused by S. aureus and P. aeruginosa. Proper management of CF lung disease however, requires appropriate antimicrobial therapy tailored to the bacterial pathogens isolated from the respiratory tract. Hence, the choice of antibiotic is based on the periodic isolation and identification of 8 pathogens from respiratory secretions and biopsy samples. Antimicrobial treatment regimes are directed against three distinct clinical stages of CF pulmonary disease (Gibson, Burns et al. 2003). In the first stage during early lung disease, CF patients receive antibiotics prophylatically to eradicate any initial infections and to delay onset of chronic colonization with P. aeruginosa. Since P. aeruginosa isolates at this stage tend to be nonmucoid, highly susceptible to antibiotics and planktonic in nature (Burns, Gibson et al. 2001), they are more amenable to eradication by antibiotic therapy, and are treated with aggressive anti-pseudomonal antibiotics. Once bacterial colonization with either S. aureus or P. aeruginosa has occurred, the second stage objective, so called maintenance therapy, is to administer prolonged antibiotic regimes to slow the progression of lung disease, and to increase the interval between pulmonary exacerbations. Where continuous anti-staphylococcal therapy was once standard practice, this practice is no longer recommended since studies found that such patients had a lower prevalence of S. aureus infections, but a significantly higher rate of P. aeruginosa acquisition (Ratjen, Comes et al. 2001; Stutman, Lieberman et al. 2002). Intermittent antistaphylococcal therapy is currently recommended. Although current practices for maintenance therapy of P. aeruginosa infections range from quarterly intravenous treatments, to inhaled antibiotics and oral quinolones (Gibson, Burns et al. 2003), P. aeruginosa maintenance therapy has been shown to stabilize pulmonary function and decrease morbidity (Ramsey, Pepe et al. 1999). Maintenance therapy practices however, have drawn concern from clinicians over the emergence of increased levels of antibiotic resistance among respiratory pathogens in CF patients. Of particular importance is the emergence of quinolone resistant strains of P. aeruginosa and S. aureus associated with longer monotherapy courses (Ball 1990; Dalhoff 1994). However, while prolonged monotherapy courses are discouraged, ciprofloxacin remains the quinolone of choice for P. aeruginosa infections in CF patients (Gibson, Burns et al. 2003). In addition to daily respiratory challenges, CF patients with chronic pulmonary infections of bacterial pathogens like P. aeruginosa, periodically experience episodes of acute exacerbation. Despite the importance of this third clinical stage, there are no defined criteria or set of standards for classifying an acute episode. Furthermore, there are no clearly outlined treatment regimes. Typically, patients are treated on an outpatient basis with either oral or inhaled antibiotics (Noone and Knowles 1999; Gibson, Burns et al. 2003), where choice of antibiotic again depends upon cultures of airway secretions. Cystic Fibrosis Foundation (CFF) guidelines recommend 9 combining two antipseudomonal antibiotics having different mechanism of action in order to achieve synergy and slow the emergence of resistance (CFF 1994; Noone and Knowles 1999). The effectiveness of therapy is measured by improvements in pulmonary function, reduction in sputum bacterial density and improved quality of life, since eradication of the pathogen or resolution of the inflammatory response rarely occurs. B. Resistance in Pseudomonas aeruginosa The difficulty in eradicating P. aeruginosa infections from the CF lung with antibiotic therapy is also a consequence of the organism itself. Like other gram-negative organisms, P. aeruginosa is intrinsically resistant to the common repertoire of antibiotics. Several factors contribute to this intrinsic resistance, beginning with an unusually restricted outer-membrane permeability (Hancock 1997; Nikaido 1998). The outer membrane (OM) of gram-negative bacteria contributes to this intrinsic resistance by serving as a general permeability barrier, slowing down diffusion of solutes and antibiotics. Hydrophilic compounds thus cross the OM barrier via water filled channels called porins. The major porin in Escherichia coli, OmpF, has a small channel size, thereby physically hindering penetration of even small antimicrobials like fluoroquinolones. Porins in the OM of P. aeruginosa are even more inefficient in diffusion than OmpF (Nikaido, Nikaido et al. 1991), further reducing the influx of solutes and antibiotics across its OM. Lipophilic compounds, in contrast, are capable of dissolving into the lipid bilayer domains of the OM, thereby crossing the OM barrier. The outer monolayer of P. aeruginosa however, is made up of highly negatively charged lipopolysaccharides (LPS), which further restrict passage of lipophilic compounds across its OM. Although the outer membrane serves as a general permeability barrier, the cytoplasmic concentration of many antibiotics reaches one half of the extracellular concentration within 30 seconds in P. aeruginosa (Nikaido 1998). Thus the outer membrane alone is not sufficient to produce significant resistance. Coupled with low OM permeability, are various secondary resistance mechanisms which take advantage of the low antibiotic exposure mediated by the low permeability. These secondary resistance mechanisms include chromosomally encoded p-lactamases and multidrug efflux pumps. For resistance to early generations of P-lactam compounds, ubiquitous expression of periplasmic P-lactamases provides resistance in synergy to that afforded by low OM permeability. For resistance to other classes of antibiotics however, expression of inducible 10 multiple drug efflux pumps like mexABoprM or mexCDoprJ, provides resistance against a broad spectrum of antibiotics, again acting in concert with the resistance afforded by low OM permeability. P. aeruginosa can also 'acquire resistance' through mutation of the bacterial genome and/or acquisition of new resistance genes through the horizontal transfer of plasmids and transposons (Davies 1997). Chromosomal mutations arise frequently in P. aeruginosa under antibiotic selective pressure (Kohler, Michea-Hamzehpour et al. 1997) and during times of bacterial stress (Foster 1993; Shapiro 1997), resulting in either target modification and thus reduced antibiotic sensitivity, or diminished antibiotic concentration from mutations in the regulation of multidrug efflux pumps. Some mutations may even produce mutator phenotypes in bacteria (LeClerc, Li et al. 1996; Shapiro 1997; Oliver, Canton et al. 2000), affording the organism an increased propensity of acquiring antibiotic resistance through mutation. Hypermutable strains have been shown to be more resistant to antibiotics, pointing to a link between high mutation rates and the evolution of antibiotic resistance (Oliver, Canton et al. 2000). Similarly, antibiotics like fluoroquinolones, induce an SOS response and increase mutability (Mamber, Kolek et al. 1993), further contributing to antibiotic resistance acquired through mutation. Mobile genetic elements like conjugative plasmids and transposons also play an important role in the transfer of antibiotic resistance determinants in P. aeruginosa (Sinclair and Holloway 1982; Vezina and Levesque 1991). Clustering of resistance gene cassettes in integrons is also commonplace in P. aeruginosa (Poirel, Lambert et al. 2001; Riccio, Docquier et al. 2003) and also contributes to acquired resistance in P. aeruginosa. In addition to intrinsic and acquired resistance mechanisms, P. aeruginosa also displays adaptive resistance. Adaptive resistance is defined as an unstable, reversible resistance that is unrelated to genetic mutation (Barclay and Begg 2001). It occurs transiently in an organism under non-lethal selective pressure, and has been well documented for aminoglycosides, as well as quinolones (Barclay and Begg 2001). Despite being a poorly understood phenomenon, it is clinically relevant (Chamberland, Malouin et al. 1990; Barclay, Begg et al. 1996). In one study, CF patients were given a single dose of tobramycin, an aminoglycoside antimicrobial. Sputum samples were taken every hour and colony forming units (CFUs) assessed. No difference in killing ability was noted over time following the single dose of tobramycin. Exposure of cultures for 90min to a second dose of tobramycin, one hour after the first dose, resulted in a transient 11 increase in CFUs, indicative of adaptive resistance in a clinical setting (Barclay, Begg et al. 1996). As previously mentioned, various physiological states of P. aeruginosa also play a role in antibiotic resistance. For example, P. aeruginosa is known to grow as a biofilm in the CF lung (Singh, Schaefer et al. 2000; Drenkard and Ausubel 2002) and many characteristics of biofilms such as slow growth rate, decreased diffusion and low metabolic activity, likely contribute to the antibiotic resistance associated with biofilm communities (Walters, Roe et al. 2003). Recent studies have identified several candidate genes that may be involved in the antibiotic resistant phenotype of biofilms (Whiteley, Bangera et al. 2001; Drenkard and Ausubel 2002; Mah, Pitts et al. 2003). Other studies of P. aeruginosa biofilms have found that while low concentrations of fluoroquinolones eliminate the majority of cells, a small fraction of persisters remains intolerant to antibiotic killing (Brooun, Liu et al. 2000; Spoering and Lewis 2001). Persisters are not mutants and appear to display adaptive resistance-like characteristics, since re-culturing of persisters produces the wild-type population (Brooun, Liu et al. 2000). Although most CF patients chronically infected with P. aeruginosa are colonized with a single genotype, many different phenotypic variants can be recovered, including the mucoid and small colony phenotypes (Zierdt and Schmidt 1964). Small colony variants (SCV) frequently arise within P. aeruginosa populations under selective pressure (Bayer, Norman et al. 1987; Haussler, Tummler et al. 1999; Drenkard and Ausubel 2002; Haussler, Ziegler et al. 2003; von Gotz, Haussler et al. 2004). These SCVs often display an enhanced ability to grow in biofilms and exhibit increased antibiotic resistance. Like persister cells, SCVs are not formed by mutation since the SCV phenotype is reversible upon growth in non-selective media (Massey, Buckling et al. 2001), again similar to adaptive resistance characteristics. Together these antibiotic resistance mechanisms and differing modes of growth contribute to the variable resistance phenotype of P. aeruginosa. Our poor understanding of adaptive resistance and the role of persisters and hypermutators in the overall resistance profile of P. aeruginosa, leaves us with a weak comprehension of the response capabilities of this organism to antimicrobial challenges in general, even for those antimicrobials for which a mechanism of action has been well defined. Yet the development of new antimicrobials and antibiotic resistance prevention methodology are all based upon the completeness of this knowledge. 12 C. Quinolone Antibiotics Of the therapies clinically recommended for maintenance of CF infections of P. aeruginosa, ciprofloxacin and the class of fluoroquinolones remain the antimicrobials of choice (Gibson, Burns et al. 2003) because they can be used on an outpatient basis without the need for further hospitalization of the patient. a. Structure and Activity Since discovery of the first quinolone, nalidixic acid, and its associated antibacterial properties in the early 1960s (Lesher, Froelich et al. 1962), much progress has been made in refining the spectrum and activity of this new class of antimicrobials. Quinolone agents exhibit a bicyclic aromatic core, which contains a carbon at the C-8 position yielding a true quinolone (Gootz 1998) (Figure 2). For antibacterial activity, quinolone agents must also contain a carboxylic acid at C-3, a ketone at C-4 and a substitution at the N - l position. A cyclic diamine is also often present attached through one of its nitrogens to C-7. These first generation quinolones displayed moderate activity against most gram-negative organisms, but often lacked useful activity against gram-positive organisms, P. aeruginosa and anaerobes (Gootz 1998). Addition of fluorine to C-6 of the basic quinolone structure significantly advanced this class of antimicrobials (Koga, Itoh et al. 1980) (Figure 2). Quinolones with this substitution, so named fluoroquinolones, have both increased quinolone activity against the target enzyme DNA gyrase and facilitated penetration into bacterial cells (Domagala, Hanna et al. 1986). Fluoroquinolones then, display improved MIC values against both gram-positive and gram-negative pathogens, with ciprofloxacin displaying the strongest gram-negative spectrum of these second generation quinolones (Wolfson and Hooper 1985). The N - l cyclopropyl group substitution of ciprofloxacin in fact, has been one of the most effective functionalities for providing broad spectrum activity (Gootz 1998). 13 A. R5 O COOH Figure 2: Structure of quinolone core (A), nalidixic acid (B) and ciprofloxacin (C). 14 b. Mechanism of Action Quinolone antimicrobials target the essential bacterial enzymes, type II DNA topoisomerases. Topoisomerases are essential enzymes for maintaining cellular DNA in the appropriate state of supercoiling in both replicating and non-replicating regions of the chromosome. This state is further important in transcription, since the transcription of many genes is sensitive to the state of DNA supercoiling (Graeme-Cook, May et al. 1989; Dorman, Bhriain et al. 1990). Topoisomerases catalyze the passage of one DNA strand through another, and are classified according to whether they introduce a transient single or double strand break for this passage. Type I topoisomerases create a single strand break for DNA passage, while type II enzymes introduce a double strand break for DNA passage. Bacteria possess four DNA topoisomerases, of which only the type II DNA topoisomerases are sensitive to inhibition by fluoroquinolones (Moreau, Robaux et al. 1990). DNA gyrase and topoisomerase IV are both type II topoisomerases. DNA gyrase introduces negative supercoils into DNA and thus relieves the topological stress that arises from the translocation of transcription and replication complexes along DNA (Drlica and Zhao 1997). Topoisomerase IV is a decatenating enzyme and works to resolve linked daughter DNA molecules following replication (Marians 1996; Drlica and Zhao 1997). DNA gyrase, first isolated from E. coli in 1976 (Gellert, Mizuuchi et al. 1976), is encoded by the unlinked gyrA and gyrB genes. It has a tetrameric A 2 B 2 structure, composed of two GyrA subunits and two GyrB subunits (Drlica and Zhao 1997). Introduction of negative supercoils into DNA involves four steps: (i) binding of gyrase to DNA containing a positive supercoil; (ii) cleavage of DNA and formation of a covalent linkage between Tyr-122 of GyrA and the 5' end of the DNA chain; (iii) active passage of DNA through the opened DNA following ATP hydrolysis by GyrB; and (iv) re-ligation of the DNA strands and release of the enzyme (Shen 1993). GyrA thus contains the active site of the DNA gyrase enzyme. Surrounding the Tyr-122 active site is the region termed the quinolone resistance determining region (QRDR) which spans Ala-67 to Gin-106, and contains the residues most frequently mutated in quinolone resistance mutants (Yoshida, Bogaki et al. 1990). The N-terminus of GyrB contains the ATP binding domain, and provides energy for the passage of double stranded DNA through the transient break, mediating introduction of a negative supercoil (Gootz 1998). 15 Quinolone antimicrobials target DNA gyrase, and stabilize the DNA-gyrase complex in its cleaved state thereby inhibiting DNA synthesis or transcription (Sugino, Peebles et al. 1977). It is uncertain whether binding of quinolones to DNA-gyrase complexes involves direct interaction with DNA. Quinolone inhibition of DNA synthesis and thus growth however, are reversible phenomena, whereas quinolone lethality is not (Goss, Deitz et al. 1965; Deitz, Cook et al. 1966). The drug-enzyme-DNA complex is reversibly formed and can dissociate upon washing of cells thereby restoring cell viability (Goss, Deitz et al. 1965; Deitz, Cook et al. 1966). As well, the concentration of quinolone required to block DNA synthesis is lower than that required to kill cells (Goss, Deitz et al. 1965; Chen, Malik et al. 1996). Cell death then, likely arises from release of the DNA ends from the drug-enzyme-DNA complex and the creation of double stranded DNA breaks (Chen, Malik et al. 1996; Drlica and Zhao 1997). Only recently in 1990, was topoisomerase IV, a homolog of DNA gyrase, discovered (Kato, Nishimura et al. 1990). Like DNA gyrase, topoisomerase IV is a tetrameric protein composed of two ParC subunits and two ParE subunits, encoded by the parC and parE genes which share significant identity with gyrA and gyrB, respectively (Kato, Nishimura et al. 1990; Peng and Marians 1993). Topoisomerase IV is the principal enzyme for separating replicated DNA molecules and like DNA gyrase, uses double strand passage of DNA to decatenate DNA molecules. Unlike DNA gyrase though, the decatenation activity of topoisomerase IV does not require DNA wrapping and thus favors intermolecular strand passage (Peng and Marians 1995). This DNA wrapping difference likely contributes to the functional differences observed between DNA gyrase and topoisomerase IV. The mechanistic similarities between DNA gyrase and topoisomerase IV however, indicated that topoisomerase IV was also a likely target for quinolone antimicrobials. Inhibition of the decatenation activity of purified E. coli topoisomerase IV enzymes however, was found to require 15 to 50 times more quinolone than inhibition of DNA gyrase negative supercoiling activity (Peng and Marians 1993; Hoshino, Kitamura et al. 1994). The primary target of quinolones and thus the response to quinolones has since been shown to vary depending on the organism. In E. coli and P. aeruginosa the primary target is DNA gyrase, whereas in S. aureus and S. pneumoniae the primary target is topoisomerase IV (Blanche, Cameron et al. 1996). As for DNA gyrase, drug-enzyme-DNA complexes are formed with quinolones and topoisomerase 16 IV. The interaction of quinolones with topoisomerase IV produces a slow inhibition of DNA synthesis in a mechanism similar to that for DNA gyrase (Hiasa, Yousef et al. 1996). c. Quinolone Resistance Mechanisms in Pseudomonas aeruginosa As with other antimicrobials, extensive use of fluoroquinolones has led to an increase in the appearance of quinolone resistance among bacterial pathogens. While intrinsic resistance mechanisms like low outer membrane permeability, play an important role in mediating resistance during the initial stages of antibiotic exposure, these mechanisms are usually not sufficient for sustained protection against prolonged exposure to the antimicrobials. Bacteria however can become resistant to quinolones by mutations in the target enzymes or by active efflux (Hooper 1999; Hooper 2001); plasmid mediated resistance has also been recently confirmed (Martinez-Martinez, Pascual et al. 1998). In P. aeruginosa, as in other gram-negative bacteria, most of the mutations which confer quinolone resistance occur in the gyrA subunit (Kohler 1998). Furthermore, most of the mutations map to the QRDR, with the most common mutations being Asp87Asn, Asp87Tyr and Thr83Ile (Yonezawa, Takahata et al. 1995). Mutations at both sites 83 and 87 confer a higher degree of resistance to fluoroquinolones compared to mutations at a single site. Resistance mapping to gyrB has also been observed in P. aeruginosa (Mouneimne, Robert et al. 1999; Le Thomas, Couetdic et al. 2001). Additional mutations have been noted in parC (Mouneimne, Robert et al. 1999), where again combinations of topoisomerase mutations were found to confer high-level quinolone resistance relative to single mutation events. Active efflux in P. aeruginosa also plays a critical role in mediating resistance to antimicrobials, including fluoroquinolones. Mutations in the regulators for multidrug efflux pumps affect resistance levels by bringing about overexpression of the respective efflux system. Furthermore, efflux pump overexpression confers cross-resistance to non-quinolone agents as well as quinolones, making development of this type of fluoroquinolone resistance in P. aeruginosa a major clinical problem. The MexAB-OprM efflux system, overexpressed in nalB mutants, confers resistance to nalidixic acid, fluoroquinolones, carbenicillin and tetracycline, and results from mutations in mexR, the repressor of the mexABoprM efflux system (Li, Nikaido et al. 1995; Ziha-Zarifi, Llanes et al. 1999). Similarly, mutants in the nfxB regulator gene 17 overexpress MexCD-OprJ and confer resistance to quinolones, erythromycin and chloramphenicol (Jakics, Iyobe et al. 1992; Poole, Gotoh et al. 1996), whereas nfxC mutants (mutants in the positive regulator, mexT) overexpress MexEF-OprN and specify resistance to all quinolones as well as chloramphenicol (Fukuda, Hosaka et al/ 1995; Kohler, Michea-Hamzehpour et al. 1997; Kohler, Epp et al. 1999). Quinolone concentration appears to affect which resistance mechanism is selected by P. aeruginosa in response to quinolone exposure. At concentrations close to the MIC, efflux type mechanisms were selected almost exclusively in the lab; gyrase type mutations appeared only at concentrations above 4x-MIC (Kohler and Pechere 2001). Conversely, quinolone selection of multigenic mutations rather than single mutation events may better explain the disparate data on the dominant resistance mechanism. Quinolone treatment is also known to induce adaptive resistance responses in P. aeruginosa (Chamberland, Malouin et al. 1990; Gould, Milne et al. 1991). Despite being a poorly understood phenomenon, it is clinically relevant (Chamberland, Malouin et al. 1990; Barclay, Begg et al. 1996). While most fluoroquinolone resistance work has focused on better understanding either the target, DNA gyrase, or the multidrug efflux mechanisms, these mutation based resistance mechanisms are obviously not contributing factors in adaptive resistance since removal of the antimicrobial challenge restores the wild type susceptibility profile of the organism (Chamberland, Malouin et al. 1990). As with other antimicrobials, the one drug one target concept is not applicable. Clearly much work needs to be conducted to better define the mechanisms of and role of adaptive resistance to quinolones, before a complete picture of fluoroquinolone resistance in P. aeruginosa can be drawn. d. Effects of Sub-inhibitory Quinolone Concentrations on Pseudomonas aeruginosa In addition to studies on the affect of quinolones on the development of resistance, numerous studies have examined the effects of sub-inhibitory fluoroquinolones on P. aeruginosa phenotypes. Sub-inhibitory concentrations of antibiotics represent a therapeutically important selective environment. Such concentrations occur at the onset of drug treatment, between dosing intervals and within the thick mucus of the CF lung (Doring, Conway et al. 2000) . With respect to CF patients chronically infected with P. aeruginosa and being treated with fluoroquinolones to 18 maintain low infection levels, each of these sub-inhibitory antibiotic environments likely occurs repeatedly over the course of therapy. Amongst the changes altered by treatment with sub-MIC ciprofloxacin are the adhesion characteristics of P. aeruginosa. Sub-MIC ciprofloxacin was found to significantly decrease the adherence of P. aeruginosa to the human larynx carcinoma cell line HEp-2 (Visser, Beumer et al. 1993) and to urinary epithelial cells (Sonstein and Burnham 1993; Zhanel, Kim et al. 1993). Similar observations have been noted for biofilm adherence (Yassien, Khardori et al. 1995). The inhibitory effect of sub-inhibitory quinolones on pili and fimbriae production in P. aeruginosa may be responsible for these observations (Sonstein and Burnham 1993). In contrast to the mucoid converting effect of supra-inhibitory concentrations of fluoroquinolones (Pina and Mattingly 1997), sub-inhibitory concentrations have been found to inhibit alginate production in P. aeruginosa (Trancassini, Brenciaglia et al. 1992; Majtan and Hybenova 1996). Furthermore, no major changes in LPS structure have been noted (Magni, Giordano et al. 1994; McKenney, Willcock et al. 1994). Sub-MIC levels of quinolones have also been found to alter the production of other virulence factors. Decreases in exotoxin A, exoenzyme S, elastase and total protease activity have all been demonstrated (Ravizzola, Pirali et al. 1987; Grimwood, To et al. 1989; Trancassini, Brenciaglia et al. 1992; Sonstein and Burnham 1993) and appear to contribute to a reduction in lung injury during chronic P. aeruginosa lung infections (Grimwood, To et al. 1989; Grimwood, To et al. 1989). Fung-Tome et al. (Fung-Tome, Kolek et al. 1993) also noted an increase in the mutation rate and resistance level of P. aeruginosa following exposure to sub-inhibitory ciprofloxacin, where the rate of resistance development depended upon the concentration and duration of exposure. Prolonged exposure to sub-inhibitory ciprofloxacin was also found to promote the development of low level resistance to structurally unrelated antimicrobial agents. Together, these findings highlight how exposure to sub-inhibitory concentrations of ciprofloxacin can influence the clinical outcome of P. aeruginosa infections in CF patients. In general, these findings seem to suggest that exposure to sub-MIC ciprofloxacin may work to limit some of the deleterious in vivo effects of P. aeruginosa, which in synergy with the mutational effects, works to promote survival of the organism in the selective environment. 19 D. Pyocins in Pseudomonas aeruginosa Bactericidal protein antibiotics are produced by many bacteria, and those produced by P. aeruginosa are termed pyocins (Jacob 1954). P. aeruginosa produces three types of pyocins: R-, F- and S-type, all of which are encoded on the chromosome rather than on plasmids as in most other bacteria. While spontaneous pyocin production or pyocinogeny is low, pyocin production can be induced upon treatment of cultures with mutagenic agents (Jacob 1954; Kageyama 1964); ciprofloxacin along with other fluoroquinolones is a mutagenic agent (Phillips, Culebras et al. 1987; Clerch, Bravo etal. 1996). Various activity and structural features distinguish each of the pyocin types (Table 1). S-type pyocins are both soluble and protease sensitive. SI, S2, AP41 and S3 pyocins have been identified and isolated from various strains of P. aeruginosa. The sequenced P. aeruginosa strain PAOl (Stover, Pham et al. 2000) contains pyocin S2 and two predicted pyocins, S4 and S5 (Parret and De Mot 2000). Al l S-type pyocins are composed of two proteins: a small protein which confers immunity protection to the host bacteria and a larger protein which possesses killing activity. Release of S-type pyocins from pyocinogenic cells occurs as an equimolar complex of the killing and immunity proteins, and is thought to occur by the lytic system encoded by the R2 and F2 pyocins (discussed in detail below) (Nakayama, Takashima et al. 2000). Before translocation across the membrane and exertion of their bactericidal activity however, pyocins must bind to specific receptors on the outer membrane of target Gram-negative cells. Based on observations that killing by pyocin SI, S2 and S3 was enhanced under iron limiting conditions, it has been proposed that iron regulated outer membrane proteins (IROMPs) may act as S-type pyocin receptors (Ohkawa, Shiga et al. 1980; Sano, Matsui et al. 1993; Duport, Baysse et al. 1995). Uptake of pyocin S3 for example, has been shown to occur through the ferripyoverdine type II receptor (Baysse, Meyer et al. 1999), and through the tolQRAB-oprL system for uptake of AP41 (Dennis, Lafontaine et al. 1996). The bactericidal activity of the S-type pyocins SI, S2, S3 and AP41 is mediated through endonuclease activity of the C-terminus of the large protein, which causes the breakdown of DNA (Duport, Baysse et al. 1995). Interestingly, the SI and S2 pyocins are also able to inhibit phospholipid synthesis under iron-limiting conditions, activity which is linked to but independent of the DNase domain and activity (Okawa, Maruo et al. 1975; Morse, Jones et al. 1980). AP41 pyocin is also able to induce pyocin 20 Table 1: Comparison of features of Pseudomonas aeruginosa pyocins. PyocinType PAOl Types Gene Organization & Homology Mechanism of Action S-type Soluble; Protease sensitive SI — — Endonuclease S2 S2 Killing protein: PA 1150 Immunity protein: PA1151 Endonuclease S3 — — Endonuclease S4 S4 Killing protein: PA3866 Immunity protein: PA3865.1 tRNase S5 S5 Killing protein: PA0985 Immunity protein: PA0984 Pore-forming AP41 — — Endonuclease R-type -> Protease insensitive; Nuclease insensitive R1-R5 R2 PA0615-PA0632 Inflexible contractile tail Homology to P2 bacteriophage Depolarize cytoplasmic membrane F-type Protease insensitive; Nuclease insensitive F1-F3 F2 PA0633 - PA0648 Flexible, non-contractile tail Homology to X bacteriophage 21 production in sensitive targets and phage in lysogenic cells, in a RecA dependent fashion (Sano and Kageyama 1981; Sano 1993). Distinct from the other S-type pyocins, the C-terminal domains of the novel pyocins are predicted to contain tRNase activity in the case of pyocin S4 and pore-forming activity in pyocin S5 (Parret and De Mot 2000). Genetic organization of the S-type pyocins is fairly well conserved amongst this group of pyocins (Figure 3) (Michel-Briand and Baysse 2002). Domain I of the larger protein incorporates the N-terminal 240 amino acids and contains the receptor binding site. Domain II is variably present in S-type pyocins and has of yet an unknown function. Translocation across the outer membrane is mediated by domain III. The C-terminal 130 amino acids of the larger protein provides the DNase and immunity protein binding function of SI, S2, AP41 and S3 pyocins, and is proposed to provide tRNase activity to S4 pyocin (Parret and De Mot 2000). Domain III of S5 pyocin is homologous to the active domain of pore forming colicins (Parret and De Mot 2000), and thus likely mediates the formation of voltage-gated ion channels in a manner similar to colicins la and lb (Wiener, Freymann et al. 1997). Closely linked to each of the large proteins is the corresponding immunity protein, providing specific immunity of the pyocinogenic strain towards its pyocin. Expression of the immunity protein is coupled to that of the larger killing protein through positioning of the ribosome binding site of the immunity protein within the C-terminal region of the killing protein. In this manner, the ribosome comes into contact with both the termination codon of the larger protein and the initiation codon of the immunity protein simultaneously, and likely couples their transcription, thereby providing protection of the pyocinogenic strain towards its own pyocin (Sano, Matsui et al. 1993). In contrast to the S-type pyocins, R-type and F-type pyocins are protease and nuclease insensitive, properties which facilitate their separation from S-type pyocins (Table 1). R-type pyocins resemble inflexible and contractile tails of P2 bacteriophage family (Nakayama, Takashima et al. 2000), and consist of a contractible sheath and core (Figure 4 - R2 and F2 pyocins) (Ishii, Nishi et al. 1965). The contracted sheath form is 46nm long (or 120nm long when extended) and 18nm wide, and exposes a core which is also 120nm long and 5.7nm wide. A baseplate is also visible at the distal end and is associated with six tail fibres, which confer R l -R5 subgroup receptor specificity (Yui-Furihata 1972; Michel-Briand and Baysse 2002). The sequenced P. aeruginosa strain PAOl contains the pyocin subgroup R2, which is situated between trpE and the F2 pyocin cluster on the chromosome (Shinomiya, Shiga et al. 1983). 22 regulatory sequences I F box S2 bp - l l f l -77 GAT(ATTGaaGTI -11 1 k i l l i n g protein transcription terminator 2060 2074 ©3GA9G GATATG 2336 2348 TAG AACAAGCCCCGT-i TTGTTCaGOGCA-1 S I S2 A P 4 1 S3 S4 S5 <tm !Da' afttn') 1 ami (Da) *a(n°l 1 ran)1: Eat aa (n°> 1 mm (Da) aafn'l 1 mm (Da) aaln*] 65.600 10,000 241 242 487 618,1 87 74.000 10,000 216 217 313 558 6B0 1 87 83,900 10,300 239 240 400 639 776 1 90 81,400 17,000 276 277 389 637 768 1 153 .81,000 .13,000 248 249 420 664 764 1 112 , .,, . lnmunity k i l l i n g protein protein Figure 3: S-type pyocin organization. Upper panel: The S2-type pyocin is constituted of an ORF1, encoding a killing protein and of an ORF2, encoding an immunity protein. The regulatory sequences constitute the P-box and may be triggered by PrtN protein. SD; Shine Dalgarno sequence, bp; base pair. Lower panel: Molecular mass (mm, Dalton) of the different S-type pyocins and amino acid number (aa) which constitutes the different domains. Reprinted with permission from reference Michel-Briand, Y. and C. Baysse 2002. 23 20nm 4 Figure 4: Schematic representation of R-type and F-type pyocins. R-type pyocin: The native form appears as an extended sheath (ES) composed of 34 annuli (a), each in turn composed of six subunits. The contracted sheath (CS) form reveals the core (C) and terminates in a base plate (BP) with tail fibres (TFi). F-type pyocin: It appears as a flexuous non-contractile rod composed of 23 annuli and a distal part (DP) from which extends a fiber part (Fi) composed of three filaments with some globular structures. Reprinted with permission from reference Michel-Briand, Y. and C. Baysse 2002. 24 Receptor sites for R-type pyocins involve core oligosaccharides of lipopolysaccharide (Meadow 1978). Following receptor adsorption and contraction of the tail-like structure (Shinomiya, Osumi et al. 1975), the bactericidal activity of R-type pyocins involves depolarization of the cytoplasmic membrane following pore formation (Uratani and Hoshino 1984) and inhibition of macromolecular synthesis (Ohsumi, Shinomiya et al. 1980). Similar to R-type pyocins, F-type pyocins resemble flexible, non-contractile tails of the X bacteriophage family (Table 1) (Nakayama, Takashima et al. 2000). F-type pyocins are also protease and nuclease insensitive. Electron microscope analysis of F-type pyocins has revealed a rod-like structure of 106nm in length, with fibre-like structures located at the distal end (Figure 4) which confer F1-F3 subgroup receptor specificity (Kuroda and Kageyama 1981). The sequenced P. aeruginosa strain PAOl contains the pyocin subgroup F2, which is also situated between the R2 pyocin gene cluster and trpGCD on the chromosome. Comparison of the nucleotide sequence of the R2/F2 pyocin encoding region from P. aeruginosa strain PAOl to those of various bacteriophage, revealed similarities in addition to the aforementioned structural similarities (Nakayama, Takashima et al. 2000). It was determined that the R-type pyocin is derived from a common ancestral origin with P2 phage, and the F-type pyocin from X phage (Nakayama, Takashima et al. 2000). Based on the absence of genes for head formation, replication and integration and the coordinated regulation of the R2/F2 and S-type pyocins however, the R2 and F2 pyocins are regarded as being phage tails that have evolutionarily specialized as bacteriocins, rather than simple defective phages (Nakayama, Takashima et al. 2000). From the results of the sequence comparison, functions for the majority of ORFs identified in the R2/F2 pyocin region have been assigned (Table 2). Of particular interest and relevance are the ORFs encoding for putative holin and lysin genes. Although release of pyocins from the pyocinogenic cell involves lysis, no genes related to lysis have previously been described for any pyocin. These ORFs then constitute the first genes for which lysis function can be attributed. PA0614 was found to be similar to holins of P2 and X phage, while PA0629 was found to have weak but significant similarity to the active domains of chitinases, which hydrolyze N-acetyl-D-glucosamine (Nakayama, Takashima et al. 2000) (Table 2). PA0630 and PA0631 were also found to be similar to the lysis control genes of P2 phage, lysB and lysC (Ziermann, Bartlett et al. 1994) (Table 2). Nakayama et al. (Nakayama, Takashima et al. 2000) further cloned and characterized the activities of PA0614 and PA0629 by expressing 25 Table 2: R2/F2 pyocin-phage operon of Pseudomonas aeruginosa. ORF a Gene Nameb Description PA0610 prtN Transcriptional activator PA0611 prtR Transcriptional repressor PA0612 « Homolog of Zn finger protein PA0613 Conserved hypothetical PA0614 hoi Holin PA0615 Conserved hypothetical PA0616 VR2 Homologous to baseplate assembly protein V PA0617 WR2 Homologous to baseplate assembly protein W PA0618 JR2 Homologous to baseplate assembly protein J PA0619 IR2 Homologous to tail protein I PA0620 HR2 Homologous to tail fibre protein H PA0621 Homologous to tail fibre assembly protein PA0622 FIR2 Homologous to contractile sheath protein Fl PA0623 FIIR2 Homologous to tail tube protein FII PA0624 Conserved hypothetical PA0625 Homologous to tail length determination protein PA0626 UR2 Homologous to tail formation protein U PA0627 XR2 Homologous to tail protein X PA0628 DR2 Homologous to tail formation protein D PA0629 lys Lytic protein; Homology to predicted chitinase PA0630 Hypothetical protein PA0631 Unique hypothetical protein PA0632 Unique hypothetical protein PA0633 Homologous to major tail protein V PA0634 Unique hypothetical protein PA0635 Conserved hypothetical protein PA0636 HF2 Homologous to tail length determination protein H PA0637 MF2 Homologous to tail fibre protein M PA0638 LF2 Homologous to tail fibre protein L PA0639 KF2 Homologous to tail assembly protein K PA0640 IF2 Homologous to tail assembly protein I PA0641 JF2 Homologous to tail fibre protein J PA0642 Hypothetical protein PA0643 Homologous to tail fibre domain protein PA0644 Hypothetical protein PA0645 Hypothetical protein PA0646 Homologous to putative tail fibre protein PA0647 Conserved hypothetical protein PA0648 Conserved hypothetical protein a - Genes are identified by ORF designation based on the Pseudomonas genome project (www.pseudomonas.com). b - Genes were named as per Table 1 of Nakayama, K. K. Takashima, et al. 2000 and reflect the homology observed by these authors to phages P2 and X. 26 each of the ORFs alone or in combination in P. aeruginosa. Their work confirmed the cytoplasmic permeabilizing action of PA0614 and lytic ability of PA0629, and indicated that PA0614 and PA0629 respectively encode a holin and lytic enzyme shared by the R2 and F2 pyocins. Furthermore, their work also suggested that this lytic system was likely to be shared by the S-type pyocins, since this group of pyocins lacks any known lysis genes (Nakayama, Takashima et al. 2000). Regulation of pyocin synthesis occurs through a P-box regulatory sequence located upstream of the ribosome binding site. The P-box consensus motif ATTGnn(n)GTnn(n), is repeated twice upstream of the R2 pyocin and four times upstream of the S-type pyocins (Matsui, Sano et al. 1993), with the exception of pyocin S5 which is likely regulated in a manner different from the other pyocins (Parret and De Mot 2000). The P-box motif serves as a binding site for the transcriptional activator protein, PrtN. PrtR protein however, is a repressor protein and functions to downregulate expression from the prtN gene, unless PrtR is inactivated by the action of activated RecA protein. RecA is an SOS responsive protein, and is therefore activated by various stresses including DNA damage induced by UV irradiation or other mutagens like ciprofloxacin (Miller and Kokjohn 1988). Activated RecA protein cleaves PrtR protein, releasing inhibition on the expression of prtN. Binding of PrtN protein to the P-box motif then initiates expression of the associated pyocins (Matsui, Sano et al. 1993) in a manner suggestive of their responsiveness to bacterial stress. E. Microarray Analysis of Gene Expression While the aforementioned work with sub-inhibitory ciprofloxacin has expanded our understanding of the many phenotypic changes induced by sub-MIC quinolones, the effect of sub-MIC quinolones on antibiotic resistance in P. aeruginosa has not been well studied. Such knowledge may improve dosing regimes for CF patients afflicted with persistent P. aeruginosa infections, or aid in the design of quinolone antibiotics which do not induce these resistance mechanisms. However, the thoroughness required of such knowledge is dependent on techniques that elucidate the underlying genetic and biochemical mechanisms on a broader level than current individual gene approaches allow. The availability of the whole P. aeruginosa genome (Stover, Pham et al. 2000) along with the development of microarray technology in the late 27 1990s (Schena, Shalon et al. 1995) has made whole genome expression profiling of the responses of P. aeruginosa to sub-inhibitory ciprofloxacin possible. At the onset of this thesis project however, commercial microarray platforms for P. aeruginosa were unavailable. Furthermore, bacterial microarrays themselves were only just gaining in popularity. Thus, standard methodologies and protocols for bacterial microarrays were not prevalent, necessitating the work outlined in the first sections of this thesis. Several other methods for transcriptome analysis surfaced around the same time (Liang and Pardee 1992; Velculescu, Zhang et al. 1995) but were not considered for this project. a. Microarray Platforms: Advantages and Disadvantages Through specific base pairing interactions, all nucleic acid strands carry the capacity to recognize their complementary sequence with high sensitivity. Complementary base pairing is the basic tenet behind the various microarray platforms: hybridization of an unknown target sample to an ordered array of immobilized DNA molecules of known sequence produces a specific hybridization pattern. Simultaneous hybridization with a known target results in competitive hybridization between the two target samples, and facilitates comparison of the differential transcriptome patterns produced from these target samples. While microarrays may vary in any number of ways from the type of solid support used to the type of DNA fragments on the array, essentially only two main microarray platforms exist: in situ synthesized oligonucleotide arrays and spotted glass slides, each with its own advantages and disadvantages. Only those features relevant to this thesis project will be discussed (for reviews in the field refer to (Bowtell 1999; Holloway, van Laar et al. 2002)). In situ synthesized oligonucleotide microarrays, such as Affymetrix GeneChips®, involve the synthesis of single stranded oligonucleotides by use of photolithographic techniques (Lockhart, Dong et al. 1996) and enable the generation of high density microarrays. Although this methodology allows for the reproducible generation of microarrays directly from information in a sequence database, it also requires complex equipment and thus tends to be restricted to commercial manufacturing. For the most part, the associated cost of producing such microarrays prohibits their accessibility for the academic user. An Affymetrix GeneChip® to the P. aeruginosa genome is commercially available, but was not used for this thesis work, even for 28 comparison purposes. Comparisons between the two microarray platforms have shown poor correlation owing to probe GC content, length, and cross hybridization (Kuo, Jenssen et al. 2002). Spotted glass slide microarrays are based on the delivery of prepared genetic material onto a glass slide. Molecules such as cDNAs, PCR products or short oligonucleotides are deposited and attached to the platform using either spotting pin or ink jet printing technologies. In addition to the associated low cost, the flexibility and versatility of such technology makes this particular microarray platform attractive to academics. The main disadvantage of spotted microarrays however is the requirement for synthesis, purification and storage of all genetic material prior to fabrication of the platform. Fabrication of spotted glass microarrays then typically involves robotic printing of purified, highly concentrated probes onto glass slides, which have low inherent fluorescence. These are typically coated with either poly-lysine or amino silane, which enhance both the hydrophobicity of the slide and the adherence of the deposited DNA, as well as limit the spread of the spotted DNA droplet on the slide (Duggan, Bittner et al. 1999). The DNA probe is then cross-linked to the glass slide by ultraviolet irradiation to achieve covalent bond formation between the thymidine residues in DNA and the positively charged group on the slide (Cheung, Morley et al. 1999). The state of the bound DNA probe is poorly defined, although differences in the DNA probe length have been found to mediate differences in hybridization results (Stillman and Tonkinson 2001; Chou, Chen et al. 2004). While shorter oligonucleotides may be more accessible for target hybridization because of their single attachment point to the microarray, their specificity due to truncated size and their variations in melting temperature due to variations in AT-GC composition, are problematic in single oligonucleotide per gene array designs. Longer PCR probes circumvent these issues and offer other advantages like specificity for homologous gene families when designed over the unique regions (Evertsz, Au-Young et al. 2001). b. Standards in Microarray Experimentation Underscoring any experiment, and in particular any microarray experiment, is the requirement for a sound biological question. Once the biological query has been established, 29 some fundamental .issues of how to design an experiment to ensure that the resulting data is amenable to statistical analysis must be considered (Churchill 2002; Yang and Speed 2002). Microarray data is inherently variable, and it is this variability which affects the significance of the microarray findings. Microarray data variation may be classified as being either experimental or systematic in nature, where each type of variation can be controlled somewhat through various design features of the microarray experiment. Experimental variation is further subdivided into biological and technical variation. Biological variation is intrinsic to all organisms but can be influenced by genetic or environmental factors. Technical variation is introduced during the extraction, labeling and hybridization of the RNA sample (Churchill 2002). With respect to the degree of variation observed from each of these sources, biological replicates introduce a larger degree of variation than technical replicates. While it is tempting to avoid including biological replicates so as to make the data appear more reproducible, inclusion of such replicates is the equivalent of generating independent experimental results. With biological replicates then, any statistically significant findings become more meaningful because they were found to occur across different biological samples. Likewise, inclusion of technical replicates affects the precision of the results, and is generally used to reduce the overall variability of the independent experiments. In general, biological replicates are used to obtain averages of independent data and to validate generalizations of conclusions, while technical replicates are used to assist in reducing the variability of these conclusions (Yang and Speed 2002). Experimental variation may also be addressed through careful consideration of the experimental design or how to pair samples for hybridization. The efficiency of an experimental design for comparing two samples is determined by the length and the number of paths connecting the two samples (Kerr and Churchill 2001; Yang and Speed 2002). For example, comparison of an unknown sample on slide A to a known sample on slide B, is less accurate than comparison of the two samples on the same slide. Hence it is most efficient to make the comparisons of greatest interest directly on the same slide, a so called direct comparison of differential expression. An all-pairs design (Yang and Speed 2002) or loop design (Kerr and Churchill 2001) is ideal, since these direct comparison designs are both efficient in using fewer slides and provide precise comparisons with low variance. These designs combined with a dye-swapping strategy, represent an effective design for the direct comparison of two samples (Kerr 30 and Churchill 2001). When comparing two samples, such a design involves two arrays, where on the first array the control sample is labeled with red fluorophore and the treatment sample is labeled with green fluorophore. On the second array, these fluorophore assignments are switched. This arrangement repeated for a given set of technical samples constitutes a repeated dye-swap experiment, and when replicated for independent biological samples constitutes a replicated dye-swap experiment (Churchill 2002). Such a replicated dye-swap experiment can account for both technical and biological variation. Systematic variation refers to the variation attributable to dust or scratches on the microarray, irregular deposition of probe on the glass slide or biases associated with different fluorescent dyes. All of these affect accurate reading of the fluorescent signals, and thus also affect the significance of the microarray data. Systematic variation is addressed by including various controls on the microarray slide and through the dye-swap experiments previously mentioned. It is well recognized that the Cy3 and Cy5 fluorescent dyes exhibit different quantum yields and are differentially sensitive to photobleaching (Churchill 2002). Furthermore, there has been indication that direct incorporation of these dyes into the cDNA introduces sequence specific artifacts (Kerr and Churchill 2001) because of the differing rates at which the bulky dyes can be incorporated into the cDNA molecule by the reverse transcriptase. Experimental designs which include dye-swaps then, are integral to addressing this type of systematic variation. Other types of systematic variation may be corrected through a process called normalization (Quackenbush 2002). Several methods for normalization exist, but the underlying principle is the requirement for a set of genes that are not affected by the experimental condition and thus remain unchanged between the experimental and control conditions. Global normalization is the most common normalization strategy and assumes that the majority of genes in a microarray experiment do not change their expression and that only a small subset of genes is differentially expressed between the two conditions. Other normalization strategies, which make use of exogenous spike-in controls, are becoming more popular (Benes and Muckenthaler 2003). These strategies facilitate assessment of signal linearity, hybridization specificity and consistency across microarray slides and slide to slide comparisons, all factors to be considered in addressing systematic variation. After establishing an appropriate experimental design for the biological query, all microarray experiments involve four main steps: sample preparation, array hybridization, detection and data handling. Sample preparation involves the isolation and fluorescent labeling of mRNA. Isolation 31 of pure RNA of high quality is the most critical factor in attaining meaningful microarray data. Because the microarray's measurement of the relative transcript abundance between two samples is dependent on high quality mRNA, factors which alter the half life or integrity of the mRNA species therefore will affect the experimental outcome. Bacteria have developed both tight transcriptional control and RNA degradation mechanisms in order to respond very rapidly to changes in the environment, (Bernstein, Khodursky et al. 2002; Hambraeus, von Wachenfeldt et al. 2003). Thus, care must be taken during RNA isolation and all subsequent RNA handling steps, to ensure the immediate cessation of transcription while simultaneously preventing RNA degradation, so that the integrity and accuracy of the mRNA population is adequately maintained. The next step in sample preparation is the fluorescent labeling of mRNA transcripts. This step however is also a major challenge in prokaryotic microarray experiments because the absence of polyadenylation on prokaryotic transcripts makes labeling of an isolated mRNA population difficult. Instead, total RNA is isolated and labeled through a reverse transcription reaction using either a pool of gene specific primers or random oligonucleotide primers. However, with respect to the commonly used random oligonucleotide priming, since rRNA and tRNA make up 95-97% of the total bacterial RNA, most of the dye is incorporated into these species rather than mRNA. The secondary structure of RNA species themselves further inhibits efficient dye incorporation. In spite of these limitations, specific hybridization of mRNA to the microarray can be achieved (de Saizieu, Certa et al. 1998). Following sample preparation, labeled cDNA is hybridized to the microarray slide. Essentially, hybridization is the competition between two sets of transcripts, the unknown and known transcriptomes, for the binding to corresponding sequences on the microarray. This competition is affected by many of the principles which affect other DNA-DNA interactions and include such factors as hybridization temperature and volume, transcript to probe specificity, length of transcript-probe specificity, ratio of probe to transcript, and length of hybridization time; many of these factors are interrelated. These factors however may be loosely divided into probe specific, transcript specific and probe-transcript specific categories. The probe specific factors such as probe sequence, probe length and probe concentration, must be considered in the design and construction of the microarray as previously noted. All transcript specific factors depend upon high quality RNA of good integrity. Ideally, transcripts will be full length and a perfect sequence match to the arrayed probe. Alterations in transcript length in addition to the 32 choice of probe, will therefore affect this sequence specificity. RNA quality and integrity furthermore affect the transcriptome competition associated with hybridization, since an RNA sample of low yield and quality will have fewer sequence specific transcripts available for competition to the sequence specific probe, giving an inaccurate account of the true state of the transcriptome for this sample. Careful consideration of hybridization temperature, volume and time further helps in the optimization of those parameters required to drive the hybridization reaction to completion. For example, larger hybridization volumes allow for more efficient mixing and thus allow for more cDNA targets to come into contact with their cognate probe. Likewise, longer hybridization times offer more opportunity for these interactions to occur. Detection of the differential expression between two samples is dependent on the efficiency of the hybridization reaction, as well as on the detection machinery. With respect to the hybridization reaction, the efficiency of the cDNA labeling reaction is most critical, since differential incorporation of the fluorophores into the two cDNA populations affects the detection results and thus accuracy of the transcriptome findings. Recently, there has been movement towards measuring fluorophore incorporation efficiencies and adjusting the amount of transcript in the hybridization reaction to compensate for any differences, rather than beginning with equal quantities of total RNA and working under the assumption that fluorophore incorporation efficiencies are equal. Binding or hybridization events between the fluorescently labeled target and the cognate probe are detected by scanning the microarray, typically using a confocal laser scanner. All slides within an experiment should be scanned using the same scanner to minimize differences in the results introduced by different scanner models. c. Statistical Analysis of Microarray Data To date, there is no consensus approach to statistical analysis, and thus microarray results tend to be analyzed in a variety of different ways (Slonim 2002). Yet a minimum set of analytical techniques should be applied to correct for the systematic variation introduced during the experiment. Starting with the initial images taken from the scanner, the probe-target interactions are detected, delineated and quantified. Background must be separated from the target signal and its contribution removed in order to generate a measurement of the intensity of hybridization for each gene element on the microarray. A number of data transformations must then be carried out 33 so to adjust the measured intensities to facilitate comparisons between sample conditions, to eliminate low quality measurements and to select genes that are significantly differentially expressed between the sample conditions (Quackenbush 2002). Normalization refers to the adjustment or balancing of the individual signal intensities, and accounts for such systematic variation as unequal quantities of starting RNA and fluorophore incorporation and detection inefficiencies. By normalizing the signal intensities within a slide, meaningful comparisons between the two sample conditions can be made. Locally weighted linear regression (lowess) analysis is the current popular strategy for normalization (Quackenbush 2002), since it can remove intensity dependent effects such as those seen for low intensity spots. Furthermore, lowess normalization is often applied in a local fashion to microarray sub-grids rather than globally to the whole microarray, because local normalization can help correct for such spatial systematic variation such as spotting pin inconsistencies, variability in the slide surface and differences in the hybridization across the microarray (Quackenbush 2002). Since the relative error increases at low signal intensity where the signal approaches that of the background, data variability increases as the measured signal intensity decreases. Data then is typically trimmed to eliminate these low quality measurements by using significance or percentage based cut-offs below which an element is filtered out. In this manner, only those elements which have signal intensities that are statistically significantly different from the background are considered. Likewise, a ceiling or maximum signal value is set to eliminate saturating elements. Identification of genes that are differentially expressed between two samples is typically the desired outcome of many microarray experiments. In early microarray experiments, a fixed fold change cut-off was often used to identify these genes. A more sophisticated approach, involves calculating the Z-score or the mean and standard deviation of the distribution of the log2 ratios, and defining a global fold-change difference and confidence (Quackenbush 2002). Once the differentially expressed genes have been identified, the data set can be further refined and filtered to better address the biological query. As well, the data set may be mined using hierarchical clustering or similar clustering techniques for expression trends or profiles exhibited by one sample and not the other. 3 4 d. Methods of Confirming Microarray Data Spending time at the onset to optimize the microarray experiment as previously described, reduces the amount of time and effort spent subsequently to invalidate erroneous expression findings. When evaluating microarray data, two important questions need to be considered. First, is whether the expression results are accurate for the biological query, and second, whether the data fundamentally describes the phenomenon being examined (Chuaqui, Bonner et al. 2002). To ascertain whether the expression results are accurate, it is important to confirm the data using an independent method of gene expression monitoring. Given the volume of data obtained from a microarray experiment however, verification of only a small but relevant subset of transcripts is feasible. There are two approaches to independent confirmation of microarray data: in silico analysis and laboratory based analysis (Chuaqui, Bonner et al. 2002). In silico analysis compares the array results to information publicly available, thereby avoiding further experimentation. Conversely, laboratory based analysis relies on additional independent experimentation to verify the observed transcriptional profile. Commonly used techniques include semi-quantitative reverse transcription PCR (RT-PCR), real-time PCR, northern blot, ribonuclease protection assay and in situ hybridization. The results from these independent validation techniques typically only show similar trends to those found by microarray analysis, since it has been demonstrated that cDNA microarrays consistently underestimate the relative transcript abundance between the known and unknown samples (Benes and Muckenthaler 2003). To ascertain whether the expression results are descriptive of the biological state being examined, a critical set of genes should be examined further in a more extensive set of experiments. For example, finding of a distinct expression profile in the treated group would warrant closer examination of those genes belonging to this profile. Closer examination might involve examining the significance of these genes in the treated group through knock-out studies or various other inhibitory assays. Overall, more extensive experimentation should lead to determination of whether these genes and their expression profiles are accurate representations of the biological state. 35 F. Rationale and Aims of this Study The influence and interaction of antimicrobials on bacterial responses, particularly antibiotic resistance, is poorly understood, yet likely contributes to the persistence of P. aeruginosa infection in chronically infected CF patients. Since the administration of the fluoroquinolone ciprofloxacin to CF patients infected with P. aeruginosa typically only results in the maintenance of low bacterial levels and not elimination of the pathogen, one of the goals of this thesis was to ascertain the effect of sub-inhibitory and inhibitory ciprofloxacin on adaptive resistance. Adaptive resistance is defined as an unstable, reversible resistance that is unrelated to genetic mutation (Barclay and Begg 2001), and occurs transiently in an organism under non-lethal selective pressure. I hypothesized that sub-MIC concentrations of ciprofloxacin could induce adaptive resistance like mechanisms in P. aeruginosa. A second goal of this work was to more closely and broadly examine the effect of sub-inhibitory and inhibitory ciprofloxacin on P. aeruginosa through use of DNA microarray technology. This required the design and construction of a custom DNA microarray to the P. aeruginosa genome and development of its associated experimental protocols and data analysis programs, since commercial platforms were unavailable at the time and bacterial microarrays in general were only beginning to be developed. From the knowledge gained through this global analysis, it was the aim of this thesis to apply these findings to further the current understanding of the effect of sub-inhibitory antibiotics, in particular ciprofloxacin, on the development of antibiotic resistance. Following upon the findings from the initial goal, I hypothesized that sub-inhibitory levels of the synthetic antimicrobial ciprofloxacin would have the ability to modify the antibiotic resistance profile of P. aeruginosa. These more detailed microarray studies revealed some unique findings, namely that sub-inhibitory ciprofloxacin up-regulated expression of the R2/F2 pyocin region in P. aeruginosa. Subsequent analysis of the susceptibility of mutants in this region found that the R2/F2 pyocin region encoded a quinolone susceptibility determinant, leading to a better understanding of the role sub-inhibitory ciprofloxacin antibiotics play in the development of resistance in P. aeruginosa. Overall, the main outcomes of this thesis work were the establishment of a P. aeruginosa custom DNA microarray and its associated protocols, and the characterization of the effect of sub-inhibitory ciprofloxacin on the transcriptional response of P. 36 aeruginosa, features of which can be applied to the design of new antimicrobials or dosing regimes which do not induce similar resistance or response profiles. 37 M A T E R I A L S A N D M E T H O D S A. Bacterial Strains and Growth Conditions All bacterial strains used in this study are described in Table 3. All strains were stored at -70°C until thawed for use. Strains were grown at 37°C in Luria-Bertani broth (LB; 1.0g/l tryptone, 0.5g/l yeast extract, 5g/l NaCl; Difco Laboratories, Detroit, MI). Transposon insertion mutants were cultured overnight in LB broth containing 50pg/ml tetracycline (Sigma-Aldrich Corp., St. Louis, MO). Mutants obtained from the University of Washington Genome Center (Seattle, WA) were grown in LB broth supplemented with 60ug/ml tetracycline. In all experiments, overnight aerobic cultures were grown with agitation in Luria-Bertani broth at 37°C, and used at a 1/1000 dilution to inoculate 50-ml portions of LB broth in 500-ml Erlenmeyer flasks. For ciprofloxacin experiments, P. aeruginosa cultures were either untreated or treated with O.lx-, 0.3x- or lx- of their respective ciprofloxacin MIC (refer to Table 5 in the Results section), and the resulting cultures grown at 37°C with agitation to the mid-logarithmic phase of growth (OD600, 0.5 to 0.6). B. Chemicals Antimicrobials were made fresh daily and were obtained from the following sources: ciprofloxacin hydrochloride (Bayer, UK) enofloxacin (Warner Lambert Co., Ann Arbor, MI), gentamycin sulfate (ICN Biomedicals Inc., Aurora, Ohio), mitomycin, nalidixic acid, norfloxacin and novobiocin (Sigma-Aldrich Corp.), ceftazidime (GlaxoSmithKline Beecham, Inc.), cefepime (Bristol Myers Squibb, New York, NY), and amikacin sulfate (Bristol Laboratories, Belleville, Ont). C. Optimization and Development of Microarray Parameters a. Genomic DNA Isolation 10ml of LB broth was inoculated with an isolated colony of PAOl-HI 03 cells and incubated 38 Table 3: Bacterial strains used in this study. Strain Genotype Abbreviation Reference H103 Wild type P. aeruginosa PAOl H103 57 PA0611 :IS/acZ/hah derivative of PAOl prtR::lSlacZ Jacobs et al., 2003 3501 PA0621 :IS/acZ/hah derivative of PAOl PA0621::IS/acZ Jacobs et al., 2003 43080 PA3617 •JSphoA/hah derivative of PAOl recA::VS>phoA Jacobs et al., 2003 PA01_ Jux _22_ E4 PA0620 •.luxCDABE derivative of HI 03; TcR ?A0620::lux This study PA01_ lux. .26. _H2 PA0641 •.luxCDABE derivative of HI 03; TcR PA0641::/MX This study PAOl. lux. .24. _A3 P A3 866 •.luxCDABE derivative of HI 03; TcR PA3866::/wx This study 39 with shaking to log phase at 37°C (OD6oo, 0.5 to 0.6). Genomic DNA was extracted twice with equal volumes of phenol/chloroform/isoamyl alcohol (25:24:1, Sigma-Aldrich Corp.) and once with an equal volume of chloroform (Sigma-Aldrich Corp.). Genomic DNA was precipitated with 2 volumes ice-cold 100% ethanol and 1/10 volume sodium acetate (3M, pH5.2, Ambion, Inc., Austin, TX). The genomic DNA was washed with ice-cold 70% ethanol, the residual ethanol evaporated and the DNA precipitate resuspended in lOOpl RNase free db^O (Ambion, Inc.). The quality of the genomic DNA was assessed by electrophoresis on a 1% agarose gel containing ethidium bromide (EtBr) in comparison to DNA size markers and visualized under a UV light box (Chemigenus2, Syngene). Samples were quantified by UV absorption at 260nm. b. PCR amplification from Genomic DNA Amplicons of various size (400, 600 and 800 bp) to oprD were generated from genomic DNA isolated from PAO-H103. Primers used here and throughout are listed in Table 4 and were designed based on the published P. aeruginosa sequence (Stover, Pham et al. 2000) (www.pseudomonas.com) using the Primer3 program (Whitehead Institute for Biomedical Research). All primers were synthesized and purified by AlphaDNA (Montreal, PQ) unless otherwise stated. Reactions contained lug genomic DNA, 3.5mM MgCb, 200uM each of dATP, dCTP, dGTP and dTTP (Invitrogen Corp., Carlsbad CA), lOOnM each primer (oprD-Rev and oprD-400, oprD-600 or opr£>-800), and 1.25U Taq DNA polymerase in lx PCR buffer (Invitrogen Corp.). Amplification proceeded on a Minicylcer (MJ Research, Inc., Reno, NV) under the following cycling conditions: 2min at 94°C, followed by 30 cycles of lmin at 94°C, lmin at 65°C and lmin at 72°C, and then 5min at 70°C. Amplicons were purified using QIAquick PCR purification kit according to the manufacturer's instructions (Qiagen Inc., Mississauga, Ont), and amplicon quality and molecular size confirmed by electrophoresis on a 1% agarose gel containing EtBr, and visualized using a UV light box. c. Synthesis of P-labeled Probe 32 Randomly P-radiolabeled DNA probes were prepared from 25ng of various oprD amplicons (400, 600 and 800bp) and used in subsequent optimization experiments. Various 40 Table 4: Nucleotide sequences of primers used in this study. Primer Name Primer Sequence (5' to 3') oprD-Rov ACGCGGTCTCGGCAACGCCGGCTT oprD-400 CAGCGAATTCGAAGGGCTCGACCTCGAGG oprD-600 GTGATGAACGACGGCAAGCCG oprD-800 AGCAGCCTCGACCTGCTGCTCCGC T7 TAATACGACTCACTATAGG pSPORT TAGGTGACACTATAGAAGAGC RUP ACAGGAAACAGCTATGACCAT GFP AGACAAGTTGGTAATGGTAGCGA oprD-Y or GTACTTGGCTTCGAGGTTGG oprD-Re\2 GCTACCTGGGCCTGAAGC rpoC-F 102 CAGCGAACGCAAGCGTCAG rpoC-F239 TCCGCGGCGTCGTTCCAG rpoC-F366 CCAGGTGGAACTCACCCAG rpoC-f'610 ACGTGTTCGAAGGCGAACAG rpoC-FlAO TCGATCCTGGCGGAAATCAG rpoC-F923 ACCGACGTACCGGCGCAG rpoC-Y\012 GCATCACCGTCAAGCGTCAG rpoC-F\230 CGAGCGCGAGCGCTACAAG rpoC-Rev TTAGTTACCGCTCGAGTTCAG rpsL-For GCAACTATCAACCAGCTGGTG rpsL-Rev GCTGTGCTCTTGCAGGTTGTG Random primer (NS)5 where N=A,T,C,G and S=C,G PA3866-For CCACTTGTCGTGACCAGAGGA PA3866-Rev CATCGACCCAGGCTCGTAA PA3617-For GTGAAGAACAAGGTTTCCCCG PA3617-Rev GAGGATCTGGAACTCGGCCT PA0610-For TAGCACTCCGATTCCACGC PA0610-Rev CCGAAGATGCGGTAGACCA PA0611-For AGCTTCAACCGCGAGGAATA PA0611-Rev CATGTCCTCCGGCGAGTACT PA0621-For TTTCCCGTCAGCAACGTAGC PA0621-Rev GCTGACTATCCCGCCATCTC PA0623-For CCGAGAAGCGCTGAATTTCT PA0623-Rev CCATTGAAAGCGCTCTGGTC PA0642-For GCTGCACCTCCTGTTCTAGC PA0642-Rev TCGAACACGAAGTCCATATCC PA0648-For GTGCAGGTGTGGAGACGGAT PA0648-Rev TCTCTTCGACCTTGGCAAGC PA4597-For TCATCGTCGATGCCGAACTAC PA4597-Rev TGTTATCCAGGGCCATGTCC rplF-For AGGTTGCTGCCGAAATTCG rp/F-Rev CTTGCCTTTGTAAGGCTCCG 41 amplicon sizes were radiolabeled to better mimic the hybridization events of subsequent gene expression analyses, where various lengths of cDNA probes would be used in hybridization experiments. All 32P-labeled probes were prepared according to kit instructions (RediPrime II, Amersham Biosciences, Baie d'Urfe, PQ). In brief, the amplicon was diluted in 45ul lOmM Tris HC1 pH 8.0; 1 mM EDTA in RNase free dH 20 and denatured by heating to 100°C for 5min and cooling on ice for 5min. The denatured amplicon was transferred to a reaction tube, and the addition of 5 pi of fresh 3 2 aP-dCTP added and mixed by pipetting. Random incorporation of radioactivity was allowed to occur for lOmin at 37°C, at which point the reaction was stopped with 5 pi of 0.2M EDTA. Radiolabeled DNA probe was stored at -20°C for up to two weeks, the first half-life for the radioisotope. d. Preparation of Nylon Membrane DNA Macroarray For all subsequent development and optimization experiments involving determination of amplicon size, concentration and hybridization temperature and solution, the specified PCR amplicons were crosslinked to nylon membranes. PCR amplicons were suspended at the desired concentration in spotting buffer (0.4M NaOH; lOmM EDTA (pH 8.2) in RNase free dH20). Samples were then heated to 100°C for lOmin to denature the DNA and immediately placed on ice. Following a brief spin at 4°C, samples were spotted onto positively charged nylon membranes (Boehringer Mannheim) at the desired volume and allowed to air dry. Dried membranes were overlaid onto filter paper soaked with alkaline denaturing solution (1.5M NaCl; 0.5M NaOH in RNase free dH20) for lOmin, and then neutralizing solution (1M NaCl; 0.5M Tris HC1 (pH 7.0) in RNase free dH20) for 5 minutes. Membranes were allowed to dry and were then baked for 30 minutes at 80°C in a Tek Star Jr. hybridization oven (Bio/Can Scientific). To crosslink the denatured amplicons to the membrane, the membranes were wrapped in transparent plastic wrap and exposed to UV light for 30 seconds (the time being calculated specifically for the UV source according to the formula in Current Protocols in Molecular Biology, 3.19.5 (Ausubel 1997)). Membranes were stored between filter papers at 4°C for up to three months. 42 e. Hybridization and Image Analysis Hybridization experiments followed the standard protocol outlined in Current Protocols in Molecular Biology, 4.9.7 (Ausubel 1997). Briefly, the DNA membrane was placed in a hybridization tube and incubated at 45°C for 3 hours with 5ml of prehybridization solution (5X SSC; 5X Denhardt solution; 50% w/v formamide; 1% w/v sodium lauryl sulfate (SDS); lOOug/ml denatured salmon sperm DNA) to block nonspecific binding. The desired 32P-labeled probe (400, 600 or 800bp oprD amplicons) was heated to 100°C for 5min and then chilled on ice for 5min to denature the probe. As per kit instructions (RediPrime II), 14pl of radiolabeled probe was added to the hybridization tube and the membrane incubated overnight at 42°C. The membrane was subsequently washed twice each in 5ml of 2X SSC/0.1%> SDS and 0.2X SSC/0.1%) SDS, while rotating at room temperature for 5min. Two further washes in 5ml of 0.2X SSC/0.1%) SDS were then performed while rotating at 45°C for 15 minutes. Membranes were wrapped in transparent plastic wrap and placed overnight in a photoimager cassette for autoradiographic imaging. A Phosphoimager SI (Molecular Dynamics) and ImageQuant vl.2 (Molecular Dynamics) software were used to visualize and quantify the hybridization signals. Local median background was subtracted from the volume of the rectangle drawn around each pair of gene spots on the membrane and divided by the rectangular area. The hybridization signals for each pair of gene spots were averaged to reduce errors in spotting technique or hybridization efficiency. The hybridization signal of one membrane was then compared to the respective spots on another membrane analyzed in an identical fashion. f. Determination of Amplicon Size, Concentration and Volume To determine which size, concentration and volume of PCR amplicon would yield the best signal when hybridized to different sizes of 32P-labeled DNA probe, 400, 600 and 800bp amplicons of oprD were PCR amplified and purified (as previously described) and spotted in duplicate onto positively charged nylon membranes. For size/concentration experiments, 1, 3, 5 and lOng of each amplicon was spotted at a 0.5 pi volume and hybridized with various sized probes (400, 600 and 800bp) in standard hybridization solution at 42°C. For size/volume experiments, 0.5, 1,2 and 5pi volumes of each amplicon was spotted at a lOng concentration and 43 hybridized with various sized probes (400, 600 and 800bp) in standard hybridization solution at 42°C. Amplicons were crosslinked to the membranes as previously described. g. Determination of Hybridization Temperature and Solution Various hybridization solutions and temperatures were examined to determine which set of conditions resulted in the best hybridization signal between 600bp 32P-labeled oprD and 0.5pl microspots containing lOng of 600bp OprD amplicon cross linked to nylon membranes (as previously described). UltraHyb (Ambion, Inc.), EpressHyb (Clontech) and a standard hybridization solution (5X SSC; 5X Denhardt solution; 50% w/v formamide; 1% w/v SDS; lOOpg/ml denatured salmon sperm DNA) were compared. For hybridization experiments with UltraHyb solution, the solution was preheated to 68°C until all precipitate dissolved. 5ml of hybridization solution was added to the hybridization tube containing a DNA membrane, and the membrane allowed to prehybridize for 30 minutes at 42°C while rotating. 14pl of 32P-labeled oprD was added to the hybridization solution and the membrane incubated overnight at 42°C while rotating. The membrane was then washed twice each in 5ml of 2X SSC/0.1% SDS and 0.1X SSC/0.1% SDS at 42°C for five minutes and 15 minutes, respectively, and analyzed as previously described. For hybridization experiments with ExpressHyb solution, the solution was preheated to 68°C until all precipitate dissolved. 5ml of hybridization solution was added to the hybridization tube containing a DNA membrane, and the membrane allowed to prehybridize for 30 minutes at 42°, 55° or 68°C while rotating. 14ul of 32P-labeled oprD was added to the hybridization solution and the membrane incubated for lhour at 42°, 55° or 68°C while rotating. The membrane was then washed thrice each in 5ml of 2X SSC/0.05% SDS and 0.1X SSC/0.1% SDS at room temperature for 15 minutes and at 50°C for 20 minutes, respectively. Hybridization experiments with the standard hybridization solution were as described in the 'Hybridization and Image Analysis' section. 44 D. Microarray Construction The Pseudomonas aeruginosa strain PAOl custom microarray described herein was made in collaboration with Chiron Co. (formerly PathoGenesis Co.; Seattle, WA). a. Primer Design Based on the publicly available sequence (Stover, Pham et al. 2000) of P. aeruginosa strain PAOl, primers were designed to every open reading frame (ORF) in the genome by Sigma-Genosys (Woodlands, TX). Care was taken to ensure the uniqueness of each primer sequence and amplicon. b. PCR Amplification, Purification and Amplicon Evaluation Each ORE in the P. aeruginosa strain PAOl genome was PCR amplified in a lOOpL reaction on Microseal skirted 96-well plates (MJ Research, Inc.) in reactions containing: 2 ng genomic DNA, 8 mM dNTP mix (Invitrogen Co.), 5% DMSO (Sigma-Aldrich Corp.), and 2.5 U Takara ExTaq polymerase (Invitrogen Co.). 40 uM primers were added last and pipette mixed. A hot-start reaction on the PTC-225 DNA Engine Tetrad cycler (MJ Research, Inc.) was cycled for: 94°C for 90sec, 10 cycles of 94°C for 30sec, 65°C for 90sec, 72°C for 90sec, and then 25 cycles of 94°C for 30sec, 60° for 90sec, 72°C for 90sec, followed by 72°C for 7min. Genomic DNA used in this set of PCR reactions was isolated from the sequenced strain PAOl, using the Bacterial Genomic DNA Purification System (Edge BioSystems; Gaithersburg, MD). Each amplicon was amplified three times in order to generate substantial volume and concentration for each of the collaborating facilities, and then pooled. Amplicons (2pl load) were checked by 1.5% agarose gel electrophoresis for yield and molecular weight, as compared to 5ul Bioline HyperLadder IV (Bioline, Randolph, MA) mass size markers. Primers yielding less than 80-lOOng product or high primer-dimer ratios or more than one amplicon, were re-synthesized by Sigma-Genosys. Amplicons were purified using the Millipore Multiscreen PCR filter plate at PathoGenesis Co. (Seattle, WA). 45 Later, amplicon yield and molecular weight were re-evaluated by capillary electrophoresis using the Caliper LabChip 90 electrophoresis system (Caliper, Hopkinton, MA). c. Sequencing of Amplicons All Amplicons were sequenced at Chiron Co. (Emeryville, WA) using an ABI 3100 Automated Capillary DNA Sequencer (Applied Biosystems, Foster City, CA). d. Re-suspension and Plate Format Transfer Following completion of the PCR amplification, purification and sequencing of the P. aeruginosa ORFs, amplicons were resuspended in 6x SSC and were transferred from a 96-well format to 384-well polystyrene plates (Nalge Nunc International, Rochester, NY) through use of the Hydra II (Robbins Scientific, Sunnyvale CA) at Xenon Genetics Inc. (Vancouver, Canada). Amplicons were diluted with distilled H2O (dH20) to achieve a final printing concentration of 25-50ng/pL in 3x SSC. For storage of source plates at -70°C, the plates were spun down and dried at 45°C and 1200 RPM in a Savant SpeedVac Plus SC210A (Telechem International, Inc., Sunnyvale, CA). e. Synthesis of Quality Control Genes A set of control genes and spots were also generated. These controls included amplicons of lysA, thrC, dapB, pheB and trpD from Bacillus subtilis (ATCC# 87482, 87483, 874844, 87485 and 87486) which were used as positive control genes on the microarray since known concentrations of complimentary spike RNA would be added to the reverse transcription reaction. These positive control amplicons were generated by PCR in reactions which contained: lOng plasmid DNA, 10 mM dNTP mix, 2% DMSO, 25pM each of T7 primer and pSPORT primer (Table 4), and 0.5U Takara ExTaq polymerase in lx PCR buffer. The PCR reaction proceeded under the following cycling conditions: 95°C for 90sec, 35 cycles of 95°C for 30sec, 55°C for 90sec and 72° for 90sec, and then 72°C for 7min. Complimentary spike transcripts were generated from plasmids isolated from overnight cultures of the ATCC clones by QIAprep 46 Miniprep columns (Qiagen, Inc.). Purified plasmids were linearized by digestion with NotI (Invitrogen Co.) and in vitro transcribed using the MEGAscript T7 kit (Ambion, Inc.) according to manufacturer's instructions. Transcribed RNA samples were treated with DNase-/ree for 15 min at 37°C to remove residual plasmid DNA and purified by RNeasy Protect Mini column (Qiagen, Inc.). The purity and quantity of RNA transcripts was evaluated by UV spectrophotometry, and RNA transcript size and integrity assessed by 2% agarose gel stained with EtBr and visualized by UV light box. Presence of RNA nucleic acid was confirmed by 2% agarose gel of samples digested with RNase I (Ambion, Inc.) according to manufacturer's instructions. As well, amplicons of the green fluorescent protein (GFP) gene were generated and utilized as sub-grid markers on the microarray since labeled GFP amplicon would be added to the labeled cDNA mix. These amplicons were generated by PCR in reactions which contained: 250ng GFP plasmid (generously donated by Gene Array Facility, Vancouver, BC), 2% DMSO, 25uM each of RUP primer and GFP primer (Table 4) and 0.5 U Takara ExTaq polymerase in lx PCR buffer. The PCR reaction proceeded under the following cycling conditions: 95°C for 2min, 35 cycles of 95°C for 30sec, 55°C for 90sec, and 72°C for 60sec, and then 72°C for 7min. GFP amplicons were purified by QIAquick PCR purification columns as above, quantified by UV spectrophotometry and resuspended for printing as for the P. aeruginosa ORF amplicons. Complimentary spike GFP amplicons were labeled using the Ready-to-Go DNA Labeling Beads dCTP kit (Amersham Pharmacia Biotech). 25ul of purified GFP amplicon at 50ng/ul was added to 24ul dH 20 and incubated for 3-5min at 95°C, at which point a labeling bead and lul of Cy-dCTP (Amersham Pharmacia Biotech) was added. The labeling reaction proceeded for lhour at 37°C, after which the labeled GFP was cleaned using a MicroSpin G-50 spin column (Amersham Pharmacia Biotech) according to the manufacturer's instructions. f. Additional Microarray Features Other controls included empty spots which served as negative controls and spots containing only 3x SSC which served as spotting solution controls. A spotting pattern was created (Figure 5) to ensure that neighbouring ORFs were not spotted adjacent to each other, and that each sub-grid was delineated at the corners by the GFP sub-grid marker. 47 1 123 123 124 124 125 125 126 126 127 127 128 128 2 2 1 116 116 117 117 118 118 119 119 120 120 121 121 122 122 1 0 108 108 109 109 110 110 111 111 112 112 113 113 114 114 115 115 100 100 101 101 102 102 103 103 104 104 105 105 106 106 107 107 92 92 93 93 94 94 95 95 96 96 97 97 98 98 99 99 84 84 85 85 86 86 87 87 88 88 89 89 90 90 91 91 76 76 77 77 78 78 79 79 80 80 81 81 82 82 83 83 69 69 70 70 71 71 72 72 73 73 1 1 74 74 75 75 61 61 62 62 63 63 64 64 65 65 66 66 67 67 68 68 54 54 55 55 56 56 0 0 57 57 58 58 59 59 60 60 46 46 47 47 48 48 49 49 50 50 51 51 52 52 53 53 39 39 40 40 41 41 42 42 0 0 43 43 44 44 45 45 31 31 32 32 33 33 34 34 35 35 36 36 37 37 38 38 23 23 24 24 25 25 26 26 27 27 28 28 29 29 30 30 16 16 17 17 0 0 18 18 19 19 20 20 21 21 22 22 1 9 9 10 10 11 11 12 12 13 13 14 14 15 15 1 Figure 5: Spotting pattern (version 3) for the P. aeruginosa custom DNA microarray. Numbers refer to quadrants on the set of 384 well plates, where each plate contains 8-48 well quadrants. 48 g. Microarray Printing and Storage All amplicons were spotted in duplicate according to the configured spotting pattern (Figure 5). Within fifteen 384-well plates and one control 384-well plate, there were a total of 122 quadrants (48-pin head areas) to be spotted (Figure 5). Using a BioRobotics Microspot 10K print tool (Genomic Solutions, Ann Arbor, MI) and BioRobotics MicroGrid II printer (Genomic Solutions) at the Gene Array Facility (Vancouver, BC), amplicons were spotted onto ArraylT superamine slides (TeleChem International Inc.). Printed slides were crosslinked in a UV Stratalinker 1800 oven (Stratagene, La Jolla, CA) set at 1200 uJoules. All printed microarray slides were stored in a cool, dry place until use. E. Microarray Quality Assessment All printed microarray slides were examined for overall grid layout, spot morphology and glass imperfections like scratches or chips, by examination of each slide under a light microscope. Slides not possessing greater than 95% spot or slide integrity were noted, and removed from the print run. This quality control process was aimed at removing any abhorrent slides from the batch prior to hybridization. a. Verification of Microarray Print Run Microarray slides were labeled with SYBR Green II (Battaglia, Salani et al. 2000) and examined under 488nm blue (argon) laser to assess the quality and quantity of the printed material. A microarray from each print batch was stained by immersion for 2min in a solution of SYBR Green II (Molecular Probes, Eugene, OR) diluted in 1:10,000 in TBE solution, pH 8.0. After staining, the slide was washed in TBE, dF^O and air dried. Scanned microarrays were examined for uniformity of spot size and printing pattern. In another assay, the spotted DNA material was terminally labeled with Cy-3 fluorophore to detect the extent, concentration and uniformity of bound material. In brief, 5x terminal transferase buffer (Roche Applied Science, Indianapolis, IN) was combined with 4pl Cy-3 dCTP (Amersham Pharmacia Biotech), 5mM CoCb, 1600U terminal transferase (added last) and dFFiO 49 to a final volume of 80pl. The entire volume was placed onto a printed microarray slide, covered with a HybriSlip hybridization cover (Grace Bio Labs, Bend, OR) and placed in a Corning hybridization chambers (Corning Life Sciences, Acton, MA) for hybridization at 37°C for 30min to lhr. Coverslips were removed by floatation in O.lx SSC, and the slide washed three times in O.lx SSC/0.1% SDS at 42°C, once in O.lx SSC at room temperature and then spun dry for 5min at 500rpm. Slides were then scanned under 543nm laser using the ScanArray Express scanner (Perkin Elmer, Wellesley, MA). b. Cross Hybridization The extent of cross hybridization was examined through hybridization of Cy-5 fluorophore (Amersham Pharmacia Biotech) labeled oprD amplicon, since oprD belongs to a large family of homologous outer membrane proteins in P. aeruginosa (Stover, Pham et al. 2000). oprD amplicons were generated by PCR from genomic DNA isolated from PAO-H103 as previously described, with oprD-For and oprD-Rev2 primers listed in Table 4. Amplicons were purified as before, and amplicon quality and molecular size confirmed as before. Amplicons were purified as previously outlined using the QIAquick PCR purification kit, and labeled in the same manner as described for GFP amplicons. Following the hybridization protocols outlined below in the 'Microarray Hybridization' section, 0.25pl, 0.5pl and lpl of labeled oprD (approximately lOng) was hybridized overnight and analyzed. F. Microarray Method Development a. Determination of RNA Isolation Methods Standard RNase free precautions were observed throughout. For comparison purposes, equivalent volumes of logarithmic phase cells were used for all total RNA isolation methodologies examined. Total RNA was isolated using SV Total RNA Isolation System (Promega), RNeasy Mini and RNeasy MIDI kits (Qiagen, Inc.) as well as phenol chloroform based methods and cesium chloride (CsCl) methods. Manufacturer's instructions were followed for all kit isolations, and total RNA eluted from the respective columns with RNase free water 50 containing ANTI-RNase RNase inhibitor (Ambion, Inc.). Total RNA samples were evaluated on a 2% agarose gel containing lOug/ml EtBr for the presence of genomic DNA contamination, RNA degradation and overall RNA yield. Samples were also evaluated for purity and quantified by UV absorption at 260nm and 280nm. For phenol based total RNA isolation, cell pellets were resuspended in Trizol and sonicated for lOsec. Following incubation at room temperature for 5min, chloroform was added, vigorously mixed and allowed to incubate a further 3min at room temperature. Isolates were centrifuged maximally for 15min at 4°C, and the aqueous phase removed. Isopropanol was added, samples incubated for lOmin at room temperature and then centrifuged at 4000rpm for 45min at 4°C. Pellets were washed with 70% EtOH, air dried and then resuspended in RNase free water containing ANTI-RNase RNase inhibitor and analyzed as noted above. For CsCl total RNA isolation, cell pellets were resuspended in 4M guanidinium isothiocyanate, containing 0.1M Tris-HCl (pH8.0) and 1% P-mercaptoethanol. The sample was layered over a cushion of 5.7M CsCl in 0.01M EDTA (pH 7.5) and centrifuged for 22hrs at 56,000 rpm at 20°C. The RNA pellet was resuspended in RNase free water, washed with 70% EtOH air dried and again resuspended in RNase free water containing ANTI-RNase RNase inhibitor and analyzed as noted above. Total RNA samples from each of the isolation methodologies were DNase treated to remove contaminating genomic DNA and then evaluated for their ability to generate RNA of high integrity. With the exception of total RNA isolated using the SV Total RNA Isolation System where genomic DNA was treated on the column prior to elution as per manufacturer's instructions, total RNA samples were treated for lhr at 37°C with DNA-free (Ambion, Inc.) according to manufacturer's recommendations, and re-evaluated by 2% agarose gel and UV spectrophotometry as above. Total RNA from each of the isolated samples was then reverse transcribed (RT) as follows: lpg total RNA and 50pmol random primer (Table 4) were heated for lOmin at 70°C; 30U ANTI-RNase (Ambion, Inc.), Ix reverse transcriptase buffer (Invitrogen, Co.), lOmM DTT, and lOmM dNTPs (Invitrogen, Co.) were added and incubated for 2min at 25°C, after which 200U Superscript™ II Reverse Transcriptase (Invitrogen, Co.) was added and the reaction allowed to proceed for lhr at 48°C and then 15min at 70°C. In a second step, varying lengths of rpoC (full transcript length is 1230bp) were amplified from the various cDNA samples to ascertain the integrity and size of the RNA transcripts recovered during sample 51 handling and RNA extraction. PCR amplification was performed as follows: RT reactions were diluted 1:2 in RNase free water, and 5ul of diluted cDNA template used with lOmM dNTPs, lOpmol each of various rpoC-Vox and rpoC-Rev primers (Table 4), 5% DMSO and 0.5U Takara ExTaq polymerase in lx PCR buffer. PCR cycling conditions were: 95°C for 90sec, 35 cycles of 95°C for 45sec, 55°C for 45sec and 72°C for 90sec, then 72°C for lOmin. Amplification was assessed by 1% agarose gel as previously described. b. Determination of Genomic DNA Treatment Contaminating genomic DNA was removed from isolated RNA samples using DNA-free (Ambion, Inc.). The length of DNase treatment was evaluated using the same concentration and sample of total RNA, and proceeded according to manufacturer's instructions except for the treatment time, which was varied to include 15min, 30min, 45min and 60min. Efficiency of genomic DNA removal and quality of remaining total RNA was evaluated by 2% agarose gel as previously described. The efficiency of the DNase treatment was also evaluated by two step RT-PCR amplification of a high abundance, ubiquitous transcript rpsL, ribosomal protein SI2. Parallel RT reactions with and without reverse transcriptase were performed as outlined for rpoC using the various DNase treated total RNA samples. The second step PCR proceeded as for rpoC except using rpsL-For and rpsL-Rev primers (Table 4) and 60°C extension temperature. RNA that is free of all genomic DNA should not yield a product when amplified in the absence of reverse transcriptase enzyme. Presence or absence of rpsL amplicons was evaluated by 1% agarose gel as previously described. c. Determination of Reverse Transcription Reaction Various amounts (10, 12 and 15 pg) of total RNA from the same extraction were assessed in parallel RT reactions for cDNA quality and yield. RT reactions proceeded as follows: DNase treated (as above) total RNA and random primer (Table 4) were incubated at 70°C for lOmin and 25°C for lOmin; lx RT buffer, lOmM DTT, lOmM dNTPs, 30 U SUPERase-IN (Ambion, Inc.) and 1500 U Superscript™ II reverse transcriptase was added and incubated for lOmin at 25°C, 37°C for lhr, 42°C for lhr and 70°C for lOmin. Residual RNA was removed from the cDNA 52 sample by treatment with 1/3 volume IN NaOH for 30min at 65°C, followed by neutralization with IN HC1. cDNA was visualized on a 1% agarose gel stained with SYBR Gold nucleic acid gel stain according to manufacturer's instructions (Molecular Probes) and quantified by UV spectrophotometry at A260. d. Determination of Labeling Method Both direct and indirect (amino-allyl) Cy-dye incorporation methodologies were evaluated for overall expression intensity. Amino-allyl labeling was further characterized using various reverse transcriptase enzymes. For direct labeling of cDNA with fluorophores by direct method #1, 12ug of DNase treated total RNA was reverse transcribed as follows: total RNA, random primer (Table 4), lx RT buffer, lOmM DTT, 6.7mM each of dATP, dTTP, dGTP, 2mM dCTP and ImM Cy-3 or Cy-5 dUTP (Amersham Pharmacia Biotech) were added and incubated at 65°C for 5min, then at 42°C for 5min. 400U Superscript™ II reverse transcriptase and 30U SUPERase-In (Ambion, Inc.) was added and the reaction incubated a further 3hrs at 42°C and then lOmin at 70°C. Residual RNA was removed through addition of 1M NaOH, incubation at 65°C for 15min and neutralization with 1M HC1 and 1M Tris-Cl (pH 7.5). Labeled cDNA samples were combined and purified from unincorporated fluorophore on a QIAquick PCR purification column as per manufacturer's instructions. Labeled GFP (sub-grid marker labeled as previously described) was spiked into the purified cDNA sample, which was then salt ethanol precipitated and resuspended in 50pl hybridization buffer #1 (50% formamide, 5x SSC, 0.1% SDS, 20ug salmon sperm DNA (Ambion, Inc.) and 25pg yeast tRNA (Ambion, Inc.) for hybridization. For direct labeling of cDNA by direct method #2, 12pg of DNase treated total RNA was reverse transcribed as follows: total RNA and random primer (Table 4) were incubated at 70°C for lOmin and 25 °C for lOmin. lx RT buffer, lOmM DTT, 3.75mM each of dATP, dTTP, dGTP, 1.875mM dCTP, O.lmM Cy-3 or Cy-5 dCTP (Amersham Pharmacia Biotech), 30U SUPERase-In and 1500U Superscript™ II reverse transcriptase were added and incubated for lOmin at 25°C, 60min at 37°C, 60min at 42°C and then lOmin at 70°C. Labeled cDNA samples were treated for residual RNA, combined and purified, spiked with labeled GFP and salt ethanol precipitated as for direct method #1. Samples from direct method #2 were resuspended in 50ul 53 hybridization buffer #2 (50% formamide, 5x SSC, 1% SDS and lOug salmon sperm DNA) for hybridization. For indirect labeling of cDNA by indirect method #3, 12pg of DNase treated total RNA was reverse transcribed in the same manner as direct method #2 except lOmM each of dATP, dCTP, dGTP, 3mM dTTP and 3mM 5-(3-aminoallyl)-dUTP (Ambion, Inc.) was substituted. Residual RNA was removed and cDNA salt ethanol precipitated. cDNA pellets were resuspended in 0.2M NaHC0 3 (pH 9.0) and labeled with 3pi of either Cy-3 or Cy-5 mono-Reactive dye (Amersham Pharmacia Biotech) resuspended in 45pi DMSO, for lhr at room temperature. Labeling reactions were stopped by incubating for 15min at room temperature in 4M hydroxylamine. Labeled cDNA samples were combined and purified by QIAquick PCR purification column as above, spiked with labeled GFP, salt ethanol precipitated and resuspended in 50pl hybridization buffer #2. In a subsequent experiment, various reverse transcriptase enzymes were evaluated in the indirect labeling method #3. Reactions followed the procedures outlined above, except for the replacement of 1500U Superscript™ II reverse transcriptase with 400U M-MuLV reverse transcriptase (New England Biolabs, Beverly, MA). In these comparison series, all labeling methods followed the same hybridization and image analysis procedures. Microarray slides were washed twice in 0.1%> SDS for 5min, five times in dld^O and air dried. Labeled probe sets were denatured for 3min at 95°C and placed on ice for 3min before addition of probe to the slide. HybriSlip hybridization covers were placed on top of the sample and slides placed in Corning hybridization chambers for overnight hybridization at 42°C. Coverslips were removed in 42°C 0.2x SSC and slides washed thrice for 5min in O.lx SSC, 0.1% SDS and thrice for 5min in O.lx SSC at room temperature. Slides were spun for 5min at 2000rpm and then scanned and analyzed as outlined below in the 'Scanning and Image Analysis' section. 54 G. Experiments on Ciprofloxacin Treated Cultures a. Determination of Minimum Inhibitory Concentrations The MIC of ciprofloxacin and the other antibiotics against the various P. aeruginosa strains was measured by the broth microdilution technique (Amsterdam 1991). Briefly, serial dilutions of each antibiotic were made in LB broth in 96-well polypropylene microtitre plates (Costar, Cambridge, MA). Wells were inoculated with 5pl LB medium containing approximately 105 CFU/ml of the test organism. In parallel, samples of the inoculum were spread on LB plates to ensure the CFU/ml range. The MIC was determined after 24hr incubation at 37°C, and was visually scored as the lowest concentration of antibiotic which inhibited bacterial growth. Results are the mode of four independent experiments. b. Growth Curve in Sub-inhibitory Ciprofloxacin For purposes of determining mid-logarithmic phase, growth curves were completed for P. aeruginosa grown in each of the ciprofloxacin treatment conditions being studied (Ox-, O.lx-, 0.3x- and lx-MIC). Overnight cultures of P. aeruginosa were sub-cultured as previously described, and UV absorption measurements at OD600 taken at half hour intervals to determine the growth dynamics of the cultures. c. Time-Kill Assay The ability of ciprofloxacin to induce adaptive resistance was evaluated using a modified time-kill assay. Briefly, P. aeruginosa strain HI03 or FA0620::luxCDABE was grown to mid-logarithmic phase in the presence or absence of their respective concentrations of ciprofloxacin (O.lx-, 0.3x- or lx-MIC). At mid-logarithmic phase, cells were pelleted by centrifugation. Supernatants were discarded and pellets washed in 10-ml phosphate buffered saline (pH 7.4) (PBS, Invitrogen, Co.; pre-warmed to 37°C). Cells were again pelleted by centrifugation and the supernatant decanted. Pellets were resuspended in 50-ml LB broth (pre-warmed to 37°C) containing the appropriate 2x-MIC ciprofloxacin, adjusted to an equivalent optical density and 55 transferred to new 500-ml Erlenmeyer flasks for incubation with shaking at 37°C. Samples were taken every 15min or 20min for 2hrs to 9hrs, and plated at various dilutions on LB agar plates for colony counts. Plates were incubated at 37°C overnight and colony forming units (CFU) counted the next day. Al l assays were repeated in quadruplicate. d. Light Microscopy From the time-kill assay cultures for strain HI03, microscope slides were inoculated with a small loop-full of bacteria taken from the untreated strain H103, and strain H103 treated with O.lx- or 0.3x-MIC ciprofloxacin. Samples were taken before and after the PBS wash, and every 15min thereafter throughout the modified time-kill assay. Cells were heat fixed and Gram stained. Cells were then examined at lOOOx magnification on Zeiss AxioLab (Zeiss Canada, Toronto, ONT), where the length of 6-10 random cells was measured in microns. e. Microarray Experimentation P. aeruginosa strain HI03 was cultured in Ox-, O.lx-, 0.3x- and lx-MIC ciprofloxacin as previously stated. Al l microarray experiments were performed as outlined below on three independent cultures and repeated at least twice per culture sample. i. RNA Isolation and Evaluation Bacterial cultures were collected at the mid-logarithmic growth phase (OD600, 0.5 to 0.6) and pelleted by centrifugation. Cells were treated as recommended by the manufacturer's mechanical disruption and lysis protocol for RNA isolation with the RNeasy MIDI columns (Qiagen, Inc.). RNA purity and quantity was assessed by UV spectrophotometry, and RNA integrity monitored by 2% agarose gel electrophoresis containing EtBr and visualized by UV light box. 56 ii . Genomic DNA Treatment and RNA Evaluation Total RNA was eluted from the RNeasy column in RNase-free water containing ANTI-RNase and purified from residual DNA by using DNA-/ree at 37°C for at least 15min (Ambion, Inc.). Total RNA was re-evaluated by UV spectrophotometry and electrophoresis to ensure high quality and quantity RNA was maintained through the genomic DNA treatment process. iii. Reverse Transcription Reaction cDNAs were generated by using random primers (Table 4) for reverse transcription. Primers were annealed at 70°C for 10 min, followed by 25°C for 10 min to lOug total RNA and to 5 exogenous transcripts (generated as noted above by in vitro transcription for ATCC# 87482, 87483, 87484, 87485 and 87486) that were added to each sample. Spike RNA transcripts served as a control for monitoring transcriptional efficiency, Cy-dye incorporation and array performance. The reverse transcription reaction proceeded at 37°C for 60 min, 42°C for 120 min, and 70°C for 10 min in a total reaction volume of 60pl containing 1500U Superscript™ II reverse transcriptase, lx Superscript™ II RT buffer, lOmM DTT, 3mM 5-(3-aminoallyl)-dUTP, 3mM dTTP, lOmM each of dATP, dCTP, dGTP, and 30U SUPERase-In. Residual RNA was removed by alkaline treatment with 1M NaOH at 65°C for 15 min, followed by neutralization with 1M HC1; cDNA was then salt-ethanol precipitated for > lhr at -70°C. iv. cDNA Labeling and Purification Precipitated cDNA was resuspended in 0.2M NaHC03 (pH 9.0) and labeled with Cy-3 or Cy-5 mono-Reactive dye resuspended in DMSO. Labeling reactions proceeded as noted above for indirect method #3. Labeled sample pairs for a given microarray experiment were combined, spiked with 0.25ul labeled GFP and purified by QIAquick PCR purification column. The combined cDNA sample was again salt-ethanol precipitated for > lhr at -70°C. 57 v. Microarray Slide Preparation and Sample Hybridization The custom P. aeruginosa microarray slides were prepared for hybridization by washing two times for 5min in 0.1% SDS and five times for lmin in water, followed by boiling for 3min in 95°C water to denature the double strand probes and air drying. The combined cDNA pellet was resuspended in 4ul lOmM EDTA, denatured for 3min at 95°C and added to 71 pi of prepared hybridization buffer #2 (Ambion, Inc.) according to the manufacturer's instructions. The entire sample volume was applied to the prepared microarrays, overlaid with a Lifter-Slip coverslip (Electron Microscopy Sciences, Fort Washington, PA) and sealed in a Corning hybridization chamber and covered in foil for overnight hybridization in a 42°C waterbath. vi. Post-hybridization Microarray Handling Following overnight hybridization, glass coverslips were removed by flotation in 42°C prewarmed 0.2x SSC. Hybridized slides were washed thrice for 5min in 25ml 0.5x SSC/0.5% SDS and thrice for 5min in 25ml 0.5x SSC, and then spun dry at 2000 rpm for 5min in 50ml conical tubes. Dried microarray slides were stored in the dark until scanning and image analysis. vii. Scanning and Image Analysis Individual microarrays were scanned on a ScanArray Express scanner (Perkin Elmer, Wellesley, MA). Al l slides for a given microarray experiment were scanned at the same laser and PMT settings. Typically this setting was 95% laser and 90%> PMT gain for Cy-3, and 95% laser and 95% PMT gain for Cy-5. Scanner images were analyzed using the spot finding program ImaGene v5.0 (BioDiscovery, Inc., Marina Del Rey, CA), and textual output data imported into either GeneSpring software v5.5 (Silicon Genetics, Redwood City, CA) or various R scripts developed by D. Aeschlimann and J. Bryan (Dept. Statistics, UBC) for further analysis. 58 viii. Background Correction and Normalization Following image and spot finding analysis, the resulting text files were subjected to background correction (R code developed by Doug Hoffart and Jochen Brumm; Gene Array Facility at VGH, Vancouver), which takes the lowest 10th percent of signal for each sub-grid as the background and subtracts this value out from each signal within that sub-grid. The background corrected output files were then compared and analyzed by various R scripts because the extensive loop design and dye swapping did not allow use of the in house ArrayPipe program. Al l microarrays were normalized using the R package vsn (Huber, Von Heydebreck et al. 2002) to correct for differences due to dye bias and other hybridization effects, and to stabilize the error variance across the range of expression levels. Pairwise comparisons were made between all genes in all treatment groups, and the baseline or untreated condition. ix. Statistical Analysis Two sample t-statistics and permutation p-values were computed for each pair. Permutation p-values shuffle the data for a given gene and compute p-values as the percentage of simulated means which are greater than the actual observed means relative to the number of simulated samples. Genes exhibiting permutation p-values < 0.05 were considered differentially expressed and subjected to further confirmatory analysis. Data was also analyzed using GeneSpring v5.5 as a secondary analysis method. The fold changes presented herein, were derived from GeneSpring analysis, and are only for those genes with statistically significant expression changes as computed from R analysis. H. Confirmatory Experiments a. Real-time PCR Differential expression of genes was examined by relative real-time PCR (rRT-PCR) using the ABI Prism 7000 Sequence Detection System (Applied Biosystems). Primers were designed using gene sequences from the Pseudomonas database (www.pseudomonas.com) and the ABI 59 Primer Express program v2.0 in the default mode. SYBR Green I dye chemistry was used for detection of PCR products. Primer sequences used for rRT-PCR are listed in Table 4. De novo cDNAs from the same RNA used in microarray target preparation were employed for all real-time PCR analyses. All cDNA samples were diluted 1/1000 fold in TE buffer, and 5ul cDNA was used for each 25pl rRT-PCR reaction (lx SYBR Master mix, Applied Biosystems; 50nM of each primer). Al l reactions were performed in triplicate using the following cycling parameters: 95°C for lOmin, and 40 cycles of 95°C for 15sec, 60°C for lmin. Dissociation curves were obtained following completion of the reaction. Raw data was analyzed by using the ABI Sequence Detection software (Applied Biosystems). Relative fold changes were calculated according to the formula: [ (GOI U nknown/ENDOunknown)/ (GOI C O ntro i /ENDOcontro i ) ] , where GOI is the gene of interest, ENDO is the endogenous control gene, and unknown and control are respective samples being assayed for expression. b. Luminescence Assays Induction and thus transcription from the various luxCDABE fusions of P. aeruginosa (Table 3) was examined in LB broth supplemented with 0, 0.01, 0.03 or O.lpg/ml ciprofloxacin. Cells were grown as previously described, and lOOpl samples removed hourly for analysis by luminescence. Luminescence was measured hourly using a SPECTRAFluorPlus luminometer (Tecan, San Jose, CA). Luminescence was corrected for growth by simultaneously monitoring the absorbance at 620 nm. The assay was repeated in triplicate. c. Twitching Assay Twitching motility was assayed on plates containing 25ml of 1% LB agar ± ciprofloxacin (0.01 or 0.03ug/ml). Following growth of P. aeruginosa strain H103 to mid-logarithmic phase in the presence or absence of ciprofloxacin as before, the culture to be tested was stab inoculated through the corresponding agar. Following overnight growth at 37°C, the zone of twitching (radiating from the point of inoculation) located between the agar and the petri dish was measured. The assay was repeated in triplicate. 60 d. Transmission Electron Microscopy Pyocin/phage like particles were isolated from strains H103 and PA0620::/WA; grown until mid-logarithmic phase in the presence or absence of their respective 0.3x-MIC ciprofloxacin concentrations. Particles were isolated as previously described (Yamamoto, Alberts et al. 1970). Briefly, cell lysis was achieved by the addition of chloroform. Cell lysates were treated with DNase I (Invitrogen, Co.) for lhr at 37°C and cell debris removed by centrifugation. Supernatants were incubated with 10% w/v PEG 6000 for lhr at 4°C to sediment out the phage particles, and then centrifuged down at 10,000g for 15min at 4°C. Pellets were resuspended in Buffer B (0.001M phenylmethyl sulfonyl fluoride, 0.1M NaCl, 0.01M phosphate pH 6.8) and insoluble material removed by centrifugation at 5,000g for 5min at 4°C. Samples were then ultracentrifuged for lhr at 70,000g at 4°C and the resulting pellets again resuspended in Buffer B. Phage particle components were retrieved from the 15% sucrose fraction of a sucrose gradient, and were concentrated by dialysis (Slide-A-Lyzer Cassette, 10,000 MWCO; Pierce Biotechnology, Inc., Rockford, IL) against Buffer B (Yamamoto, Alberts et al. 1970). Samples were prepared on Formvar (Marivac, Halifax, N.S.) carbon stabilized grids, and negatively stained by 1% phosphotungstate. Prepared samples were examined by transmission electron microscopy (Hitachi H7600 TEM) under standard operating conditions at the UBC Biosciences E M lab facility. e. Serial Selection of Genomic Loss of Pyocin/Phage Region Overnight L B cultures of P. aeruginosa strain HI 03 and ArecA were sub-cultured at 1/1000 dilution into 5ml fresh L B in the presence or absence of their respective 0.3x-MIC ciprofloxacin concentrations. Cultures were grown overnight at 37°C with shaking. The dilution and treatment condition was repeated for 5 consecutive days, after which lOpl of culture was spread plated onto their respective lOx-MIC ciprofloxacin L B plates. Plates were incubated overnight at 37°C and the resulting colonies assessed for the presence or absence of the R2/F2 pyocin region by colony PCR. Colony PCR was performed for PA0623, PA0642 and PA0648 genes as follows: lx PCR buffer (Invitrogen Co.), 2mM M g C l 2 , 4% DMSO, 200pM each of dATP, dCTP, dGTP and dTTP (Invitrogen Co.), 100 nM each primer (Table 4), and 1.25U Tag D N A polymerase 61 (Invitrogen Co.). Template was added by dipping a pipette tip into the colony and pipette mixing into the PCR mixture. Amplification proceeded on a Minicylcer (MJ Research, Inc.) under the following cycling conditions: 5min at 94°C, followed by 50 cycles of 30sec at 94°C, 30sec at 55°C and 30sec at 72°C, and then lOmin at 72°C. Presence or absence of amplicon was checked by electrophoresis on a 1% agarose gel containing EtBr, and visualized using a UV light box. 62 R E S U L T S CHAPTER ONE: Design and Construction of a Pseudomonas Custom Microarray A. Introduction Completion of the genome sequence for P. aeruginosa strain PAOl (Stover, Pham et al. 2000) afforded researchers the unparalleled opportunity for high throughput and global screening of the organism at the gene or protein level. Around the same time, DNA and oligonucleotide microarrays (Schena, Shalon et al. 1995; Lockhart, Dong et al. 1996) were being developed and becoming increasingly popular as a method of transcriptional analysis. Since commercial P. aeruginosa GeneChips® from Affymetrix were not yet available, and were predicted to be prohibitively costly, it was decided that our own research interests would be best addressed by manufacturing our own custom DNA microarray for P. aeruginosa. The experiments in this chapter describe the strategy and approach to the design and construction of the custom DNA microarray to P. aeruginosa. Work in the latter part of this chapter was done in collaboration with the P. aeruginosa group at PathoGenesis Co. (subsequently acquired by Chiron Co., Seattle, WA). B. Microarray Design B . l . Optimization of Amplicon Size, Concentration and Volume The extent of DNA-DNA hybridization depends upon the length of molecules being hybridized, since shorter DNA probes are less available for hybridization and since a minimum number of bases are required to interact with the microarray surface (Stillman and Tonkinson 2001) . To determine the size of amplicon best suited for generating intense hybridization signals, various lengths of oprD were spotted on nylon membranes; volume and hybridization temperature were kept constant, while concentration of the spotted probe and length of labeled probe were varied. A notable increase in signal intensity was observed for spotted probes of 600bp and 800bp within the concentration range of 3-10ng (Figure 6). Since concentration of 63 • 400bp spot + 400bp probe 3 400bp spot + 600bp probe • 400bp spot + 800bp probe • 600bp spot + 400bp probe • 600bp spot + 600bp probe • 600bp spot + 800bp probe • 800bp spot + 400bp probe B 800bp spot + 600bp probe B 800bp spot + 800bp probe I 3 5 Concentration (ng) 10 Figure 6: Effect of varying concentration and amplicon size on hybridization signal. 0.5ul of 400, 600 and 800bp oprD amplicons at varying concentrations were crosslinked to positively charged nylon membrane and hybridized to 400, 600 or 800bp oprD probes at 42°C in standard hybridization solution. 64 spotted DNA material also affects the hybridization reaction by altering the rate of hybridization and number of available binding sites for the hybridizing probe (Stillman and Tonkinson 2001), various spotting concentrations were simultaneously evaluated. Intense hybridization signals were observed for 600bp and 800bp spotted probes spotted in lOng amounts (Figure 6). lOng as a spotting amount was thus selected for further study. The extent of DNA-DNA hybridization also depends on the volume spotted onto the array platform. Larger spotting volumes present problems with respect to spot bleeding and uniformity, and reduced microarray probe density. To determine the volume of amplicon best suited for generating intense hybridization signals, various volumes of oprD amplicon were spotted on nylon membranes; concentration and hybridization temperature were kept constant, while volume of spotted probe and length of probe were varied. As the volume of spot was increased from 0.5(0.1 to 5pl, the hybridization signal decreased (Figure 7). As well, spot uniformity was found to decline with increasing spot size (data not shown). A spotting volume of 0.5pi was thus selected for further study since this volume would also allow for a high density of spots on the microarray. It was again noted that signal intensities were high for spotted probes of 600bp and 800bp length in comparison to the signal from 400bp length spotted probes (Figure 7). Overall, signal intensity was consistently highest for 600bp spotted probes whether these probes were hybridized to labeled probes of equal or greater length. Spotted probes of 600bp length were chosen for all subsequent analyses. B.2. Optimization of Hybridization Temperature and Hybridization Solution Base composition has a known effect on hybridization given the lower stability of A-T versus G-C pairs (Southern, Mir et al. 1999; Herwig, Schmitt et al. 2000; Stomakhin, Vasiliskov et al. 2000). GC-content of the amplicon is thus an important factor to consider in determining hybridization temperature because probes with an increased GC-content will have an increased hybridization stability due to the increased number of these stable base pairs. Hybridization temperature was thus varied to determine the optimal temperature for the generation of intense hybridization signal; volume, concentration and spotted probe length were kept constant, while hybridization temperature and solution were varied. Strong and consistent signals were detected 65 • 400bp spot H 400bp spot H 400bp spot • 600bp spot • 600bp spot • 600bp spot • 800bp spot B 800bp spot B 800bp spot + 400bp probe + 600bp probe + 800bp probe + 400bp probe + 600bp probe + 800bp probe + 400bp probe + 600bp probe + 800bp probe 0.5 1 2 Volume (ul) Figure 7: Effect of volume and amplicon size on hybridization signal. lOng of 400, 600 and 800bp oprD amplicons at varying volumes were crosslinked to positively charged nylon membranes and hybridized to 400, 600 or 800bp oprD probes at 42°C in standard hybridization solution. 66 from those hybridizations carried out at 42°C in comparison to hybridizations at either 55°C or 68°C (Figure 8). Hybridization at 42°C was selected for all subsequent microarray experiments. Similarly, the composition of the hybridization solution, in particular the stringency or salt concentration (Southern, Mir et al. 1999), affects the molecular interactions on the microarray. In order to find the hybridization solution which would generate the best hybridization signal, various formulations of hybridization solution were compared; again, volume, concentration and spotted probe length were kept constant, while hybridization solution and temperature were varied. The standard hybridization solution and UltraHyb solution (Ambion, Inc.) were found to perform optimally at 42°C (Figure 8). Ambion's UltraHyb solution was selected as the optimal hybridization solution because it performed marginally better with labeled probes of shorter length and did not require technical composition, thereby eliminating the introduction of further technical errors into the microarray experiment. C. Microarray Construction and Validation C. l . Evaluation of Amplicon Integrity Using primers uniquely designed against each ORE in the P. aeruginosa strain PAOl genome and designed such that each ORF had approximately the same GC content (designed in collaboration with PathoGenesis Co. and synthesized by Sigma-Genosys), PCR amplification conditions were worked out to allow for amplification of each ORF in a high-throughput fashion. Amplification was completed in triplicate for each ORF and pooled before analysis of integrity. Pooled amplicons were also purified and re-evaluated. i. Agarose Gel Analysis of Amplicon Size and Uniqueness To ascertain the quality of the PCR reactions, pooled PCR reactions were analyzed by gel electrophoresis. Amplified products were required to be of the correct predicted size, be present as a predominant single band on the gel and fall above the minimum concentration of 35ng/ul in order to be considered successful. Given the high-throughput nature of the project, if the amplification reactions within a 96-well plate fell below an 80% first pass success rate (<77 of 67 Figure 8: Effect of temperature and hybridization solution on hybridization signal. Positively charged nylon membranes crosslinked with 0.5pl of lOng 600bp oprD amplicon were hybridized to 400, 600 or 800bp oprD probes at various hybridization temperatures and with various hybridization solutions. 68 96 wells), then the entire set of 96-wells was either reamplified or the failed ORFs targeted for primer redesign. A sample 96-well electrophoresis gel image displaying 100% amplification success is shown in Figure 9. Overall, the success rate of first pass PCR was 95.58%, with only a total of 248 ORFs requiring primer redesign and amplification. With respect to amplicon size and uniqueness, 111 ( 1 . 9 8 % ) ORFs did not amplify on first pass, 7 (0.12%) PCR reactions contained more than one band and 11 (0.20%) PCR reactions were of the wrong molecular size. Data tracking of the PCR amplification uniqueness, size and concentration for each ORF is summarized in Appendix I. ii. Sequencing Analysis Following purification of the pooled PCR reactions, amplicons were analyzed by sequencing and sequences analyzed by BLAST (Altschul, Gish et al. 1990) to the published P. aeruginosa strain PAOl genome (Stover, Pham et al. 2000) (work completed in collaboration with PathoGenesis Co. and Chiron Co.). Amplicons which failed to produce sequence that corresponded to the appropriate ORF were sent for primer redesign. Overall, the success rate of the sequencing reactions was 99.61%), with only 22 (0.39%) ORFs requiring primer redesign and amplification. Data tracking of the sequencing reactions is also summarized in Appendix I. C.2. Evaluation of Amplicon Concentration i. Agarose Gel Analysis In addition to evaluating amplicon size and uniqueness, agarose gels were also used to estimate the concentration by comparison to a mass size marker (Figure 9; details of the specific concentrations for each marker band are available from Bioline). A minimum concentration of 35ng/ul was required for the PCR reaction to be considered successful. The majority of PCR reactions (98.27%) fell well above this minimum cut-off, with only 97 (1.73%) PCR reactions falling below the minimum yield requirements (refer to Appendix I). 69 Figure 9: Sample agarose gel for evaluation of 96 well P C R and corresponding molecular ladder size range. All PCR reactions performed in 96 well plate format were evaluated for yield, uniqueness and molecular size by 1.5% agarose gel electrophoresis stained with EtBr in comparison to ladder (L) molecular size markers. 70 ii . Capillary Electrophoresis Analysis Following transfer of 96 well plates to 384 well plate format suitable for printing (performed at Xenon Genetics, Vancouver, BC) and following several initial print runs (Gene Array Facility, Vancouver, BC), concentrations were re-evaluated by capillary electrophoresis. This assay uses nl volumes of sample to electronically compute a representative agarose gel from which concentration is estimated (Figure 10). Complete results of this analysis are summarized in Appendix II. In comparison to the agarose gel electrophoresis data, where 95.58% of the ORFs netted a yield greater than the required spotting concentration of lOng/pl (Appendix I), the capillary electrophoresis data found only 73.32%> of ORFs had a concentration greater than lOng/pl (Appendix II). Because maintaining a high concentration of probe for spotting is necessary to achieve good quality microarray results, loss of probe concentration over successive print runs was deemed unacceptable. At this point, microarray printing was switched to a different facility (Microarray Centre, University of Toronto, Toronto, ONT) and the back-up set of 384 well plates employed. C.3. Evaluation of Print Run In light of some of the many factors which may affect microarray quality (Eisen and Brown 1999), the uniformity of DNA deposition on the microarray plays a key role in array performance and as such, evaluation of the printed microarray has become an important point of control. To assess the quality and quantity of the printed material, microarrays were labeled with SYBR Green (Battaglia, Salani et al. 2000). Spots were found to be of uniform size and concentration, and evenly spaced with minimal tailing effect (Figure 11). Later, microarrays were terminal transferase labeled with Cy-3 fluorophore and again evaluated for quality and quantity of the printed material. Spots were again found to be of uniform size, concentration and distribution (Figure 12). In general, demonstration of high quality print runs was taken to be indicative of their capability of providing high quality data. 71 L 1 2 3 4 5 6 7 8 9 10 11 12 Figure 10: Sample capillary electrophoresis digital agarose gel. Digital interpretation of the raw data output for the first 12 columns from the first row of 384 well plate C4PS01-P3-0001-BL008-001 from the Caliper LabChip 90 electrophoresis system. 7 2 Figure 11: SYBR Green analysis of upper left sub-grids of the P. aeruginosa microarray. Microarray printed at Gene Array Facility (Vancouver, BC). 7 3 t m 61 * • « • • • r • -a .* • a * • • p 41 • * i: • • * i p ; •.. - A ' P p n *> • • A • p 0 • *• r # • • e • ». * • • • • » a 0 * * 0 5" C* * » * • • -» # « < ... c • i & • i 4 # -is- ft ;» n "J •s r *= 0 e • • 0 c -* ? «» fc e 1 :*> i e v.- <e G fe • • • o Q • p • • • - * # • •• Figure 1 2 : Terminal transferase Cy-3 label analysis of two sub-grids of the aeruginosa microarray. Microarray printed at the Microarray Centre (Toronto, ONT). 7 4 D. Summary In this chapter, many of the parameters in the design and construction of a microarray which affect the performance of the microarray itself were evaluated. Length of amplicon, concentration and volume of spotted amplicon, as well as hybridization temperature and hybridization solution were investigated. It was found that 0.5ul of a lOng amplicon solution were ideal spotting parameters. 600bp amplicon fragments were also found to produce intense hybridization signals, findings in agreement with recently published work (Stillman and Tonkinson 2001). With respect to hybridization, the UltraHyb spotting solution (Ambion, Inc.) at a hybridization temperature of 42°C was found to produce optimal signal intensities, parameters similar to those subsequently used by Affymetrix P. aeruginosa GeneChip® users. These findings were used in the subsequent construction of the custom P. aeruginosa strain PAOl DNA microarray. Several quality control checks were implemented throughout the construction of the DNA microarray. These included evaluation of the amplicon size, uniqueness and concentration, as well as validation of the amplicon sequence, and assessment of the print run. The success rate of the high-throughput strategy for PCR amplification and evaluation of the amplicons was extremely good (95.58%) for the first pass. Amplicons were found to be amplified in sufficient quantity (>10ng/pl) for subsequent microarray printing, and were found by sequence analysis to be correctly amplified from the desired target ORF. Microarray print runs were found to contain uniform and clean amplicon spots, arranged in an organized fashion, eliminating the microarray itself as a source of error in subsequent microarray experimentation. The design and construction of a custom DNA microarray is a constantly changing field. Standards in design, printing and quality assessment of custom microarrays are continuously evolving (Hessner, Wang et al. 2003; Gordon and Sensen 2004; Rimour, Hill et al. 2004) and where possible these new techniques have been incorporated into the overall P. aeruginosa microarray design. However, because these parameters are still under flux, they will need to be periodically re-evaluated to maintain the integrity and quality of the current P. aeruginosa custom DNA microarray. 75 CHAPTER TWO: Method Development for the Pseudomonas Custom Microarray A. Introduction The complimentary methods for conducting microarray experiments using bacterial transcripts were poorly defined or lacking at the start of this project. At the time, protocols for only Helicobacter pylori, Mycobacterium tuberculosis and Bacillus subtilis microarray experimentation were available (Wilson, DeRisi et al. 1999; Salama, Guillemin et al. 2000; Ye, Tao et al. 2000) and were used as guidelines in the development of microarray protocols for P. aeruginosa. Protocols for the commercial P. aeruginosa GeneChip® were available on a beta-testing basis, but were not relevant here because of the significant differences in platform technology between Affymetrix and custom DNA microarrays. Furthermore, given the differences between eukaryotic and prokaryotic RNA species, many of the procedures implemented in mammalian microarray studies (particularly reverse transcription with oligo dT) were not applicable to P. aeruginosa. As well, several of the microarray procedures were only novel and not standard in the microarray field at the time (Manduchi, Scearce et al. 2002), and thus required evaluation and comparison to the proven techniques. The experiments in this chapter thus describe the development of the methodology for conducting custom DNA microarray experiments using P. aeruginosa. B. Comparison of RNA Isolation Methods Since separation of mRNA from rRNA within bacterial species cannot be achieved through use of the poly-A tail on mRNA as is the case for mammalian cells, total RNA must be isolated and used with random or gene specific priming to generate cDNA. The low abundance (<2%) of mRNA relative to the total RNA species is a further limitation on bacterial microarray studies when considering total RNA reverse transcription with random primers, since the ratio of mRNA:rRNA will be maintained in the resulting cDNA population reducing the likelihood of mRNA hybridization to the spotted probe. While amplification of mRNA species has been possible for mammalian transcripts for some time, methodologies for amplification of bacterial mRNA have only recently been developed (Ambion 2004). Thus, the aim of the RNA isolation 76 method comparison was to achieve consistent isolation of high yield total RNA. Total RNA was isolated from P. aeruginosa cultures grown to mid-logarithmic phase using a variety of isolation methodologies. Various isolation strategies were evaluated for their ability to generate high yields of total RNA of good purity and integrity, as well as for consistency and ease of use, and were evaluated by agarose gel analysis, UV spectrophotometry and amplification of full length rpoC transcript. RNA isolation using the CsCl method proved difficult, time consuming and inconsistent (Figure 13, lane 2). RNA isolated using either the ProMega or Qiagen kits was easy, quick and yielded consistent results (Figure 13, lanes 3-5). The phenol based method also gave high yields and included isolation of small RNA species (tRNA, 5S rRNA) which were absent from the column based methods (Figure 13, lanes 6-7), but was found to be difficult to achieve consistent results. Of the isolation strategies evaluated, the Qiagen QMIDI RNeasy kit was chosen for all subsequent RNA isolations because it consistently yielded abundant total RNA of good quality and integrity (as noted below). The ProMega kit which also produced high yields of good quality total RNA, was not chosen as the RNA isolation method because the genomic DNA treatment step prior to elution from the column did not always produce consistent genomic DNA free total RNA samples (as noted below in the 'Evaluation of Genomic DNase Treatments' section). To better evaluate the quality and integrity of total RNA isolated by the various strategies, total RNA samples were reverse transcribed to cDNA and the transcript length of rpoC assessed by PCR. If the RNA isolation strategy was robust then the full length of a long transcript like rpoC should be attainable; the full transcript length of rpoC is 1323bp. For all isolation strategies, except the CsCl method which was not evaluated in this regard because it gave poor RNA (Figure 13, lane 2), transcripts of 1230bp were generated (Figure 14), demonstrating that all RNA isolation methods evaluated were robust in their ability to produce RNA of high integrity. C. Evaluation of Genomic DNase Treatments Presence of genomic DNA in the total RNA sample would affect both the RT reaction by binding the nucleotides and primers required for cDNA synthesis, and the microarray hybridization reaction by competing for binding to the complimentary spotted probes, thus 77 Figure 13: Comparison of RNA isolation methods. Logarithmic cultures of P. areuginosa isolated by: Lane 1 - lOObp ladder; Lane 2 - CsCl; Lane 3 - ProMega SV Total RNA Isolation System; Lane 4 - Qiagen RNeasy QMIDI Kit pre-genomic DNase treatment; Lane 5 - Qiagen RNeasy QMIDI Kit post-genomic DNase treatment; Lane 6 -Trizol method pre-genomic DNase treatment; and Lane 7 - Trizol method post-genomic DNase treatment. RNA samples were run in a 2% agarose gel containing lOpg/ml EtBr for 45min at 90V. 78 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Figure 14: Comparison of RNA isolation methods with respect to rpoC transcript length. Total RNA was treated with DNase to remove genomic DNA. reverse transcribed into cDNA and various lengths of rpoC transcript PCR amplified. Panel A - Trizol method; Panel B -ProMega SV Total RNA Isolation System; and Panel C - Qiagen QMIDI RNeasy kit. For all panels lanes are organized as follows: 1, lOObp ladder; 2, rpoC 102bp; 3, rpoC 239bp; 4, rpoC 366bp; 5, rpoC 610bp; 6, rpoC 740bp; 7, rpoC 923bp; 8, rpoC 1072bp; and 9. rpoC 1230bp. PCR products were run on a 1% agarose gel containing lOug/ml EtBr for 45min at 90V. 79 limiting binding of the labeled cDNA and accurate measurement of the expression level. Contaminating genomic DNA was thus removed from total RNA preparations by treatment with DNase I. Various lengths of treatment time were evaluated to ascertain the minimum time required, as evaluated by the inability to PCR amplify the 3OS ribosomal subunit protein rpsL from the RNA preparation post-genomic DNA treatment. Based on agarose gel electrophoresis analysis of DNase treated samples, 15min appeared to be sufficient for complete removal of genomic DNA contamination (Figure 15). This observation was confirmed by the absence of rpsL products when DNase treated total RNA samples were then used as template for amplification of rpsL (data not shown as the agarose gel is blank). Thus, a minimum of 15min was required for removal of genomic DNA contamination from total RNA preparations. D. Comparison of Reverse Transcription Reactions Since GC base pairs form more stable base pairing interactions (Southern, Mir et al. 1999; Herwig, Schmitt et al. 2000), it was hypothesized that the reverse transcriptase enzyme may have more difficulty in synthesizing cDNA from the P. aeruginosa total RNA template which has a GC content of 66.6%. Incomplete denaturation of RNA secondary structure during cDNA synthesis step thus could halt the polymerase, resulting in shorter cDNA copies of target mRNA (Gupta, Cherkassky et al. 2003). Thus, a greater amount of total RNA reverse transcribed would not necessarily yield a greater amount of cDNA from the RT reaction. To ascertain these possibilities, various amounts of total RNA were used as template in the reverse transcription reaction. From SYBR Gold stained agarose gel electrophoresis analysis of the cDNA product, it appeared that equal amounts of cDNA were produced from lOug and 12pg total RNA, while 15ug total RNA yielded visibly more cDNA (Figure 16). cDNA prepared from 15ug of total RNA also seemed to contain longer transcripts than either of the other preparations. To confirm these observations, cDNA preparations were quantified by UV spectrophotometry at A260. From lOpg of total RNA, the cDNA yield was 364ng/ul, from 12ug it was 480ng/ul and from 15pg it was 650ng/pl. Thus, as the starting amount of template was increased, the amount of cDNA synthesized by the RT reaction was also increased irrespective of the high GC content of P. aeruginosa transcripts, and was most significant for 15pg total RNA. While 15pg of total RNA would be the ideal choice for RT reactions, this amount and concentration of RNA was not 80 Figure 15: Comparison of various Ambion DNase-/ree treatment times for removal of genomic DNA contamination. Genomic D N A treatments were all performed on total RNA isolated from P. aeruginosa and were conducted at 37°C for the following time frames: lanes 1, lOObp ladder; 2, 15min treatment; 3, 30min treatment; 4, 45min treatment; 5, 60 min treatment; and 6, untreated total RNA. RNA samples were run on a 2% agarose gel containing lOpg/ml EtBr for 45min at 90V. H I Figure 16: Comparison of cDNA preparations from different initial amounts of total RNA. Various amounts of total RNA isolated from P. aeruginosa was reverse transcribed into cDNA as follows: lanes 1, lOObp ladder; 2, lOug total RNA; 3, 12pg total RNA; and 4, 15pg total RNA. cDNA samples were run on a 1% agarose gel for 45min at 90V and stained with SYBR Gold nucleic acid gel stain. 82 consistently achievable from the RNA isolation strategies employed. Furthermore, when such concentrations were achieved, the total amount of RNA isolated from logarithmic cultures was not sufficient to allow for technical repetition of the microarray experiment or downstream confirmatory analyses. Thus, a minimum of lOug of total RNA was used for all subsequent reverse transcription reactions, since this amount and concentration could consistently be attained by the RNA isolation method and was sufficient to allow for further experimentation. E. Comparison of cDNA Labeling Methods For a microarray experiment to be successful in defining an expression pattern, it must be both sensitive and accurate. Efficient and accurate incorporation into or attachment of a detectable label to the cDNA is thus of critical importance. Of the available cDNA labeling methods (Nimmakayalu, Henegariu et al. 2000; Stears, Getts et al. 2000; Karsten, Van Deerlin et al. 2002; Ramakrishnan, Dorris et al. 2002; Richter, Schwager et al. 2002; Francois, Bento et al. 2003), the direct and indirect fluorescent labeling methods were not prohibitively costly and were compared herein. In the direct labeling method, a Cy3 or Cy5 fluorescently labeled nucleotide is directly incorporated during cDNA synthesis. Critics of this method highlight its main shortcoming, enzyme introduced labeling and sequence bias (Gupta, Cherkassky et al. 2003). This limitation is addressed by the indirect fluorescent labeling method, where amino-allyl modified nucleotides rather than fluorescently labeled nucleotides, are incorporated during cDNA synthesis. In a subsequent step, Cy3 or Cy5 dyes are coupled to the reactive amine groups incorporated in the cDNA. Overall, comparison of the direct and indirect labeling methods found that the indirect method provided superior labeling. As is evident from the hybridization images (Figure 17), direct labeling method #1 yielded negligible amounts of labeled cDNA, whereas direct labeling method #2 yielded more intensely labeled cDNA spots. Although the slide image shown had high background, the indirect labeling method #3 produced the most uniformly labeled spots of the methods tested (Figure 17). The indirect labeling method was therefore pursued further, and different reverse transcriptase enzymes evaluated. When total RNA was reverse transcribed using the indirect labeling method with Superscript™ II RT, better hybridization images were observed than when total RNA was reverse transcribed with M-MuLV (Figure 18); hybridized 83 Figure 17: Comparison of direct and indirect cDNA labeling methods. Total RNA from P. aeruginosa was reverse transcribed and labelled using: A, direct labelling method #1; B, direct labelling method #2; and C, indirect labelling method #3. The same four subgrids are visible in each image. 84 A - Superscript™ II R T ^^^^M B - M - M u L V R T . ^ ^ ^ B ^ ^ l . - ' • , * * . , - * - • • • > • • • . . . . . . . C -" " « ^ • # » • • . - J * • • . . ! . ' ." r -»**•.'• • - t - . . « • • ' * ; . . . . - . , - * • ,.-*'*""•> • . • . . 4 ^ « , . . . . . . . . . . . . . . - . * . . • - -* - * • . r ... - | : #• . ; •* m in . . a | • ' V * • • -. * • . . . . . . . « • • . • . . . , . » . J, » . . . . * • • - . . . . . . , . . . . „ • * * • Figure 18: Comparison of reverse transcriptase enzymes in the indirect cDNA labeling method. Total RNA isolated from P. aeruginosa was reverse transcribed using: A, Superscript™ II reverse transcriptase; and B, M-MuLV reverse transcriptase. 85 spots were more intense, uniform and distributed. Since M - M u L V also contains RNase H activity, indirect labeling using Superscript™ II (M-MuLV RNase H ) reverse transcriptase was used for all subsequent microarray experiments, and the techniques practiced and fine tuned until the standard operating protocols described in the 'Methods' section were finalized. F. Summary In this chapter many of the methodological details of microarray experimentation were evaluated. These details affect the performance of the microarray experiment overall and included evaluation of R N A isolation strategies, RNA quality assessment, genomic D N A treatment, and cDNA synthesis and labeling strategies. Of the R N A isolation protocols examined, total R N A isolated using the QMIDI RNeasy kit (Qiagen) proved to be the most consistent with respect to yield, purity and integrity, and as well, was easy and quick to use. Although total R N A isolation was not specific for R N A over D N A , contaminating genomic D N A was found to be effectively removed by treatment with DNase-free (Ambion, Inc.) for a minimum of 15min at 37°C. While higher amounts of total R N A template in the reverse transcription reaction were found to yield better amounts of cDNA, the inability of the RNA isolation strategy to consistently yield amounts of total R N A large enough to complete a set of microarray and confirmatory experiments, prohibited use of template amounts of total RNA greater than lOug. lOug of total RNA however is fairly standard amongst other bacterial microarray studies (Ye, Tao et al. 2000; Schembri, Kjaergaard et al. 2003). cDNA labeling for microarray hybridization was best achieved through indirect labeling methods using the Superscript™ II reverse transcriptase enzyme, rather than through direct fluorophore incorporation methods. Similarly, indirect labeling methods with Superscript™ II RT have also become standard in bacterial microarray experimentation (Postier, Wang et al. 2003). 86 CHAPTER THREE: Quinolone Induction of Adaptive Resistance A. Introduction Microarray experiments provide a global perspective on the response of an organism to particular environmental stimuli, knowledge which is not achievable by more focused studies. Such transcriptome analyses thus provide a more holistic view of how an organism responds to its environment and allows for more directed follow-up research. While exposure to antimicrobials is known to induce bacterial resistance mechanisms (Daikos, Jackson et al. 1990; Giwercman, Lambert et al. 1990; Eung-Tomc, Kolek et al. 1993), the full extent of bacterial responses to such stimuli are only beginning to be investigated (Betts, McLaren et al. 2003; Shaw, Miller et al. 2003; Utaida, Dunman et al. 2003; Hutter, Schaab et al. 2004). Of particular importance to CF patients chronically infected with P. aeruginosa is the response of the organism to antibiotic challenges, which occur continuously over the course of CF therapy. Fluoroquinolones, namely ciprofloxacin, are commonly prescribed to CF patients on an outpatient basis to help maintain low levels of P. aeruginosa infection. It was hypothesized that exposure to ciprofloxacin would induce resistance in P. aeruginosa, particularly adaptive resistance-like mechanisms since isolates of P. aeruginosa from CF patients during ciprofloxacin therapy display a resistance level not seen in isolates either before or after the therapy (Chamberland, Malouin et al. 1990). Adaptive resistance is defined as an unstable, reversible resistance that is unrelated to genetic mutation (Barclay and Begg 2001). It occurs transiently in an organism under non-lethal selective pressure, and has been well documented for aminoglycosides, as well as quinolones (Gould, Milne et al. 1991; Barclay and Begg 2001). The effect of sub-inhibitory and inhibitory concentrations of ciprofloxacin on the development of adaptive resistance in P. aeruginosa was thus examined on a global scale. Custom DNA microarrays were thus used to identify the response signature of P. aeruginosa to ciprofloxacin. B. Determination of the Minimum Inhibitory Concentration to Ciprofloxacin Prior to characterizing the effect of various concentrations of ciprofloxacin on the expression response of P. aeruginosa, it was necessary to define the inhibitory concentration so that sub-87 inhibitory concentrations could be appropriately calculated. The minimum inhibitory concentration (MIC) for ciprofloxacin was found to be O.lpg/ml (Table 5; Table 5 also includes the MIC values for various other strains). This value was comparable to other ciprofloxacin sensitive strains (Fung-Tome, Kolek et al. 1993) and was used to compute the 0.3x-MIC and 0.lx-MIC concentrations of 0.03 and 0.01 ug/ml, respectively. C. Growth Profile in Presence of Ciprofloxacin Before conducting global analyses on the effect of ciprofloxacin on P. aeruginosa, it was necessary to determine the point in ciprofloxacin exposure which would be best suited for this analysis. This point should show no appreciable affect on the growth ability of the organism across the various concentrations of ciprofloxacin being examined, so that the expression profile would not be confounded by expression changes due to growth differences. Cultures of P. aeruginosa were therefore grown in various concentrations of ciprofloxacin to ascertain their effect on the growth profile. The growth rate of P. aeruginosa was not appreciably affected by either 0.01 or 0.03pg/ml ciprofloxacin (Figure 19). As expected, the inhibitory concentration (O.lpg/ml) of ciprofloxacin severely limited the growth of strain HI03, although it did not completely result in killing even after 7.5 hours. At 2.5 hours however, the growth rate of P. aeruginosa in the presence of lx-MIC ciprofloxacin was analogous to the growth rates of P. aeruginosa cultured with O.lx- and 0.3x-MIC ciprofloxacin (Figure 19). This time point was therefore taken for all further analyses. Cultures at this time point were also in the mid-logarithmic phase of growth, and thus particularly amenable to microarray analysis since transcript levels would be high. D. Time-Kill Assays and Microscopy D. 1. Observation of Adaptive Resistance to Sub-Inhibitory Ciprofloxacin Sub-inhibitory concentrations of antibiotic occur at the onset of drug therapy, between dosing intervals and within the thick mucus of the CF lung (Doring, Conway et al. 2000). The effect of sub-inhibitory concentrations of antibiotic on P. aeruginosa however has not been well 88 Table 5: Minimum inhibitory concentrations of various antimicrobials against various strains of P. aeruginosa. Cultures were grown in LB medium and represent the mode of 4 independent experiments. Cipro, ciprofloxacin; Eno, enofloxacin; Nor, norfloxacin; Nal, naldixic acid; Ami, amikacin, Genta, gentamicin and Cefe, cefepime. ND, not determined. MIC (ug/ml) Strain Cipro Eno Nor Nal Ami Genta Cefe H103 0.06 0.3 0.15 15.6 0.4 1.75 0.6 PA0620::/wx 0.5 1.56 >2.5 250 0.4 0.875 2.5 PA0621::IS/acZ 0.5 >5 >2.5 250 0.2 0.875 0.3 PA0641::/wx 0.5 1.56 1.25 125 0.4 1.75 1.25 PA3866::/wx 0.06 0.3 0.15 31.25 0.4 ND 0.6 prtRvlSlacZ 0.06 0.3 0.15 31.25 0.4 0.875 0.6 recAr.ISlacZ 0.5 3.125 >2.5 250 0.4 0.875 0.6 89 10 -, - • - H I 03 -B-H103 + 0.0lug/ml ciprofloxacin Hours Figure 19: Growth curve of P. aeruginosa grown with various concentrations ciprofloxacin. 90 characterized apart from studies evaluating various phenotypes like adherence (Sonstein and Burnham 1993; Visser, Beumer et al. 1993; Zhanel, Kim et al. 1993) or virulence factor production (Ravizzola, Pirali et al. 1987; Grimwood, To et al. 1989; Trancassini, Brenciaglia et al. 1992; Sonstein and Burnham 1993). Fung-Tome et al. (Fung-Tome, Kolek et al. 1993) also noted an increase in the mutation rate and resistance level of P. aeruginosa following exposure to sub-inhibitory ciprofloxacin. No studies however have addressed the adaptive resistance response of P. aeruginosa to sub-inhibitory concentrations of fluoroquinolones. Given the highly compartmentalized lung environment, the thick mucus of the CF lung, and the propensity for P. aeruginosa to covert to mucoidy and grow in biofilms, it was hypothesized that sub-inhibitory ciprofloxacin mediates adaptive resistance in P. aeruginosa. To test this hypothesis, the time-to-kill assay was modified to include a pre-incubation step prior to the time-to-kill curve. CFUs were measured as they more accurately reflect viable cells rather than OD measures. In this manner, the effect of exposure to various concentrations of ciprofloxacin on the survival or adaptation of P. aeruginosa to subsequent supra-inhibitory ciprofloxacin could be examined. Cells pretreated with sub-inhibitory concentrations of ciprofloxacin, washed and then resuspended in 2x-MIC ciprofloxacin, grew better than untreated P. aeruginosa cells (Figure 20). In particular, cells pretreated with 0.3x-MIC ciprofloxacin outperformed all other pretreatment groups by approximately one logio of growth. This difference in survival became most distinctive at 75 minutes post supra-inhibitory exposure. P. aeruginosa cultures pretreated with lx-MIC ciprofloxacin are not displayed because these cells were not observed at detectable levels in subsequent exposure to 2x-MIC ciprofloxacin over the time frame examined (but did survive if left overnight in 2x-MIC ciprofloxacin; data not shown). Thus, exposure of P. aeruginosa to sub-inhibitory concentrations of ciprofloxacin to the mid-logarithmic phase induces mechanisms which aid the organism's survival in supra-inhibitory concentrations of ciprofloxacin. The adaptive resistance assay was also followed for an extended time course beginning at 2 hours post exposure to 2x-MIC ciprofloxacin (Figure 21). As the length of time cultures were exposed to 2x-MIC ciprofloxacin increased, the difference in survival noted for sub-inhibitory pretreated cultures compared to untreated cultures remained (Figure 21). Thus adaptive resistance in P. aeruginosa is sustained for the duration of exposure to supra-inhibitory ciprofloxacin. Similar prolonged adaptive resistance observations have been noted in both in 91 . 0 0 E + 0 6 15 30 45 60 75 Time (minutes) 90 105 120 Figure 20: Survival ability of P. aeruginosa in 2x-MIC ciprofloxacin following growth in sub-inhibitory concentrations of ciprofloxacin. Cultures of P. aeruginosa strain HI 03 were grown in the presence or absence of sub-inhibitory concentrations of ciprofloxacin until mid-logarithmic phase, washed and then exposed to 0.2pg/ml ciprofloxacin. Colony forming units were counted every 15 minutes for 2 hours. Data representative of three separate experiments are shown. 92 l.OOE+05 i g 1.00E+04 DD O J l.OOE+03 - • - H I 0 3 - Untreated -B-H103 +0.01ug/ml ciprofloxacin —A—H103 +0.03ug/ml ciprofloxacin Time (hrs) Figure 21: Survival ability of pretreated cultures of P. ciprofloxacin over a longer time frame. aeruginosa in 2x-MIC Cultures of P. aeruginosa strain HI03 were grown in the presence or absence of sub-inhibitory concentrations of ciprofloxacin until mid-logarithmic phase, washed and then exposed to 0.2pg/ml ciprofloxacin. Colony forming units were counted for 7 hours. Data representative of two separate experiments are shown. 93 vitro (Barclay, Begg et al. 1992) and in vivo (Daikos, Lolans et al. 1991) assays assessing adaptive resistance in P. aeruginosa elicited by aminoglycosides. D.2. Verification of Adaptive Resistance to Sub-Inhibitory Ciprofloxacin Ciprofloxacin at supra-inhibitory concentrations is known to induce conversion of P. aeruginosa to a mucoid phenotype (Pina and Mattingly 1997). Although sub-inhibitory concentrations have not been found to induce similar phenotypic switches (Trancassini, Brenciaglia et al. 1992; Majtan and Hybenova 1996) or result in major changes in LPS structure (Magni, Giordano et al. 1994; McKenney, Willcock et al. 1994), mucoid growth was also examined to eliminate the possibility that the increase in CFUs for pretreated cultures resulted from mechanical disruption of mucoid cells. In comparison to the shiny, mucoid appearance of P. aeruginosa continuously grown in supra-inhibitory ciprofloxacin (Pina and Mattingly 1997), P. aeruginosa cultures exposed to sub-inhibitory ciprofloxacin were not found to be mucoid. Thus, the higher bacterial counts observed during the adaptive resistance assay were not a result of the spreading of ciprofloxacin converted-mucoid cells. Gram stains of untreated and pretreated cultures before and during the adaptive resistance assay however showed that pretreatment of strain HI03 with sub-inhibitory ciprofloxacin only slightly reduced cell division (Figure 22, panels A. and B.). For comparison purposes, the dramatic inhibition resultant from pretreatment of P. aeruginosa cells with inhibitory ciprofloxacin is shown (Figure 22, panels A.4. and B.4.). Cell lengths were measured to be 2.8, 3.0, 3.6 and 6.5 microns in panel A and 4.8, 5.4, 7.0 and 11.3 microns in panel B, for untreated, O.lx-, 0.3x- and lx-MIC ciprofloxacin treated cultures, respectively. In contrast to the effect of mucoidy on CFUs, inhibition of cell division instead indicates that the adaptive resistance time-kill assay underestimated the colony counts for the ciprofloxacin pretreated groups since cells were not dividing as rapidly as the untreated group. Thus, the overall difference in bacterial counts observed during the adaptive resistance assay would be larger if inhibition of cell division is taken into account. Following completion of the adaptive resistance assay, culturing of P. aeruginosa cells in drug free media resulted in the return of wild type cell morphology (Figure 22, panel C) , illustrating the transiency of the effect. Cell lengths for panel C were measured to 94 Ciprofloxacin Untreated 0.01 ug/ml 0.03ug/ml O.lpg/ml A . l . ' 3 i i 2. i * 3. 4. :> - ^ v it B.l. I 2. uS f _ 3. X* i - "f^"*' 4. it •-I- JS 3 © o JL * u £ A oa p 2 •= a «5 i II — — fed CJ = * S iL Q. W 2 5 a. z 1 C.l . 2. 3. Figure 22: Gram stains of P. aeruginosa before, during and after the adaptive resistance assay. Ciprofloxacin interferes with P. aeruginosa cell division as evident from Gram stains prior to (A) and post 1.5hrs (B) growth in 2x MIC ciprofloxacin. Cells cultured for an additional 3hrs in drug free media (C) were not inhibited. (1) PAO-H103 untreated; (2) PAO-H103 pre-treated in 0.01pg/ml ciprofloxacin; (3) PAO-H103 pre-treated in 0.03pg/ml ciprofloxacin; and (4) PAO-H103 pre-treated in 0.1 pg/ml ciprofloxacin. A l l micrographs are at lOOOx magnification. Cell lengths were measured to be 2.8, 3.0, 3.6 and 6.5 microns in panel A , 4.8, 5.4, 7.0 and 11.3 microns in panel B, and 3.6, 3.0 and 3.3 microns in panel C for untreated, O.lx-, 0.3x- and lx-MIC ciprofloxacin treated cultures, respectively. 95 be 3.6, 3.0 and 3.3 microns for untreated, O.lx-, 0.3x- and lx-MIC ciprofloxacin treated cultures, respectively. E. Microarray Studies E. 1. Expression Responses following Treatment with Sub-Inhibitory Ciprofloxacin Having established that sub-inhibitory concentrations of ciprofloxacin induce adaptive resistance in P. aeruginosa, custom DNA microarrays to P. aeruginosa were used to better define how sub-inhibitory ciprofloxacin affects the expression response of this organism in an attempt to define the molecular mechanisms behind adaptive resistance. Microarrays were performed on cultures of P. aeruginosa following growth to the mid-logarithmic phase in the presence or absence of various concentrations of ciprofloxacin as before. The complete set of genes exhibiting significant (p< 0.05) expression changes is provided in Appendix III, where they have been separated into up-regulated and down-regulated gene lists, and sorted according to functional classification. Magnitudes of transcript inhibition or activation expressed as fold change were calculated using untreated cultures as baseline; fold changes in gene expression are rounded to two significant figures. Changes in the control spike transcript intensities were found to be less than two-fold for all comparisons, indicating that the efficiency of cDNA synthesis and labeling was similar from sample to sample. Overall, as the concentration of ciprofloxacin was increased, the number of genes exhibiting significant expression changes increased in parallel (Appendix III and summarized in Table 6). Exposure to 0.lx-MIC ciprofloxacin resulted in significant expression changes in 566 genes. This number increased to 941 and 1230 following exposure to 0.3x-MIC and lx-MIC ciprofloxacin, respectively. Overall though, many of the expression changes were fairly subtle, with the majority of significant expression changes falling below 2-fold (Appendix III and Table 6). Each of the gene lists was further analyzed with respect to magnitude of fold change. Of the 566 genes identified as being affected by 0.lx-MIC ciprofloxacin, only 148 genes (26%) showed increases or decreases of more that 2-fold (Table 6). This value of 26%> did not change with the increase in ciprofloxacin concentration (24% and 29%> for 0.3x- and lx-MIC ciprofloxacin, respectively). Comparable numbers of genes exhibiting >2-fold and >5-fold expression changes, 96 Table 6: Summary of custom DNA microarray findings and trends. Ciprofloxacin O.lx-MIC 0.3x-MIC lx-MIC Total # Genes with Significant Expression Changes (p < 0.05) Total # Genes Up-regulated Total # Genes Down-regulated 566 332 234 941 554 387 1230 743 487 # Genes with Fold Change -2> % < 2 418 717 870 # Genes with Fold Change > +2 (inclusive of 5-, 10- and 15-fold changes) # Genes with Fold Change > -2 (inclusive of 5-, 10- and 15-fold changes) 99 49 133 91 207 153 # Genes with Fold Change > +5 (inclusive of 10- and 15-fold changes) # Genes with Fold Change > -5 (inclusive of 10- and 15-fold changes) 7 0 8 8 55 8 # Genes with Fold Change > +10 (inclusive of 15-fold changes) # Genes with Fold Change > -10 (inclusive of 15-fold changes) 3 0 2 2 30 4 # Genes with Fold Change > +15 # Genes with Fold Change > -15 1 0 1 1 16 4 # Genes with Hypothetical/ Putative Classification # Hypo. Genes |> 5| Fold Change # Hypo. Genes > 10| Fold Change # Hypo. Genes |> 15| Fold Change 278 5 3 1 432 12 3 2 579 20 7 6 97 however, have been observed after imipenem treatment of P. aeruginosa by another microarray user (Bagge, Schuster et al. 2004). While the percentage of genes exhibiting greater than 2-fold expression changes did not vary across the treatment conditions, the extent of the expression change did vary. At the inhibitory concentration of ciprofloxacin, the number of genes having fold changes larger than 5-fold, 10-fold or 15-fold increased substantially in comparison to the O.lx- and 0.3x-MIC treatment groups, which had comparable numbers of genes in each of these categories (Table 6). Whereas the O.lx- and 0.3x-MIC groups had 7 or 16 genes respectively with expression changes greater than 5-fold, the lx-MIC group had 63 genes with expression changes greater than 5-fold. Thus, as the concentration of ciprofloxacin was increased from sub-inhibitory to inhibitory, the expression changes in P. aeruginosa became more intense in response to the increasingly challenging growth environment. With few exceptions, where expression changes were noted in more than one treatment group, the trend in expression was the same for both groups, such that if inhibition of expression was noted for an ORF in the 0.3x-MIC group, the extent of inhibition of expression remained the same or was enhanced in the lx-MIC group (Appendix III). As observed by other P. aeruginosa microarrays users (Affymetrix), where genes formed an operon such as the trp cluster (PA0649 to PA0651), the changes for the first gene in the operon were much greater than the changes for the downstream genes (Appendix III) (Ochsner, Wilderman et al. 2002; Schuster, Lostroh et al. 2003; Wagner, Bushnell et al. 2003). As well, as noted by others, not all genes within an operon had statistically significant changes in transcript expression. For example, at least three genes involved in motility and attachment, pilA, pilB, and pilC, exhibited statistically significant decreases in transcript expression across the sub-inhibitory concentrations of ciprofloxacin, but other genes in this operon did not change. Separating the genes exhibiting significant expression changes on the basis of functional classification also revealed some interesting results. In general, as the concentration of ciprofloxacin was increased, the number of genes within each functional category also increased (Appendix III). As well, it was found that approximately half of all differentially expressed genes in all treatment groups were functionally classified as either 'hypothetical, unclassified protein' or 'putative enzyme' (Appendix III and Table 6). Furthermore, the expression changes for these ORFs tended to be fairly large. These genes, rather than the annotated genes, constituted many of the genes previously noted as exhibiting expression changes greater than 5-fold, 10-fold or 15-98 fold (Table 6). Together these findings illustrate that a large proportion of genes responsive to ciprofloxacin remain to be fully characterized. Their large expression changes also underscore the importance of these uncharacterized ORFs in the cellular response of P. aeruginosa to ciprofloxacin challenge. Other noteworthy functional categories include the 'DNA replication, recombination, modification and repair' and 'chaperones and heat shock proteins' categories (Appendix III). These categories were of particular interest because they include the targets for ciprofloxacin, and function in the SOS DNA-repair response induced by fluoroquinolone antimicrobials (Phillips, Culebras et al. 1987; Lewin, Howard et al. 1989). Genes with ciprofloxacin-modified expression from these functional categories included lexA, recA, recN, mutS, mutL and mutY, as well as groEL and groES among others (Appendix III and Table 7). Expression levels of many of these DNA repair genes were up-regulated with increasing ciprofloxacin concentration. Each concentration of ciprofloxacin contained a number of ORFs differentially expressed only at that particular concentration (Appendix III and Figure 23). Again, the number of such genes was found to increase with increasing concentration of ciprofloxacin. Of interest were the ORFs present only in the 0.3x-MIC treatment group, since these genes may be responsible for the observed adaptive resistance response induced by pretreatment with 0.3x-MIC ciprofloxacin. However, the descriptions for these 271 ORFs did not provide any significant leads, although subtle expression changes were noted for several members of the multidrug efflux pumps including mexD, oprJ and mexB and opmB (Table 7). Changes in these genes however were inconsistent (i.e. showing increased expression of oprJ but decreased expression of mexD) and not amenable to any theory of resistance. With specific reference to the genes known to be altered by sub-inhibitory ciprofloxacin, like genes involved in pili and fimbriae production (Sonstein and Burnham 1993) and thus mediating bacterial-host cell adherence (Sonstein and Burnham 1993; Visser, Beumer et al. 1993; Zhanel, Kim et al. 1993), many of the ORFs functionally classified as being involved in 'motility and attachment' were found to be down-regulated with increasing concentration of ciprofloxacin (Appendix III and Table 7). With reference to the specific virulence factors known to be down-regulated by sub-inhibitory ciprofloxacin (Ravizzola, Pirali et al. 1987; Grimwood, To et al. 1989; Trancassini, Brenciaglia et al. 1992; Sonstein and Burnham 1993), only exoenzyme S, exoS, the secretory component of phospholipase C,plcR and hydrogen cyanide synthase B, hcnB 99 Table 7: Expression changes in various genes induced by O.lx-, 0.3x- and lx-MIC ciprofloxacin. ORF Gene name Fold change in PAO-H103 Ciprofloxacin O.lx MIC 0.3xMIC lxMIC Functional Classification PA4385 groEL 1.30 Chaperones & heat shock proteins PA4386 groES 2.03 2.12 Chaperones & heat shock proteins PA3617 recA 1.86 2.83 4.55 DNA replication, recombination and repair PA3620 mutS 1.82 1.49 DNA replication, recombination and repair PA4763 recN 1.59 2.11 DNA replication, recombination and repair PA4946 mutL 1.51 DNA replication, recombination and repair PA5147 mutY 1.20 1.48 DNA replication, recombination and repair PA3007 lexA 1.60 2.55 3.58 Translation, post-trans.modification,degrad PA0426 mexB 1.40 Antibiotic resistance and susceptibility PA4598 mexD -1.36 Antibiotic resistance and susceptibility PA0004 gyrB -1.61 DNA replication, recombination and repair PA4596 1.79 1.53 Transcriptional regulators PA0427 oprM 2.51 Transport of small molecules PA2525 opmB -2.58 Transport of small molecules PA3522 -1.57 Transport of small molecules PA4208 opmD 1.34 1.88 Transport of small molecules PA4597 oprJ 1.47 Transport of small molecules PA2194 hcnB -1.56 -1.62 -3.35 Central intermediary metabolism PA0843 plcR -3.80 Secreted Factors (toxins, enzymes,alginate) P A3 841 exoS -1.80 -1.79 Secreted Factors (toxins, enzymes,alginate) PA0396 pilU -1.45 Motility & Attachment PA0408 pilG -1.91 -1.57 -1.84 Motility & Attachment PA0410 pill -1.27 Motility & Attachment PA0994 -4.34 Motility & Attachment PA4525 pilA -2.81 Motility & Attachment PA4526 pilB -1.42 Motility & Attachment PA4527 pilC -1.69 Motility & Attachment PA0763 mucA 2.11 2.16 Secreted Factors (toxins, enzymes,alginate) PA0762 algU 2.06 Transcriptional regulators PA4002 rodA 1.77 Cell wall / LPS / capsule PA4003 pbpA 1.46 Cell wall / LPS / capsule PA4418 ftsl 1.06 Cell wall / LPS / capsule 100 O.lx-MIC 0.3x-MIC Figure 23: Relationship of genes with significant expression changes induced by 0.1 0.3x- and lx-MIC ciprofloxacin in P. aeruginosa. 101 were found to be inhibited by increasing concentrations of ciprofloxacin (Appendix III and Table 7). In corroboration with the above findings on mucus production, many of the genes involved in the regulation and production of alginate were not found to have significant expression changes compared to the untreated condition. Although expression of the main alternative sigma factor algU was increased with lx-MIC ciprofloxacin, the ability of this gene to induce alginate production is modulated by the corresponding increase in expression of the anti-sigma factor mucA, data which mirrors the known mechanisms of alginate biosynthesis regulation (Schurr, Yu et al. 1996). As well, in corroboration with the above findings on filamentation, a few of the genes known to be involved in cell division (Errington, Daniel et al. 2003) were found to be up-regulated only at the inhibitory ciprofloxacin condition, namely ftsl, rodA and pbpA (Appendix III and Table 7). Many of the other key components in cell division however, were not detected. It is important to note though that the inhibition of cell division was not visibly noticeable until approximately 2 hours post-exposure to 2x-MIC ciprofloxacin, rather than after growth to the mid-logarithmic phase in sub-inhibitory ciprofloxacin; slight inhibition of cell division was only visible for the inhibitory ciprofloxacin condition at this point in time (Figure 22, panel A.4.). So, more dramatic changes in cell division genes would be expected from transcriptional analysis of cultures following the adaptive resistance time-to-kill assay. Of the genes known to contribute to ciprofloxacin resistance, no significant expression changes were observed (Appendix III and Table 7). A modest decrease in expression of gyrB was noted but only for the 0.lx-MIC ciprofloxacin treatment condition. Significant expression differences were not noted for the other topoisomerases, parC or parE. Comparatively, more changes were noted in the multidrug efflux mechanisms. Expression of nfxB, the negative regulatory gene for the resistance-nodule-cell division (RND) multidrug efflux system mexCDoprJ was undetected, although PA4596, which is 74% similar to nfxB of P. aeruginosa, was modestly increased in the O.lx- and 0.3x-MIC treatment conditions, with an average fold change of 1.7. Expression of the outer membrane protein oprJ increased only in the presence of 0.3x-MIC ciprofloxacin, and only by 1.5 fold. Other members of the OprM family of outer membrane proteins were also altered including oprM, opmD and opmB, but again only in the 0.3x-MIC condition. 102 The region between PA0609 and PA0651 was most striking because every gene was affected (Appendix III and Table 8). PA0613 to PA0648 represents a region of the P. aeruginosa genome which is related to various bacteriophage (Nakayama, Takashima et al. 2000). This region is described in more detail in the following chapter. F. Microarray Confirmation Assays Confirmation of the expression changes in the R2/F2 pyocin region are discussed in more detail in the next chapter. The involvement of type IV pili in twitching motility was investigated since many of the genes in type IV fimbriae biogenesis (Aim and Mattick 1997) were found to be down-regulated by custom DNA microarray analysis (Appendix III). The twitching motility of P. aeruginosa cultures exposed to sub-inhibitory concentrations of ciprofloxacin was found to be inhibited compared to untreated cultures (Figure 24), confirming the microarray findings. G. Summary The effects of sub-inhibitory concentrations of ciprofloxacin were examined in this chapter. Since these concentrations commonly occur at the onset of drug therapy, between dosing intervals and within the thick compartmentalized mucoid environment of the CF lung (Doring, Conway et al. 2000), they likely play an important role in priming P. aeruginosa to its environment. Sub-inhibitory concentrations of ciprofloxacin were not found to appreciably affect the growth ability of P. aeruginosa, although as expected inhibitory concentrations of ciprofloxacin were found to dramatically affect the growth of the organism. Similar instances of incomplete killing with inhibitory concentrations of ciprofloxacin have been previously noted and have been proposed to result from the presence of persister cells (Keren, Kaldalu et al. 2004). Pretreatment of P. aeruginosa cultures with sub-MIC ciprofloxacin was found to enhance the survival of the organism when subsequently challenged with supra-inhibitory ciprofloxacin. The reduction in kill was approximately one logio for 0.3x-MIC exposed cultures, a value consistent with previous work (Gould, Milne et al. 1991). In contrast to previous work (Gould, Milne et al. 103 Table 8: R2/F2 Pyocin-phage operon and related genes induced by 0.3x- and lx-MIC ciprofloxacin3. Fold Fold ORF Gene change in change in Description Nameb 0.3x-MIC lx-MIC PA0610 prtN 2.0 6.8 Transcriptional activator PA0611 prtR 1.3 3.0 Transcriptional repressor PA0612 e 8.5 Homolog of Zn finger protein PA0613 4.0 19.3 Conserved hypothetical PA0614 hoi 3.3 15.3 Holin PA0615 1.7 7.2 Conserved hypothetical PA0616 VR2 2.9 13.0 Homologous to baseplate assembly protein V PA0617 wR2 3.0 13.9 Homologous to baseplate assembly protein W PA0618 JR.2 2.1 10.0 Homologous to baseplate assembly protein J PA0619 IR2 3.2 14.7 Homologous to tail protein I PA0620 HR2 2.7 12.1 Homologous to tail fibre protein H PA0621 4.3 16.7 Homologous to tail fibre assembly protein PA0622 FIR2 2.4 10.3 Homologous to contractile sheath protein Fl PA0623 FIIR2 2.9 12.8 Homologous to tail tube protein FII PA0624 2.7 12.8 Conserved hypothetical PA0625 2.5 14.1 Homologous to tail length determination protein PA0626 UR2 1.8 9.8 Homologous to tail formation protein TJ PA0627 XR2 2.3 10.8 Homologous to tail protein X PA0628 DR2 1.8 10.0 Homologous to tail formation protein D PA0629 lys 2.2 8.7 Lytic protein; Homology to predicted chitinase PA0630 1.6 10.6 Hypothetical protein PA0631 2.4 16.2 Unique hypothetical protein PA0632 3.4 21.2 Unique hypothetical protein PA0633 VF2 4.3 16.4 Homologous to major tail protein V PA0634 3.2 15.8 Unique hypothetical protein PA0635 2.0 15.9 Conserved hypothetical protein PA0636 HF2 2.1 10.0 Homologous to tail length determination protein H PA0637 MF2 2.7 11.0 Homologous to tail fibre protein M PA0638 LF2 — 8.5 Homologous to tail fibre protein L PA0639 KF2 3.0 15.9 Homologous to tail assembly protein K PA0640 h2 3.1 16.4 Homologous to tail assembly protein I PA0641 JF2 2.4 9.8 Homologous to tail fibre protein J PA0642 2.7 12.9 Hypothetical protein PA0643 1.9 8.9 Homologous to tail fibre domain protein PA0644 4.0 18.0 Hypothetical protein PA0645 1.8 15.8 Hypothetical protein PA0646 3.1 13.4 Homologous to putative tail fibre protein PA0647 3.8 23.0 Conserved hypothetical protein PA0648 — 8.2 Conserved hypothetical protein 104 PA0985 pys5 4.5 18.1 Pyocin S5 PA1150 pys2 — 5.4 Pyocin S2 PA3617 recA 2.8 4.6 Recombinase for DNA recombination and repair P A3 866 pys4 13.9 51.3 Pyocin S4 a - Only genes identified as being affected by sub-inhibitory (0.3 ug/ml) or inhibitory (1 ug/ml) ciprofloxacin relative to untreated strain HI03, and that were relevant to pyocin/phage are included. A list of all genes identified as being affected by these concentrations of ciprofloxacin is available in Appendix III. Genes are identified by ORF designation, gene name or alternative gene name and homology description based on the Pseudomonas genome project (www.pseudomonas.com). b - Genes were named as per Table 1 of Nakayama, K., K. Takashima et al. 2000 and reflect the homology observed by these authors to phages P2 and X. c - "—" means no significant change in expression detected. 105 2.50 -, PAO-H103 Untreated PAO-H103 + O.lx-MIC PAO-H103 + 0.3x-MIC ciprofloxacin ciprofloxacin Figure 24: Analysis of twitch motility of P. aeruginosa cells cultured in the presence or absence of ciprofloxacin. Twitch zones were measured where they occurred between the plate and agar interface. Results are the average of three independent experiments. 106 1991) however, inhibitory ciprofloxacin was not found to induce adaptive resistance in P. aeruginosa. This is likely due to the short two hour time frame used in the kill assay; others have shown adaptive resistance with supra-inhibitory ciprofloxacin but only after two hours (Gould, Milne et al. 1991). Indeed, inhibitory ciprofloxacin pretreated cultures were viable after overnight growth (data not shown). Adaptive resistance in P. aeruginosa to 0.3x-MIC ciprofloxacin was observed to occur for more than 7 hours, a time frame comparable to but longer than that noted for adaptive resistance in response to 8x-MIC ciprofloxacin (Gould, Milne et al. 1991) and may be related to the lower second exposure concentration (2x-MIC compared to 8x-MIC). Thus, although adaptive resistance is defined as unstable, in the continued presence of drug, adaptive resistance was enhanced and prolonged (Daikos, Jackson et al. 1990). No effect of sub-inhibitory ciprofloxacin on mucus production was found, in contrast to the known effects of inhibitory ciprofloxacin on conversion of P. aeruginosa to mucoidy (Pina and Mattingly 1997). Mucus production therefore was not responsible for this measured enhancement. It was found however that ciprofloxacin inhibits cell division, findings which indicate that the extent of adaptive resistance observed for the O.lx- and 0.3x-MIC ciprofloxacin pretreatment groups relative to the untreated pretreatment group, was likely being underestimated. From custom DNA microarray analysis of ciprofloxacin exposed P. aeruginosa cultures, it was discovered that as the concentration of ciprofloxacin was increased, the number of genes exhibiting significant differential expression increased in parallel, as did their level of expression. Although changes in the first gene of an operon were often larger than changes in downstream genes, other microarray studies have observed similar results (Ochsner, Wilderman et al. 2002; Schuster, Lostroh et al. 2003; Wagner, Bushnell et al. 2003). RNA degradation from the 3' end of the transcript is one explanation for this phenomenon. It is also possible that such an operon is under coordinated expression, whereby transcription of the first gene occurs prior to transcription of downstream genes, as is the case for type III secretion systems (Hueck 1998). Also noted by other microarray users was the inconsistency of all genes within an operon to have statistically significant changes in transcript expression (Ochsner, Wilderman et al. 2002; Wagner, Bushnell et al. 2003). This observation has several explanations. While this discrepancy may be partially explained by the absence of some genes on the microarray (due to an inability to amplify the genes to a high enough concentration for spotting), differences in transcript levels 107 between replicates were also observed. Thus, biological variation and/or RNA degradation may have also increased the standard deviation of some operon genes, thereby forcing their exclusion from the gene lists. As well, the intensities of poorly expressed transcripts were often close to the background intensity calculated for the associated sub-grid, and thus these transcripts were defined as absent, again excluding such operon genes from further analysis. Of all the functional classifications observed to change following exposure to sub-inhibitory ciprofloxacin, the greatest numbers of genes were designated as hypothetical or putative, and their expression changes tended to be fairly large compared to the remaining annotated genes. Since none of the known ciprofloxacin resistance mechanisms were found to exhibit significant differential expression changes, these uncharacterized ORFs may contribute in as yet unidentified ways to the observed response of the P. aeruginosa to ciprofloxacin. In accordance with the ability of fluoroquinolones to provoke the SOS DNA-repair response, ciprofloxacin treatment resulted in expression changes in many DNA repair genes (Appendix III). Induction of the SOS response is known to involve the genes for RecA, RecBCD and LexA; many of these genes were found to exhibit increased expression levels. Others have also shown the ability of ciprofloxacin but not novobiocin, a gyrase B subunit inhibitor, to elicit the SOS response (Gmuender, Kuratli et al. 2001). These results thus highlight the SOS DNA-repair system in the response of P. aeruginosa to the DNA damaging effects of ciprofloxacin. Many of the genes previously identified as being affected by sub-inhibitory ciprofloxacin were confirmed by the microarray study. These included genes involved in filamentation and alginate production, as well as virulence factors and motility/attachment genes. The inhibition of type IV fimbriae gene expression was further confirmed by a reduction in twitch zones for P. aeruginosa cultures exposed to sub-inhibitory ciprofloxacin. Together, this confirmation of earlier phenotypic work on sub-inhibitory ciprofloxacin treated P. aeruginosa cultures was taken as validation of the design and construction of this custom DNA microarray, and its associated methods and analysis programs. Although expression changes in many of the known ciprofloxacin resistance genes were not observed, interestingly, this may be a result of the concentrations of ciprofloxacin studied since quinolone concentration appears to affect which resistance mechanism is selected by P. aeruginosa in response to quinolone exposure. At concentrations close to the MIC, efflux type mechanisms were selected almost exclusively in the lab with gyrase type mutations appearing 108 only at concentrations above 4x-MIC (Kohler and Pechere 2001). Since only the MIC and lower concentrations of ciprofloxacin were examined here, the absence of up-regulation of classical resistance genes may be related to the dose. Similar absentia observations have been made by others characterizing the response of Haemophilus influenzae to lx- and lOx-MIC ciprofloxacin (Gmuender, Kuratli etal. 2001). The striking expression changes in the R2/F2 pyocin region are discussed in the next chapter. The role of this region in the response of P. aeruginosa to ciprofloxacin however is unknown. 109 CHAPTER FOUR: Involvement of Pyocin/Phage Expression in Quinolone Resistance A. Introduction Pyocins are bactericidal compounds produced by P. aeruginosa (Jacob 1954). While spontaneous pyocin production or pyocinogeny is low, pyocin production can be induced upon treatment of cultures with mutagenic agents (Jacob 1954; Kageyama 1964); ciprofloxacin along with other fluoroquinolones, is a mutagenic agent (Phillips, Culebras et al. 1987; Clerch, Bravo et al. 1996). Thus, up-regulation of pyocin expression in response to ciprofloxacin exposure was not unexpected. P. aeruginosa produces three types of pyocins: R-, F- and S-type, all encoded on the chromosome, rather than on plasmids as in other bacteria. From the microarray findings, the most striking results were seen for the R2/F2 pyocin region, where the entire region was up-regulated in response to 0.3x- and lx-MIC ciprofloxacin treatment (data presented and discussed in this chapter). While up-regulation of this pyocin region was anticipated, its role in the response of P. aeruginosa to ciprofloxacin and its potential role in mediating the observed adaptive resistance response to sub-inhibitory ciprofloxacin, remain unclear. These roles were examined more extensively in this chapter. B. Microarray Studies B . l . Expression Responses of the R2/F2 Pyocin Region to Sub-inhibitory Ciprofloxacin Of all the expression changes observed in response to ciprofloxacin challenge, the region between PA0609 and PA0651 was most striking because every gene was affected (Appendix III and Table 8). PA0613 to PA0648 represents a region of the P. aeruginosa genome which is related to various bacteriophages (Nakayama, Takashima et al. 2000). The genes flanking this region correspond to the anthranilate synthesis genes (PA0609, PA0649, PA0650 and PA0651), and were independently affected by ciprofloxacin. Other notable genes relating to regulation of expression of this region include prtR, prtN (PA0610, 0611) and recA (PA3617) (Matsui, Sano et al. 1993). These genes were affected appropriately with respect to regulation of the pyocin 110 genes, as were other pyocins regulated by the same mechanisms, pyocin S2 (pys2, PA1150), pyocin S4 (PA3866) and pyocin S5 (PA0985). Expression of these genes and regions all increased in parallel with increases in the concentration of ciprofloxacin (Appendix III and Table 8). C. Microarray Confirmation Assays CA. Real-time PCR Analysis of Expression Expression changes for a variety of pyocin/phage genes were confirmed by real time PCR. Overall, the expression changes identified by relative real-time PCR followed the same trends as those found by custom DNA microarray analysis (Table 9). Expression changes determined by relative real-time PCR however tended to be of a higher magnitude than that found by microarray analysis, data which is common since real-time PCR in general is more sensitive to transcript levels (Yuen, Wurmbach et al. 2002; Park, Cao et al. 2004). C.2. Luminescence Analysis of Expression Expression changes for the R2/F2 pyocin region were also confirmed by following the luminescence from several mini-Tn5 luxCDABE transcriptional fusions (Table 3) over a three hour time course in the presence of various concentrations of ciprofloxacin (Figure 25). Luminescence, indicative of transcription from lux fusions, was found to increase over the time period examined and with increasing ciprofloxacin concentration. These expression trends were similar to those noted by custom DNA microarray analysis for mid-logarithmic phase P. aeruginosa cells treated with ciprofloxacin (Appendix III and Table 8). For example, expression of PA0620 was induced 3-fold and 2.7-fold following exposure to 0.3x-MIC ciprofloxacin for 3 hours, as detected by the luminescence assay and microarray, respectively. PA0641 was induced 2-fold and 2.4-fold after exposure to 0.3x-MIC ciprofloxacin for 3 hours, as detected by the luminescence assay and microarray, respectively. I l l Table 9: Comparison of expression changes for various genes as analyzed by relative real-time PCR and custom P. aeruginosa microarray. ORF Gene Name Fold change in PAO-H103 0.3x-MIC Ciprofloxacin Custom Microarray Real-time PCR PA0610 prtN 2.0 10.0 PA0611 prtR 1.3 11.4 PA0621 4.3 3.0 PA3617 recA 2.8 3.9 PA3866 pys3 13.9 22.5 112 A. 4000 3000 s S •c a o g 2000 s -OX 1000 • PA0620::lux Untreated • PA0620::lux + 0.1 x-MIC ciprofloxacin • PA0620::lux + 0.3x-MIC ciprofloxacin B. Time (hours) 1000 500 • PA0641 • PA0641 • PA0641 lux Untreated lux + 0.1 x-MIC ciprofloxacin lux + 0.3x-MIC ciprofloxacin Time (hours) Figure 25: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of ciprofloxacin. PA0620::luxCDABE (A) and PA0641-.-.luxCDABE (B) were grown for three hours in the presence of O.Olpg/ml and 0.03pg/ml ciprofloxacin. Expression is defined as the average luminescence of three wells corrected for culture density measured at OD6oo- Results are the average of 4 independent experiments. 113 D. Analysis of Pyocin/Phage Expression DA. Luminescent Analysis of Pyocin/Phage Expression As presented above, the ability of ciprofloxacin and other DNA gyrase inhibitors to induce expression of the R2/F2 pyocin region was investigated using luxCDABE transcriptional fusions to various genes in the R2/F2 operon (Table 3). Other DNA damaging agents and antibiotics were used for comparison. As previously demonstrated for ciprofloxacin (Figure 25), these fusions allow for the measurement of expression through the production of light. i. Novobiocin Dose and Time Course Like quinolones, DNA gyrase is the target of the coumarin group of antibacterials, of which novobiocin is a member. Coumarins are known to inhibit the ATPase activity of gyrase by binding to the 24-kDa N-terminal sub-domain of gyrase B protein (Ali, Jackson et al. 1993; Maxwell 1993; Gilbert and Maxwell 1994) whereas fluoroquinolones like ciprofloxacin are known to interact with gyrase A protein. The effect of novobiocin on pyocin expression was therefore examined to determine if different gyrase subunit inhibitors could induce similar expression changes. For comparison purposes to ciprofloxacin, O.lx- and 0.3x-MIC novobiocin (see Table 10) were examined. These concentrations of novobiocin were found to not appreciably inhibit the growth of P. aeruginosa (Figure 26). Expression levels increased 2.5-fold for both the R2 and F2 pyocin mutants after 3 hour exposure to 0.3x-MIC novobiocin (Figure 27); in comparison, expression levels increased 3-fold and 2-fold for PA0620::/«x and PA0641::/ux, respectively, after 3 hour exposure to 0.3x-MIC ciprofloxacin (Figure 25). Thus, increasing concentrations of novobiocin were found to stimulate transcription from the pyocin/phage lux fusions as measured by increases in luminescence. Expression was also found to increase with length of exposure to novobiocin, from 700 or 300 units after 1 hour exposure to 2800 or 700 units for PA0620::/w;t and PA0641::/wx mutants respectively, after 3 hour exposure (Figure 27). Thus, gyrase B subunit inhibitors like novobiocin also induce expression of the R2/F2 pyocin region. 114 HI03 - Untreated H103 + 0.lx-MIC novobiocin HI03 + 0.3x-MIC novobiocin 3 4 Time (hrs) Figure 26: Growth curve of P. aeruginosa in various concentrations of novobiocin. 115 A. 6000 4000 5 o .-_ E s -2000 • PA0620::lux - Untreated • PA0620::lux + 0 . lx -MIC novobiocin • PA0620::lux + 0.3x-MIC novobiocin B. Time (hours) 1000 o 0 500 • PA0641 ::lux - Untreated • PA0641:: lux + 0.1 x-MIC novobiocin • PA0641::lux + 0.3x-MIC novobiocin Time (hours) Figure 27: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of novobiocin. PA0620::luxCDABE (A) and PA0641 y.luxCDABE (B) were grown for three hours in the presence of 50pg/ml and 150pg/ml novobiocin. Expression is defined as the average luminescence of three wells corrected for culture density measured at OD6oo- Results are the average of 4 independent experiments. 116 ii. Mitomycin Dose and Time Course Since ciprofloxacin is a known mutagen (Phillips, Culebras et al. 1987; Clerch, Bravo et al. 1996), the ability of this antimicrobial to induce expression of the R2/F2 pyocin region was compared to the potent DNA damaging agent mitomycin. Exposure of P. aeruginosa to O.lx- or 0.3x-MIC mitomycin (Table 5) was not found to affect the growth capabilities of the organism, at least not over the three hour time frame examined in the luminescence assay (Figure 28). Exposure of the lux transcriptional fusions to O.lx- and 0.3x-MIC mitomycin however, was found to induce extensive expression responses from PA0620::/wx and PA0641::/«x (Figure 29). These responses were considerably larger than those induced by ciprofloxacin at both the O.lx-and 0.3x-MIC concentration. For example, after 3 hour exposure to 0.3x-MIC, relative light units for PA0620::/wx were 6300 and 2300 for mitomycin and ciprofloxacin, respectively; for PA0641::/wx relative light units were 1400 and 500 for mitomycin and ciprofloxacin, respectively. The untreated response was similar for both mitomycin and ciprofloxacin for both mutants (900 and 300 units respectively for PA0620::/wx and PA0641::/«x; Figures 25 and 29). Overall, expression responses from the PA0641::/wx fusion, a mutant in the F2 pyocin, were not as large as those from PA0620::/wx, a mutant in the R2 pyocin, regardless of induction by mitomycin, ciprofloxacin or novobiocin (Figures 25, 27 and 29). Thus, expression of the R2/F2 pyocin region is responsive to DNA damage. iii. Ceftazidime Dose and Time Course To confirm that expression of the R2/F2 pyocins was in response to DNA damage as would be the result from gyrase inhibition, ceftazidime, a P-lactam antibiotic not known to inhibit DNA gyrase or cause DNA damage, was used in the luminescence assay. No effect on growth ability was seen for either O.lx- or 0.3x-MIC ceftazidime (Figure 30). No differences in expression from either PA0620::/wx or PA0641::/«x were observed with either increasing concentration of ceftazidime or over length of treatment (Figure 31). Basal expression levels were similar to those previously observed for all other agents (700 or 300 units for PA0620::/wx and PA0641::/wx, respectively). Thus, the expression changes observed for the R2/F2 pyocins are 117 10 o o vo O 1 BID O J 0.1 - • - H I 0 3 - Untreated - B - H103 + 0.1 x-MIC mitomycin - H103 + 0.3x-MIC mitomycin 3 4 Time (hrs) Figure 28: Growth curve of P. aeruginosa in various concentrations of mitomycin. 118 A. 10000 8000 • PA0620::lux - Untreated • PA0620::Iux + 0.1x-MIC mitomycin • PA0620::lux + 0.3x-MIC mitomycin 6000 2 C E = -J 4000 2000 • Time (hours) B. 2000 1500 Q O > 1000 500 • PA0641 ::lux - Untreated • PA0641 ::lux + 0.1 x-MIC mitomycin • PA0641 ::lux + 0.3x-MIC mitomycin CD Time (hours) Figure 29: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of mitomycin. PA0620:\luxCDABE (A) and ?A064\::luxCDABE (B) were grown for three hours in the presence of 0.25pg/ml and 0.75pg/ml mitomycin. Expression is defined as the average luminescence of three wells corrected for culture density measured at OD6oo- Results are the average of 4 independent experiments. 119 -H103 - Untreated -H103 + 0.lx-MIC ceftazidime •HI03 + 0.3x-MIC ceftazidime 3 4 Time (hours) Figure 30: Growth curve of P. aeruginosa in various concentrations of ceftazidime. 120 6000 -i 4000 -e S c c E 2000 • PA0620::lux - Untreated • PA0620::lux + 0.1 x-MIC ceftazidime • PA0620::lux + 0.3x-MIC ceftazidime • r i i Time (hours) B. 1000 500 • PA0641::Iux - Untreated • PA0641::lux + 0.1 x-MIC ceftazidime • PA0641 ::lux + 0.3x-MIC ceftazidime Time (hours) Figure 31: Analysis of luminescence from luxCDABE transcriptional fusions grown in the presence of various concentrations of ceftazidime. ?A0620::luxCDABE (A) and PA0641 v.luxCDABE (B) were grown for three hours in the presence of O.lpg/ml and 0.3pg/ml ceftazidime. Expression is defined as the average luminescence of three wells corrected for culture density measured at OD6oo- Results are the average of 4 independent experiments. 121 in response to DNA damaging agents like mitomycin, ciprofloxacin and novobiocin and not antimicrobial treatment in general.-E. Electron Microscope Analysis of Pyocin/Phage Release Increases in transcript abundance for the R2/F2 pyocins, is not necessarily indicative of increased pyocin proteins. Studies have shown that mRNA expression levels do not correlate with protein expression even in bacteria (Gmuender, Kuratli et al. 2001). Isolation and purification of the R2/F2 pyocins was completed on the basis of their structural similarity to phage tails (Yamamoto, Alberts et al. 1970; Nakayama, Takashima et al. 2000). While spontaneous pyocin production occurs in P. aeruginosa at a low level (Jacob 1954; Kageyama 1964), transmission electron microscopy of untreated and 0.3x-MIC ciprofloxacin treated P. aeruginosa clearly differentiated the presence and increase of phage tail proteins in the extracts from sub-inhibitory ciprofloxacin treated P. aeruginosa cells (Figure 32A and B). These particles were absent from PA0620::/ux cultures either untreated or treated with ciprofloxacin (Figure 32C and D), indicating that exposure to sub-inhibitory ciprofloxacin was both inducing transcriptional and translational expression changes. Long filamentous-like structures were also visible in both the P. aeruginosa and PA0620::/ux cultures. F. Determination of the Minimum Inhibitory Concentrations for Pyocin/Phage Mutants The striking expression changes in the R2/F2 pyocin region following exposure to ciprofloxacin and other DNA damaging agents but not other antimicrobials, indicated that this region possibly played an important role in some response of the organism to these environmental challenges. These findings therefore were followed up with further susceptibility analyses of various mutants in this region (Table 3). MIC values to ciprofloxacin, as well as other fluoroquinolones, antimicrobials and DNA damaging agents were re-evaluated for all mutants in this region. Interestingly, all mutants in the R2/F2 pyocin region exhibited at least an 8-fold higher MIC for all fluoroquinolones but not other antimicrobials as compared to wild type P. aeruginosa strain HI 03 (Table 5). Consistent with this trend, mutation of the repressor protein PrtR (prtRy.lSlacZ) had no effect on the ciprofloxacin MIC (Table 5), since this strain would still 122 U n t r e a t e d + 0 .3x -MIC c i p r o f l o x a c i n o o < ON D Figure 32: Electron micrographs of supernatants from untreated and 0.3x-MIC ciprofloxacin treated P. aeruginosa strain H103 and PA0620::/«* cells. (A) Untreated strain HI03 cells; (B) 0.3x-MIC ciprofloxacin treated strain HI03 cells; (C) Untreated strain ?A0620::luxCDABE cells; and (D) 0.3x-MIC ciprofloxacin treated ?A0620::luxCDABE cells showing absence of R-type pyocin tail structures. Structures in A and C are likely F-type pyocins or flagella. Arrows point to the presence of pyocin/phage tail structures. 123 be capable of producing viable pyocin/phage. Mutation of the regulatory component recA::lSlacZ resulted in resistance to fluoroquinolones (Table 5), data which is also consistent with the regulatory control of R2/F2 pyocin expression (Matsui, Sano et al. 1993). Interestingly, mutation of the S-type pyocin encoded by PA3866 did not result in a similar resistance profile as for mutation of the R2/F2 pyocin region (Table 5). These findings then suggest that induction of the R2/F2 pyocin region but not other pyocin regions, contribute to the fluoroquinolone susceptibility of P. aeruginosa at the MIC. In parallel to these MIC trends for ciprofloxacin, the pyocin/phage and recAy.ISlacZ mutants exhibited increased MICs to the DNA damaging agent mitomycin (Table 10). MIC values to novobiocin were more variable but also tended to be increased; MICs were unchanged for ceftazidime, a P-lactam with no known gyrase inhibition or DNA damaging effects (Table 10). Together, these results indicate that the R2/F2 pyocin region not only contributes to the fluoroquinolone susceptibility of P. aeruginosa, but to the general susceptibility of the organism to DNA damage. G. Time-Kill Assays for Pyocin/Phage Mutants If the R2/F2 pyocin region were important in mediating the adaptive resistance response to sub-inhibitory ciprofloxacin through its role as a fluoroquinolone or DNA damage susceptibility determinant, it would be predicted that mutants in the R2/F2 pyocin region would not be capable of eliciting the same adaptive resistance response as previously observed for wild type P. aeruginosa. To test this hypothesis the PA0620::/wx transcriptional fusion mutant was used in the adaptive resistance assay and compared to wild type P. aeruginosa strain HI03. The concentrations of ciprofloxacin used in the assay were calculated based on the strain respective MIC (Table 5). As previously demonstrated (Figure 20), wild type P. aeruginosa cultures exposed to 0.3x-MIC ciprofloxacin, exhibit enhanced survival ability compared to untreated cultures (Figure 33). Interestingly, both the untreated and sub-inhibitory ciprofloxacin treated mutant cultures grew better than the 0.3x-MIC ciprofloxacin treated P. aeruginosa HI03 culture in supra-inhibitory ciprofloxacin, even recovering growth ability after initial killing in supra-inhibitory ciprofloxacin (Figure 33). This enhanced survival was not attributable to any significant growth advantage of the PA0620::/wx mutant over wild type P. aeruginosa (Figure 124 Table 10: Minimum inhibitory concentrations of various antimicrobials against various strains of P. aeruginosa. Cultures were grown in LB medium and represent the mode of four independent experiments. Novo, novobiocin; Ceft, ceftazidime; Mito, mitomycin; Cipro, ciprofloxacin. MIC (pg/ml) Strain Novo Ceft Mito Cipro H103 500 1 2.5 0.06 prtRy.lSlacZ 500 1 1.25 0.06 PA0620::/wx >1600 2 20 0.5 PA0621:: IS/acZ 250 0.5 20 0.5 PA0641::/ux >1600 2 20 0.5 recAv.lSlacZ 250 1 10 0.5 125 1.00E+08 1.00E+03 -i 1 . 1 1 1 1 1 1 0 20 40 60 80 100 120 160 Time (minutes) Figure 33: Survival ability of P. aeruginosa strain H103 and PA0620::/«JC in 2x-MIC ciprofloxacin following growth in 0.3x-MIC ciprofloxacin. Cultures of P. aeruginosa strain H103 and PA0620::/wx were grown in the presence or absence of 0.3x-MIC ciprofloxacin until mid-logarithmic phase, washed and then exposed to 2x-MIC ciprofloxacin. Concentrations of ciprofloxacin were calculated based on the respective strain's MIC. Colony forming units were counted every 20min for >2hours. Data representative of three separate experiments are shown. 126 34). There was no appreciable difference in growth between untreated and 0.3x-MIC ciprofloxacin treated mutant cultures, indicating that the mutants no longer had the capacity to exhibit differential survival or adaptive resistance. Thus, the R2/F2 pyocin region appears to play a role in mediating the adaptive resistance response of P. aeruginosa to ciprofloxacin. Absence of R2/F2 pyocin expression instead appears to allow the organism to eventually recover growth in supra-inhibitory ciprofloxacin. H. Serial Selection of Pyocin/Phage Loss with Sub-inhibitory Ciprofloxacin Up-regulation of the R2/F2 pyocin region was observed following exposure to sub-inhibitory ciprofloxacin, conditions also found to induce an adaptive resistance-like response in P. aeruginosa. Yet strains lacking expression from the R2/F2 pyocin region were found not to exhibit adaptive resistance-like responses following exposure to sub-inhibitory ciprofloxacin. Furthermore, the R2/F2 pyocin region was found to be involved in fluoroquinolone and DNA damage susceptibility; its absence conferred fluoroquinolone and mitomycin resistance. These findings are disparate; up-regulation of the R2/F2 pyocins results in resistance in one case, while absence of expression results in resistance in another. If expression of the R2/F2 pyocin region does contribute to the antimicrobial susceptibility of P. aeruginosa, then ubiquitous absence of this operon would be expected. Interestingly, the pyocin/phage operon has been reported to be absent in the vast majority of isolates of P. aeruginosa, including a wide variety of clinical and environmental isolates (Ernst, D'Argenio et al. 2003; Wolfgang, Kulasekara et al. 2003). Studies of clinical isolates have also shown progressive loss of pyocin production and inhibition of phage replication (Holloway, Rossiter et al. 1973; Romling, Fiedler er al. 1994). Given that clinical and environmental isolates of P. aeruginosa often lack the R2/F2 pyocin region and that similar mutant strains used herein were found to be more resistant to fluoroquinolones, it was hypothesized that exposure to sub-inhibitory ciprofloxacin induces expression of the R2/F2 pyocins, which in turn select for the loss of this region from the genome and thus increased fluoroquinolone resistance. There is precedence for colicin, the E. coli equivalent of pyocins, involvement in such microbial dominance. After seven days of co-culture, flasks initially containing a mixed population of colicin-sensitive, -producing and -resistant 127 Time (hours) Figure 34: Growth curve of P. aeruginosa strain H103 and PA0620::/«x mutant. 128 strains were only found to contain the colicin-resistant strain (Kerr, Riley et al. 2002; Kirkup and Riley 2004). Although not examined on a mutational level, these findings at least demonstrate a selection pressure in favor of colicin-resistant strains. To test this hypothesis, P. aeruginosa strains were serially cultured for five days in 0.3x-MIC ciprofloxacin and then plated on lOx-MIC ciprofloxacin. Only untreated wild type P. aeruginosa strain HI03 was found to not grow on lOx-MIC ciprofloxacin plates; all other conditions predictably produced colonies on lOx-MIC plates (data not shown). Colony PCR of the resulting colonies found that there was ubiquitous presence of both the R2 and F2 pyocins within the genomes of all P. aeruginosa strains regardless of sub-inhibitory ciprofloxacin exposure condition (data not shown). It was anticipated however that either only one or none of the pyocin regions would be found in the genome of sub-inhibitory ciprofloxacin treated wild type P. aeruginosa strain HI 03 colonies. I. Summary While the overall response of P. aeruginosa to sub-inhibitory ciprofloxacin appears to be a collection of subtle expression changes, expression of the entire R2/F2 pyocin region was dramatically up-regulated in response to 0.3x- and lx-MIC ciprofloxacin. This striking result was confirmed by real-time PCR, as well as expression analysis from lux transcriptional fusions to ORFs in this region cultured under similar sub-inhibitory ciprofloxacin conditions. The role of the R2/F2 pyocin region in the response of P. aeruginosa to ciprofloxacin however is unknown. In an attempt to better understand the R2/F2 pyocin expression response to sub-inhibitory ciprofloxacin, a series of experiments was conducted on various lux transcriptional fusions to genes within this region. Firstly, the responsiveness of the R2/F2 pyocin region to different classes of gyrase inhibitors was compared to general DNA damaging agents and antimicrobials not known to either inhibit DNA gyrase or cause DNA damage,,to ascertain the specificity of this response. Analysis of the transcriptional response over a time and dose course of treatment revealed that the expression responses of the R2/F2 pyocin region were related to DNA damage and not antimicrobial challenge in general. This finding confirms previous work showing pyocin induction by mutagenic agents (Jacob 1954; Kageyama 1964). The analysis also showed that the extent of DNA damage induced by the gyrase inhibitors was much less than that observed for 129 mitomycin. Expression responses were also found to increase with both time and concentration of the drug. Differences in expression level were observed however between the two transcriptional fusions analyzed. This difference in expression level between PA0620::/ux and PA0641::/wx likely reflects the genetic organization and regulation of the R2/F2 pyocin region. Induction of RecA cleavage of PrtR by DNA damage, releases the repression of PrtR on prtN expression (Matsui, Sano et al. 1993). Binding of PrtN to the P-box motif located upstream of PA0613 and PA0633 (according to Nakayama et al. (Nakayama, Takashima et al. 2000) or within PA0632 according to analyses conducted herein) promotes expression of the corresponding R2 and F2 pyocin operons. The consensus sequence for the P-box motif (ATTGnn(n)GTnn(n)) is known to be repeated twice for the R2 pyocin and four times for the S-type pyocins (Matsui, Sano et al. 1993; Sano, Matsui et al. 1993). However, this consensus sequence is repeated only once for the F2 pyocin (personal communication with W. Hsiao and F. Brinkman, SFU) and is located within the sequence for PA0632. Since the P-box is responsible for the activation of the promoters of the pyocin genes by the prtN gene, but is not crucial for the promoter activity itself (Matsui, Sano et al. 1993), the P-box motif within PA0632 may either not be effective at promoter activation or may not be a strong binding motif compared to the multiplicity of sequences seen for the S-type or R2 pyocins. Hence the difference in P-box motif between R2 and F2 pyocins may contribute to the differential expression of the R2 and F2 pyocins. It is also possible that the lack of substantial expression changes in the lux mutants, particularly the PA0641::/wx mutant, reflects use of the incorrect ciprofloxacin concentration, since the concentration used to induce expression in the lux mutant strains in this assay were relevant to PAO-H103, rather than the mutant concentrations computed later to be much higher than PAO-H103. Because transcriptional changes do not always correlate with changes at the protein level, pyocin expression and production was confirmed by TEM. Comparison of the supernatants from wild type P. aeruginosa and the pyocin mutant PA0620::/«x clearly demonstrated induction of pyocin expression by exposure of wild type P. aeruginosa to sub-inhibitory ciprofloxacin. Similar induction responses in the pyocin mutant were not observed, indicating that the transcriptional response of P. aeruginosa to sub-inhibitory ciprofloxacin was being translated into changes at the protein level. The long filamentous fibres also noted by TEM are likely the F2 pyocin, since expression of this pyocin would not be affected by transposon insertion into 130 PA0620, a Pv2 pyocin gene, and since spontaneous pyocin production does occur at a low level (Jacob 1954; Kageyama 1964). These structures may also represent flagella cleaved off from the bacterial surface, although this is not likely given the isolation and purification procedure (Yamamoto, Alberts et al. 1970). Since sub-inhibitory ciprofloxacin exposure to the mid-logarithmic phase induced both expression changes in the R2/F2 pyocin region and was found to elicit enhanced growth in the adaptive resistance assay, a possible role of the R2/F2 pyocins in the adaptive resistance response was investigated further. Following treatment with sub-inhibitory ciprofloxacin, the ability of pyocin mutants to survive in supra-inhibitory ciprofloxacin was compared to wild type P. aeruginosa. Interestingly, it was found that the mutant strain grew better than wild type P. aeruginosa in supra-inhibitory ciprofloxacin, the PA0620::/wx mutant exhibiting a delay before resuming growth. Thus, while exposure to sub-inhibitory ciprofloxacin no longer produced differences in survival to supra-inhibitory ciprofloxacin in the pretreated versus untreated mutants, the pyocin/phage mutant was better able to cope with and resist the supra-inhibitory ciprofloxacin challenge, suggesting a novel role for the R2/F2 pyocin region in mediating ciprofloxacin susceptibility/resistance in P. aeruginosa. Whether secondary mutations, such as those likely to occur through ciprofloxacin induction of SOS dependent mutagenesis (Power and Phillips 1993), contributed to the eventual ability of the mutant strains to overcome the supra-inhibitory ciprofloxacin condition was not determined. Given that ciprofloxacin in the previous chapter was demonstrated to induce the SOS response, it seems that the SOS response either through induction of the R2/F2 pyocin region or through its own mutagenic capabilities, plays an important role in the response of P. aeruginosa to ciprofloxacin. The pyocin transcriptional mutants were also evaluated with respect to their antimicrobial susceptibility pattern. Surprisingly, all of the mutants examined displayed 8-fold or higher MICs to fluoroquinolones but not to other classes of antibiotics; the susceptibility patterns of mutants in the R2/F2 pyocin regulatory elements were consistent with these findings. Moreover, the pyocin/phage mutants also displayed higher MICs to mitomycin and novobiocin, other DNA damaging agents. Together, these results indicate a novel role for the R2/F2 pyocin region in the fluoroquinolone and/or DNA damage susceptibility of P. aeruginosa. But these results also present a contradiction in roles: the R2/F2 pyocins seem to be important in mediating both adaptive resistance and susceptibility to fluoroquinolones. Recent 131 studies on P. aeruginosa offer some insights into how these two roles may be related. It has been shown that the genomes of various clinical and environmental isolates of P. aeruginosa do not always contain the R2/F2 pyocin region (Ernst, D'Argenio et al. 2003; Wolfgang, Kulasekara et al. 2003). Other work has shown a role for colicins in promoting microbial diversity (Kerr, Riley et al. 2002; Kirkup and Riley 2004). While this work does not demonstrate colicin selection of mutations that produce colicin-resistant strains, it does demonstrate a selection pressure in favor of colicin-resistant strains, a selection pressure similar to that of sub-inhibitory ciprofloxacin. In the adaptive resistance assay, the pyocin mutant strains were able to grow better than wild type P. aeruginosa in supra-inhibitory ciprofloxacin (calculated from the respective strain MIC), findings which support this selection pressure theory, since pyocin mutants are able to outcompete pyocin producing strains on growth advantage. Examination of the ability of serial exposures to sub-inhibitory ciprofloxacin to select for loss of the R2/F2 pyocin region from the genome and thus for increased resistance, however, was unsuccessful. However, other options resulting in diminished pyocin expression after serial selection are also possible, such as mutations in the regulatory region affecting expression of the pyocin region. It is also possible that the ability of the strains to grow on lOx-MIC ciprofloxacin plates was attributable to mutations leading to other resistance mechanisms, such as mutations in either gyrase or the regulatory elements of multidrug efflux pumps, mutations which are also known to confer fluoroquinolone resistance. Whether the selection methodology was inappropriate or there were compensatory secondary resistance mechanisms or other mechanisms such as regulatory mutations leading to reduced pyocin expression, remains to be determined. 132 DISCUSSION That antibiotic challenge eventually leads to the development of resistance is well established. Studies of bacterial resistance mechanisms however, tend to focus on either intrinsic or acquired resistance to the exclusion of adaptive resistance. Yet adaptive resistance is also clinically relevant, having been observed for both aminoglycoside and quinolone antimicrobials (Chamberland, Malouin et al. 1990; Gould, Milne et al. 1991; Barclay, Begg et al. 1996). Defined as an unstable, reversible resistance that is unrelated to genetic mutation (Barclay and Begg 2001), it occurs transiently in an organism under non-lethal selective pressure. The molecular mechanisms responsible for the development of adaptive resistance responses to aminoglycosides and quinolones however, are poorly understood at best. The aim of this project was to address these knowledge deficits for the quinolone ciprofloxacin from a global perspective using DNA microarray technology to P. aeruginosa. A. Sub-inhibitory ciprofloxacin induces adaptive resistance in P. aeruginosa While previous work on quinolone induced adaptive resistance has established that such resistance occurs following first exposure to 8x-MIC ciprofloxacin (Gould, Milne et al. 1991), achieving this concentration in the CF lung seems unrealistic given the mucoid and compartmentalized conditions of the CF lung. The ability of sub-inhibitory ciprofloxacin to elicit adaptive resistance in P. aeruginosa was therefore examined herein, since these concentrations are more likely to accurately reflect drug levels in the CF lung. Sub-inhibitory ciprofloxacin (0.3x-MIC) was found to induce one logio difference in survival compared to untreated cultures challenged with supra-inhibitory ciprofloxacin. In addition to the findings of Gould et al. (Gould, Milne et al. 1991), this work shows that a full range of ciprofloxacin concentrations induce adaptive resistance-like mechanisms in P. aeruginosa. Sub-inhibitory first exposure concentrations were furthermore found to induce prolonged adaptive resistance responses. Taken together with previous work, these findings underscore the importance of ciprofloxacin at any concentration in eliciting extended adaptive resistance responses, and stress the need to better understand this type of resistance. This has obvious and important implications in the clinic and in the design of dosing regimes, since exposure of the pathogen to any concentration of 133 ciprofloxacin can induce mechanisms which allow the organism to survive subsequent ciprofloxacin challenges. Such resistance has already been observed in the clinic; strains of P. aeruginosa isolated from a patient receiving oral ciprofloxacin therapy exhibited transient resistance to ciprofloxacin as measured by MIC (Chamberland, Malouin et al. 1990). B. Sub-inhibitory ciprofloxacin induces expression of R2/F2 pyocins Although the importance of adaptive resistance in the clinic and dosing regimes is appreciated, it is unclear how exposure to ciprofloxacin elicits adaptive resistance. To date, no research has sufficiently addressed the molecular mechanisms underlying this type of resistance. Given the transient nature of adaptive resistance (Barclay, Begg et al. 1996), transcriptomes of P. aeruginosa were analyzed by custom DNA microarray analysis in an attempt to better define adaptive resistance at a molecular yet global level. Exposure of P. aeruginosa to sub-inhibitory and inhibitory concentrations of ciprofloxacin to the mid-logarithmic phase was found to induce numerous expression changes. Overall, many of the expression changes were fairly subtle, the majority of significant expression changes falling below 2-fold. Included in this category were many of the genes expected to change following treatment with ciprofloxacin, including the target DNA gyrase and various multidrug efflux pumps. Expression changes in these genes were expected since mutations in these genes or their regulatory elements confer resistance to ciprofloxacin (Li, Nikaido et al. 1995; Poole, Gotoh et al. 1996; Kohler, Epp et al. 1999; Mouneimne, Robert et al. 1999; Le Thomas, Couetdic et al. 2001). Absence of large expression changes in these known ciprofloxacin resistance genes may be related to the effective concentration within the cell. The intrinsically low outer membrane permeability of P. aeruginosa serves to decrease the rate of uptake of antibiotics into the cell, thereby allowing secondary resistance mechanisms like antibiotic efflux or degradation enzymes, to work effectively in eliminating the antibiotic from within the cell (Hancock and Speert 2000). Thus, the concentration of ciprofloxacin available within the cell may be considerably smaller than the original dose, to the extent that it has no discernible effect on the organism beyond a myriad of subtle expression changes. Certainly as the concentration of ciprofloxacin was increased, the extent and number of expression changes was found to increase correspondingly, 134 with SOS response genes figuring more prominently than resistance genes with increasing concentration (Appendix III). Absence of known ciprofloxacin resistance genes may also be a result of the concentrations of ciprofloxacin studied, since concentration affects which resistance mechanism is selected by P. aeruginosa in response to quinolone exposure. At concentrations close to the MIC, efflux type mechanisms were selected almost exclusively in the lab; gyrase type mutations appeared only at concentrations above 4x-MIC (Kohler and Pechere 2001). Since only the MIC and not higher concentrations of ciprofloxacin were examined, absence of resistance genes is also likely related to the dose. Absence of large expression changes in the known ciprofloxacin resistance genes may also be reflective of the early time point examined. Resistance responses may require longer exposure times than the 2.5 hours examined and may follow induction of SOS responses. Similar absentia results were observed by Gmuender et al. in their evaluation of gene expression changes triggered by H. influenzae in response to ciprofloxacin (Gmuender, Kuratli et al. 2001). lx-MIC ciprofloxacin conditions were likewise not found to induce expression changes in either the target or other known resistance mechanisms following 1 hour of exposure. Only when lOx-MIC ciprofloxacin conditions were evaluated did expression of the DNA gyrase A and B subunits change, and even then only by <2.0 fold after 1 hour of exposure. SOS response genes however were rapidly induced following 30min exposure to inhibitory ciprofloxacin (Gmuender, Kuratli et al. 2001). These results support the notion that P. aeruginosa responses to ciprofloxacin are time sensitive and hierarchal with respect to SOS and resistance. Since higher doses of ciprofloxacin are also known to elicit adaptive resistance responses (Gould, Milne et al. 1991), it may be argued that such concentrations should have been examined instead. But in addition to the biological interest of low dose ciprofloxacin and early time points with respect to adaptive resistance, transcriptome analysis of sub-inhibitory ciprofloxacin treated cells at mid-log reduced the amount of complicating microarray data resulting from inhibition of secondary targets within the cell and downstream effects resulting from inhibition of the primary target (Shaw and Morrow 2003). Of the significantly differentially expressed ORFs, no molecular mechanism of adaptive resistance was apparent. Outside of the aforementioned known quinolone resistance genes, not even general resistance genes were observed to exhibit noteworthy expression changes. Again, similar absentia results were noted by others (Gmuender, Kuratli et al. 2001). The lack of an 135 apparent molecular mechanism underlying adaptive resistance is further compounded by the observation that most ORFs which did exhibit large fold changes were classified as either 'hypothetical protein' or 'putative enzyme' (Table 6 and Appendix III). Thus, many of the genes likely to be important in mediating the adaptive resistance response in P. aeruginosa, as indicated by their large expression response, remain to be fully characterized. Certainly the study of these ORFs offers unprecedented potential for discovering novel resistance mechanisms and drug targets. Nonetheless, some interesting results were identified through the microarray studies. Striking expression changes were observed for all genes between PA0613 and PA0648 (Table 8), a region of the genome which encodes the R2/F2 pyocins (Nakayama, Takashima et al. 2000). As the concentration of ciprofloxacin was increased, the number of ORFs within this region and their fold changes also increased. Based on the mutagenic nature of ciprofloxacin (Phillips 1987; Debnath, de Compadre et al. 1992; Fort 1992), expression changes in this region and in other pyocins was expected since regulatory elements of the pyocins are known to be responsive to cellular stress through the SOS response (Matsui, Sano et al. 1993). The correlation between pyocin expression and the adaptive resistance response of P. aeruginosa to sub-inhibitory ciprofloxacin however was not obvious. The dramatic up-regulation of R2/F2 pyocins by sub-inhibitory ciprofloxacin suggested at least, that the mechanism of action of ciprofloxacin may not be as well defined as previously thought. Downstream effects such as up-regulation of R2/F2 pyocin expression resulting from inhibition of the primary DNA gyrase target are apparent. Noteworthy expression changes were also observed for genes involved in the SOS response of the cell to ciprofloxacin induced DNA damage. Induction of the SOS response itself may also play a role in mediating the adaptive resistance response to ciprofloxacin, since induction of the SOS response is known to have two consequences (Drlica and Zhao 1997). One is quinolone induced mutagenesis which could be very important if it contributes to the acquisition of quinolone resistance mutations (Drlica and Zhao 1997). Given the nature of adaptive resistance however, such mutagenesis would result in stable resistance rather than transient adaptive resistance. The other consequence is enhanced survival in the presence of fluoroquinolones, since mutations that block SOS induction have been shown to increase the susceptibility of cells to fluoroquinolones (Howard, Pinney et al. 1993). Thus, ciprofloxacin induction of the SOS 136 system may promote enhanced survival of P. aeruginosa in supra-inhibitory ciprofloxacin conditions (Figure 35). Overall, the microarray findings presented here provide a snapshot signature of the expression response of P. aeruginosa at the mid-logarithmic phase to sub-inhibitory and inhibitory ciprofloxacin. Understanding the organism's physiological changes on a global level can assist in defining the mechanism of resistance (Shaw and Morrow 2003). As well, understanding the organism's responses can help better define the mechanism of action of an inhibitor. As seen in this study for example, observations of changes to non-target genes or pathways, as noted above for the R2/F2 pyocins, can yield information that might not have been discovered by single gene molecular approaches. C. R2/F2 pyocin induction is related to DNA damage If activating expression of pyocins is a novel component in the mechanism of action for ciprofloxacin, this ability was hypothesized to be attributable to ciprofloxacin's mutagenic activity rather than secondary target activity, since the prtR and prtN pyocin regulatory elements are known to be responsive to the SOS system (Matsui, Sano et al. 1993). Indeed, transcriptional fusions to genes in the R2/F2 pyocin region showed that expression was inducible by both the mutagenic activity of novobiocin, another class of DNA gyrase inhibitors acting on the gyrase B subunit, and mitomycin, a potent DNA damaging agent. No induction of R2/F2 pyocin expression was observed for ceftazidime, a P-lactam with no known gyrase activity or DNA damage activity. Thus, pyocin expression was shown to be sensitive to the DNA damage resultant from inhibition of DNA gyrase, and not a secondary target of ciprofloxacin. A connection between the DNA damage induced SOS response and R2/F2 pyocin expression though, suggested that such expression changes in response to sub-inhibitory ciprofloxacin may be related to the organism's overall stress response and not just DNA damage stress. This however does not seem likely since other P. aeruginosa microarray experiments examining various other stress inducing conditions like treatment with imipenem (Bagge, Schuster et al. 2004), did not find similar changes in the R2/F2 pyocin region. Expression changes in the R2/F2 pyocin region thus seem to be directly related to the response of P. aeruginosa to ciprofloxacin and its DNA damaging effects. 137 '5K^v D N A + D N A g y r a s e Quinolone (Ciprofloxacin) I rf^-k dsDNA Breaks 1 Activation of RecA I s SOS Response I. Quinolone Induced Mutagenesis -> Acquire quinolone resistance mutations 2. recA dependent recombination & SOS induction -> Enhance survival in quinolones ^ R 2 / F 2 Pyocins ^Mutation events -> Alter genetic control (prtRN) to ^/R2/¥2 -> QuinoloneR 1s SOS -> Transient mutators -> Persister Cells -> Quinolone tolerance A D A P T I V E R E S I S T A N C E -> Enhanced survival in Ciprofloxacin Sustained Challenge 9 • 1sSOS -> f Mutators alleles (i.e. ^mutS) -> 1s mutation rate i^SOS -> /hR2/F2 -> Colicin/pyocin selection of R2/F2 resistant strains M U T A T I O N A L R E S I S T A N C E Figure 35: Potential relationships between ciprofloxacin and the development of resistance. 138 D. R2/F2 pyocins have a role in adaptive resistance to sub-inhibitory ciprofloxacin Whether the expression changes in the R2/F2 pyocin region played a role in mediating the adaptive resistance response seen upon subsequent exposure of cultures to supra-inhibitory ciprofloxacin however was not yet clear. Mutants in the R2/F2 pyocin region were thus re-evaluated in the adaptive resistance assay. Care was taken to calculate the sub-inhibitory and supra-inhibitory concentrations of ciprofloxacin appropriately using the strain specific MIC values. Mutants exposed to sub-inhibitory ciprofloxacin, however, were no longer found to exhibit a difference in survival between sub-MIC ciprofloxacin treated and untreated cultures upon exposure to supra-inhibitory ciprofloxacin, implying a potential role for the R2/F2 pyocin in adaptive resistance. Interestingly, growth of the mutants in supra-inhibitory ciprofloxacin regardless of the pretreatment condition was found to be quicker than either of the wild type strain conditions, and to eventually even overcome the supra-inhibitory condition. This difference was not attributable to any growth advantage afforded by an increased ciprofloxacin resistance profile, since supra-inhibitory concentrations for each strain were calculated from the strain specific MIC. Instead, the growth advantage is likely due to the improved fitness of mutants, which no longer have to contend with lytic events related to the release of pyocins. While the transcriptional units of the R2 and F2 pyocins have not been fully elucidated, based on the location of the P-box motifs required for prtN activation of pyocin expression upstream of PA0613 (R2 pyocin) and within PA0632 (F2 pyocin) (Matsui, Sano et al. 1993; Nakayama, Takashima et al. 2000), disruption of PA0620 would introduce a polar mutation in the operon. Since there appears to be only one lytic system responsible for release of all pyocins in P. aeruginosa encoded by PA0614 and PA0629 (Nakayama, Takashima et al. 2000), such a polar mutation would render the lytic system incomplete, and would prevent the cell lysis associated with pyocin release. Thus without cell lysis, mutants in the R2/F2 pyocin region would exhibit a growth advantage over wild type strains containing both the pyocin and functional lytic system. Whether secondary mutations contributed to the ability of the mutant strains to eventually overcome the supra-inhibitory ciprofloxacin condition was not determined. How expression of the R2/F2 pyocin region would contribute to the adaptive resistance response of P. aeruginosa is not obvious. Expression of the R2/F2 pyocin region and the 139 resulting cell lysis and pyocin release would reduce the overall cell population, rather than enhance it or allow for its survival. Thus, expression of the R2/F2 pyocin region would not be expected to result in an adaptive resistance response as was observed. While the adaptive resistance response was observed for wild type P. aeruginosa producing pyocin, and not for pyocin mutant strains, the enhanced survival of wild type cells in supra-inhibitory ciprofloxacin may be explained if expression of the R2/F2 pyocin region were down-regulated or if a genetic mutation in wild type P. aeruginosa resulting in either a pyocin or lytic mutant strain had occurred. Such an event would be predicted to yield growth advantages parallel to those seen for the pyocin mutant strain in supra-inhibitory ciprofloxacin conditions. Bacteria possess mechanisms and strategies for responding to the constant changes in their environment. These include phenotypic acclimation, by which an individual organism modifies some aspect of its behavior, morphology or metabolism in response to environmental change, and genetic adaptation, whereby the genetic composition of a population may change as a result of natural selection (Moxon, Rainey et al. 1994). Acclimation is achieved by such mechanisms as feedback loops and two-component sensory systems, which ultimately regulate gene expression in response to some environmental change. When confronted with a persisting unfavorable environment for which classical regulation of gene expression is inadequate however, bacteria must adapt genetically by natural selection or face extinction (Moxon, Rainey etal 1994). It is important to note that increased expression of the R2/F2 pyocins was measured with microarrays at the mid-logarithmic phase, prior to supra-inhibitory ciprofloxacin exposure. R2/F2 pyocin expression was not measured during the adaptive resistance assay (i.e. at 90min), thus proof of decreased expression of the R2/F2 pyocins resulting from such events is not available. However, extension of the transcriptional fusion exposure assays showed that expression of the R2/F2 pyocins decreased after 4 hours of exposure to 0.3x-MIC ciprofloxacin (data not shown). This concentration did not substantially alter the growth ability of the organism even after 4 hours, and was the same concentration for which adaptive resistance was observed. These observations support the theory that down-regulation of the initial R2/F2 pyocin gene expression during prolonged exposure to ciprofloxacin could be important in the adaptive resistance assay. 140 Outside of gene regulation of R2/F2 pyocin expression, mutation events in the R2/F2 pyocins are not only possible but prevalent under selective conditions such as antimicrobial environments. Recent work has pointed to the existence of transient mutators, whose mutator alleles revert to normal mutator status through reversion (Oliver, Baquero et al. 2002) or recombination (Brown, LeClerc et al. 2001). Such transient mutators are thought to exist within persister populations (Chopra, O'Neill et al. 2003). Persister populations exhibit tolerance, meaning that cells do not grow in the presence of antibiotics but do not die either (Keren, Kaldalu et al. 2004). In conditions where P. aeruginosa is non-dividing or growing slowly, such elevated mutation frequencies have been observed (Alonso, Campanario et al. 1999). These growth conditions are similar to the adaptive resistance assay, since exposure to supra-inhibitory ciprofloxacin would not be conducive to growth. Moreover, it has been suggested that fluoroquinolones, like ciprofloxacin, likely create transient mutator phenotypes through induction of the SOS response (Livermore 2003), a response prominent in the cultures studied herein. Furthermore, Fung-Tome et al. have demonstrated that sub-inhibitory ciprofloxacin exposure increases the mutation rate and thus resistance of P. aeruginosa (Fung-Tome, Kolek et al. 1993). The microarray findings herein corroborate this data and show diminished expression of mutS with increasing ciprofloxacin concentration, a gene whose inactivation leads to a mutator phenotype (Oliver, Baquero et al. 2002). Other researchers have also shown that low concentrations of antibiotics, as in sub-inhibitory ciprofloxacin, contribute to the problem of elevated mutation frequencies by selecting mutator alleles (Giraud, Matic et al. 2002; Negri, Morosini et al. 2002). Taken together, the observed adaptive resistance response of 0.3x-MIC ciprofloxacin pretreated cultures may be attributable to a transient hypermutator state affording mutation of wild type P. aeruginosa producing the R2/F2 pyocins into a pyocin mutant strain (Figure 35). E. R2/F2 pyocin region is a fluoroquinolone susceptibility determinant Whatever the mechanism underlying decreased expression of the R2/F2 pyocins, there seemed to be no apparent logic for decreased expression, beyond the adaptive resistance response of a moderately improved growth profile or eventual survival in supra-inhibitory ciprofloxacin. Re-evaluation of the susceptibility profiles of the R2/F2 pyocin mutants however 141 revealed some very interesting and novel findings in this regard. Al l of the R27F2 pyocin mutants exhibited at least an 8-fold increase in fluoroquinolone resistance; MIC values to various other antimicrobials were unchanged. Corresponding to the known regulation of R2/F2 pyocin expression (Matsui, Sano et al. 1993), a recA mutant but not a mutant in the negative regulator prtR, exhibited similar altered susceptibility profiles. Similar observations with respect to fluoroquinolone MIC for recA have been noted by others (Walters, Piddock et al. 1989). Furthermore, similar susceptibility trends were not seen for mutants in other pyocins like the S-type pyocins. The R2/F2 pyocin mutants also exhibited increases in mitomycin and novobiocin resistance, other DNA damaging agents. Together these findings suggest that the R2/F2 pyocin region plays a role in mediating fluoroquinolone and/or DNA damage susceptibility at the MIC in P. aeruginosa. Furthermore, these findings indicate why wild type P. aeruginosa producing pyocin would want to decrease expression of the R2/F2 pyocins or convert into a non-producing phenotype. In addition to the moderate enhancement in growth profile afforded by lack of R2/F2 pyocin production, such strains of P. aeruginosa would also be resistant to the DNA damage induced by fluoroquinolones, and thus better able to tolerate growth in the face of continued ciprofloxacin/DNA damaging exposure. F. R2/F2 pyocins play a role in microbial diversity Having established the R2/F2 pyocin region as a fluoroquinolone/DNA damage susceptibility determinant in P. aeruginosa, it was hypothesized that prolonged induction of the adaptive resistance response with ciprofloxacin could lead to or select for more stable resistance types, like mutational resistance. Thus, it was hypothesized that sustained exposure to ciprofloxacin could lead to mutation events in the R2/F2 pyocin region, resulting in either loss of expression or lysis or deletion of the region itself, and conferring fluoroquinolone resistance to the organism. Certainly such transient mutator states are possible both within P. aeruginosa and with the antimicrobial challenges explored herein, as noted previously (Fung-Tome, Kolek et al. 1993; Livermore 2003). Furthermore, recent genome analyses have reported the absence of the R2/F2 pyocin region in the vast majority of clinical and environmental isolates of P. aeruginosa (Ernst, D'Argenio et al. 2003; Wolfgang, Kulasekara et al. 2003). Other studies of clinical isolates have also shown progressive loss of pyocin production and inhibition of phage replication (Holloway, 142 Rossiter et al. 1973; Romling, Fiedler et al. 1994). Together these findings support the notion that the R2/F2 pyocin region is a region easily mutated or expended from the genome under certain conditions, possibly such as ciprofloxacin selective conditions. Expression of the R2/F2 pyocins themselves may play a further role in selecting for such mutational resistance. Recent work on colicins (pyocins of E. coli) and their role in biodiversity, have identified colicins as having the ability to promote microbial diversity in static or localized environments, in a manner similar to the game rock-paper-scissors (Kerr, Riley et al. 2002; Kirkup and Riley 2004). In this model, strains that produce colicin kill sensitive strains, which outcompete resistant strains, which in turn, outcompete colicin producing strains on the basis of growth-rate advantage. However, in well-mixed or large spatial environments, the resistant strain predominates over time. Although not examined on a mutational level, these findings at least demonstrate a selection pressure in favor of colicin resistant strains based on growth advantage. Such selection pressure is akin to that of the sub-inhibitory to supra-inhibitory ciprofloxacin environment shift of the adaptive resistance assay. Furthermore, similar growth and survival advantage for non-pyocin producing strains has already been observed in the adaptive resistance assay. Thus, a sustained adaptive resistance response and/or up-regulation of R2/F2 pyocins induced by continuous ciprofloxacin challenge, may lead to mutational resistance (Figure 35). Genetic analysis of the R2/F2 pyocin region of P. aeruginosa cultures grown under successive sub-inhibitory ciprofloxacin conditions and then exposed to supra-inhibitory ciprofloxacin however, did not reveal loss of either the R2 or F2 pyocin region. Evaluation of the susceptibility profiles of various clinical and environmental isolates variably containing the R2/F2 pyocin region (Ernst, D'Argenio et al. 2003; Wolfgang, Kulasekara et al. 2003) was also inconclusive with respect to correlating absence of a pyocin region and fluoroquinolone resistance (data not shown). While unfortunate, this result is not conclusive in that other types of mutation events were not explored. For example, mutation events in either the positive regulator prtN, resulting in a dysfunctional transcriptional activator, or mutation of the negative regulator prtR, resulting in constitutive repression, could also abolish expression of the R2/F2 pyocins without affecting the genomic content. Similarly, mutation events in the lysis genes or introduction of a premature stop codon in any of the pyocin ORFs could result in lack of pyocin release or functional pyocin production, respectively. 143 Nonetheless, it is intriguing to speculate that the absence of the R2/F2 pyocin region in clinical isolates of P. aeruginosa results from selection and survival of the pyocin resistant (non-producing) strain. For an isolate from a CF patient receiving fluoroquinolone antibiotics, selection and survival of the pyocin resistant strain would be equivalent to inheriting ciprofloxacin resistance. As mentioned above, evidence of development, survival and existence of such pyocin non-producing strains has been well documented (Holloway, Rossiter et al. 1973; Romling, Fiedler et al. 1994; Ernst, D'Argenio et al. 2003; Wolfgang, Kulasekara et al. 2003). Thus, sub-inhibitory concentrations of ciprofloxacin play an important role in both the adaptive resistance response of P. aeruginosa and possibly in mediating the development of more stable fluoroquinolone resistance through expression and/or genomic changes in the fluoroquinolone susceptibility determining R2/F2 pyocin region. Sub-inhibitory concentrations of antimicrobials, in particular ciprofloxacin, then present an important therapeutic challenge not adequately addressed in current clinical settings. G. Future directions Microbial genomics has revolutionized our capacity for antimicrobial drug discovery and resistance research. Based on the wealth of genomic information, DNA microarray technology has allowed researchers to confirm the validity of existing drug targets, to identify new targets and to detect the cellular responses to treatment, yielding some very interesting findings. Using the information from these transcriptional profiling studies, we are only now beginning to understand the complexity of responses of bacterial pathogens to antimicrobials. This study and others have indicated that there is no single target or mechanism of action for a given antimicrobial and as a result, have opened up new target areas for research. Several obvious future research questions arise from the data presented in this thesis. These include: (1) identification of the expression signature for the adaptive resistance response of P. aeruginosa and comparison to the expression signatures obtained herein, (2) conclusive confirmation that the up-regulation of the R2/F2 pyocin region can lead to the loss of either R2/F2 pyocin expression or the region's genomic content altogether, and (3) investigation of the mechanism by which such loss could occur. Further microarray studies could be used for comparative analysis of the present expression profiles with those taken during the adaptive 144 resistance response, to ascertain the signature of adaptive resistance as induced by sub-inhibitory ciprofloxacin conditions. Work with various R2/F2 pyocin mutant constructs could be used to characterize the mechanisms underlying the regulatory or functional loss of the R2/F2 pyocin region. Some exciting new prospects for how this regulation or loss might be achieved come from the area of bacteriophage research. While it is known that DNA gyrase cleaves at sequence specific sites in the genome (Fisher, Mizuuchi et al. 1981), gyrase specific sites are also important for transposition of certain bacteriophage. For example, the bacteriophage Mu genome contains a centrally located strong gyrase binding site (SGS) that is required for efficient transposition of the phage (Pato, Howe et al. 1990). A consensus SGS sequence has been established, and preliminary analysis of the P. aeruginosa genome shows that this site is present in the middle of the R2/F2 pyocin region at PA0626 (data not shown). Whether similar transposition events occur in the R2/F2 pyocin region are not known, but may be possible given the homology of this region to various phage (Nakayama, Takashima et al. 2000). Such transposition events present an intriguing mechanism for loss of the R2/F2 region from the P. aeruginosa genome and require further investigation. 145 R E F E R E N C E S Abman, S. H., J. W. Ogle, et al. (1991). "Early bacteriologic, immunologic, and clinical courses of young infants with cystic fibrosis identified by neonatal screening." J Pediatr 119(2): 211-7. Ali , J. A., A. P. Jackson, et al. (1993). "The 43-kilodalton N-terminal fragment of the DNA gyrase B protein hydrolyzes ATP and binds coumarin drugs." Biochemistry 32(10): 2717-24. Aim, R. A. and J. S. Mattick (1997). "Genes involved in the biogenesis and function of type-4 fimbriae in Pseudomonas aeruginosa." Gene 192(1): 89-98. Alonso, A., E. Campanario, et al. (1999). "Emergence of multidrug-resistant mutants is increased under antibiotic selective pressure in Pseudomonas aeruginosa." Microbiology 145(Pt 10): 2857-62. Altschul, S. F., W. Gish, et al. (1990). "Basic local alignment search tool." J Mol Biol 215(3): 403-10. Ambion, I. (2004). Bacterial Whole Genome Array Analysis. www.ambion.eom/techlib/tn/114/8.html Amsterdam, D. (1991). Susceptibility testing of antimicrobials in liquid media. Antibiotics in laboratory medicine. V. Lorian. Baltimore, Md., Williams & Wilkins: 72-78. Andersen, D. H. (1938). "Cystic fibrosis of the pancreas and its relation to celiac disease: a clinical and pathologic study." Am J Pis Child 56: 344-399. Ausubel, F. M. , Roger Brent, Robert E. Kingston, David D. Moore, J.G. Seidman, John A. Smith and Kevin Struhl, Ed. (1997). Current protocols in molecular biology. New York, Greene Pub. Associates ; Wiley-Interscience. Bagge, N. , M . Schuster, et al. (2004). "Pseudomonas aeruginosa biofilms exposed to imipenem exhibit changes in global gene expression and beta-lactamase and alginate production." Antimicrob Agents Chemother 48(4): 1175-87. Ball, P. (1990). "Emergent resistance to ciprofloxacin amongst Pseudomonas aeruginosa and Staphylococcus aureus: clinical significance and therapeutic approaches." J Antimicrob Chemother 26(Suppl F): 165-79. Barclay, M. L. and E. J. Begg (2001). "Aminoglycoside adaptive resistance: importance for effective dosage regimens." Drugs 61(6): 713-21. Barclay, M . L., E. J. Begg, et al. (1992). "Adaptive resistance following single doses of gentamicin in a dynamic in vitro model." Antimicrob Agents Chemother 36(9): 1951-7. 146 Barclay, M . L., E. J. Begg, et al. (1996). "Adaptive resistance to tobramycin in Pseudomonas aeruginosa lung infection in cystic fibrosis." J Antimicrob Chemother 37(6): 1155-64. Battaglia, C , G. Salani, et al. (2000). "Analysis of DNA microarrays by non-destructive fluorescent staining using SYBR green II." Biotechniques 29(1): 78-81. Bayer, A. S., D. C. Norman, et al. (1987). "Characterization of impermeability variants of Pseudomonas aeruginosa isolated during unsuccessful therapy of experimental endocarditis." Antimicrob Agents Chemother 31(1): 70-5. Baysse, C , J. M . Meyer, et al. (1999). "Uptake of pyocin S3 occurs through the outer membrane ferripyoverdine type II receptor of Pseudomonas aeruginosa." J Bacteriol 181(12): 3849-51. Benes, V. and M . Muckenthaler (2003). "Standardization of protocols in cDNA microarray analysis." Trends Biochem Sci 28(5): 244-9. Bernstein, J. A., A. B. Khodursky, et al. (2002). "Global analysis of mRNA decay and abundance in Escherichia coli at single-gene resolution using two-color fluorescent DNA microarrays." Proc Natl Acad Sci U S A 99(15): 9697-702. Betts, J. C , A. McLaren, et al. (2003). "Signature gene expression profiles discriminate between isoniazid-, thiolactomycin-, and triclosan-treated Mycobacterium tuberculosis." Antimicrob Agents Chemother 47(9): 2903-13. Birrer, P., N . G. McElvaney, et al. (1994). "Protease-antiprotease imbalance in the lungs of children with cystic fibrosis." Am J Respir Crit Care Med 150(1): 207-13. Blanche, F., B. Cameron, et al. (1996). "Differential behaviors of Staphylococcus aureus and Escherichia coli type II DNA topoisomerases." Antimicrob Agents Chemother 40(12): 2714-20. Bonfield, T. L., J. R. Panuska, et al. (1995). "Inflammatory cytokines in cystic fibrosis lungs [published erratum appears in Am J Respir Crit Care Med 1996 Oct; 154(4 Pt l):following 12171." Am J Respir Crit Care Med 152(6 Pt 1): 2111-8. Bowtell, D. D. (1999). "Options available—from start to finish—for obtaining expression data by microarray." Nat Genet 21(1 Suppl): 25-32. Brooun, A., S. Liu, et al. (2000). "A dose-response study of antibiotic resistance in Pseudomonas aeruginosa biofilms." Antimicrob Agents Chemother 44(3): 640-6. Brown, E. W., J. E. LeClerc, et al. (2001). "Phylogenetic evidence for horizontal transfer of mutS alleles among naturally occurring Escherichia coli strains." J Bacteriol 183(5): 1631-44. 147 Bryan, R., D. Kube, et al. (1998). "Overproduction of the CFTR R domain leads to increased levels of asialoGMl and increased Pseudomonas aeruginosa binding by epithelial cells." Am J Respir Cell Mol Biol 19(2): 269-77. Burns, J. L., R. L. Gibson, et al. (2001). "Longitudinal assessment of Pseudomonas aeruginosa in young children with cystic fibrosis." J Infect Pis 183(3): 444-52. CCFF (2003). 2003 Canadian Cystic Fibrosis Foundation Annual Report. Toronto, ONT, Canadian Cystic Fibrosis Foundation: 1-12. CCFF (2003a). Facts about CF, Canadian Cystic Fibrosis Foundation. http://www.cysticfibrosis.ca/page.asp?id=21 CFF (1994). Microbiology and infectious disease in cystic fibrosis. Cystic Fibrosis Foundation: Concepts of care. Bethesda, MP, Cystic Fibrosis Foundation. V section 1: 1-26. CFF (2003). Cystic Fibrosis Foundation Patient Registry Annual Pata Report 2003. Bethesda, MP, Cystic Fibrosis Foundation. Chamberland, S., F. Malouin, et al. (1990). "Persistence of Pseudomonas aeruginosa during ciprofloxacin therapy of a cystic fibrosis patient: transient resistance to quinolones and protein F-deficiency." J Antimicrob Chemother 25(6): 995-1010. Chen, C. R., M . Malik, et al. (1996). "PNA gyrase and topoisomerase IV on the bacterial chromosome: quinolone-induced PNA cleavage." J Mol Biol 258(4): 627-37. Cheung, V. G., M . Morley, et al. (1999). "Making and reading microarrays." Nat Genet 21(1 Suppl): 15-9. Chmiel, J. F., M . Berger, et al. (2002). "The role of inflammation in the pathophysiology of CF lung disease." Clin Rev Allergy Immunol 23(1): 5-27. Chopra, I., A. J. O'Neill, et al. (2003). "The role of mutators in the emergence of antibiotic-resistant bacteria." Prug Resist Updat 6(3): 137-45. Chou, C. C , C. H. Chen, et al. (2004). "Optimization of probe length and the number of probes per gene for optimal microarray analysis of gene expression." Nucleic Acids Res 32(12): e99. Chuaqui, R. F., R. F. Bonner, et al. (2002). "Post-analysis follow-up and validation of microarray experiments." Nat Genet 32(Suppl): 509-14. Churchill, G. A. (2002). "Fundamentals of experimental design for cONA microarrays." Nat Genet 32(Suppl): 490-5. 148 Clerch, B., J. M . Bravo, et al. (1996). "Analysis of the ciprofloxacin-induced mutations in Salmonella typhimurium." Environ Mol Mutagen 27(2): 110-5. Coleman, F. T., S. Mueschenborn, et al. (2003). "Hypersusceptibility of cystic fibrosis mice to chronic Pseudomonas aeruginosa oropharyngeal colonization and lung infection." Proc Natl Acad Sci U S A 100(4): 1949-54. Costerton, J. W., P. S. Stewart, et al. (1999). "Bacterial biofilms: a common cause of persistent infections." Science 284(5418): 1318-22. Daikos, G. L., G. G. Jackson, et al. (1990). "Adaptive resistance to aminoglycoside antibiotics from first-exposure down-regulation." J Infect Pis 162(2): 414-20. Daikos, G. L., V. T. Lolans, et al. (1991). "First-exposure adaptive resistance to aminoglycoside antibiotics in vivo with meaning for optimal clinical use." Antimicrob Agents Chemother 35(1): 117-23. Dalhoff, A. (1994). "Quinolone resistance in Pseudomonas aeruginosa and Staphylococcus aureus. Development during therapy and clinical significance." Infection 22(Suppl 2): SI 11-21. Davies, J. C , M . Stern, et al. (1997). "CFTR gene transfer reduces the binding of Pseudomonas aeruginosa to cystic fibrosis respiratory epithelium." Am J Respir Cell Mol Biol 16(6): 657-63. Davies, J. E. (1997). "Origins, acquisition and dissemination of antibiotic resistance determinants." Ciba Found Symp 207: 15-27; discussion 27-35. de Bentzmann, S., P. Roger, et al. (1996). "Asialo GM1 is a receptor for Pseudomonas aeruginosa adherence to regenerating respiratory epithelial cells." Infect Immun 64(5): 1582-8. de Saizieu, A., TJ. Certa, et al. (1998). "Bacterial transcript imaging by hybridization of total RNA to oligonucleotide arrays." Nat Biotechnol 16(1): 45-8. Debnath, A. K., R. L. de Compadre, et al. (1992). "Mutagenicity of quinolines in Salmonella typhimurium TA100. A QSAR study based on hydrophobicity and molecular orbital determinants." Mutat Res 280(1): 55-65. Deitz, W. H., T. M . Cook, et al. (1966). "Mechanism of action of nalidixic acid on Escherichia coli. 3. Conditions required for lethality." J Bacteriol 91(2): 768-73. Dennis, J. J., E. R. Lafontaine, et al. (1996). "Identification and characterization of the tolQRA genes of Pseudomonas aeruginosa." J Bacteriol 178(24): 7059-68. 149 Deretic, V., D. W. Martin, et al. (1993). "Conversion to mucoidy in Pseudomonas aeruginosa." Biotechnology (N Y) 11(10): 1133-6. Deretic, V., M . J. Schurr, et al. (1994). "Conversion of Pseudomonas aeruginosa to mucoidy in cystic fibrosis: environmental stress and regulation of bacterial virulence by alternative sigma factors." J Bacteriol 176(10): 2773-80. DiMango, E., H. J. Zar, et al. (1995). "Diverse Pseudomonas aeruginosa gene products stimulate respiratory epithelial cells to produce interleukin-8." J Clin Invest 96(5): 2204-10. Domagala, J. M. , L. D. Hanna, et al. (1986). "New structure-activity relationships of the quinolone antibacterials using the target enzyme. The development and application of a DNA gyrase assay." J Med Chem 29(3): 394-404. Donaldson, S. H., E. G. Poligone, etal. (2002). "CFTR regulation of ENaC." Methods Mol Med 70: 343-64. Doring, G., S. P. Conway, et al. (2000). "Antibiotic therapy against Pseudomonas aeruginosa in cystic fibrosis: a European consensus." Eur Respir J 16(4): 749-67. Dorman, C. J., N . N . Bhriain, et al. (1990). "DNA supercoiling and environmental regulation of virulence gene expression in Shigella flexneri." Nature 344(6268): 789-92. Drenkard, E. and F. M . Ausubel (2002). "Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation." Nature 416(6882): 740-3. Drlica, K. and X. Zhao (1997). "DNA gyrase, topoisomerase IV, and the 4-quinolones." Microbiol Mol Biol Rev 61(3): 377-92. Duggan, D. J., M . Bittner, et al. (1999). "Expression profiling using cDNA microarrays." Nat Genet 21(1 Suppl): 10-4. Duport, C , C. Baysse, et al. (1995). "Molecular characterization of pyocin S3, a novel S-type pyocin from Pseudomonas aeruginosa." J Biol Chem 270(15): 8920-7. Egan, M. , T. Flotte, et al. (1992). "Defective regulation of outwardly rectifying CI- channels by protein kinase A corrected by insertion of CFTR." Nature 358(6387): 581-4. Eisen, M . B. and P. O. Brown (1999). "DNA arrays for analysis of gene expression." Methods Enzvmol 303: 179-205. Emerson, J., M . Rosenfeld, et al. (2002). "Pseudomonas aeruginosa and other predictors of mortality and morbidity in young children with cystic fibrosis." Pediatr Pulmonol 34(2): 91-100. 150 Ernst, R. K., D. A. D'Argenio, et al. (2003). "Genome mosaicism is conserved but not unique in Pseudomonas aeruginosa isolates from the airways of young children with cystic fibrosis." Environ Microbiol 5(12): 1341-9. Ernst, R. K., E. C. Y i , etal. (1999). "Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa." Science 286(5444): 1561-5. Errington, J., R. A. Daniel, et al. (2003). "Cytokinesis in bacteria." Microbiol Mol Biol Rev 67(1): 52-65, table of contents. Evertsz, E. M. , J. Au-Young, et al. (2001). "Hybridization cross-reactivity within homologous gene families on glass cDNA microarrays." Biotechniques 31(5): 1182, 1184, 1186 passim. Fisher, L. M. , K. Mizuuchi, etal. (1981). "Site-specific interaction of DNA gyrase with DNA." Proc Natl Acad Sci U S A 78(7): 4165-9. Fort, F. L. (1992). "Mutagenicity of quinolone antibacterials." Drug Saf 7(3): 214-22. Foster, P. L. (1993). "Adaptive mutation: the uses of adversity." Annu Rev Microbiol 47: 467-504. Francois, P., M . Bento, et al. (2003). "Comparison of fluorescence and resonance light scattering for highly sensitive microarray detection of bacterial pathogens." J Microbiol Methods 55(3): 755-62. Fukuda, H., M . Hosaka, et al. (1995). "nfxC-type quinolone resistance in a clinical isolate of Pseudomonas aeruginosa." Antimicrob Agents Chemother 39(3): 790-2. Fung-Tome, J., B. Kolek, et al. (1993). "Ciprofloxacin-induced, low-level resistance to structurally unrelated antibiotics in Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus." Antimicrob Agents Chemother 37(6): 1289-96. Gellert, M. , K. Mizuuchi, et al. (1976). "DNA gyrase: an enzyme that introduces superhelical turns into DNA." Proc Natl Acad Sci U S A 73(11): 3872-6. Gibson, R. L., J. L. Burns, et al. (2003). "Pathophysiology and management of pulmonary infections in cystic fibrosis." Am J Respir Crit Care Med 168(8): 918-51. Gilbert, E. J. and A. Maxwell (1994). "The 24 kDa N-terminal sub-domain of the DNA gyrase B protein binds coumarin drugs." Mol Microbiol 12(3): 365-73. Giraud, A., I. Matic, et al. (2002). "Mutator bacteria as a risk factor in treatment of infectious diseases." Antimicrob Agents Chemother 46(3): 863-5. 151 Giwercman, B., P. A. Lambert, et al. (1990). "Rapid emergence of resistance in Pseudomonas aeruginosa in cystic fibrosis patients due to in-vivo selection of stable partially derepressed beta-lactamase producing strains." J Antimicrob Chemother 26(2): 247-59. Gmuender, H., K. Kuratli, et al. (2001). "Gene expression changes triggered by exposure of Haemophilus influenzae to novobiocin or ciprofloxacin: combined transcription and translation analysis." Genome Res 11(1): 28-42. Goldman, M. J., G. M . Anderson, et al. (1997). "Human beta-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis." Cell 88(4): 553-60. Gootz, T. D., Brighty, K. E. (1998). Chemistry and Mechanism of Action of the Quinolone Antibacterials. The Quinolones. V. T. Andriole. San Diego, Academic Press: 29-80. Gordon, P. M . and C. W. Sensen (2004). "Osprey: a comprehensive tool employing novel methods for the design of oligonucleotides for DNA sequencing and microarrays." Nucleic Acids Res 32(17): el33. Goss, W. A., W. H. Deitz, et al. (1965). "Mechanism of action of nalidixic acid on Escherichia coli. II. Inhibition of deoxyribonucleic acid synthesis." J Bacteriol 89: 1068-74. Gould, I. M. , K. Milne, et al. (1991). "Ionic binding, adaptive resistance and post-antibiotic effect of netilmicin and ciprofloxacin." J Antimicrob Chemother 27(6): 741-8. Graeme-Cook, K. A., G. May, et al. (1989). "Osmotic regulation of porin expression: a role for DNA supercoiling." Mol Microbiol 3(9): 1287-94. Grimwood, K., M . To, et al. (1989). "Inhibition of Pseudomonas aeruginosa exoenzyme expression by subinhibitory antibiotic concentrations." Antimicrob Agents Chemother 33(1): 41-7. Grimwood, K., M . To, et al. (1989). "Subinhibitory antibiotics reduce Pseudomonas aeruginosa tissue injury in the rat lung model." J Antimicrob Chemother 24(6): 937-45. Gupta, V., A. Cherkassky, et al. (2003). "Directly labeled mRNA produces highly precise and unbiased differential gene expression data." Nucleic Acids Res 31(4): el3. Hambraeus, G., C. von Wachenfeldt, et al. (2003). "Genome-wide survey of mRNA half-lives in Bacillus subtilis identifies extremely stable mRNAs." Mol Genet Genomics 269(5): 706-14. Hancock, R. E.W. and D. P. Speert (2000). "Antibiotic resistance in Pseudomonas aeruginosa: mechanisms and impact on treatment." Drug Resist Updat 3(4): 247-255. Hancock, R. E. W. (1997). "The bacterial outer membrane as a drug barrier." Trends Microbiol 5(1): 37-42. 152 Hancock, R. E. W., L. M . Mutharia, et al. (1983). "Pseudomonas aeruginosa isolates from patients with cystic fibrosis: a class of serum-sensitive, nontypable strains deficient in lipopolysaccharide O side chains." Infect Immun 42(1): 170-7. Haussler, S., B. Tummler, et al. (1999). "Small-colony variants of Pseudomonas aeruginosa in cystic fibrosis." Clin Infect Pis 29(3): 621-5. Haussler, S., I. Ziegler, et al. (2003). "Highly adherent small-colony variants of Pseudomonas aeruginosa in cystic fibrosis lung infection." J Med Microbiol 52(Pt 4): 295-301. Henry, R. L., C. M . Mellis, et al. (1992). "Mucoid Pseudomonas aeruginosa is a marker of poor survival in cystic fibrosis." Pediatr Pulmonol 12(3): 158-61. Herwig, R., A. O. Schmitt, et al. (2000). "Information theoretical probe selection for hybridisation experiments." Bioinformatics 16(10): 890-8. Hessner, M . J., X. Wang, et al. (2003). "Use of a three-color cPNA microarray platform to measure and control support-bound probe for improved data quality and reproducibility." Nucleic Acids Res 31(11): e60. Hiasa, H., P. O. Yousef, et al. (1996). "PNA strand cleavage is required for replication fork arrest by a frozen topoisomerase-quinolone-PNA ternary complex." J Biol Chem 271(42): 26424-9. Holloway, A. J., R. K. van Laar, et al. (2002). "Options available—from start to finish—for obtaining data from PNA microarrays II." Nat Genet 32(Suppl): 481-9. Holloway, B. W., H. Rossiter, et al. (1973). "Aeruginocin tolerant mutants of Pseudomonas aeruginosa." Genet Res 22(3): 239-53. Hooper, P. C. (1999). "Mechanisms of fluoroquinolone resistance." Prug Resist Updat 2(1): 38-55. Hooper, P. C. (2001). "Emerging mechanisms of fluoroquinolone resistance." Emerg Infect Pis 7(2): 337-41. Hoshino, K., A. Kitamura, et al. (1994). "Comparison of inhibition of Escherichia coli topoisomerase IV by quinolones with PNA gyrase inhibition." Antimicrob Agents Chemother 38(11): 2623-7. Howard, B. M. , R. J. Pinney, et al. (1993). "Function of the SOS process in repair of PNA damage induced by modern 4-quinolones." J Pharm Pharmacol 45(7): 658-62. Huber, W., A. Von Heydebreck, et al. (2002). "Variance stabilization applied to microarray data calibration and to the quantification of differential expression." Bioinformatics 18(Suppl 1): S96-S104. 153 Hueck, C. J. (1998). "Type III protein secretion systems in bacterial pathogens of animals and plants." Microbiol Mol Biol Rev 62(2): 379-433. Hutchison, M . L. and J. R. Govan (1999). "Pathogenicity of microbes associated with cystic fibrosis." Microbes Infect 1(12): 1005-14. Hutter, B., C. Schaab, et al. (2004). "Prediction of mechanisms of action of antibacterial compounds by gene expression profiling." Antimicrob Agents Chemother 48(8): 2838-44. Imundo, L., J. Barasch, et al. (1995). "Cystic fibrosis epithelial cells have a receptor for pathogenic bacteria on their apical surface [published erratum appears in Proc Natl Acad Sci U S A 1995 Nov 21 ;92(24): 11322]." Proc Natl Acad Sci U S A 92(7): 3019-23. Ishii, S. I., Y. Nishi, etal. (1965). "The fine structure of apyocin." J Mol Biol 13(2): 428-31. Isles, A., I. Maclusky, et al. (1984). "Pseudomonas cepacia infection in cystic fibrosis: an emerging problem." J Pediatr 104(2): 206-10. Jacob, F. (1954). "[Induced biosynthesis and mode of action of a pyocine, antibiotic produced by Pseudomonas aeruginosa.}." Ann Inst Pasteur (Paris) 86(2): 149-60. Jacobs, M.A., A. Alwood, et al. (2003). "Comprehensive transposon mutant library of Pseudomonas aeruginosa." Proc Natl Acad Sci U S A 100(24): 14339-44. Jakics, E. B., S. Iyobe, et al. (1992). "Occurrence of the nfxB type mutation in clinical isolates of Pseudomonas aeruginosa." Antimicrob Agents Chemother 36(11): 2562-5. Kageyama, M . (1964). "Studies of a pyocin. I. Physical and chemical properties." J Biochem (Tokyo) 55: 49-53. Karsten, S. L., V. M . Van Deerlin, et al. (2002). "An evaluation of tyramide signal amplification and archived fixed and frozen tissue in microarray gene expression analysis." Nucleic Acids Res 30(2): E4. Kato, J., Y. Nishimura, et al. (1990). "New topoisomerase essential for chromosome segregation in E. coli." CeU 63(2): 393-404. Kerem, B., J. M. Rommens, et al. (1989). "Identification of the cystic fibrosis gene: genetic analysis." Science 245(4922): 1073-80. Keren, I., N. Kaldalu, et al. (2004). "Persister cells and tolerance to antimicrobials." FEMS Microbiol Lett 230(1): 13-8. Kerr, B., M . A. Riley, et al. (2002). "Local dispersal promotes biodiversity in a real-life game of rock-paper-scissors." Nature 418(6894): 171-4. 154 Kerr, M . K. and G. A. Churchill (2001). "Experimental design for gene expression microarrays." Biostatistics 2(2): 183-201. Khan, T. Z., J. S. Wagener, et al. (1995). "Early pulmonary inflammation in infants with cystic fibrosis." Am J Respir Crit Care Med 151(4): 1075-82. Kirkup, B. C. and M . A. Riley (2004). "Antibiotic-mediated antagonism leads to a bacterial game of rock-paper-scissors in vivo." Nature 428(6981): 412-4. Koga, H., A. Itoh, et al. (1980). "Structure-activity relationships of antibacterial 6,7- and 7,8-disubstituted l-alkyl-l,4-dihydro-4-oxoquinoline-3-carboxylic acids." J Med Chem 23(12): 1358-63. Kohler, T., S. F. Epp, etal. (1999). "Characterization of MexT, the regulator of the MexE-MexF-OprN multidrug efflux system of Pseudomonas aeruginosa." J Bacteriol 181(20): 6300-5. Kohler, T., M . Michea-Hamzehpour, et al. (1997). "Characterization of MexE-MexF-OprN, a positively regulated multidrug efflux system of Pseudomonas aeruginosa." Mol Microbiol 23(2): 345-54. Kohler, T., M. Michea-Hamzehpour, et al. (1997). "Differential selection of multidrug efflux systems by quinolones in Pseudomonas aeruginosa." Antimicrob Agents Chemother 41(11): 2540-3. Kohler, T. and C. Pechere (2001). "In vitro selection of antibiotic resistance in Pseudomonas aeruginosa." Clin Microbiol Infect 7(Suppl 5): 7-10. Kohler, T. a. P., J. C. (1998). Bacterial Resistance to Quinolones. The Quinolones. V. T. Andriole. San Diego, CA, Academic Press: 117-142. Kuo, W. P., T. K. Jenssen, et al. (2002). "Analysis of matched mRNA measurements from two different microarray technologies." Bioinformatics 18(3): 405-12. Kuroda, K. and M . Kageyama (1981). "Comparative study of F-type pyocins of Pseudomonas aeruginosa." J Biochem (Tokyo) 89(6): 1721-36. Le Thomas, I., G. Couetdic, et al. (2001). "In vivo selection of a target/efflux double mutant of Pseudomonas aeruginosa by ciprofloxacin therapy." J Antimicrob Chemother 48(4): 553-5. LeClerc, J. E., B. L i , et al. (1996). "High mutation frequencies among Escherichia coli and Salmonella pathogens." Science 274(5290): 1208-11. Lesher, G. Y., E. J. Froelich, etal. (1962). "1,8-Naphthyridine derivatives. A new class of chemotherapeutic agents." J Med Pharm Chem 91: 1063-5. 155 Lewin, C. S., B. M . Howard, etal. (1989). "4-quinolones and the SOS response." JMed Microbiol 29(2): 139-44. Li , X. Z., H. Nikaido, et al. (1995). "Role of mexA-mexB-oprM m antibiotic efflux in Pseudomonas aeruginosa." Antimicrob Agents Chemother 39(9): 1948-53. Liang, P. and A. B. Pardee (1992). "Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction." Science 257(5072): 967-71. Livermore, D. M . (2003). "Bacterial resistance: origins, epidemiology, and impact." Clin Infect PJs36(Suppl 1): SI 1-23. Lockhart, D. J., H. Dong, et al. (1996). "Expression monitoring by hybridization to high-density oligonucleotide arrays." Nat Biotechnol 14(13): 1675-80. Lucas, A. M. a. L. C. D. (1994). "Principles underlying ciliary activity in the respiratory tract. II. A comparison of nasal clearance in man, monkey and other mammals." Arch Otolaryngol 230: 518-541. Luzar, M . A. and T. C. Montie (1985). "Avirulence and altered physiological properties of cystic fibrosis strains of Pseudomonas aeruginosa." Infect Immun 50(2): 572-6. Luzar, M . A., M. J. Thomassen, et al. (1985). "Flagella and motility alterations in Pseudomonas aeruginosa strains from patients with cystic fibrosis: relationship to patient clinical condition." Infect Immun 50(2): 577-82. Magni, A., A. Giordano, et al. (1994). "Effects of continuous exposure to ciprofloxacin on the outer membrane of P. aeruginosa." New Microbiol 17(4): 307-12. Mah, T. F., B. Pitts, et al. (2003). "A genetic basis for Pseudomonas aeruginosa biofdm antibiotic resistance." Nature 426(6964): 306-10. Majtan, V. and D. Hybenova (1996). "Inhibition of Pseudomonas aeruginosa alginate expression by subinhibitory concentrations of antibiotics." Folia Microbiol (Praha) 41(1): 61-4. Mamber, S. W., B. Kolek, et al. (1993). "Activity of quinolones in the Ames Salmonella TA102 mutagenicity test and other bacterial genotoxicity assays." Antimicrob Agents Chemother 37(2): 213-7. Manduchi, E., L. M . Scearce, et al. (2002). "Comparison of different labeling methods for two-channel high-density microarray experiments." Physiol Genomics 10(3): 169-79. Marians, K. J. (1996). Replication fork propogation. E.coli and Salmonella Cellular and Molecular Biology. F. C. Neidhardt. Washington, D C , American Society for Microbiology. 1: 749-763. 156 Martinez-Martinez, L., A. Pascual, et al. (1998). "Quinolone resistance from a transferable plasmid." Lancet 351(9105): 797-9. Massey, R. C , A. Buckling, etal. (2001). "Phenotypic switching of antibiotic resistance circumvents permanent costs in Staphylococcus aureus." Curr Biol 11(22): 1810-4. Matsui, H., B. R. Grubb, et al. (1998). "Evidence for periciliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic fibrosis airways disease." Cell 95(7): 1005-1015. Matsui, H., Y. Sano, et al. (1993). "Regulation of pyocin genes in Pseudomonas aeruginosa by positive (prtN) and negative (prtR) regulatory genes." J Bacteriol 175(5): 1257-63. Maxwell, A. (1993). "The interaction between coumarin drugs and DNA gyrase." Mol Microbiol 9(4): 681-6. McKenney, D., L. Willcock, et al. (1994). "Effect of sub-MIC antibiotics on the cell surface and extracellular virulence determinants of Pseudomonas cepacia." J Appl Bacteriol 76(2): 190-5. Meadow, P. M . and P.L. Wells. (1978). "Receptor sites for R-type pyocins and bacteriophage E79 in the core part of the lipopolysaccharide of Pseudomonas aeruginosa PAC1." Journal of General Microbiology 108: 339-343. Michel-Briand, Y. and C. Baysse (2002). "The pyocins of Pseudomonas aeruginosa." Biochimie 84(5-6): 499-510. Miller, R. V. and T. A. Kokjohn (1988). "Expression of the recA gene of Pseudomonas aeruginosa PAO is inducible by DNA-damaging agents." J Bacteriol 170(5): 2385-7. Moreau, N . J., H. Robaux, et al. (1990). "Inhibitory effects of quinolones on pro- and eucaryotic DNA topoisomerases I and II." Antimicrob Agents Chemother 34(10): 1955-60. Morse, S. A., B. V. Jones, etal. (1980). "Pyocin inhibition ofNeisseria gonorrhoeae: mechanism of action." Antimicrob Agents Chemother 18(3): 416-23. Mouneimne, H., J. Robert, et al. (1999). "Type II topoisomerase mutations in ciprofloxacin-resistant strains of Pseudomonas aeruginosa." Antimicrob Agents Chemother 43(1): 62-6. Moxon, E. R., P. B. Rainey, et al. (1994). "Adaptive evolution of highly mutable loci in pathogenic bacteria." Curr Biol 4(1): 24-33. Mulvey, M . A., Y. S. Lopez-Boado, etal. (1998). "Induction and evasion of host defenses by type 1-piliated uropathogenic Escherichia coli." Science 282(5393): 1494-7. 157 Nakayama, K., K. Takashima, et al. (2000). "The R-type pyocin of Pseudomonas aeruginosa is related to P2 phage, and the F-type is related to lambda phage." Mol Microbiol 38(2): 213-31. Negri, M. C , M . I. Morosini, et al. (2002). "Very low cefotaxime concentrations select for hypermutable Streptococcus pneumoniae populations." Antimicrob Agents Chemother 46(2): 528-30. Nikaido, H. (1998). "The role of outer membrane and efflux pumps in the resistance of gram-negative bacteria. Can we improve drug access?" Drug Resist Updat 1: 93-98. Nikaido, H., K. Nikaido, et al. (1991). "Identification and characterization of porins in Pseudomonas aeruginosa." J Biol Chem 266(2): 770-9. Nimmakayalu, M. , O. Henegariu, et al. (2000). "Simple method for preparation of fluor/hapten-labeled dUTP." Biotechniques 28(3): 518-22. Nixon, G. M. , D. S. Armstrong, et al. (2001). "Clinical outcome after early Pseudomonas aeruginosa infection in cystic fibrosis." J Pediatr 138(5): 699-704. Noone, P. G. and M . Knowles (1999). Standard therapy of CF lung disease. Cystic fibrosis in adults. J. Yankaskas and M . Knowles. Philadelphia, PA, Lippincott-Raven: 145-175. Ochsner, U. A., P. J. Wilderman, et al. (2002). "GeneChip expression analysis of the iron starvation response in Pseudomonas aeruginosa: identification of novel pyoverdine biosynthesis genes." Mol Microbiol 45(5): 1277-87. Ohkawa, I., S. Shiga, et al. (1980). "Effect of iron concentration in the growth medium on the sensitivity of Pseudomonas aeruginosa to pyocin S2." J Biochem (Tokyo) 87(1): 323-31. Ohsumi, M. , T. Shinomiya, et al. (1980). "Comparative study on R-type pyocins of Pseudomonas aeruginosa." J Biochem (Tokyo) 87(4): 1119-25. Okawa, I., B. Maruo, etal. (1975). "Preferential inhibition of lipid synthesis by the bacteriocin pyocin S2." J Biochem (Tokyo) 78(1): 213-23. Oliver, A., F. Baquero, et al. (2002). "The mismatch repair system (mutS, mutL and uvrD genes) in Pseudomonas aeruginosa: molecular characterization of naturally occurring mutants." Mol Microbiol 43(6): 1641-50. Oliver, A., R. Canton, et al. (2000). "High frequency of hypermutable Pseudomonas aeruginosa in cystic fibrosis lung infection." Science 288(5469): 1251-4. Park, P. J., Y. A. Cao, et al. (2004). "Current issues for DNA microarrays: platform comparison, double linear amplification, and universal RNA reference." J Biotechnol 112(3): 225-45. 158 Parret, A. and R. De Mot (2000). "Novel bacteriocins with predicted tRNase and pore-forming activities in Pseudomonas aeruginosa PAOl." Mol Microbiol 35(2): 472-3. Pato, M. L., M . M . Howe, et al. (1990). "A DNA gyrase-binding site at the center of the bacteriophage Mu genome is required for efficient replicative transposition." Proc Natl Acad Sci U S A 87(22): 8716-20. Pedersen, S. S., N . Hoiby, et al. (1992). "Role of alginate in infection with mucoid Pseudomonas aeruginosa in cystic fibrosis." Thorax 47(1): 6-13. Peng, H. and K. J. Marians (1993). "Escherichia coli topoisomerase IV. Purification, characterization, subunit structure, and subunit interactions." J Biol Chem 268(32): 24481-90. Peng, H. and K. J. Marians (1995). "The interaction of Escherichia coli topoisomerase IV with DNA." J Biol Chem 270(42): 25286-90. Phillips, I. (1987). "Bacterial mutagenicity and the 4-quinolones." J Antimicrob Chemother 20(6): 771-3. Phillips, I., E. Culebras, et al. (1987). "Induction of the SOS response by new 4-quinolones." J Antimicrob Chemother 20(5): 631-8. Pier, G. B. (1985). "Pulmonary disease associated with Pseudomonas aeruginosa in cystic fibrosis: current status of the host-bacterium interaction." J Infect Pis 151(4): 575-80. Pier, G. B., M . Grout, et al. (1996). "Role of mutant CFTR in hypersusceptibility of cystic fibrosis patients to lung infections." Science 271(5245): 64-7. Pina, S. E. and S. J. Mattingly (1997). "The role of fluoroquinolones in the promotion of alginate synthesis and antibiotic resistance in Pseudomonas aeruginosa." Curr Microbiol 35(2): 103-8. Poirel, L., T. Lambert, et al. (2001). "Characterization of Class 1 integrons from Pseudomonas aeruginosa that contain the bla(VIM-2) carbapenem-hydrolyzing beta-lactamase gene and of two novel aminoglycoside resistance gene cassettes." Antimicrob Agents Chemother 45(2): 546-52. Poole, K., N. Gotoh, et al. (1996). "Overexpression of the mexC-mexD-oprJ efflux operon in nfxB-type multidrug-resistant strains of Pseudomonas aeruginosa." Mol Microbiol 21(4): 713-24. Poschet, J. F., J. C. Boucher, et al. (2001). "Molecular basis for defective glycosylation and Pseudomonas pathogenesis in cystic fibrosis lung." Proc Natl Acad Sci U S A 98(24): 13972-7. 159 Postier, B. L., H. L. Wang, et al. (2003). "The construction and use of bacterial DNA microarrays based on an optimized two-stage PCR strategy." BMC Genomics 4(1): 23. Power, E. G. and I. Phillips (1993). "Correlation between umuC induction and Salmonella mutagenicity assay for quinolone antimicrobial agents." FEMS Microbiol Lett 112(3): 251-4. Prince, A. (1992). "Adhesins and receptors of Pseudomonas aeruginosa associated with infection of the respiratory tract." Microb Pathog 13(4): 251-60. Proctor, D. F. (1977). "The upper airways. I. Nasal physiology and defense of the lungs." Am Rev Respir Pis 115(1): 97-129. Quackenbush, J. (2002). "Microarray data normalization and transformation." Nat Genet 32(Suppl): 496-501. Ramakrishnan, R., P. Porris, et al. (2002). "An assessment of Motorola CodeLink microarray performance for gene expression profiling applications." Nucleic Acids Res 30(7): e30. Ramsey, B. W., M . S. Pepe, et al. (1999). "Intermittent administration of inhaled tobramycin in patients with cystic fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group." N Engl J Med 340(1): 23-30. Ratjen, F., G. Comes, et al. (2001). "Effect of continuous antistaphylococcal therapy on the rate of P. aeruginosa acquisition in patients with cystic fibrosis." Pediatr Pulmonol 31(1): 13-6. Ravizzola, G., F. Pirali, et al. (1987). "Reduced virulence in ciprofloxacin-resistant variants of Pseudomonas aeruginosa strains." J Antimicrob Chemother 20(6): 825-9. Riccio, M . L., J. P. Pocquier, et al. (2003). "Novel 3-N-aminoglycoside acetyltransferase gene, aac(3)-Ic, from a Pseudomonas aeruginosa integron." Antimicrob Agents Chemother 47(5): 1746-8. Richter, A., C. Schwager, et al. (2002). "Comparison of fluorescent tag PNA labeling methods used for expression analysis by PNA microarrays." Biotechniques 33(3): 620-8, 630. Rimour, S., P. Hill, et al. (2004). "GoArrays: highly dynamic and efficient microarray probe design." Bioinformatics 5: 5. Riordan, J. R., J. M . Rommens, et al. (1989). "Identification of the cystic fibrosis gene: cloning and characterization of complementary PNA." Science 245(4922): 1066-73. Romling, U., B. Fiedler, et al. (1994). "Epidemiology of chronic Pseudomonas aeruginosa infections in cystic fibrosis." J Infect Pis 170(6): 1616-21. 160 Rommens, J. M. , M . C. Iannuzzi, et al. (1989). "Identification of the cystic fibrosis gene: chromosome walking and jumping." Science 245(4922): 1059-65. Saiman, L. and A. Prince (1993). "Pseudomonas aeruginosa pili bind to asialoGMl which is increased on the surface of cystic fibrosis epithelial cells." J Clin Invest 92(4): 1875-80. Salama, N., K. Guillemin, et al. (2000). "A whole-genome microarray reveals genetic diversity among Helicobacter pylori strains." Proc Natl Acad Sci U S A 97(26): 14668-73. Sano, Y. (1993). "The inherent DNase of pyocin AP41 causes breakdown of chromosomal DNA." J Bacteriol 175(3): 912-5. Sano, Y. and M . Kageyama (1981). "Purification and properties of an S-type pyocin, pyocin AP41." J Bacteriol 146(2): 733-9. Sano, Y., H. Matsui, et al. (1993). "Molecular structures and functions of pyocins SI and S2 in Pseudomonas aeruginosa." J Bacteriol 175(10): 2907-16. Schembri, M . A., K. Kjaergaard, et al. (2003). "Global gene expression in Escherichia coli biofilms." Mol Microbiol 48(1): 253-67. Schena, M. , D. Shalon, et al. (1995). "Quantitative monitoring of gene expression patterns with a complementary DNA microarray." Science 270(5235): 467-70. Schroeder, T. H., N. Reiniger, et al. (2001). "Transgenic cystic fibrosis mice exhibit reduced early clearance of Pseudomonas aeruginosa from the respiratory tract." J Immunol 166(12): 7410-8. Schurr, M . J., H. Yu, et al. (1996). "Control of AlgU, a member of the sigma E-like family of stress sigma factors, by the negative regulators MucA and MucB and Pseudomonas aeruginosa conversion to mucoidy in cystic fibrosis." J Bacteriol 178(16): 4997-5004. Schuster, M. , C. P. Lostroh, et al. (2003). "Identification, timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis." J Bacteriol 185(7): 2066-79. Shapiro, J. A. (1997). "Genome organization, natural genetic engineering and adaptive mutation." Trends Genet 13(3): 98-104. Shaw, K. J., N . Miller, et al. (2003). "Comparison of the changes in global gene expression of Escherichia coli induced by four bactericidal agents." J Mol Microbiol Biotechnol 5(2): 105-22. Shaw, K. J. and B. J. Morrow (2003). "Transcriptional profiling and drug discovery." Curr Opin Pharmacol 3(5): 508-12. 161 Shen, L. L. (1993). Quinolone-DNA Interaction. Quinolone Antimicrobial Agents. D. C. a. W. Hooper, J. S. Washington, DC, American Society for Microbiology: 77-95. Shinomiya, T., M . Osumi, et al. (1975). "Defective pyocin particles produced by some mutant strains of Pseudomonas aeruginosa." J Bacteriol 124(3): 1508-21. Shinomiya, T., S. Shiga, et al. (1983). "Genetic determinant of pyocin R2 in Pseudomonas aeruginosa PAO. I. Localization of the pyocin R2 gene cluster between the trpCD and trpE genes." Mol Gen Genet 189(3): 375-81. Sinclair, M. I. and B. W. Holloway (1982). "A chromosomally located transposon in Pseudomonas aeruginosa." J Bacteriol 151(2): 569-79. Singh, P. K., A. L. Schaefer, et al. (2000). "Quorum-sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms.".Nature 407(6805): 762-764. Slonim, D. K. (2002). "From patterns to pathways: gene expression data analysis comes of age." Nat Genet 32(Suppl): 502-8. Smith, J. J., P. H. Karp, et al. (1994). "Defective fluid transport by cystic fibrosis airway epithelia." J Clin Invest 93(3): 1307-11. Smith, J. J., S. M. Travis, et al. (1996). "Cystic fibrosis airway epithelia fail to kill bacteria because of abnormal airway surface fluid." Cell 85(2): 229-36. Sonstein, S. A. and J. C. Burnham (1993). "Effect of low concentrations of quinolone antibiotics on bacterial virulence mechanisms." Diagn Microbiol Infect Pis 16(4): 277-89. Southern, E., K. Mir, etal. (1999). "Molecular interactions on microarrays." Nat Genet 21(1 Suppl): 5-9. Spencer, D. H., A. Kas, et al. (2003). "Whole-genome sequence variation among multiple isolates of Pseudomonas aeruginosa." J Bacteriol 185(4): 1316-25. Spoering, A. L. and K. Lewis (2001). "Biofilms and planktonic cells of Pseudomonas aeruginosa have similar resistance to killing by antimicrobials." J Bacteriol 183(23): 6746-51. Stears, R. L., R. C. Getts, et al. (2000). "A novel, sensitive detection system for high-density microarrays using dendrimer technology." Physiol Genomics 3(2): 93-9. Stillman, B. A. and J. L. Tonkinson (2001). "Expression microarray hybridization kinetics depend on length of the immobilized DNA but are independent of immobilization substrate." Anal Biochem 295(2): 149-57. 162 Stomakhin, A. A., V. A. Vasiliskov, et al. (2000). "DNA sequence analysis by hybridization with oligonucleotide microchips: MALDI mass spectrometry identification of 5mers contiguously stacked to microchip oligonucleotides." Nucleic Acids Res 28(5): 1193-8. Stover, C. K., X. Q. Pham, et al. (2000). "Complete genome sequence of Pseudomonas aeruginosa PAOl, an opportunistic pathogen." Nature 406(6799): 959-64. Stutman, H. R., J. M . Lieberman, et al. (2002). "Antibiotic prophylaxis in infants and young children with cystic fibrosis: a randomized controlled trial." J Pediatr 140(3): 299-305. Stutts, M. J., B. C. Rossier, et al. (1997). "Cystic fibrosis transmembrane conductance regulator inverts protein kinase A-mediated regulation of epithelial sodium channel single channel kinetics." J Biol Chem 272(22): 14037-40. Sugino, A., C. L. Peebles, et al. (1977). "Mechanism of action of nalidixic acid: purification of Escherichia coli nalA gene product and its relationship to DNA gyrase and a novel nicking-closing enzyme." Proc Natl Acad Sci U S A 74(11): 4767-71. Taylor, R. F., M . E. Hodson, et al. (1992). "Auxotrophy of Pseudomonas aeruginosa in cystic fibrosis." FEMS Microbiol Lett 71(3): 243-6. Taylor, R. F., M . E. Hodson, et al. (1993). "Adult cystic fibrosis: association of acute pulmonary exacerbations and increasing severity of lung disease with auxotrophic mutants of Pseudomonas aeruginosa." Thorax 48(10): 1002-5. Thomas, S. R., A. Ray, et al. (2000). "Increased sputum amino acid concentrations and auxotrophy of Pseudomonas aeruginosa in severe cystic fibrosis lung disease." Thorax 55(9): 795-7. Thomassen, M . J., C. A. Demko, et al. (1987). "Cystic fibrosis: a review of pulmonary infections and interventions." Pediatr Pulmonol 3(5): 334-51. Thomassen, M. J., C. A. Demko, et al. (1985). "Pseudomonas cepacia colonization among patients with cystic fibrosis. A new opportunist." Am Rev Respir Pis 131(5): 791-6. Trancassini, M. , M . I. Brenciaglia, et al. (1992). "Modification of Pseudomonas aeruginosa virulence factors by sub-inhibitory concentrations of antibiotics." J Chemother 4(2): 78-81. Tsui, L. C , M . Buchwald, et al. (1985). "Cystic fibrosis locus defined by a genetically linked polymorphic PNA marker." Science 230(4729): 1054-7. Uratani, Y. and T. Hoshino (1984). "Pyocin R l inhibits active transport in Pseudomonas aeruginosa and depolarizes membrane potential." J Bacteriol 157(2): 632-6. 163 Utaida, S., P. M . Dunman, et al. (2003). "Genome-wide transcriptional profiling of the response of Staphylococcus aureus to cell-wall-active antibiotics reveals a cell-wall-stress stimulon." Microbiology 149(Pt 10): 2719-32. Velculescu, V. E., L. Zhang, et al. (1995). "Serial analysis of gene expression." Science 270(5235): 484-7. Vezina, G. and R. C. Levesque (1991). "Molecular characterization of the class II multiresistance transposable element Tnl403 from Pseudomonas aeruginosa." Antimicrob Agents Chemother 35(2): 313-21. Visser, M . R., H. Beumer, et al. (1993). "Changes in adherence of respiratory pathogens to HEp-2 cells induced by subinhibitory concentrations of sparfloxacin, ciprofloxacin, and trimethoprim." Antimicrob Agents Chemother 37(4): 885-8. von Gotz, F., S. Haussler, et al. (2004). "Expression analysis of a highly adherent and cytotoxic small colony variant of Pseudomonas aeruginosa isolated from a lung of a patient with cystic fibrosis." J Bacteriol 186(12): 3837-47. Wagner, V. E., D. Bushnell, et al. (2003). "Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effects of growth phase and environment." J Bacteriol 185(7): 2080-95. Wainwright, B. J., P. J. Scambler, et al. (1985). "Localization of cystic fibrosis locus to human chromosome 7cen-q22." Nature 318(6044): 384-5. Walters, M . C , 3rd, F. Roe, et al. (2003). "Contributions of antibiotic penetration, oxygen limitation, and low metabolic activity to tolerance of Pseudomonas aeruginosa biofilms to ciprofloxacin and tobramycin." Antimicrob Agents Chemother 47(1): 317-23. Walters, R. N. , L. J. Piddock, et al. (1989). "The effect of mutations in the SOS response on the kinetics of quinolone killing." J Antimicrob Chemother 24(6): 863-73. West, S. E., L. Zeng, et al. (2002). "Respiratory infections with Pseudomonas aeruginosa in children with cystic fibrosis: early detection by serology and assessment of risk factors." Jama 287(22): 2958-67. White, R., S. Woodward, et al. (1985). "A closely linked genetic marker for cystic fibrosis." Nature 318(6044): 382-4. Whiteley, M. , M . G. Bangera, et al. (2001). "Gene expression in Pseudomonas aeruginosa biofilms." Nature 413(6858): 860-4. Wiener, M. , D. Freymann, et al. (1997). "Crystal structure of colicin la." Nature 385(6615): 461-4. 164 Wilmott, R. W., J. T. Kassab, et al. (1990). "Increased levels of interleukin-1 in bronchoalveolar washings from children with bacterial pulmonary infections." Am Rev Respir Pis 142(2): 365-8. Wilson, M. , J. PeRisi, et al. (1999). "Exploring drug-induced alterations in gene expression in Mycobacterium tuberculosis by microarray hybridization." Proc Natl Acad Sci U S A 96(22): 12833-8. Wolfgang, M . C , B. R. Kulasekara, et al. (2003). "Conservation of genome content and virulence determinants among clinical and environmental isolates of Pseudomonas aeruginosa." Proc Natl Acad Sci U S A 100(14): 8484-9. Wolfson, J. S. and P. C. Hooper (1985). "The fluoroquinolones: structures, mechanisms of action and resistance, and spectra of activity in vitro." Antimicrob Agents Chemother 28(4): 581-6. Yamamoto, K. R., B. M . Alberts, et al. (1970). "Rapid bacteriophage sedimentation in the presence of polyethylene glycol and its application to large-scale virus purification." Virology 40(3): 734-44. Yang, Y. H. and T. Speed (2002). "Pesign issues for cPNA microarray experiments." Nat Rev Genet 3(8): 579-88. Yassien, M. , N . Khardori, et al. (1995). "Modulation of biofilms of Pseudomonas aeruginosa by quinolones." Antimicrob Agents Chemother 39(10): 2262-8. Ye, R. W., W. Tao, et al. (2000). "Global gene expression profiles of Bacillus subtilis grown under anaerobic conditions." J Bacteriol 182(16): 4458-65. Yonezawa, M. , M. Takahata, et al. (1995). "PNA gyrase gyrA mutations in quinolone-resistant clinical isolates of Pseudomonas aeruginosa." Antimicrob Agents Chemother 39(9): 1970-2. Yoshida, H., M . Bogaki, et al. (1990). "Quinolone resistance-determining region in the PNA gyrase gyrA gene of Escherichia coli." Antimicrob Agents Chemother 34(6): 1271-2. Yuen, T., E. Wurmbach, et al. (2002). "Accuracy and calibration of commercial oligonucleotide and custom cPNA microarrays." Nucleic Acids Res 30(10): e48. Yui-Furihata, C. (1972). "Structure of pyocin R. II. Subunits of sheath." J Biochem (Tokyo) 72(1): 1-10. Zabner, J., J. J. Smith, et al. (1998). "Loss of CFTR chloride channels alters salt absorption by cystic fibrosis airway epithelia in vitro." Mol Cell 2(3): 397-403. 165 Zhanel, G. G., S. O. Kim, et al. (1993). "Effect of subinhibitory concentrations of Ciprofloxacin and gentamicin on the adherence of Pseudomonas aeruginosa to Vero cells and voided uroepithelial cells." Chemotherapy 39(2): 105-11. Zierdt, C. H. and P. J. Schmidt (1964). "Dissociation in Pseudomonas aeruginosa." J Bacteriol 87(5): 1003-10. Ziermann, R., B. Bartlett, etal. (1994). "Functions involved in bacteriophage P2-induced host cell lysis and identification of a new tail gene." J Bacteriol 176(16): 4974-84. Ziha-Zarifi, I., C. Llanes, et al. (1999). "In vivo emergence of multidrug-resistant mutants of Pseudomonas aeruginosa overexpressing the active efflux system MexA-MexB-OprM." Antimicrob Agents Chemother 43(2): 287-91. 166 A P P E N D I X I: Sample data tracking of amplicon uniqueness, size and concentration for all reading frames used in P. aeruginosa custom DNA microarray 167 UBC_PlateNumber UBC_BarCode MasterCoordinate ORF_Name ProductLength GelName LaneNumber ExcludedWell ExclusionCriteria UBC 01 AR0060 A01 PA0001 747 PG01CL13 2 FALSE UBC 01 AR0060 A02 PA0011 797 PG01CL13 3 FALSE UBC 01 AR0060 A03 PA0021 709 PG01CL13 4 FALSE UBC_01 AR0060 A04 PA0032 711 PG01CL13 5 FALSE UBC_01 AR0060 A05 PA0046 354 PG01CL13 6 FALSE UBC 01 AR0060 A06 PA0065 645 PG01CL13 7 FALSE UBC 01 AR0060 A07 PA0075 718 PG01CL13 8 FALSE UBC 01 AR0060 A08 PA0090 715 PG01CL13 9 FALSE UBC_01 AR0060 A09 PA0101 710 PG01CL13 10 FALSE UBC 01 AR0060 A10 PA0111 579 PG01CL13 11 FALSE UBC_01 AR0060 A11 PA0123 734 PG01CL13 12 FALSE UBC 01 AR0060 A12 PA0143 656 PG01CL13 13 FALSE UBC_01 AR0060 B01 PA0002 717 PG01CL13 14 FALSE UBC_01 AR0060 B02 PA0012 267 PG01CL13 15 FALSE UBC_01 AR0060 B03 PA0023 667 PG01CL13 16 FALSE UBC_01 AR0060 B04 PA0033 366 PG01CL13 17 FALSE UBC 01 AR0060 B05 PA0048 384 PG01CL13 18 FALSE UBC 01 AR0060 B06 PA0066 543 PG01CL13 19 FALSE UBC 01 AR0060 B07 PA0078 791 PG01CL13 20 FALSE UBC 01 AR0060 B08 PA0092 285 PG01CL13 21 FALSE UBC 01 AR0060 B09 PA0102 631 PG01CL13 22 FALSE UBC 01 AR0060 B10 PA0112 608 PG01CL13 23 FALSE UBC_01 AR0060 B11 PA0124 282 PG01CL13 24 FALSE UBC_01 AR0060 B12 PA0146 775 PG01CL13 25 FALSE UBC_01 AR0060 C01 PA0003 771 PG01CL13 28 FALSE UBC_01 AR0060 C02 PA0014 273 PG01CL13 29 FALSE UBC 01 AR0060 C03 PA0024 630 PG01CL13 30 FALSE UBC 01 AR0060 C04 PA0035 750 PG01CL13 31 FALSE UBC_01 AR0060 C05 PA0049 697 PG01CL13 32 FALSE UBC_01 AR0060 C06 PA0067 671 PG01CL13 33 FALSE UBC 01 AR0060 C07 PA0079 797 PG01CL13 34 FALSE UBC_01 AR0060 C08 PA0093 774 PG01CL13 35 FALSE UBC 01 AR0060 C09 PA0103 733 PG01CL13 36 FALSE UBC_01 AR0060 C10 PA0113 773 PG01CL13 37 FALSE UBC_01 AR0060 C11 PA0127 501 PG01CL13 38 FALSE UBC_01 AR0060 C12 PA0147 765 PG01CL13 39 FALSE UBC_01 AR0060 D01 PA0004 768 PG01CL13 40 FALSE UBC_01 AR0060 D02 PA0015 318 PG01CL13 41 FALSE UBC 01 AR0060 D03 PA0026 772 PG01CL13 42 FALSE UBC_01 AR0060 D04 PA0036 800 PG01CL13 43 FALSE UBC_01 AR0060 D05 PA0050 141 . PG01CL13 44 FALSE UBC 01 AR0060 D06 PA0069 788 PG01CL13 45 FALSE UBC_01 AR0060 D07 PA0081 722 PG01CL13 46 FALSE UBC_01 AR0060 D08 PA0094 435 PG01CL13 47 FALSE UBC_01 AR0060 D09 PA0105 750 PG01CL13 48 FALSE UBC 01 AR0060 D10 PA0114 610 PG01CL13 49 FALSE UBC 01 AR0060 D11 PA0130 800 PG01CL13 50 FALSE UBC 01 AR0060 D12 PA0148 724 PG01CL13 51 FALSE UBC_01 AR0060 E01 PA0006 537 PG01CL13 54 FALSE UBC_01 AR0060 E02 PA0016 730 PG01CL13 55 FALSE UBC_01 AR0060 E03 PA0027 751 PG01CL13 56 TRUE UBC_01 AR0060 E04 PA0038 216 PG01CL13 57 FALSE UBC_01 AR0060 E05 PA0051 793 PG01CL13 58 FALSE UBC 01 AR0060 E06 PA0070 762 PG01CL13 59 FALSE UBC 01 AR0060 E07 PA0082 729 PG01CL13 60 FALSE UBC_01 AR0060 E08 PA0097 719 PG01CL13 61 FALSE UBC 01 AR0060 E09 PA0106 773 PG01CL13 62 FALSE UBC_01 AR0060 E10 PA0117 718 PG01CL13 63 FALSE UBC 01 AR0060 E11 PA0132 613 PG01CL13 64 FALSE UBC_01 AR0060 E12 PA0149 546 PG01CL13 65 FALSE UBC 01 AR0060 F01 PA0007 775 PG01CL13 66 FALSE UBC_01 AR0060 F02 PA0018 719 PG01CL13 67 FALSE UBC_01 AR0060' F03 PA0029 764 PG01CL13 68 FALSE UBC 01 AR0060 F04 PA0041 779 PG01CL13 69 FALSE UBC_01 AR0060 F05 PA0053 255 PG01CL13 70 FALSE UBC_01 AR0060 F06 PA0071 764 PG01CL13 71 FALSE UBC_01 AR0060 F07 PA0084 641 PG01CL13 72 FALSE UBC_01 AR0060 F08 PA0098 790 PG01CL13 73 FALSE UBC_01 AR0060 F09 PA0108 726 PG01CL13 74 FALSE UBC 01 AR0060 F10 PA0119 695 PG01CL13 75 FALSE UBC 01 AR0060 F11 PA0133 718 PG01CL13 76 TRUE UBC 01 AR0060 F12 PA0150 786 PG01CL13 77 FALSE UBC 01 AR0060 G01 PA0008 707 PG01CL13 80 FALSE UBC_01 AR0060 G02 PA0019 507 PG01CL13 81 FALSE UBC_01 AR0060 G03 PA0030 644 PG01CL13 82 FALSE UBC 01 AR0060 G04 PA0042 396 PG01CL13 83 FALSE UBC_01 AR0060 G05 PA0054 549 PG01CL13 84 FALSE UBC_01 AR0060 G06 PA0072 723 PG01CL13 85 FALSE UBC_01 AR0060 G07 PA0088 694 PG01CL13 86 FALSE UBC 01 AR0060 G08 PA0099 620 PG01CL13 87 FALSE UBC_01 AR0060 G09 PA0109 210 PG01CL13 88 FALSE UBC_01 AR0060 G10 PA0120 675 PG01CL13 89 FALSE UBC_01 AR0060 G11 P A0136 635 PG01CL13 90 FALSE UBC 01 AR0060 G12 PA0151 685 PG01CL13 91 FALSE UBC_01 AR0060 H01 PA0009 701 PG01CL13 92 FALSE UBC 01 AR0060 H02 PA0020 620 PG01CL13 93 FALSE UBC 01 AR0060 H03 PA0031 774 PG01CL13 94 FALSE UBC 01 AR0060 H04 PA0045 687 PG01CL13 95 FALSE UBC 01 AR0060. H05 PA0063 624 PG01CL13 96 FALSE UBC_01 AR0060 H06 PA0074 709 PG01CL13 97 FALSE UBC_01 AR0060 H07 PA0089 696 PG01CL13 98 FALSE UBC_01 AR0060 H08 PA0100 708 PG01CL13 99 FALSE UBC_01 AR0060 H09 PA0110 785 PG01CL13 100 FALSE LOW PRODUCT YIELD (<50ng/ul) LOW PRODUCT YIELD (<50ng/ul) o UBC_01 AR0060 H10 PA0121 721 PG01CL13 101 FALSE UBC_01 AR0060 H11 PA0137 744 PG01CL13 102 FALSE UBC_01 AR0060 H12 PA0153 701 PG01CL13 103 FALSE UBC_02 AR0071 A01 PA0155 743 UB02CL13 2 FALSE UBC 02 AR0071 A02 PA0172 668 UB02CL13 3 FALSE UBC_02 AR0071 A03 PA0187 800 UB02CL13 4 FALSE UBC_02 AR0071 A04 PA0202 627 UB02CL13 5 FALSE UBC_02 AR0071 A05 PA0214 766 UB02CL13 6 FALSE UBC_02 AR0071 A06 PA0226 763 UB02CL13 7 FALSE UBC_02 AR0071 A07 PA0239 667 UB02CL13 8 FALSE UBC_02 AR0071 A08 PA0252 288 UB02CL13 9 FALSE UBC_02 AR0071 A09 PA0261 498 . UB02CL13 10 FALSE UBC_02 AR0071 A10 PA0276 516 UB02CL13 11 FALSE UBC_02 AR0071 A11 PA0290 798 UB02CL13 12 FALSE UBC_02 AR0071 A12 PA0305 758 UB02CL13 13 FALSE UBC 02 AR0071 B01 PA0156 655 UB02CL13 14 FALSE UBC_02 AR0071 B02 PA0175 757 UB02CL13 15 FALSE UBC_02 AR0071 B03 PA0190 725 UB02CL13 16 FALSE UBC 02 AR0071 B04 PA0203 628 UB02CL13 17 FALSE UBC 02 AR0071 B05 PA0215 405 UB02CL13 18 FALSE UBC 02 AR0071 B06 PA0227 783 UB02CL13 19 FALSE UBC 02 AR0071 B07 PA0240 779 UB02CL13 20 TRUE UBC 02 AR0071 B08 PA0254 729 UB02CL13 21 FALSE UBC 02 AR0071 B09 PA0262 693 UB02CL13 22 FALSE UBC_02 AR0071 B10 PA0277 695 UB02CL13 23 FALSE UBC 02 AR0071 B11 PA0291 797 UB02CL13 24 FALSE UBC 02 AR0071 B12 PA0307 612 UB02CL13 25 FALSE UBC 02 AR0071 C01 PA0159 732 UB02CL13 28 FALSE UBC_02 AR0071 C02 PA0176 785 UB02CL13 29 FALSE UBC_02 AR0071 C03 PA0194 657 UB02CL13 30 FALSE UBC 02 AR0071 C04 PA0204 734 UB02CL13 31 FALSE UBC_02 AR0071 C05 PA0216 716 UB02CL13 32 FALSE UBC 02 AR0071 C06 PA0228 686 UB02CL13 33 FALSE UBC_02 AR0071 C07 PA0241 673 UB02CL13 34 TRUE UBC_02 AR0071 C08 PA0255 611 UB02CL13 35 FALSE UBC_02 AR0071 C09 PA0265 606 UB02CL13 36 FALSE UBC_02 AR0071 C10 PA0278 691 UB02CL13 37 FALSE UBC 02 AR0071 C11 PA0293 660 UB02CL13 38 FALSE UBC_02 AR0071 C12 PA0308 715 UB02CL13 39 FALSE UBC_02 AR0071 D01 PA0161 153 UB02CL13 40 FALSE UBC 02 AR0071 D02 PA0178 684 UB02CL13 41 FALSE UBC 02 AR0071 D03 PA0195 746 UB02CL13 42 FALSE UBC_02 AR0071 D04 PA0205 609 UB02CL13 43 FALSE UBC_02 AR0071 D05 PA0217 747 UB02CL13 44 FALSE UBC_02 AR0071 D06 PA0230 701 UB02CL13 45 FALSE UBC_02 AR0071 D07 PA0243 669 UB02CL13 46 TRUE UBC_02 AR0071 D08 PA0256 799 UB02CL13 47 FALSE BLANK LANE INCORRECT BAND SIZE BLANK LANE UBC 02 AR0071 D09 PA0268 UBC 02 AR0071 D10 PA0282 UBC 02 AR0071 D11 PA0297 UBC_02 AR0071 D12 PA0309 UBC_02 AR0071 E01 PA0162 UBC_02 AR0071 E02 PA0180 UBC 02 AR0071 E03 PA0196 UBC 02 AR0071 E04 PA0207 UBC_02 AR0071 E05 PA0219 UBC_02 AR0071 E06 PA0231 UBC_02 AR0071 E07 PA0244 UBC_02 AR0071 E08 PA0257 UBC_02 AR0071 E09 PA0269 UBC_02 AR0071 E10 PA0284 UBC 02 AR0071 E11 PA0298 UBC_02 AR0071 E12 PA0313 UBC 02 AR0071 F01 PA0164 UBC 02 AR0071 F02 PA0182 UBC_02 AR0071 F03 PA0197 UBC_02 AR0071 F04 PA0208 UBC_02 AR0071 F05 PA0222 UBC_02 AR0071 F06 PA0233 UBC_02 AR0071 F07 PA0246 UBC_02 AR0071 F08 PA0258 UBC_02 AR0071 F09 PA0272 UBC_02 AR0071 F10 PA0285 UBC_02 AR0071 F11 PA0299 UBC_02 AR0071 F12 PA0314 UBC_02 AR0071 G01 PA0165 UBC 02 AR0071 G02 PA0183 UBC_02 AR0071 G03 PA0198 UBC_02 AR0071 G04 PA0209 UBC_02 AR0071 G05 PA0223 UBC_02 AR0071 G06 PA0234 UBC_02 AR0071 G07 PA0247 UBC_02 AR0071 G08 PA0259 UBC_02 AR0071 G09 PA0274 UBC_02 AR0071 G10 PA0286 UBC_02 AR0071 G11 PA0300 UBC 02 AR0071 G12 PA0316 UBC_02 AR0071 H01 PA0169 UBC_02 AR0071 H02 PA0184 UBC 02 AR0071 H03 PA0200 UBC_02 AR0071 H04 PA0211 UBC_02 AR0071 H05 PA0224 UBC_02 AR0071 H06 PA0237 UBC_02 AR0071 H07 PA0248 762 UB02CL13 48 FALSE 658 UB02CL13 49 FALSE 687 UB02CL13 50 FALSE 655 UB02CL13 51 FALSE 793 UB02CL13 54 FALSE 752 UB02CL13 55 FALSE 780 UB02CL13 56 FALSE 769 UB02CL13 57 FALSE 749 UB02CL13 58 FALSE 771 UB02CL13 59 FALSE 693 UB02CL13 60 FALSE 774 UB02CL13 61 FALSE 438 UB02CL13 62 FALSE 183 UB02CL13 63 FALSE 770 UB02CL13 64 FALSE 668 UB02CL13 65 FALSE 800 UB02CL13 66 FALSE 753 UB02CL13 67 FALSE 710 UB02CL13 68 TRUE 780 UB02CL13 69 FALSE 703 UB02CL13 70 FALSE 601 UB02CL13 71 FALSE 650 UB02CL13 72 FALSE 186 UB02CL13 73 FALSE 800 UB02CL13 74 FALSE 691 UB02CL13 75 FALSE 703 UB02CL13 76 FALSE 627 UB02CL13 77 FALSE 702 UB02CL13 80 FALSE 643 UB02CL13 81 FALSE 720 UB02CL13 82 FALSE 767 UB02CL13 83 FALSE 712 UB02CL13 84 FALSE 763 UB02CL13 85 FALSE 800 UB02CL13 86 FALSE 749 UB02CL13 87 FALSE 606 UB02CL13 88 FALSE 793 UB02CL13 89 FALSE 792 UB02CL13 90 FALSE 724 UB02CL13 91 FALSE 644 UB02CL13 92 FALSE 757 UB02CL13 93 FALSE 213 UB02CL13 94 FALSE 778 UB02CL13 95 FALSE 783 UB02CL13 96 FALSE 707 UB02CL13 97 FALSE 715 UB02CL13 98 FALSE LOW PRODUCT YIELD (<50ng/ul) —1 to UBC_02 AR0071 H08 PA0260 755 UB02CL13 99 FALSE UBC_02 AR0071 H09 PA0275 661 UB02CL13 100 FALSE UBC_02 AR0071 H10 PA0287 669 UB02CL13 101 FALSE UBC 02 AR0071 H11 PA0304 705 UB02CL13 102 FALSE UBC 02 AR0071 H12 PA0317 643 UB02CL13 103 FALSE UBC 03 AR0009 A01 PA0318 625 PG03CL13 2 FALSE UBC_03 AR0009 A02 PA0331 659 PG03CL13 3 FALSE UBC_03 AR0009 A03 PA0342 740 PG03CL13 4 FALSE UBC_03 AR0009 A04 PA0357 645 PG03CL13 5 FALSE UBC 03 AR0009 A05 PA0368 760 PG03CL13 6 FALSE UBC_03 AR0009 A06 PA0383 653 PG03CL13 7 FALSE UBC 03 AR0009 A07 PA0399 623 PG03CL13 8 FALSE UBC_03 AR0009 A08 PA0411 710 PG03CL13 9 FALSE UBC 03 AR0009 A09 PA0429 716 PG03CL13 10 FALSE UBC_03 AR0009 A10 PA0437 778 PG03CL13 11 FALSE UBC 03 AR0009 A11 PA0453 668 PG03CL13 12 FALSE UBC_03 AR0009 A12 PA0470 679 PG03CL13 13 FALSE UBC_03 AR0009 B01 PA0320 351 PG03CL13 14 FALSE UBC 03 AR0009 B02 PA0334 695 PG03CL13 15 FALSE UBC 03 AR0009 B03 PA0343 798 PG03CL13 16 FALSE UBC 03 AR0009 B04 PA0359 345 PG03CL13 17 FALSE UBC 03 AR0009 B05 PA0370 597 PG03CL13 18 FALSE UBC 03 AR0009 B06 PA0385 324 PG03CL13 19 FALSE UBC_03 AR0009 B07 PA0400 682 PG03CL13 20 FALSE UBC_03 AR0009 B08 PA0413 607 PG03CL13 21 FALSE UBC 03 AR0009 B09 PA0430 780 PG03CL13 22 FALSE UBC 03 AR0009 B10 PA0438 773 PG03CL13 23 FALSE UBC 03 AR0009 B11 PA0455 687 PG03CL13 24 FALSE UBC_03 AR0009 B12 PA0471 749 PG03CL13 25 FALSE UBC_03 AR0009 C01 PA0322 734 PG03CL13 28 FALSE UBC 03 AR0009 C02 PA0336 480 PG03CL13 29 FALSE UBC_03 AR0009 C03 PA0346 363 PG03CL13 30 FALSE UBC 03 AR0009 C04 PA0360 640 PG03CL13 31 FALSE UBC_03 AR0009 C05 PA0372 800 PG03CL13 32 FALSE UBC_03 AR0009 C06 PA0388 420 PG03CL13 33 FALSE UBC.03 AR0009 C07 PA0401 702 PG03CL13 34 FALSE UBC_03 AR0009 C08 PA0414 777 PG03CL13 35 FALSE UBC 03 AR0009 C09 PA0431 555 PG03CL13 36 FALSE UBC_03 AR0009 C10 PA0439 613 PG03CL13 37 FALSE UBC_03 AR0009 C11 PA0456 210 PG03CL13 38 FALSE UBC 03 AR0009 C12 PA0472 519 PG03CL13 39 FALSE UBC 03 AR0009 D01 PA0323 796 PG03CL13 40 FALSE UBC_03 AR0009 D02 PA0337 643 PG03CL13 41 FALSE UBC_03 AR0009 D03 PA0347 727 PG03CL13 42 FALSE UBC_03 AR0009 D04 PA0361 774 PG03CL13 43 FALSE UBC_03 AR0009 D05 PA0373 773 PG03CL13 44 FALSE UBC_03 AR0009 D06 PA0390 642 PG03CL13 45 FALSE UBC_03 AR0009 D07 PA0402 UBC_03 AR0009 D08 PA0416 UBC_03 AR0009 D09 PA0432 UBC_03 AR0009 D10 PA0440 UBC_03 AR0009 D11 PA0461 UBC_03 AR0009 D12 PA0473 UBC_03 AR0009 E01 PA0326 UBC 03 AR0009 E02 PA0338 UBC 03 AR0009 E03 PA0348 UBC_03 AR0009 E04 PA0363 UBC 03 AR0009 E05 PA0375 UBC_03 AR0009 E06 PA0393 UBC 03 AR0009 E07 PA0403 UBC_03 AR0009 E08 PA0424 UBC 03 AR0009 E09 PA0433 UBC_03 AR0009 E10 PA0441 UBC_03 AR0009 E11 PA0462 UBC 03 AR0009 E12 PA0477 UBC_03 AR0009 F01 PA0327 UBC_03 AR0009 F02 PA0339 UBC_03 AR0009 F03 PA0349 UBC_03 AR0009 F04 PA0364 UBC_03 AR0009 F05 PA0378 UBC_03 AR0009 F06 PA0395 UBC_03 AR0009 F07 PA0406 UBC_03 AR0009 F08 PA0425 UBC_03 AR0009 F09 PA0434 UBC_03 AR0009 F10 PA0442 UBC_03 AR0009 F11 PA0463 UBC 03 AR0009 F12 PA0478 UBC_03 AR0009 G01 PA0328 UBC 03 AR0009 G02 PA0340 UBC_03 AR0009 G03 PA0353 UBC 03 AR0009 G04 PA0366 UBC_03 AR0009 G05 PA0381 UBC_03 AR0009 G06 PA0396 UBC_03 AR0009 G07 PA0407 UBC_03 AR0009 G08 PA0426 UBC_03 AR0009 G09 PA0435 UBC_03 AR0009 G10 PA0443 UBC_03 AR0009 G11 PA0464 UBC_03 AR0009 G12 PA0479 UBC_03 AR0009 H01 PA0329 UBC_03 AR0009 H02 PA0341 UBC 03 AR0009 H03 PA0356 UBC_03 AR0009 H04 PA0367 UBC_03 AR0009 H05 PA0382 756 PG03CL13 46 FALSE 696 PG03CL13 47 FALSE 800 PG03CL13 48 FALSE 773 PG03CL13 49 FALSE 698 PG03CL13 50 FALSE 637 PG03CL13 51 FALSE 650 PG03CL13 54 FALSE 800 PG03CL13 55 FALSE 698 PG03CL13 56 FALSE 480 PG03CL13 57 FALSE 644 PG03CL13 58 FALSE 680 PG03CL13 59 FALSE 513 PG03CL13 60 FALSE 444 PG03CL13 61 FALSE 435 PG03CL13 62 FALSE 718 PG03CL13 63 FALSE 705 PG03CL13 64 FALSE 785 PG03CL13 65 FALSE 788 PG03CL13 66 FALSE 704 PG03CL13 67 FALSE 722 PG03CL13 68 FALSE 757 PG03CL13 69 FALSE 691 PG03CL13 70 FALSE 788 PG03CL13 71 FALSE 791 PG03CL13 72 FALSE 800 PG03CL13 73 FALSE 783 PG03CL13 74 FALSE 117 PG03CL13 75 TRUE 653 PG03CL13 76 FALSE 477 PG03CL13 77 FALSE 800 PG03CL13 80 FALSE 800 PG03CL13 81 FALSE 758 PG03CL13 82 FALSE 611 PG03CL13 83 FALSE 772 PG03CL13 84 FALSE 638 PG03CL13 85 FALSE 687 PG03CL13 86 FALSE 797 PG03CL13 87 FALSE 624 PG03CL13 88 FALSE 714 PG03CL13 89 FALSE 617 PG03CL13 90 FALSE 714 PG03CL13 91 FALSE 333 PG03CL13 92 FALSE 766 PG03CL13 93 FALSE 730 PG03CL13 94 FALSE 639 PG03CL13 95 FALSE 672 PG03CL13 96 FALSE B L A N K L A N E UBC_03 AR0009 H06 PA0397 753 UBC_03 AR0009 H07 PA0408 408 UBC_03 AR0009 H08 PA0428 800 UBC_03 AR0009 H09 PA0436 621 UBC_03 AR0009 H10 PA0452 707 UBC_03 AR0009 H11 PA0465 659 UBC_03 AR0009 H12 PA0480 785 UBC_04 AR0010 A01 PA0482 730 UBC 04 AR0010 A02 PA0495 602 UBC_04 AR0010 A03 PA0509 771 UBC 04 AR0010 A04 PA0529 701 UBC_04 AR0010 A05 PA0543 689 UBC 04 AR0010 A06 PA0554 342 UBC_04 AR0010 A07 PA0568 459 UBC 04 AR0010 A08 PA0581 570 UBC_04 AR0010 A09 PA0595 780 UBC_04 AR0010 A10 PA0606 792 UBC 04 AR0010 A11 PA0618 795 UBC_04 AR0010 A12 PA0631 258 UBC_04 AR0010 B01 PA0483 444 UBC_04 AR0010 B02 PA0499 695 UBC_04 AR0010 B03 PA0510 672 UBC_04 AR0010 B04 PA0531 673 UBC_04 AR0010 B05 PA0544 719 UBC 04 AR0010 B06 PA0555 800 UBC 04 AR0010 B07 PA0572 635 UBC_04 AR0010 B08 PA0582 354 UBC_04 AR0010 B09 PA0596 703 UBC_04 AR0010 B10 PA0607 625 UBC 04 AR0010 B11 PA0619 534 UBC_04 AR0010 B12 PA0632 231 UBC_04 AR0010 C01 PA0484 516 UBC_04 AR0010 C02 PA0500 772 UBC_04 AR0010 C03 PA0516 697 UBC_04 AR0010 C04 PA0532 507 UBC 04 AR0010 C05 PA0546 763 UBC_04 AR0010 C06 PA0557 622 UBC 04 AR0010 C07 PA0573 336 UBC_04 AR0010 C08 PA0584 705 UBC_04 AR0010 C09 PA0598 638 UBC 04 AR0010 C10 PA0608 764 UBC_04 AR0010 C11 PA0620 757 UBC_04 AR0010 C12 PA0635 255 UBC 04 AR0010 D01 PA0487 729 UBC_04 AR0010 D02 PA0501 772 UBC_04 AR0010 D03 PA0518 315 UBC 04 AR0010 D04 PA0533 771 PG03CL13 97 FALSE PG03CL13 98 FALSE PG03CL13 99 FALSE PG03CL13 100 FALSE PG03CL13 101 FALSE PG03CL13 102 FALSE PG03CL13 103 FALSE UB04CL13 2 TRUE UB04CL13 3 FALSE UB04CL13 4 TRUE UB04CL13 5 FALSE UB04CL13 6 FALSE UB04CL13 7 FALSE UB04CL13 8 FALSE UB04CL13 9 FALSE UB04CL13 10 FALSE UB04CL13 11 FALSE UB04CL13 12 FALSE UB04CL13 13 FALSE UB04CL13 14 FALSE UB04CL13 15 FALSE UB04CL13 16 FALSE UB04CL13 17 FALSE UB04CL13 18 FALSE UB04CL13 19 FALSE UB04CL13 20 FALSE UB04CL13 21 FALSE UB04CL13 22 FALSE UB04CL13 23 FALSE UB04CL13 24 FALSE UB04CL13 25 FALSE UB04CL13 28 TRUE UB04CL13 29 FALSE UB04CL13 30 FALSE UB04CL13 31 FALSE UB04CL13 32 FALSE UB04CL13 33 FALSE UB04CL13 34 FALSE UB04CL13 35 FALSE UB04CL13 36 FALSE UB04CL13 37 FALSE UB04CL13 38 FALSE UB04CL13 39 FALSE UB04CL13 40 TRUE UB04CL13 41 FALSE UB04CL13 42 FALSE UB04CL13 43 FALSE L O W P R O D U C T YIELD (<50ng/ul) L O W P R O D U C T YIELD (<50ng/ul) L O W P R O D U C T YIELD (<50ng/ul) BORDERLINE YIELD (50 to 60ng/ul) - HAND R E J E C T UBC_04 AR0010 D05 PA0547 UBC 04 AR0010 D06 PA0558 UBC_04 AR0010 D07 PA0574 UBC_04 AR0010 D08 PA0587 UBC_04 • AR0010 D09 PA0599 UBC_04 AR0010 D10 PA0609 UBC_04 AR0010 D11 PA0621 UBC 04 AR0010 D12 PA0636 UBC_04 AR0010 E01 PA0490 UBC_04 AR0010 E02 PA0502 UBC_04 AR0010 E03 PA0519 UBC_04 AR0010 E04 PA0536 UBC_04 AR0010 E05 PA0548 UBC_04 AR0010 E06 PA0560 UBC_04 AR0010 E07 PA0575 UBC_04 AR0010 E08 PA0588 UBC 04 AR0010 E09 PA0600 UBC_04 AR0010 E10 PA0611 UBC_04 AR0010 E11 PA0622 UBC_04 AR0010 E12 PA0638 UBC_04 AR0010 F01 PA0491 UBC_04 AR0010 F02 PA0504 UBC_04 AR0010 F03 PA0520 UBC_04 AR0010 F04 PA0539 UBC_04 AR0010 F05 PA0549 UBC_04 AR0010 F06 PA0562 UBC_04 AR0010 F07 PA0577 UBC_04 AR0010 F08 PA0590 UBC 04 AR0010 F09 PA0602 UBC_04 AR0010 F10 PA0614 UBC_04 AR0010 F11 PA0625 UBC_04 AR0010 F12 PA0639 UBC_04 AR0010 G01 PA0492 UBC_04 AR0010 G02 PA0506 UBC 04 AR0010 G03 PA0523 UBC_04 AR0010 G04 PA0540 UBC 04 AR0010 G05 PA0550 UBC_04 AR0010 G06 PA0564 UBC 04 AR0010 G07 PA0578 UBC_04 AR0010 G08 PA0593 UBC_04 AR0010 G09 PA0604 UBC_04 AR0010 G10 PA0615 UBC_04 AR0010 G11 PA0626 UBC_04 AR0010 G12 PA0640 UBC_04 AR0010 H01 PA0494 UBC 04 AR0010 H02 PA0507 UBC 04 AR0010 H03 PA0525 775 UB04CL13 44 FALSE 755 UB04CL13 45 FALSE 680 UB04CL13 46 FALSE 800 UB04CL13 47 FALSE 800 UB04CL13 48 FALSE 753 UB04CL13 49 FALSE 459 UB04CL13 50 FALSE 642 UB04CL13 51 FALSE 294' UB04CL13 54 FALSE 702 UB04CL13 55 FALSE 744 UB04CL13 56 TRUE 696 UB04CL13 57 FALSE 635 UB04CL13 58 FALSE 489 UB04CL13 59 FALSE 720 UB04CL13 60 FALSE 727 UB04CL13 61 FALSE 765 UB04CL13 62 FALSE 620 UB04CL13 63 FALSE 700 UB04CL13 64 FALSE 637 UB04CL13 65 FALSE 742 UB04CL13 66 FALSE 687 UB04CL13 67 FALSE 741 UB04CL13 68 FALSE 780 UB04CL13 69 FALSE 765 UB04CL13 70 FALSE 634 UB04CL13 71 FALSE 768 UB04CL13 72 FALSE 679 UB04CL13 73 FALSE 770 UB04CL13 74 FALSE 450 UB04CL13 75 FALSE 686 UB04CL13 76 FALSE 625 UB04CL13 77 FALSE 682 UB04CL13 80 FALSE 674 UB04CL13 81 FALSE 441 UB04CL13 82 FALSE 387 UB04CL13 83 FALSE 785 UB04CL13 84 FALSE 752 UB04CL13 85 FALSE 450 UB04CL13 86 FALSE 619 UB04CL13 87 FALSE 624 UB04CL13 88 FALSE 516 UB04CL13 89 FALSE 785 UB04CL13 90 FALSE 603 UB04CL13 91 FALSE 749 UB04CL13 92 FALSE 784 UB04CL13 93 FALSE 682 UB04CL13 94 FALSE L O W PRODUCT YIELD (<50ng/ul) ON UBC_58 AR0126 C07 PA1984 743 PG58CL13A 34 FALSE UBC 58 AR0126 C08 PA3993 766 PG58CL13A 35 FALSE UBC_58 AR0126 C09 PA4277 755 PG58CL13A 36 FALSE UBC_58 AR0126 C10 PA1280 450 PG58CL13A 37 FALSE UBC_58 AR0126 C11 PA2458 675 PG58CL13A 38 TRUE UBC_58 AR0126 C12 PA5343 671 PG58CL13A 39 FALSE UBC 58 AR0126 D01 PA5453 798 PG58CL13A 40 FALSE UBC 58 AR0126 D02 PA5477 773 PG58CL13A 41 FALSE UBC_58 AR0126 D03 PA5501 675 PG58CL13A 42 FALSE UBC_58 AR0126 D04 PA5545 742 PG58CL13A 43 FALSE UBC_58 AR0126 D05 PA5567 800 PG58CL13A 44 FALSE UBC_58 AR0126 D06 PA1901 754 PG58CL13A 45 FALSE UBC_58 AR0126 D07 PA2291 742 PG58CL13A 46 FALSE UBC_58 AR0126 D08 PA4022 743 PG58CL13A 47 FALSE UBC_58 AR0126 D09 PA4625 688 PG58CL13A 48 FALSE UBC.58 AR0126 D10 PA1605 758 PG58CL13A 49 FALSE UBC 58 AR0126 D11 PA2460 227 PG58CL13A 50 TRUE UBC_58 AR0126 D12 PA0497 606 PG58CL13A 51 FALSE UBC_58 AR0126 E01 PA5456 696 PG58CL13A 54 FALSE UBC_58 AR0126 E02 PA5482 162 PG58CL13A 55 FALSE UBC_58 AR0126 E03 PA5505 762 PG58CL13A 56 FALSE UBC_58 AR0126 E04 PA5546 742 PG58CL13A 57 FALSE UBC_58 AR0126 E05 PA0040 786 PG58CL13A 58 FALSE UBC_58 AR0126 E06 PA1902 624 PG58CL13A 59 FALSE UBC_58 AR0126 E07 PA2319 766 PG58CL13A 60 FALSE UBC_58 AR0126 E08 PA4212 754 PG58CL13A 61 FALSE UBC_58 AR0126 E09 PA4797 766 PG58CL13A 62 FALSE UBC_58 AR0126 E10 PA1632 55 PG58CL13A 63 TRUE UBC 58 AR0126 E11 PA4125 335 PG58CL13A 64 FALSE UBC_58 AR0126 E12 PA0498 505 PG58CL13A 65 FALSE UBC_58 AR0126 F01 PA5459 739 PG58CL13A 66 FALSE UBC_58 AR0126 F02 PA5486 594 PG58CL13A 67 FALSE UBC_58 AR0126 F03 PA5508 794 PG58CL13A 68 FALSE UBC_58 AR0126 F04 PA5548 739 PG58CL13A 69 FALSE UBC_58 AR0126 F05 PA0263 519 PG58CL13A 70 FALSE UBC_58 AR0126 F06 PA1903 733 PG58CL13A 71 FALSE UBC 58 AR0126 F07 PA2463 786 PG58CL13A 72 FALSE UBC_58 AR0126 F08 PA4213 624 PG58CL13A 73 FALSE UBC 58 AR0126 F09 PA0167 631 PG58CL13A 74 FALSE UBC_58 AR0126 F10 PA1968 98 PG58CL13A 75 TRUE UBC_58 AR0126 F11 PA4190 669 PG58CL13A 76 FALSE UBC_58 AR0126 F12 PA0886 172 PG58CL13A 77 FALSE UBC_58 AR0126 G01 PA5461 318 PG58CL13A 80 FALSE UBC_58 AR0126 G02 PA5489 614 PG58CL13A 81 FALSE UBC_58 AR0126 G03 PA5515 501 PG58CL13A 82 FALSE UBC 58 AR0126 G04 PA5551 510 PG58CL13A 83 FALSE UBC 58 AR0126 G05 PA0445 766 PG58CL13A 84 FALSE BLANK LANE BLANK LANE BLANK LANE L O W P R O D U C T YIELD (<50ng/ul) APPENDIX II: Sample capillary electrophoresis data for all open reading frames used in P. aeruginosa custom DNA microarray 177 J B 384 Well Plate Barcode Sample C4PS01-P3-0001-BL008-001 A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 A16 A17 A18 A19 ! A20 A21 A22 A23 A24 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 Size(bp) Concentration(ng/ul) x4 384WellPlatelD ORF_Nam 718 15.3 61.2 384-01 PA0001 773 7.6 30.4 384-01 PA0155 845 8.8 35.2 384-01 PA0011 663 13.2 52.8 384-01 PA0172 700 20.7 82.8 384-01 PA0021 791 14.3 57.2 384-01 PA0187 690 21.9 87.6 384-01 PA0032 677 11.6 46.4 384-01 PA0202 405 19.1 76.4 384-01 PA0046 800 15.9 63.6 384-01 PA0214 677 20.5 82 384-01 PA0065 745 20.2 80.8 384-01 PA0226 782 18.3 73.2 384-01 PA0075 680 17.5 70 384-01 PA0239 693 14.1 56.4 384-01 PA0090 327 10.1 40.4 384-01 PA0252 670 22.2 88.8 384-01 PA0101 547 8.3 33.2 384-01 PA0261 623 16.4 65.6 384-01 PA0111 570 15.9 63.6 384-01 PA0276 718 16.8 67.2 384-01 PA0123 736 16 64 384-01 PA0290 660 20.3 81.2 384-01 PA0143 755 7.4 29.6 384-01 PA0305 661 8.6 34.4 384-01 PA0318 745 3.6 14.4 384-01 PA0482 688 6.8 27.2 384-01 PA0331 676 9.6 38.4 384-01 PA0495 755 12.3 49.2 384-01 PA0342 782 2.9 11.6 384-01 PA0509 688 4.4 17.6 384-01 PA0357 700 5.7 22.8 384-01 PA0529 855 11.2 44.8 384-01 PA0368 718 9.7 38.8 384-01 PA0543 718 9.3 37.2 384-01 PA0383 389 5.2 20.8 384-01 PA0554 B13 669 12.5 B14 500 7 B15 709 12.6 B16 618 7.8 B17 736 9.1 B18 809 2.3 B19 818 12.9 B20 800 13.2 B21 736 14.3 B22 882 4.5 B23 782 10.2 B24 295 4.1 C1 747 16.3 C2 675 14.9 C3 299 5.4 C4 747 5.6 C5 683 12.4 C6 765 10.6 C7 419 12.6 C8 719 12.6 C9 433 13 C10 457 12.3 C11 596 11.9 C12 747 9.9 C13 812 14.2 C14 -C15 320 12 C16 774 12 C17 700 16.7 C18 747 11 C19 667 16 C20 719 8.5 C21 316 5.2 C22 849 7.2 C23 802 14.5 C24 654 10.3 D1 400 6.8 50 384-01 PA0399 28 384-01 PA0568 50.4 384-01 PA0411 31.2 384-01 PA0581 36.4 384-01 PA0429 9.2 384-01 PA0595 51.6 384-01 PA0437 52.8 384-01 PA0606 57.2 384-01 PA0453 18 384-01 PA0618 40.8 384-01 PA0470 16.4 384-01 PA0631 65.2 384-01 PA0002 59.6 384-01 PA0156 21.6 384-01 PA0012 22.4 384-01 PA0175 49.6 384-01 PA0023 42.4 384-01 PA0190 50.4 384-01 PA0033 50.4 384-01 PA0203 52 384-01 PA0048 49.2 384-01 PA0215 47.6 384-01 PA0066 39.6 384-01 PA0227 56.8 384-01 PA0078 0 384-01 PA0240 48 384-01 PA0092 48 384-01 PA0254 66.8 384-01 PA0102 44 384-01 PA0262 64 384-01 PA0112 34 384-01 PA0277 20.8 384-01 PA0124 28.8 384-01 PA0291 58 384-01 PA0146 41.2 384-01 PA0307 27.2 384-01 PA0320 D2 483 2.1 D3 696 11.8 D4 784 8 D5 821 8.2 D6 691 10.7 D7 398 6.1 D8 747 4.8 D9 649 9 D10 774 10.4 D11 376 7.8 D12 858 9.7 D13 737 8.7 D14 709 12.1 D15 653 11 D16 407 7 D17 821 10.6 D18 821 9.1 D19 821 7.2 D20 666 5.3 D21 747 10.8 D22 602 6.2 D23 867 8 D24 275 3.2 E1 795 12.1 E2 824 16.7 E3 311 4.1 E4 767 11.9 E5 665 17.9 E6 710 14.4 E7 729 14.3 E8 776 11 E9 729 10.8 E10 767 10.7 E11 687 16.9 E12 696 15.4 E13 805 13.5 E14 1373 4 8.4 384-01 PA0483 47.2 384-01 PA0334 32 384-01 PA0499 32.8 384-01 PA0343 42.8 384-01 PA0510 24.4 384-01 PA0359 19.2 384-01 PA0531 36 384-01 PA0370 41.6 384-01 PA0544 31.2 384-01 PA0385 38.8 384-01 PA0555 34.8 384-01 PA0400 48.4 384-01 PA0572 44 384-01 PA0413 28 384-01 PA0582 42.4 384-01 PA0430 36.4 384-01 PA0596 28.8 384-01 PA0438 21.2 384-01 PA0607 43.2 384-01 PA0455 24.8 384-01 PA0619 32 384-01 PA0471 12.8 384-01 PA0632 48.4 384-01 PA0003 66.8 384-01 PA0159 16.4 384-01 PA0014 47.6 384-01 PA0176 71.6 384-01 PA0024 57.6 384-01 PA0194 57.2 384-01 PA0035 44 384-01 PA0204 43.2 384-01 PA0049 42.8 384-01 PA0216 67.6 384-01 PA0067 61.6 384-01 PA0228 54 384-01 PA0079 16 384-01 PA0241 E 1 5 7 6 7 1 3 . 7 E 1 6 6 6 1 5 .6 E 1 7 7 6 7 1 7 . 5 E 1 8 6 5 2 1 0 . 3 E 1 9 8 6 2 1 1 . 7 E 2 0 7 3 8 6 . 4 E 2 1 5 3 9 1 1 . 6 E 2 2 6 9 1 2 . 5 E 2 3 8 1 4 1 4 . 5 E 2 4 7 3 8 6 F 1 7 8 8 1 8 . 4 F 2 5 9 3 2 . 7 F 3 5 5 3 7 . 5 F 4 8 1 7 9 . 9 F 5 4 1 8 4 . 5 F 6 7 4 9 6 . 2 F 7 6 8 7 7 . 7 F 8 5 6 7 1 0 . 8 F 9 8 7 6 1 2 . 8 F 1 0 8 1 7 10 .1 F 1 1 4 6 8 1 2 . 3 F 1 2 6 7 3 9 . 3 F 1 3 7 4 9 1 3 . 3 F 1 4 3 9 9 6 . 8 F 1 5 8 3 7 1 3 . 8 F 1 6 7 9 8 1 6 . 4 F 1 7 6 3 3 1 2 F 1 8 7 0 0 7 .6 F 1 9 6 9 6 4 . 5 F 2 0 8 2 7 1 2 . 4 F 2 1 2 4 7 2 . 2 F 2 2 8 9 5 2 . 8 F 2 3 5 8 4 1 2 . 2 F 2 4 2 9 0 2 . 7 G 1 8 2 7 1 5 . 4 G 2 1 9 1 5 . 8 G 3 3 6 7 6 . 7 5 4 . 8 3 8 4 - 0 1 P A 0 0 9 3 2 2 . 4 3 8 4 - 0 1 P A 0 2 5 5 7 0 3 8 4 - 0 1 P A 0 1 0 3 4 1 . 2 3 8 4 - 0 1 P A 0 2 6 5 4 6 . 8 3 8 4 - 0 1 P A 0 1 1 3 2 5 . 6 3 8 4 - 0 1 P A 0 2 7 8 4 6 . 4 3 8 4 - 0 1 P A 0 1 2 7 1 0 3 8 4 - 0 1 P A 0 2 9 3 5 8 3 8 4 - 0 1 P A 0 1 4 7 2 4 3 8 4 - 0 1 P A 0 3 0 8 7 3 . 6 3 8 4 - 0 1 P A 0 3 2 2 1 0 . 8 3 8 4 - 0 1 P A 0 4 8 4 3 0 3 8 4 - 0 1 P A 0 3 3 6 3 9 . 6 3 8 4 - 0 1 P A 0 5 0 0 1 8 3 8 4 - 0 1 P A 0 3 4 6 2 4 . 8 3 8 4 - 0 1 P A 0 5 1 6 3 0 . 8 3 8 4 - 0 1 P A 0 3 6 0 4 3 . 2 3 8 4 - 0 1 P A 0 5 3 2 5 1 . 2 3 8 4 - 0 1 P A 0 3 7 2 4 0 . 4 3 8 4 - 0 1 P A 0 5 4 6 4 9 . 2 3 8 4 - 0 1 P A 0 3 8 8 3 7 . 2 3 8 4 - 0 1 P A 0 5 5 7 5 3 . 2 3 8 4 - 0 1 P A 0 4 0 1 2 7 . 2 3 8 4 - 0 1 P A 0 5 7 3 5 5 . 2 3 8 4 - 0 1 P A 0 4 1 4 6 5 . 6 3 8 4 - 0 1 P A 0 5 8 4 4 8 3 8 4 - 0 1 P A 0 4 3 1 3 0 . 4 3 8 4 - 0 1 P A 0 5 9 8 1 8 3 8 4 - 0 1 P A 0 4 3 9 4 9 . 6 3 8 4 - 0 1 P A 0 6 0 8 8 . 8 3 8 4 - 0 1 P A 0 4 5 6 1 1 . 2 3 8 4 - 0 1 P A 0 6 2 0 4 8 . 8 3 8 4 - 0 1 P A 0 4 7 2 1 0 . 8 3 8 4 - 0 1 P A 0 6 3 5 6 1 . 6 3 8 4 - 0 1 P A 0 0 0 4 2 3 . 2 3 8 4 - 0 1 P A 0 1 6 1 2 6 . 8 3 8 4 - 0 1 P A 0 0 1 5 G4 739 17.5 G5 837 17.5 G6 837 12.2 G7 856 14.5 G8 686 13.9 G9 181 7.9 G10 837 11.2 G11 837 13.7 G12 720 8.4 G13 798 9.5 G14 G15 477 9.6 G16 856 6.4 G17 788 12.4 G18 885 11.6 G19 664 12.2 G20 686 13.7 G21 885 13.7 G22 739 6.9 G23 768 13.9 G24 691 12.9 H1 846 9.3 H2 788 1.4 H3 686 4.9 H4 827 7.5 H5 807 11.8 H6 360 4.9 H7 817 10 H8 856 10.3 H9 846 8.8 H10 837 9.3 H11 690 11.4 H12 856 7.2 H13 895 12.5 H14 749 7.6 H15 720 10 H16 846 12.3 70 384-01 P A0178 70 384-01 PA0026 48.8 384-01 PA0195 58 384-01 PA0036 55.6 384-01 PA0205 31.6 384-01 PA0050 44.8 384-01 PA0217 54.8 384-01 PA0069 33.6 384-01 PA0230 38 384-01 PA0081 0 384-01 PA0243 38.4 384-01 PA0094 25.6 384-01 PA0256 49.6 384-01 PA0105 46.4 384-01 PA0268 48.8 384-01 PA0114 54.8 384-01 PA0282 54.8 384-01 PA0130 27.6 384-01 PA0297 55.6 384-01 PA0148 51.6 384-01 PA0309 37.2 384-01 PA0323 5.6 384-01 PA0487 19.6 384-01 PA0337 30 384-01 PA0501 47.2 384-01 PA0347 19.6 384-01 PA0518 40 384-01 PA0361 41.2 384-01 PA0533 35.2 384-01 PA0373 37.2 384-01 PA0547 45.6 384-01 PA0390 28.8 384-01 PA0558 50 384-01 PA0402 30.4 384-01 PA0574 40 384-01 PA0416 49.2 384-01 PA0587 H 1 7 876 11 H18 885 11.6 H 1 9 866 11.8 H 2 0 837 6.4 H21 788 10.1 H 2 2 498 3.3 H 2 3 676 5.9 H 2 4 695 8 11 590 14.5 12 880 15.5 13 750 7.9 14 800 8.8 15 61 1.2 16 810 10.9 17 256 4.6 18 840 14.2 19 910 12.5 110 740 12.9 111 860 12.1 112 890 9.1 113 730 14.9 114 760 14.3 115 750 15.9 116 960 6.3 117 820 13.6 118 492 9.6 119 770 11.2 I20 223 5.2 121 667 12.4 I22 820 9.2 I23 595 14.3 I24 700 8 J1 685 12.3 J 2 347 5.8 J 3 833 13.6 J4 731 7.7 J 5 772 9.8 44 384-01 P A 0 4 3 2 46.4 384-01 P A 0 5 9 9 47.2 384-01 P A 0 4 4 0 25.6 384-01 P A 0 6 0 9 40.4 384-01 PA0461 13.2 384-01 PA0621 23.6 384-01 P A 0 4 7 3 32 384-01 P A 0 6 3 6 58 384-01 P A 0 0 0 6 62 384-01 P A 0 1 6 2 31.6 384-01 P A 0 0 1 6 35.2 384-01 P A 0 1 8 0 4.8 384-01 P A 0 0 2 7 43.6 384-01 P A 0 1 9 6 18.4 384-01 P A 0 0 3 8 56.8 384-01 P A 0 2 0 7 50 384-01 PA0051 51.6 384-01 P A 0 2 1 9 48.4 384-01 P A 0 0 7 0 36.4 384-01 PA0231 59.6 384-01 P A 0 0 8 2 57.2 384-01 P A 0 2 4 4 63.6 384-01 P A 0 0 9 7 25.2 384-01 P A 0 2 5 7 54.4 384-01 P A 0 1 0 6 38.4 384-01 P A 0 2 6 9 44.8 384-01 P A 0 1 1 7 20.8 384-01 P A 0 2 8 4 49.6 384-01 P A 0 1 3 2 36.8 384-01 P A 0 2 9 8 57.2 384-01 P A 0 1 4 9 32 384-01 P A 0 3 1 3 49.2 384-01 P A 0 3 2 6 23.2 384-01 P A 0 4 9 0 54.4 384-01 P A 0 3 3 8 30.8 384-01 P A 0 5 0 2 39.2 384-01 P A 0 3 4 8 J6 782 2.1 J7 520 12.4 J8 772 8.1 J9 695 9.4 J10 710 9.8 J11 762 12.1 J12 563 3.9 J13 568 7.3 J14 721 9.1 J15 494 7.6 J16 772 9.3 J17 484 8.9 J18 864 10.1 J19 803 10.6 J20 690 10.8 J21 772 9.5 J22 762 5.3 J23 885 13.9 J24 700 3.7 K1 864 20.4 K2 895 11.9 K3 782 18.3 K4 813 11.6 K5 885 20 K6 K7 864 12 K8 874 13.1 K9 288 7.2 K10 772 14.9 K11 813 8.7 K12 680 10.2 K13 671 12.9 K14 731 14.5 K15 844 13.9 K16 223 4.9 K17 792 11 K18 864 12 8.4 384-01 PA0519 49.6 384-01 PA0363 32.4 384-01 PA0536 37.6 384-01 PA0375 39.2 384-01 PA0548 48.4 384-01 PA0393 15.6 384-01 PA0560 29.2 384-01 PA0403 36.4 384-01 PA0575 30.4 384-01 PA0424 37.2 384-01 PA0588 35.6 384-01 PA0433 40.4 384-01 PA0600 42.4 384-01 PA0441 43.2 384-01 PA0611 38 384-01 PA0462 21.2 384-01 PA0622 55.6 384-01 PA0477 14.8 384-01 PA0638 81.6 384-01 PA0007 47.6 384-01 PA0164 73.2 384-01 PA0018 46.4 384-01 PA0182 80 384-01 PA0029 0 384-01 PA0197 48 384-01 PA0041 52.4 384-01 PA0208 28.8 384-01 PA0053 59.6 384-01 PA0222 34.8 384-01 PA0071 40.8 384-01 • PA0233 51.6 384-01 PA0084 58 384-01 PA0246 55.6 384-01 PA0098 19.6 384-01 PA0258 44 384-01 PA0108 48 384-01 PA0272 K19 772 11.9 K20 741 9.6 K21 864 8.7 K22 710 8 K23 844 6.9 K24 656 6.5 L1 870 13.4 L2 830 4.3 L3 810 17.8 L4 790 3.9 L5 820 12.4 L6 860 6.9 L7 850 15.8 L8 870 10.2 L9 780 12.2 L10 860 9 L11 810 11.5 L12 720 10.4 L13 850 12.5 L14 880 10.4 L15 870 11.6 L16 770 10.5 L17 910 11.3 L18 820 8.6 L19 L20 494 9.1 L21 720 13.6 L22 760 3.4 L23 530 4.6 L24 685 7.3 M1 813 11 M2 792 5.2 M3 556 9.3 M4 695 6.4 M5 679 14.3 M6 782 7.7 M7 464 7.5 47.6 384-01 PA0119 38.4 384-01 PA0285 34.8 384-01 PA0133 32 384-01 PA0299 27.6 384-01 PA0150 26 384-01 PA0314 53.6 384-01 PA0327 17.2 384-01 PA0491 71.2 384-01 PA0339 15.6 384-01 PA0504 49.6 384-01 PA0349 27.6 384-01 PA0520 63.2 384-01 PA0364 40.8 384-01 PA0539 48.8 384-01 PA0378 36 384-01 PA0549 46 384-01 PA0395 41.6 384-01 PA0562 50 384-01 PA0406 41.6 384-01 PA0577 46.4 384-01 PA0425 42 384-01 PA0590 45.2 384-01 PA0434 34.4 384-01 PA0602 0 384-01 PA0442 36.4 384-01 PA0614 54.4 384-01 PA0463 13.6 384-01 PA0625 18.4 384-01 PA0478 29.2 384-01 PA0639 44 384-01 PA0008 20.8 384-01 PA0165 37.2 384-01 PA0019 25.6 384-01 PA0183 57.2 384-01 PA0030 30.8 384-01 PA0198 30 384-01 PA0042 023 732 6.6 26.4 384-15 PA4936 024 569 4.3 17.2 384-15 PA2954 P1 0 384-15 PA3458 P2 0 384-15 BLANK P3 665 2 8 384-15 PA3955 P4 0 384-15 BLANK P5 606 5.1 20.4 384-15 PA4399 P6 0 384-15 BLANK P7 808 1.2 4.8 384-15 PA5323 P8 0 384-15 BLANK P9 405 0.57 2.28 384-15 PA5169 P10 0 384-15 BLANK P11 0 384-15 BLANK P12 0 384-15 BLANK P13 0 384-15 BLANK P14 0 384-15 BLANK P15 0 384-15 BLANK P16 0 384-15 BLANK P17 0 384-15 BLANK P18 0 384-15 BLANK P19 0 384-15 BLANK P20 0 384-15 BLANK P21 0 384-15 BLANK P22 0 384-15 BLANK P23 0 384-15 BLANK P24 0 384-15 BLANK "#ofOs=" 640 152 "=Blanks" "Total# genes=" 5608 A P P E N D I X III: Genes up- and down-regulated in P. aeruginosa in response to various concentrations of ciprofloxacin 187 Genes which are Up-regulated in Response to Ciprofloxacin Fold change In PAO-H103 ORF Gene name Ciprofloxacin Function Description 0.1XMIC 0.3x MIC 1xMIC PA0595 OStA 1.24 1.38 Adaptation, protection 46% similar to organic solvent tolerance Protein OstA [E. coli]. PA 1008 bcp 3.02 2.96 3.33 Adaptation, protection 52% similar to bacterioferritin comigratory Protein [E. coli] PA 1432 lasl 1.19 1.43 Adaptation, protection 100% identical to Autoinducer synthesis Protein Lasl [P.aeruginosa] PA2622 cspD 2.09 1.37 Adaptation, protection 71% similar to cspD gene product of [E. coli]. PA3625 surE 1.38 Adaptation, protection 64% similar to SurE Protein [E. coli] PA4235 bfrA 1.52 Adaptation, protection 88% similar to bacterioferritin [P. putida] PA4743 rbfA 1.42 Adaptation, protection 68% similar to ribosome-binding factor A[E. coli] PA4760 dnaJ 1.34 1.37 Adaptation, protection 78% similar to dnaJ gene product [E. coli]. PA5241 PPX 1.47 2.11 Adaptation, protection 60% similar to ppx gene product of [E. coli] PA5338 spoT 1.42 1.65 Adaptation, protection 73% similar to (p)ppGpp 3'-pyrophosphohydrolase [E. coli] PA0331 ilvA1 1.81 1.61 Amino acid biosynthesis and metabolism 70% similar to threonine deaminase llvA [E. coli]. PA0390 metX 1.49 Amino acid biosynthesis and metabolism 67% similar to homoserine O-acetyltransferase MetX [Leptospira meyeri] PA0609 trpE 1.65 Amino acid biosynthesis and metabolism 100% identical to anthranilate synthase component I [P.aeruginosa] PA0649 trpG 5.95 Amino acid biosynthesis and metabolism 99% similar to anthranilate synthase component II TrpG [P.aeruginosa] PA0650 trpD 2.68 Amino acid biosynthesis and metabolism PA1684 1.59 1.41 Amino acid biosynthesis and metabolism 73% similar to E-2/E-2' Protein of [Klebsiella oxytoca] PA1750 1.30 1.21 Amino acid biosynthesis and metabolism 73% similar to E. coli aroF gene product. PA1756 cysH 1.27 Amino acid biosynthesis and metabolism 99% similar to PA CysH; 73% similar to A P S reductase [A. thaliana] PA2943 1.25 Amino acid biosynthesis and metabolism 69% similar to E. coli aroF gene product. PA3175 1.45 1.72 Amino acid biosynthesis and metabolism PA3537 argF 2.61 2.58 2.71 Amino acid biosynthesis and metabolism 64% similar to ornithine carbamoyltransferase [B. subtilis] PA4695 ilvH 1.34 1.42 Amino acid biosynthesis and metabolism 84% similar to Acetolactate synthase isozyme III [S. typhimurium] PA4756 carB 1.20 Amino acid biosynthesis and metabolism PA4759 dapB 1.58 1.27 1.41 Amino acid biosynthesis and metabolism 92% similar to P. syringae dihydrodipicolinate reductase. PA4930 air 1.66 Amino acid biosynthesis and metabolism 57% similar to S. typhimurium air biosynthetic alanine racemase. PA4976 aspC 1.32 Amino acid biosynthesis and metabolism 57% similar to Asp aminotransferase [Thermus aquaticus thermophilus] PA5013 ilvE 1.40 Amino acid biosynthesis and metabolism 68% similar to branched-chain amino acid aminotransferase (ilvE) [E. coli] PA5035 gito 1.54 Amino acid biosynthesis and metabolism 100% identical to gltD from P A 0 1 ; 80% similar to gltD from E. coli PA5036 gltB 1.27 1.38 Amino acid biosynthesis and metabolism 99% identical to gltB from P A 0 1 ; 75% similar to gltB from E. coli PA5039 aroK 1.18 1.41 Amino acid biosynthesis and metabolism 71% similar to E. coli shikimate kinase I (AroK) PA5066 hisl 1.98 2.58 Amino acid biosynthesis and metabolism 94% similar to N-terminal of phosphoribosyl c-AMP hydrolase [A.chroococ PA5141 hisA 1.27 1.47 Amino acid biosynthesis and metabolism 49% similar to E. coli hisA gene product. PA5142 hisM 1.74 Amino acid biosynthesis and metabolism 50% similar to hisH gene product of [Azospirillum brasilense] PA5172 arcB 1.60 Amino acid biosynthesis and metabolism PA5302 dadX 1.13 Amino acid biosynthesis and metabolism 65% similar to catabolic alanine racemase of [E. coli]. PA5323 argB 1.37 1.60 Amino acid biosynthesis and metabolism 61% similar to B. stearothermophilus acetylglutamate kinase. PA5413 ItaA 1.21 Amino acid biosynthesis and metabolism 94% similar to low specificity L-threonine aldolase [P. sp.]. PA0426 mexB 1.40 Antibiotic resistance and susceptibility 99% similar to MexB [P.aeruginosa] PA0350 folA 1.46 Biosynthesis of cofactors, prosthetic groups and carriers 61 % similar to dihydrofolate reductase type I [E. coli] PA0381 thiG 1.47 Biosynthesis of cofactors, prosthetic groups and carriers 64% similar to thiamin biosynthesis, thiazole moiety ThiG [E. coli] PA0551 epd 1.47 1.66 Biosynthesis of cofactors, prosthetic groups and carriers 72% similar to gapB gene product of E. coli PA1505 moaA2 1.20 Biosynthesis of cofactors, prosthetic groups and carriers 56% similar to molybdopterin co-factor synthesis Protein MoaA [A.nicotin PA1546 hemN 1.63 -1.78 Biosynthesis of cofactors, prosthetic groups and carriers 99% similar to oxygen-independent coproporphyrinogen III dehyrogenase PA1674 folE2 3.00 3.14 3.47 Biosynthesis of cofactors, prosthetic groups and carriers 70% similar to G T P cyclohydrolase I [Rattus norvegicus] oo vo PA1758 pabB 1.71 1.51 1.68 Biosynthesis of cofactors, prosthetic groups and carriers 60% similar to p-aminobenzoate synthetase, component I PabB [S. typh PA1796 folD 1.15 1.23 Biosynthesis of cofactors, prosthetic groups and carriers 80% similar to E. coli folD gene product PA2908 cbiD 1.64 Biosynthesis of cofactors, prosthetic groups and carriers 57% similar to putative cobalamin biosynthesis Protein CbiD [B.mega PA2909 1.53 1.84 Biosynthesis of cofactors, prosthetic groups and carriers 55% similar to putative precorrin-6x reductase CobK [R.erythropolis] PA2947 1.43 Biosynthesis of cofactors, prosthetic groups and carriers 49% similar to cobE Protein [P. denitrificans]. PA3029 moaB2 1.32 Biosynthesis of cofactors, prosthetic groups and carriers 73% similar to molybdopterin biosynthesis. Protein B [E. coli]. PA3030 1.24 1.19 Biosynthesis of cofactors, prosthetic groups and carriers 47% similar to molybdoterin-guanine dinucleotide biosynthesis Protein A PA3627 ygbB 2.84 Biosynthesis of cofactors, prosthetic groups and carriers 79% similar to ygbB gene product of [E. coli] PA3650 dxr 1.89 3.68 3.37 Biosynthesis of cofactors, prosthetic groups and carriers 71 % similar to yaeM gene product of [E. coli] PA3915 moaB1 1.22 Biosynthesis of cofactors, prosthetic groups and carriers 72% similar to E. coli moaB gene product PA3916 moaE 8.36 7.90 7.69 Biosynthesis of cofactors, prosthetic groups and carriers 70% similar to molybdopterin converting factor subunit 2 [E. coli] PA3976 thiE 4.34 2.04 3.17 Biosynthesis of cofactors, prosthetic groups and carriers 52% similar to thiamine-phosphate pyrophosphorylase ThiC [B. subtilis]. PA3977 hemL 1.57 1.67 Biosynthesis of cofactors, prosthetic groups and carriers PA3997 lipB 1.54 1.52 1.76 Biosynthesis of cofactors, prosthetic groups and carriers 70% similar to lipoate biosynthesis Protein B [E. coli] PA4007 proA 1.94 Biosynthesis of cofactors, prosthetic groups and carriers 68% similar to gamma-glutamylphosphate reductase [E. coli] PA4047 ribA 1.46 Biosynthesis of cofactors, prosthetic groups and carriers 76% similar to GTP cyclohydrolase II of [E. coli] PA4280 birA 1.70 Biosynthesis of cofactors, prosthetic groups and carriers 56% similar to bifunctional Protein BirAJE. coli]. PA4529 coaE 1.31 Biosynthesis of cofactors, prosthetic groups and carriers 83% similar to yacE DNA repair Protein [P. putida] PA4561 ribF 1.30 1.44 Biosynthesis of cofactors, prosthetic groups and earners 83% similar to Riboflavin kinase (flavokinase) [P. fluorescens] PA4893 ureG 1.16 Biosynthesis of cofactors, prosthetic groups and earners 83% similar to urease accessory Protein G [Klebsiella aerogenes]. PA4919 pncB1 1.33 Biosynthesis of cofactors, prosthetic groups and carriers 69% similar to pncB gene product of S. typhimurium PA4920 nadE 1.17 Biosynthesis of cofactors, prosthetic groups and carriers 65% similar to nadE gene product of E. coli PA5063 ubiE 1.18 1.31 Biosynthesis of cofactors, prosthetic groups and carriers 84% similar to ubiE gene product of [E. coli] PA5065 ubiB 1.69 Biosynthesis of cofactors, prosthetic groups and carriers 70% similar to putative ubiquinone biosynthesis Protein AarF [E. coli] PA5223 ubiH 1.19 Biosynthesis of cofactors, prosthetic groups and carriers 56% similar to ubiH gene product of [E. coli] PA5243 hemB 1.28 Biosynthesis of cofactors, prosthetic groups and carriers 77% similar to hemB gene product of [Bradyrhizobium japonicum] PA5516 pdxY 1.60 Biosynthesis of cofactors, prosthetic groups and carriers 72% similar to pyridoxamine kinase [E. coli]. PA0208 mdcA 1.21 Carbon compound catabolism 85% similar to malonate decarboxylase alpha subunit MdcA [K. pneumon PA0214 1.49 Carbon compound catabolism 73% similar to malonate decarboxylase transacylase component MdcH PA0228 pcaF 1.29 Carbon compound catabolism 91% similar to PcaF [P. putida]. PA0230 pcaB 1.69 Carbon compound catabolism 87% similar to 3-carboxy-cis,cis-muconate cycloisomerase [P. putida]. PA0231 pcaD 1.34 Carbon compound catabolism 63% similar to B-ketoadipate enol-lactone hydrolase [B. japonicum]. PA0555 fda 1.22 1.24 1.26 Carbon compound catabolism 97% similar to fda gene product of [P. stutzeri] PA0608 1.55 Carbon compound catabolism 57% similar to phosphoglycolate phosphatase [E. coli]. PA0792 prpD 1.42 Carbon compound catabolism 73% similar to prpD gene products of E. coli and S. typhimurium PA0810 2.55 Carbon compound catabolism 41% similar to Had1, 2-haloalkanoic acid dehalogenase I [P. sp.] PA1950 rbsK 1.43 Carbon compound catabolism Similar to other ribokinase genes from multiple genera PA2098 1.27 Carbon compound catabolism 56% similar to acetylhydrolase [Streptomyces hygroscopicus] PA2321 1.36 1.38 Carbon compound catabolism 57% similarity to gntK; 55% similarity to gntV gene products of [E. coli] PA2343 mtlY 1.88 Carbon compound catabolism 76% similar to xylulose kinase [P. fluorescens]. PA2509 catB 1.44 1.54 1.76 Carbon compound catabolism 83% similar to muconate cycloisomerase I [P. putida]. PA2512 antA 1.68 Carbon compound catabolism 85% similar to anthranilate dioxygenase large subunit [Acinetobacter sp.] PA2517 xylY 1.67 1.70 Carbon compound catabolism 89% similar to toluate 1,2-dioxygenase beta subunit [Plasmid pWWO]. PA4901 mdIC 1.39 Carbon compound catabolism 76% similar to benzoylformate decarboxylase [P. putida] PA4905 vanB 1.84 1.79 1.91 Carbon compound catabolism 69% similar to vanillate O-demethylase oxidoreductase [P. sp.]. PA5057 phaD 1.41 Carbon compound catabolism 100% similar to PHA-depolymerase [P.aeruginosa] PA5350 1.33 1.77 1.82 Carbon compound catabolism 86% similar to rubredoxin of [Acinetobacter calcoaceticus] PA5416 soxB 1.72 1.61 1.40 Carbon compound catabolism 75% similar to sarcosine oxidase beta subunit [Corynebacterium sp.]. VO o PA0857 bolA 1.82 1.86 Cell division 64% similar to BolA Protein [E. coli] PA4479 mreD 2.92 Cell division 54% similar to mreD [E. coli] PA4751 ftsH 1.22 1.43 Cell division 81 % similar to FtsH gene product of E. coli PA4752 ftsJ 1.52 Cell division 74% similar to FtsJ gene product of E. coli PA4941 hftC 1.79 1.76 2.61 Cell division 59% similar to HfIC [E. coli]. PA4942 hflK 1.51 1.94 2.61 Cell division 59% similar to Protein HflK [E. coli]. PA5565 gidA 1.52 Cell division 81% similar to glucose-inhibited division Protein A [E. coli] PA3160 wzz 1.51 1.61 Cell wall / LPS / capsule PA3999 dacC 1.36 Cell wall / LPS / capsule 66% similar to D-alanyl-D-alanine carboxypeptidase; PBP6 (DacC) PA4002 rodA 1.77 Cell wall / LPS / capsule 73% similar to the rod-shape-determining Protein of [E. coli]. PA4003 pbpA 1.46 Cell wall / LPS / capsule 63% similar to the Penicillin-binding Protein 2 of [E. coli]. PA4418 ftsl 1.06 Cell wall / LPS / capsule 45% identical, 62% similar to E. coli pbp3 PA4545 comL 1.81 Cell wall / LPS / capsule 62% similar to peptidoglycan-linked lipoprotein precursor ComL [N.gono PA4662 murl 1.37 1.44 Cell wall / LPS / capsule 58% similar to glutamate racemase[E. coli] PA4700 mrcB 2.09 2.33 2.15 Cell wall / LPS / capsule 61% similar to peptidoglycan synthetase; penicillin-binding Protein 1B PA4947 amiB 1.51 Cell wall / LPS / capsule 56% similar to N-acetylymuramoyl-l-alanine amidase AmiC precursor PA4988 waaA 1.29 Cell wall / LPS / capsule 67% similar to Kdo transferase WaaA (former kdtA) [E. coli] PA5009 waaP 1.40 1.49 Cell wall / LPS / capsule 72% similar to waaP (rfaP) gene product of [E. coli]. PA5010 waaG 1.58 1.91 2.16 Cell wall / LPS / capsule 66% similar to rfaG gene product of [E. coli]. PA5011 waaC 1.33 1.42 Cell wall / LPS / capsule 99% similar to heptosyl transferase I [P.aeruginosa]. PA5012 waaF 2.78 Cell wall / LPS / capsule 100% similar to heptosyl transferase II [P.aeruginosa]. PA5077 mdoH 2.48 2.32 2.16 Cell wall / LPS / capsule 76% similar to mdoH [E.coli] PA5161 rmlB 1.32 1.60 Cell wall / LPS / capsule 73% similar to rfbB gene product of [E. coli] PA5163 rmlA 1.18 1.33 Cell wall / LPS / capsule 87% similar to rfbA gene product of [E. coli] PA5164 rmIC 1.20 1.34 Cell wall / LPS / capsule 75% similar to rfbC gene product of [E. coli] PA5276 IppL 3.31 Cell wall / LPS / capsule PA0654 speD 1.70 Central intermediary metabolism 81 % similar to S-adenosylmethionine decarboxylase SpeD [E. coli]. PA2356 msuD 1.56 Central intermediary metabolism PA2393 1.75 Central intermediary metabolism PA3471 1.99 Central intermediary metabolism 80% similar to putative NAD-linked malate dehydrogenase (malic enzyme PA4068 1.93 Central intermediary metabolism 60% similar to hypothetical UDP-glucose 4-epimerase of [P. horikoshii] PA4626 hprA 1.37 1.45 Central intermediary metabolism 61% similarity to hprA gene product of [Methylobacterium extorquens]. PA4864 ureD 2.20 2.18 Central intermediary metabolism 51% similar to urease accessory Protein [Klebsiella aerogenes]. PA4868 ureC 1.55 Central intermediary metabolism 83% similar to urease subunit C [Klebsiella aerogenes]. PA5060 phaF 1.13 1.29 Central intermediary metabolism 100% identical to fragment of hypothetical Protein 4 (phaC2 3' region) PA5421 fdhA 1.46 Central intermediary metabolism 92% similar to glutathione-independent formaldehyde dehydrogenase PA5435 1.27 1.23 Central intermediary metabolism 75% similar to L.pneumophila oadA homolog PA5436 1.36 Central intermediary metabolism 71 % similar to biotin carboxylase [B. subtilis] PA0538 dsbB 2.18 Chaperones & heat shock proteins 51 % similar to disulfide bond formation Protein DsbB [E. coli]. PA1068 2.77 2.50 2.71 Chaperones & heat shock proteins 47% similar to heat shock Protein HtpG, chaperone Hsp90, heat shock PA2477 1.89 1.68 Chaperones & heat shock proteins C-terminal portion 48% similar to putative thiol PA2614 lolA 1.31 Chaperones & heat shock proteins 51% similar to lipoprotein carrier Protein precursor LolA [E. coli]. PA3221 csaA 2.05 2.17 5.11 Chaperones & heat shock proteins 68% similar to molecular chaperonin CsaA [B. subtilis]. PA4083 1.36 Chaperones & heat shock proteins 57% similar to FimB chaperone-like Protein [Bordetella pertussis]. PA4176 ppiC2 1.55 1.58 Chaperones & heat shock proteins 67% similar to ppiC gene product [E. coli] PA4385 groEL 1.30 Chaperones & heat shock proteins PA4386 groES 2.03 2.12 Chaperones & heat shock proteins PA4761 dnaK 2.09 1.81 2.59 Chaperones & heat shock proteins 87% similar to hsp70 Protein (dnaK gene) [E. coli] PA4762 grpE 1.29 2.12 Chaperones & heat shock proteins 63% similar to grpE gene product [E. coli] PA4845 dipZ 1.42 Chaperones & heat shock proteins 94% similar to PA dipZ gene product. PA5053 hsIV 1 7 7 Chaperones & heat shock proteins 100% similar to heat shock Protein hslV[E. coli] PA5054 hsIU 1.93 2.03 3.27 Chaperones & heat shock proteins 84% similar to heat shock Protein HsIU [E. coli]. PA5195 2.15 2.19 2.13 Chaperones & heat shock proteins 67% similar to no-longer-hypothetical yrfH gene product [E. coli] PA5256 dsbH 1.28 1.54 Chaperones & heat shock proteins 51% similar to dsbB gene product (disulfide oxidoreductase) of [B.cepa PA0180 1.35 Chemotaxis 43% similar to C-terminus of aerotaxis sensor receptor [E. coli] PA0415 1.43 Chemotaxis PA2788 1.38 1.33 Chemotaxis 56% similar to hypothetical transducer [P.aeruginosa]. PA4844 1.47 1.28 Chemotaxis 53% similar to chemotactic transducer [P.aeruginosa] PA4954 motA 1.54 Chemotaxis 67% similar to MotA [S. typhimurium] PA0382 micA 1.25 DNA replication, recombination, modification and repair 74% similar to [micA] gene products [E. coli] PA0577 dnaG 2.10 DNA replication, recombination, modification and repair 80% similar to DnaG gene product of [P. putida] PA1534 recR 1.30 DNA replication, recombination, modification and repair 79% similar to recombination Protein RecR [E. coli] PA1686 alkA 1.44 DNA replication, recombination, modification and repair 52% similar to (inducible) 3-methyl-adenine D N A glycosylase II [E. coli]. PA1815 rnhA 1.75 DNA replication, recombination, modification and repair 76% similar to ribonuclease H [E. coli] PA3617 recA 1.86 2.83 4.55 DNA replication, recombination, modification and repair PA3620 mutS 1.82 1.49 DNA replication, recombination, modification and repair 93% similar to DNA mismatch repair Protein MutS [Azotobacter vinelandi PA3640 dnaE 1.19 DNA replication, recombination, modification and repair 73% similar to dnaE gene product of [E. coli] PA3777 xseA 1.48 DNA replication, recombination, modification and repair 61 % similar to xseA gene product [E. coli] PA3989 holA 1.25 DNA replication, recombination, modification and repair 50% similar to holA gene product of [E. coli] PA4232 ssb 1.99 DNA replication, recombination, modification and repair PA4234 uvrA 1.47 DNA replication, recombination, modification and repair 87% similar to E. coli uvrA gene product. PA4316 sbcB 1.15 DNA replication, recombination, modification and repair 66% similar to exonuclease I [E. coli] PA4763 recN 1.59 2.11 DNA replication, recombination, modification and repair 61% similar to recN gene product [E. coli] PA4931 dnaB 1.33 1.37 DNA replication, recombination, modification and repair 79% similar to dnaB gene product of [E. coli]. PA4946 mutL 1.51 D N A replication, recombination, modification and repair 60% similar to mismatchrepair Protein mutL [S. typhimurium] PA5050 priA 2.00 2.22 D N A replication, recombination, modification and repair 64% similar to priA gene product (primosomal Protein N') of [E. coli] PA5147 mutY 1.20 1.48 D N A replication, recombination, modification and repair 67% similar to mutY gene product of [E. coli] PA5296 rep 1.61 D N A replication, recombination, modification and repair 71 % similar to rep gene product of [E. coli] PA5319 radC 1.60 1.55 D N A replication, recombination, modification and repair 68% similar to DNA repair Protein RadC of [E. coli] PA0330 rpiA 1.45 1.90 Energy metabolism 78% similar to rpiA gene product of [E. coli] PA0362 fdxl 2.39 Energy metabolism 85% similar to ferredoxin [Chromatium vinosum] PA0511 nirJ 1.32 Energy metabolism PA0512 1.21 Energy metabolism 79% similar to NirH Protein [P. stutzeri]. PA0519 nirS 1.30 -1.54 Energy metabolism 100% identical to NITRITE R E D U C T A S E P R E C U R S O R [P.aeruginosa] PA0525 1.33 Energy metabolism 100% identical to hypothetical Protein downstream of norCB [P.aerugino PA0552 pgk 1.37 Energy metabolism 82% similar to pgk gene product of E. coli PA0607 rpe 1.27 1.35 1.89 Energy metabolism 84% similar to the rpe gene product of [E. coli] PA1481 ccmG 2.44 Energy metabolism 83% similar to Thiol PA1482 ccmH 1.56 Energy metabolism 74% similar to Cytochrome C-type biogenesis Protein CycL [P. fluoresc PA1483 cycH 2.04 Energy metabolism 70% similar to Cytochrome C-type biogenesis Protein CycH [P. fluoresc PA1551 1.27 1.24 Energy metabolism 56% similar to FixG Protein [Bradymizobium japonicum] PA1555 1.06 -2.02 Energy metabolism 51 % similar to C c o P [Paracoccus denitrificans] PA1556 1.63 1.89 Energy metabolism 64% similar to fixO Protein - Rhizobium meliloti PA1557 • '1.25 -1.60 Energy metabolism 78% similar to the CytN gene product of [Azospirillum brasilense]. PA1581 sdhC 1.48 1.58 1.30 Energy metabolism 72% similar to sdhC gene product of [E. coli] PA1582 sdhD 1.52 1.50 1.54 Energy metabolism 61 % similar to sdhD gene product of [E. coli] to PA1583 sdhA 1.24 Energy metabolism 81% similar to sdhA gene product of [E. coli]. PA1584 sdhB 1.26 Energy metabolism 83% similar to sdhB gene product of [E. coli]. PA1587 IpdG 1.29 -1.40 Energy metabolism -100% identity with lipoamide dehydrogenase from P. fluorescens PA1883 1.00 1.11 Energy metabolism 73% similar to NADH dehydrogenase I chain A NuoA [E. coli]. PA2290 gcd 1.59 Energy metabolism 65% similar to gcd gene product of [E. coli] PA2297 3.70 4.54 Energy metabolism PA2637 nuoA 1.66 2.02 Energy metabolism 72% simialr to NADH dehydrogenase I chain A [E. coli] PA2640 nuoE 1.42 Energy metabolism 81% similar to NADH dehydrogenase I chain E [E. coli] PA2644 nuol 1.15 1.62 Energy metabolism 85% similar to NADH dehydrogenase I chain I [E. coli] PA2649 nuoN 1.79 Energy metabolism 81% similar to NADH dehydrogenase I chain N [E. coli]. PA2995 nqrE 1.06 Energy metabolism 84% similar to Na-translocating NADH PA2999 nqrA 3.02 2.01 2.46 Energy metabolism 72% similar to NADH PA3491 2.85 2.54 Energy metabolism 58% similar to RnfC Protein [Rhodobacter capsulatus] PA3873 narj 2 7 8 3.60 Energy metabolism 63% SIMILAR T O E. COLI narJ gene product. PA3875 narG 3.63 3.21 Energy metabolism 83% similar to E. coli narG gene product. PA4061 1.58 Energy metabolism 52% similar to hypothetical ybbN gene product of [E. coli] PA4571 1.27 Energy metabolism 62% similar to cytochrome c of [Acetobacter europaeus] PA4640 mqoB 1.12 Energy metabolism 69% similar to L-malate dehydrogenase (acceptor) [C. glutamicum] PA4748 tpiA 1.28 1.82 Energy metabolism 67% similar to Tpi gene product of [E. coli] PA4771 lldD 1.37 Energy metabolism 92% similar to L-lactate dehydrogenase [E. coli] PA4809 fdhE 1.59 Energy metabolism 55% similar to fdhE Protein [E. coli] PA4810 fdnl 1.21 Energy metabolism 65% similar to formate dehydrogenase-N gamma subunit [E. coli] PA4812 fdnG 1.20 Energy metabolism 79% similar to formate dehydrogenase-O, major subunit [E. coli] PA4922 azu 1.40 Energy metabolism 100% similar to azu gene product of P.aeruginosa PA4971 aspP 1.61 Energy metabolism 66% similar to adenosine diphosphate sugar pyrophosphatase [E. coli] PA5015 aceE 1.37 1.60 Energy metabolism PA5016 aceF 1.29 Energy metabolism PA5300 cycB 1.46 1.72 Energy metabolism 80% similar to cytochrome c5 of [Azotobacter vinelandii]. PA5332 crc 1.25 1.60 Energy metabolism PA5490 cc4 1.40 1.47 Energy metabolism 82% similar to cytochrome c4 precursor [P. stutzeri]. PA5491 1.61 3.00 Energy metabolism 56% similar to cytochrome c5 [Azotobacter vinelandii]. PA5559 atpE 1.45 1.24 Energy metabolism 79% similar to atpE gene product of E. coli PA5560 atpB 1.24 Energy metabolism 66% similar to atpB gene product of E. coli PA5561 atpl 1.63 Energy metabolism 79% similar to atpl homolog of P. putida; only 53% similar to E. coli atpl PA0005 1.39 Fatty acid and phospholipid metabolism 56% similar to putative 1-acyl-sn-glycerol-3-phosphate acyltransferase PA0286 1.89 Fatty acid and phospholipid metabolism 85% similar to stearoyl-CoA desaturase [Azotobacter vinelandii] PA3013 foaB 1.36 Fatty acid and phospholipid metabolism 95% similar to fatty-acid oxidation complex beta-subunit [P. fragi]. PA3673 pIsB 1.93 Fatty acid and phospholipid metabolism 58% similar to E. coli PIsB, sn-glycerol-3-phosphate acyltransferase PA4813 HpC 2.00 1.88 Fatty acid and phospholipid metabolism 100% identical to lipase LipC [P.aeruginosa] PA0030 5.52 5.99 5.61 Hypothetical, unclassified, unknown PA0046 1.38 1.87 Hypothetical, unclassified, unknown PA0053 1.43 1.74 Hypothetical, unclassified, unknown PA0054 1.52 Hypothetical, unclassified, unknown 74% similar to hypothetical Protein Yjil [E. coli]. PA0055 1.34 Hypothetical, unclassified, unknown PA0061 2.25 Hypothetical, unclassified, unknown PA0064 1.32 Hypothetical, unclassified, unknown PA0065 1.64 Hypothetical, unclassified, unknown 54% similar to hypothetical Protein Rv2233 [M. tuberculosis] PA0068 1.60 Hypothetical, unclassified, unknown VO PA0070 1.95 Hypothetical unclassified, unknown PA0071 2.05 Hypothetical unclassified, unknown PA0072 1.28 Hypothetical unclassified, unknown PA0076 1.41 1.67 Hypothetical unclassified, unknown PA0077 2.13 Hypothetical unclassified, unknown PA0081 1.40 Hypothetical unclassified, unknown PA0082 1.53 Hypothetical unclassified, unknown PA0083 3.47 Hypothetical unclassified, unknown PA0084 1.81 Hypothetical unclassified, unknown PA0085 1.48 2.30 Hypothetical unclassified, unknown PA0086 1.24 2.03 Hypothetical unclassified, unknown PA0087 2.04 Hypothetical unclassified, unknown PA0092 1.28 1.99 Hypothetical unclassified, unknown PA0093 1.61 Hypothetical unclassified, unknown PA0094 2.30 Hypothetical unclassified, unknown PA0095 1.75 3.23 Hypothetical unclassified, unknown PA0104 1.59 Hypothetical unclassified, unknown PA0126 1.40 Hypothetical unclassified, unknown PA0127 1.33 Hypothetical unclassified, unknown PA0160 1.42 1.50 1.54 Hypothetical unclassified, unknown PA0201 1.36 Hypothetical unclassified, unknown PA0222 1.43 Hypothetical unclassified, unknown PA0250 2.93 3.57 Hypothetical unclassified, unknown PA0255 1.45 1.80 Hypothetical unclassified, unknown PA0277 2.11 3.62 Hypothetical unclassified, unknown PA0290 1.21 1.20 Hypothetical unclassified, unknown PA0309 3.31 2.84 3.60 Hypothetical unclassified, unknown PA0315 1.39 Hypothetical unclassified, unknown PA0317 2.17 2.21 2.40 Hypothetical unclassified, unknown PA0319 1.74 Hypothetical unclassified, unknown PA0332 1.41 1.81 Hypothetical unclassified, unknown PA0338 1.44 1.56 Hypothetical unclassified, unknown PA0340 1.46 1.44 1.55 Hypothetical unclassified, unknown PA0344 1.14 Hypothetical unclassified, unknown PA0351 1.57 Hypothetical unclassified, unknown PA0387 1.29 3.84 Hypothetical unclassified, unknown PA0392 1.12 1.63 Hypothetical unclassified, unknown PA0394 1.48 Hypothetical unclassified, unknown PA0422 1.42 1.73 Hypothetical unclassified, unknown PA0423 1.28 Hypothetical unclassified, unknown PA0446 1.41 Hypothetical unclassified, unknown PA0451 1.27 Hypothetical unclassified, unknown PA0454 1.59 1.74 Hypothetical unclassified, unknown PA0457 1.41 Hypothetical unclassified, unknown PA0460 1.30 1.61 Hypothetical unclassified, unknown PA0462 1.38 Hypothetical unclassified, unknown PA0466 1.37 1.55 Hypothetical unclassified, unknown PA0541 1.85 2.02 2.17 Hypothetical unclassified, unknown 42% similar to a region of endo-xylanase XylR [B. stearothermophilus]. 40% similar to IcmF Protein [Legionella pneumophila]. 41% similar to a region of circumsporozoite Protein [Plasmodium berghei]. 61% similar to putative 19.5 kDa Protein [Edwardsiella ictaluri]. 66% similar to putative 54.5 kDa Protein [Edwardsiella ictaluri]. 47% similar to putative 17.8 kDa Protein [Edwardsiella ictaluri] 45% similar to putative 18.8 kDa Protein [Edwardsiella ictaluri]. 50% similar to VgrE and VgrG Proteins [E. coli]. 49% similar to hypothetical Protein [Clostridium beijerinckii]. 50% similar to putative mannopine-binding periplasmic Protein MotA 60% similar to hypothetical Protein Rv2406c [M. tuberculosis]. 65% similar to hypothetical Protein [Sinorhizobium meliloti] 74% similar to hypothetical Proteins YcaL [E. coli] and YggG (E. coli]. 54% similar to C-terminus of PleD [Caulobacter crescentus] 53% similarity to Aip2 gene product of Saccharomyces cerevisiae. 50% similar to pleD gene product of [Synechocystis sp.] 57% similar to H. influenzae hypothetical Protein HI0902. 57% similar to conserved Protein [Methanobacterium thermoautotrophic 70% similar to putative ribosomal Protein YggV [E. coli) 65% similar to hypothetical Protein [Vibrio alginolyticus] 100% identical to hypothetical 24.5kD Protein; PilT [P.aeruginosa] 78% similar to putative cytochrome [E. coli] 84% similar to hypothetical Protein Ycel [E. coli] 68% similar to an unknown Protein from [Sphingomonas aromaticivorans] 65% similar to NfeD gene product of [Rhizobium etli] 58% similar to hypothetical Protein YccS [E. coli]. 4^ PA0554 1.97 Hypothetical unclassified, unknown PA0559 1.28 1.51 Hypothetical unclassified, unknown 70% similar to YhiN gene products of E. coli and H. influenzae PA0563 1.95 6.50 Hypothetical unclassified, unknown 64% similarity to YhaH gene product of E. coli PA0565 1.62 Hypothetical unclassified, unknown 79% similar to conserved hypothetical Protein HI1053 [H. influenzae Rd]. PA0587 1.73 Hypothetical unclassified, unknown 75% similar to hypothetical Protein YeaH [E. coli] PA0596 1.17 Hypothetical unclassified, unknown 41% similar to orfT [Chlorobium tepidum]. PA0612 8.45 Hypothetical unclassified, unknown 59% similar to hypothetical Protein Ybil [E. coli] PA0613 2.06 3.99 19.29 Hypothetical unclassified unknown PA0614 3.32 15.33 Hypothetical unclassified unknown PA0615 1.72 7.21 Hypothetical unclassified unknown PA0653 2.65 Hypothetical unclassified unknown 69% similar to hypothetical Protein YhfA [E. coli]. PA0663 1.49 Hypothetical unclassified unknown PA0670 1.21 1.96 Hypothetical unclassified unknown PA0671 1.52 Hypothetical unclassified unknown 52% similar to cell division inhibitor SulA [Serratia marcescens]. PA0673 1.17 1.47 Hypothetical unclassified unknown PA0737 1.70 6.14 Hypothetical unclassified unknown PA0790 1.73 Hypothetical unclassified unknown 42% similar to hypothetical Protein [Haemophilus influenzae Rd]. PA0808 2.44 Hypothetical unclassified unknown PA0819 1.47 1.37 1.72 Hypothetical unclassified unknown PA0822 2.15 Hypothetical unclassified unknown PA0907 1.87 Hypothetical unclassified unknown PA0908 3.39 Hypothetical unclassified unknown PA0909 1.65 7.14 Hypothetical unclassified unknown 57% similar to orflO [P.aeruginosa phage phi CTX] PA0910 5.83 Hypothetical unclassified unknown PA0911 6.18 Hypothetical unclassified unknown PA0912 1.74 7.30 Hypothetical unclassified unknown PA0921 1.52 Hypothetical unclassified unknown PA0940 2.28 Hypothetical unclassified unknown PA0947 4.84 Hypothetical unclassified unknown 74% similar to putative D N A replication factor [E. coli] PA1055 1.49 Hypothetical unclassified unknown 71% similar to phaC [Sinorhizobium meliloti] PA1058 2.23 1.88 Hypothetical unclassified unknown 74% similar to phaF [Sinorhizobium meliloti]. PA1061 1.99 Hypothetical unclassified unknown 72% similar to hypothetical Protein Rv3684 [M. tuberculosis] PA1063 2.22 2.15 Hypothetical unclassified unknown PA1064 1.72 Hypothetical unclassified unknown PA1065 1.23 Hypothetical unclassified unknown 69% similar to hypothetical gene product yeaO of [E. coli] PA1069 1.69 1.51 Hypothetical unclassified unknown PA1091 2.11 Hypothetical unclassified unknown 53% similar to C-terminal fragment of RbfC [Riftia pachyptila endosymbio PA1096 2.61 Hypothetical unclassified unknown 67%) similarity to putative Protein (Orf4) [P.aeruginosa strain DG1] PA1152 3.90 Hypothetical unclassified unknown PA1154 1.60 Hypothetical unclassified unknown 65% similar to hypothetical Protein YafM [E. coli] PA1166 1.50 Hypothetical unclassified unknown 50% similar to small basic Protein SbpA [Legionella pneumophila] PA1213 1.39 Hypothetical unclassified unknown PA1307 1.39 Hypothetical unclassified unknown 75% similar to hypothetical Protein YafJ [E. coli] PA1374 1.55 Hypothetical unclassified unknown 51% similar to hypothetical Protein Rv3095 [M. tuberculosis]. PA1415 1.38 Hypothetical unclassified unknown PA1433 2.21 Hypothetical unclassified unknown 43% similar to c-di-GMP phosphodiesterase A [Acetobacter xylinus] PA1494 2.13 Hypothetical unclassified unknown 43% similar to hypothetical Protein AF068721 [Caenorhabditis elegans] PA1499 2.74 2.84 3.08 Hypothetical unclassified unknown 65% similar to putative hydroxypyruvate reductase; inducible by tartrate VO PA1509 1.26 Hypothetical unclassified, unknown PA1530 1.59 Hypothetical unclassified, unknown PA1536 3.08 3.27 Hypothetical unclassified, unknown PA1539 1.85 1.50 1.69 Hypothetical unclassified, unknown PA1540 1.64 Hypothetical unclassified, unknown PA1548 1.77 Hypothetical unclassified, unknown PA1550 1.38 1.42 Hypothetical unclassified, unknown PA1573 1.23 1.39 1.48 Hypothetical unclassified, unknown PA1577 3.00 2.73 2.24 Hypothetical unclassified, unknown PA1612 1.76 1.83 Hypothetical unclassified, unknown PA1645 1.68 Hypothetical unclassified, unknown PA1652 1.50 Hypothetical unclassified, unknown PA1657 1.68 Hypothetical unclassified, unknown PA1699 4.88 3.14 Hypothetical unclassified, unknown PA1702 1.93 Hypothetical unclassified, unknown PA1711 3.02 Hypothetical unclassified, unknown PA1761 1.42 Hypothetical unclassified, unknown PA1765 2.16 1.74 2.14 Hypothetical unclassified, unknown PA1767 1.46 1.45 Hypothetical unclassified, unknown PA1769 1.67 Hypothetical unclassified, unknown PA1774 1.49 1.81 1.76 Hypothetical unclassified, unknown PA1789 1.34 Hypothetical unclassified, unknown PA1791 1.65 2.03 Hypothetical unclassified, unknown PA1824 1.74 1.92 Hypothetical unclassified, unknown PA1825 1.36 Hypothetical unclassified, unknown PA1889 1.47 Hypothetical unclassified, unknown PA1891 1.15 Hypothetical unclassified, unknown PA1892 3.76 2.35 Hypothetical unclassified, unknown PA1893 1.16 Hypothetical unclassified, unknown PA1907 2.19 1.33 1.72 Hypothetical unclassified, unknown PA2077 2.06 1.91 Hypothetical unclassified, unknown PA2112 11.70 8.83 Hypothetical unclassified, unknown PA2120 1.84 1.91 Hypothetical unclassified, unknown PA2175 2.49 Hypothetical unclassified, unknown PA2207 2.59 Hypothetical unclassified, unknown PA2209 3.15 Hypothetical unclassified, unknown PA2243 1.59 Hypothetical unclassified, unknown PA2269 1.81 2.00 Hypothetical unclassified, unknown PA2274 11.95 6.56 10.50 Hypothetical unclassified, unknown PA2285 1.28 Hypothetical unclassified, unknown PA2287 1.85 2.23 2.70 Hypothetical unclassified, unknown PA2288 1.76 Hypothetical unclassified, unknown PA2292 3.52 2.98 3.98 Hypothetical unclassified, unknown PA2304 1.33 1.18 Hypothetical unclassified, unknown PA2353 1.66 Hypothetical unclassified, unknown PA2358 1.56 Hypothetical unclassified, unknown PA2361 2.51 2.04 Hypothetical unclassified, unknown PA2365 1.33 Hypothetical unclassified, unknown 55% similar to hypothetical Protein Rv2295 [M. tuberculosis]. 53% similar to a region of hypothetical Protein YijC [E. coli]. 80% similar to putative ethidium bromide resistance Protein [E. coli]. 64% similar to fixS gene product [Bradymizobium japonicum] 77% similar to hypothetical gene product yijF of [E. coli] 68% similar to unknown Protein [E. coii] 69% similar to Yop secretion and targeting control Protein [Y. pestis] 60% similar to YscY [Yersinia enterocolitica] 68% similar to hypothetical Protein YdiA [E. coli] 51% similar (with gaps) to hypothetical gene product ydaA of [E. coli] 54% similar to putative structural Protein YfcA [E. coli]. 41 % similar to penicillin acylase precursor [P. sp.]. 63% similar to an unknown O R F of [Bordetella pertussis] 57% similar to hypothetical Protein [Agrobacterium vitis]. 49% similar to hypothetical Protein Rv3537 [M. tuberculosis]. 66% similar to hypothetical Protein SC9B10.25c of [S.coelicolor]. 51% similar to syringomycin biosynthesis enzyme [P. syringae]. 51% similar to hypothetical Protein YhcM [E. coli]. 71% similar to putative 19.5 kDa Protein Eip20 [Edwardsiella ictaluri]. VO ON PA2380 2.38 2.32 1.93 Hypothetical unclassified, unknown PA2389 2.16 2.03 Hypothetical unclassified, unknown PA2404 2.12 1.79 Hypothetical unclassified, unknown PA2405 1.73 Hypothetical unclassified, unknown PA2406 1.38 Hypothetical unclassified, unknown PA2418 1.22 Hypothetical unclassified, unknown PA2455 2.94 Hypothetical unclassified, unknown PA2471 1.34 Hypothetical unclassified, unknown PA2569 2.24 2.18 2.31 Hypothetical unclassified, unknown PA2576 1.79 Hypothetical unclassified. unknown PA2604 1.41 Hypothetical unclassified, unknown PA2609 1.22 1.20 Hypothetical unclassified, unknown PA2625 1.27 Hypothetical unclassified unknown PA2632 5.12 Hypothetical unclassified unknown PA2685 1.56 Hypothetical unclassified unknown PA2708 3.22 Hypothetical unclassified unknown PA2720 1.64 Hypothetical unclassified unknown PA2728 1.60 1.65 Hypothetical unclassified unknown PA2729 1.23 Hypothetical unclassified unknown PA2746 2.54 Hypothetical unclassified unknown PA2761 1.85 Hypothetical unclassified unknown PA2765 1.57 Hypothetical unclassified unknown PA2772 1.45 1.41 1.97 Hypothetical unclassified unknown PA2773 1.91 Hypothetical unclassified unknown PA2782 1.32 Hypothetical unclassified unknown PA2785 1.31 Hypothetical unclassified unknown PA2786 1.21 Hypothetical unclassified unknown PA2792 1.36 Hypothetical unclassified unknown PA2793 1.14 Hypothetical unclassified unknown PA2797 1.32 Hypothetical unclassified unknown PA2808 1.92 Hypothetical unclassified unknown PA2880 1.78 Hypothetical unclassified unknown PA2883 2.23 Hypothetical unclassified unknown PA2898 1.65 Hypothetical unclassified unknown PA2902 1.77 2.00 Hypothetical unclassified unknown PA2916 1.27 Hypothetical unclassified unknown PA2919 1.67 Hypothetical unclassified unknown PA2937 2.48 Hypothetical unclassified unknown PA2980 5.78 Hypothetical unclassified unknown PA3008 2.77 5.22 6.50 Hypothetical unclassified unknown PA3009 3.15 4.20 Hypothetical unclassified unknown PA3010 1.33 Hypothetical unclassified unknown PA3017 1.53 Hypothetical unclassified unknown PA3023 1.09 Hypothetical unclassified unknown PA3033 1.28 Hypothetical unclassified unknown PA3040 1.40 Hypothetical unclassified unknown PA3066 1.74 1.82 Hypothetical unclassified unknown PA3084 1.27 Hypothetical unclassified unknown 51% similar to putative membrane Protein [E. coli]. 52% similar to nuclear Protein pirin [Homo sapiens]. 69% similar to hypothetical Protein [Sphingomonas sp. RW5] 48% similar to hypothetical Protein YxxF [B. subtilis]. 69% similar to putative carrier/transport Protein YccA [E. coli]. 49% similar to phosphoribosylanthranilate transferase [Aquifex aeolicus] 55% similar to hypothetical Protein YGL067w [E. coli). 52% similar to VgrG Protein [E. coli]; 51% similar to VgrE Protein [E. coli] 46% similar to a region of D N A helicase related Protein [M. thermoauto 52% similar to a region of hypothetical Protein [E. coli]. 73% similar to hypothetical Protein YozG [B. subtilis]. 42% similar to amino acid A B C transporter(yckK) [H. pylori 26695] 43% similar to hypothetical Protein YfiK [E. coli]. 75% similar to hypothetical Protein YcaR [E. coli]. C-terminal 120 amino acids 50% similar to sulA of [S. marcescens] 60% similar to hypothetical Protein 146 [Coxiella burnetii]. 59% similar to hypothetical Protein YegS [E. coli]. 60% similar to hypothetical Protein YqjD [E. coli]. 66% similar to hypothetical yecD gene product [E. coli] PA3085 1.86 1.62 Hypothetical unclassified, unknown PA3091 1.98 1.94 1.84 Hypothetical unclassified, unknown PA3178 2.10 Hypothetical unclassified, unknown PA3180 1.41 1.62 Hypothetical unclassified, unknown PA3196 1.42 Hypothetical unclassified, unknown PA3203 1.43 Hypothetical unclassified, unknown PA3208 1.37 Hypothetical unclassified, unknown PA3275 1.34 Hypothetical unclassified, unknown PA3306 1.28 Hypothetical unclassified, unknown PA3322 1.77 Hypothetical unclassified, unknown PA3369 1.53 Hypothetical unclassified, unknown PA3373 1.47 Hypothetical unclassified, unknown PA3379 1.49 Hypothetical unclassified, unknown PA3400 1.57 Hypothetical unclassified, unknown PA3413 1.62 2.08 4.18 Hypothetical unclassified, unknown PA3414 1.47 Hypothetical unclassified, unknown PA3432 2.25 Hypothetical unclassified, unknown PA3474 1.71 1.46 Hypothetical unclassified, unknown PA3499 1.27 Hypothetical unclassified, unknown PA3505 5.55 Hypothetical unclassified, unknown PA3530 3.00 Hypothetical unclassified, unknown PA3552 1.53 Hypothetical unclassified, unknown PA3580 1.72 Hypothetical unclassified, unknown PA3611 1.14 1.43 Hypothetical unclassified, unknown PA3616 2.05 3.09 Hypothetical unclassified, unknown PA3623 1.61 Hypothetical unclassified, unknown PA3634 1.67 1.57 1.26 Hypothetical unclassified, unknown PA3649 1.27 Hypothetical unclassified, unknown PA3661 1.34 Hypothetical unclassified, unknown PA3663 1.71 Hypothetical unclassified, unknown PA3716 1.25 Hypothetical unclassified, unknown PA3722 2.07 Hypothetical unclassified, unknown PA3726 1.20 Hypothetical unclassified, unknown PA3730 1.24 1.54 Hypothetical unclassified, unknown PA3731 1.49 1.42 1.52 Hypothetical unclassified, unknown PA3732 3.36 Hypothetical unclassified, unknown PA3748 1.26 Hypothetical unclassified, unknown PA3750 1.47 1.56 Hypothetical unclassified, unknown PA3756 1.78 2.02 Hypothetical unclassified, unknown PA3773 1.53 1.59 Hypothetical unclassified, unknown PA3775 1.33 1.55 Hypothetical unclassified, unknown PA3779 1.60 Hypothetical unclassified, unknown PA3784 1.34 1.59 Hypothetical unclassified, unknown PA3787 1.34 Hypothetical unclassified, unknown PA3789 1.37 Hypothetical unclassified, unknown PA3794 2.22 Hypothetical unclassified, unknown PA3796 1.34 1.77 Hypothetical unclassified, unknown PA3799 1.23 Hypothetical unclassified, unknown 72% similar to a region of hypothetical Protein Y e a C [E. coli]. 62% similar to E. coli ydjA hypothetical gene product. 56% similar to hypothetical Protein YnfA [E. coli]. 58% similar to HisP-like nucleotide binding Protein PhnN [E. coli] 56% similar to phnH gene product [E. coli] 45% similar to conserved hypothetical integral membrane Protein [H. pylor 55% similar to hypothetical Protein YebG [E. coli] 57% similar to hypothetical Protein YohJ [E. coli]. 70% similar to hypothetical Protein YigM [E. coli]. 48% similar to a conserved Protein of [M.thermoautotrophicum] 66% similar to Bacterioferritin-associated ferTedoxin Bfd [E. coli] 80% similar to hypothetical Protein YfbE [E. coli] 100% identical to OrfX of [PA01] 82% similar to RecX [P. fluorescens]. 56% similar to lipoprotein D precursor [E. coli] 55% similar to hypothetical ygbQ gene product of [H. influenzae] 63% similar to hypothetical yaeL gene product of [E. coli] 53% similar to hypothetical Protein YaeQ [E. coli]. 57% similar to putative alpha helical Protein YjfJ [E. coli]. 54% similar to hypothetical Protein Yjfl [E. coli]. 64% similar to hypothetical YfjD gene product [E. coli] 45% similar to hypothetical 28.0 KD Protein in GmhA-DinJ intergenic 51% similar to hypothetical 36.0 KD Protein Y iaO [E. coli] 49-52% similar to putative Protein from [Aquifex aeolicus] 52% similar to hypothetical 13.8 KD Protein YidB in IbpA-GyrB intergenic 76% similar to hypothetical gene product yfgK[E. coli]. VO oo PA3806 1.54 Hypothetical, unclassified unknown PA3819 1.76 Hypothetical, unclassified unknown PA3850 1.20 1.66 Hypothetical, unclassified unknown PA3854 1.26 Hypothetical, unclassified unknown PA3881 1.21 Hypothetical, unclassified unknown PA3882 2.49 Hypothetical, unclassified unknown PA3884 1.75 1.90 Hypothetical, unclassified unknown PA3886 1.66 Hypothetical, unclassified unknown PA3892 1.70 Hypothetical, unclassified unknown PA3904 1.25 1.76 Hypothetical unclassified unknown PA3905 1.94 2.50 2.92 Hypothetical unclassified unknown PA3911 1.46 Hypothetical unclassified unknown PA3912 1.59 1.70 Hypothetical, unclassified unknown PA3934 1.21 1.33 Hypothetical unclassified unknown PA3944 1.54 Hypothetical unclassified unknown PA3949 1.45 1.41 Hypothetical unclassified unknown PA3952 1.25 1.41 Hypothetical, unclassified unknown PA3953 1.50 Hypothetical unclassified unknown PA3955 2.15 Hypothetical unclassified unknown PA3958 1.47 1.50 Hypothetical, unclassified unknown PA3962 1.09 1.11 Hypothetical unclassified unknown PA3966 1.71 2.08 2.78 Hypothetical unclassified unknown PA3967 1.54 Hypothetical, unclassified unknown PA3971 1.70 1.63 1.60 Hypothetical unclassified unknown PA3979 2.18 2.27 3.13 Hypothetical, unclassified unknown PA3980 1.14 1.30 Hypothetical, unclassified unknown PA3981 1.48 1.70 1.77 Hypothetical, unclassified unknown PA3982 1.32 1.51 Hypothetical, unclassified unknown PA3983 2.13 2.75 Hypothetical, unclassified unknown PA3992 1.82 Hypothetical, unclassified unknown PA3998 1.46 Hypothetical, unclassified unknown PA4004 2.28 Hypothetical, unclassified unknown PA4005 2.22 2.77 Hypothetical, unclassified unknown PA4013 1.24 1.38 Hypothetical, unclassified unknown PA4028 1.58 Hypothetical unclassified unknown PA4046 1.29 1.58 Hypothetical, unclassified unknown PA4058 1.35 Hypothetical unclassified unknown PA4065 2.48 Hypothetical unclassified unknown PA4075 2.51 2.85 Hypothetical unclassified unknown PA4090 2.83 Hypothetical, unclassified unknown PA4093 1.41 1.44 Hypothetical unclassified unknown PA4099 3.03 2.83 Hypothetical unclassified unknown PA4149 1.54 Hypothetical unclassified unknown PA4163 1.23 Hypothetical unclassified unknown PA4164 1.77 Hypothetical unclassified unknown PA4181 1.91 Hypothetical unclassified unknown PA4182 1.25 Hypothetical unclassified unknown PA4183 2.06 Hypothetical unclassified unknown 73% similar to 43.1 kDa hypothetical Protein (YfgB) in ndk-gcpE interge 78% similar to hypothetical Protein YcfJ [E. coli] 54% similar to 11.3kb hypothetical Protein YhbQ in SohA-Mtr intergenic 47% similar to Picea glauca emb34 gene product 43% similar to hypothetical Protein [Rickettsia prowazekii]. 61% similar to putative membrane Protein YdhJ [E. coli]. 65% similar to hypothetical Protein YhbT [E. coli]. 65% similar to hypothetical Protein YhbV [E. coli]. 69% similar to hypothetical Protein [Haemophilus influenzae Rd]. 68% similar to probable acetyl transferase of [Proteus mirabilis] 45% similar to hypothetical Protein YtfP [B. subtilis] 55% similar to hypothetical Protein YrdC [B. subtilis] 80% similar to putative tRNA-thiotransferase MiaB [S. typhimurium]. 76% similar to the PHOH-LIKE Protein of [E. coli] 73% similar to the hypothetical Protein HI0004 of [H. influenzae]. 71% similar to the hypothetical 33.3 KD Protein in cutE-asnB intergenic 48-50% similarity to portions of lytic transglycosylases 58% similar to YbeD, hypothetical 9.8 KD Protein in lipB-dacA intergenic 77% similar to conserved hypothetical Protein ybeA of [E. coli] 73% similar to hypothetical Protein ybeB of [E. coli] 63% similar to hypothetical yohK gene product of [E. coli] 41% similar to conserved hypothetical integral membrane Protein [T. palli 43% similar to hypothetical Protein Rv1597 [M. tuberculosis] 36% similar to porin OprD [P.aeruginosa] 80% similar to hypothetical acoX gene product of [P. putida]. 38% similar to putative amidase [Streptomyces coelicolor] 51% similar to hypothetical Protein YKL070w [S.cerevisiae]. 50% similar to a region of hypothetical Protein SC1 E6.02c [S. coelicolor]. PA4205 4.66 2.77 4.69 Hypothetical unclassified, unknown PA4220 1.69 Hypothetical unclassified, unknown PA4279 1.65 1.43 Hypothetical unclassified, unknown PA4308 1.48 Hypothetical unclassified, unknown PA4317 2.11 Hypothetical unclassified, unknown PA4318 1.91 Hypothetical unclassified, unknown PA4319 1.61 2.24 2.68 Hypothetical unclassified, unknown PA4323 1.21 Hypothetical unclassified, unknown PA4326 1.31 Hypothetical unclassified, unknown PA4335 1.41 Hypothetical unclassified, unknown PA4340 1.47 1.44 1.87 Hypothetical unclassified, unknown PA4359 2.50 2.02 2.14 Hypothetical unclassified, unknown PA4392 2.35 Hypothetical unclassified, unknown PA4399 1.31 Hypothetical unclassified, unknown PA4405 1.83 Hypothetical unclassified, unknown PA4451 2.07 Hypothetical unclassified, unknown PA4459 2.00 Hypothetical unclassified, unknown PA4473 1.93 Hypothetical unclassified, unknown PA4486 1.44 Hypothetical unclassified, unknown PA4488 1.36 1.57 Hypothetical unclassified, unknown PA4489 1.26 Hypothetical unclassified, unknown PA4490 1.49 Hypothetical unclassified, unknown PA4491 1.37 1.81 Hypothetical unclassified, unknown PA4492 1.44 Hypothetical unclassified, unknown PA4510 1.26 Hypothetical unclassified, unknown PA4517 1.65 1.63 1.67 Hypothetical unclassified, unknown PA4521 1.90 1.94 2.00 Hypothetical unclassified, unknown PA4523 1.42 1.55 1.53 Hypothetical unclassified, unknown PA4532 1.36 1.47 Hypothetical unclassified, unknown PA4534 2.24 Hypothetical unclassified, unknown PA4535 1.41 1.52 Hypothetical unclassified, unknown PA4536 1.54 1.49 1.46 Hypothetical unclassified, unknown PA4541 1.49 1.77 Hypothetical unclassified, unknown PA4573 2.29 1.84 Hypothetical unclassified, unknown PA4578 1.12 1.15 Hypothetical unclassified, unknown PA4582 1.15 Hypothetical unclassified, unknown PA4601 1.13 Hypothetical unclassified, unknown PA4608 1.71 Hypothetical unclassified, unknown PA4617 1.53 Hypothetical unclassified, unknown PA4620 1.60 1.74 1.76 Hypothetical unclassified, unknown PA4627 1.14 1.47 Hypothetical unclassified, unknown PA4631 1.53 1.78 Hypothetical unclassified, unknown PA4634 2.17 2.42 Hypothetical unclassified, unknown PA4636 1.17 1.33 Hypothetical unclassified, unknown PA4637 1.64 1.84 Hypothetical unclassified, unknown PA4638 2.24 Hypothetical unclassified, unknown PA4639 1.67 1.87 Hypothetical unclassified, unknown PA4642 2.03 1.73 Hypothetical unclassified, unknown 100% similar to probable fptB Protein [P.aeruginosa] 42% similar to a region of Bvg accessory factor [Bordetella pertussis]. 67% similar to hypothetical Protein YjgR [E. coli] 50% similar to hypothetical Protein Rv3695 [M. tuberculosis] 45% similar to hypothetical Protein[Synechocystis sp.]. 50% similar to hypothetical Protein[Synechocystis sp.]. 80% similar to FeoA (iron(ll) transport system Protein) [E. coli] 63% similar to hypothetical Protein YbaZ [E. coli]. 60% similar to hypothetical Protein YvqK [B. subtilis). 92% similar to toluene tolerance Protein Ttg2F [P. putida] 42% similar to hypothetical Protein YrbK [E. coli]. 59% similar to yjgA gene product [E. coli] 66% similar to 4-carboxymuconolactone decarboxylase PcaC [A. calco 64% similar to hypothetical Protein [E. coli] 71% similar to b2228 (putative membrane Protein) [E. coli] 90% similar to hypothetical O R F b2229 [E. coli] (mature Protein). 61% similar to yfaA (b2230 57% similar to hypothetical O R F b2225 [E. coli] (mature Protein). 58% similar to hypothetical Protein [Pyrococcus horikoshii]. 74% similar to hypothetical Protein YijP [E. coli] 40% similar to signalling Protein AmpE [E. coli]. 33% similar to P E - P G R S (glycine rich Protein) [M. tuberculosis] 70% similar to hypothetical Protein [Anabaena variabilis]. 61 % similar to putative nitrogen fixation positive activator Protein [Synec 58% similar to hypothetical Protein YgjO [E. coli]. 52% similar to 4-Hydroxybenzoyl-CoA reductase gamma-subunit [T. aro 53% similar to putative enzyme YjjT [E. coli]. 44%> similar to putative dihydroflavonol 4-reductase [Synechocystis sp.]. 48% similar to hypothetical Protein Rv3226c [M. tuberculosis]. PA4643 1.41 Hypothetical unclassified, unknown PA4650 2.98 Hypothetical unclassified, unknown PA4652 2.01 Hypothetical unclassified, unknown PA4656 1.35 Hypothetical unclassified, unknown PA4681 1.41 Hypothetical unclassified, unknown PA4685 1.43 Hypothetical unclassified, unknown PA4686 1.47 1.35 1.41 Hypothetical unclassified, unknown PA4689 1.39 1.48 Hypothetical unclassified, unknown PA4690 1.76 1.93 Hypothetical unclassified, unknown PA4692 1.80 2.95 Hypothetical unclassified, unknown PA4697 1.45 Hypothetical unclassified, unknown PA4701 1.21 Hypothetical unclassified, unknown PA4704 1.35 Hypothetical unclassified, unknown PA4714 1.42 1.57 Hypothetical unclassified, unknown PA4716 1.54 1.50 1.51 Hypothetical unclassified, unknown PA4717 1.31 1.46 1.54 Hypothetical unclassified, unknown PA4746 3.19 Hypothetical unclassified, unknown PA4774 1.33 1.48 1.56 Hypothetical unclassified, unknown PA4779 1.22 Hypothetical unclassified, unknown PA4780 1.29 Hypothetical unclassified, unknown PA4782 1.66 Hypothetical unclassified, unknown PA4791 1.75 Hypothetical unclassified, unknown PA4798 1.17 Hypothetical unclassified, unknown PA4799 1.26 Hypothetical unclassified, unknown PA4800 1.24 Hypothetical unclassified, unknown PA4801 1.27 Hypothetical unclassified, unknown PA4834 1.57 1.60 Hypothetical unclassified, unknown PA4841 1.22 Hypothetical unclassified, unknown PA4842 2.24 2.28 2.71 Hypothetical unclassified, unknown PA4857 2.02 Hypothetical unclassified, unknown PA4872 1.55 1.65 1.73 Hypothetical unclassified, unknown PA4874 1.92 Hypothetical unclassified, unknown PA4877 1.52 Hypothetical unclassified, unknown PA4879 1.23 1.78 Hypothetical unclassified, unknown PA4883 2.06 Hypothetical unclassified, unknown PA4884 1.73 Hypothetical unclassified, unknown PA4916 1.84 1.96 Hypothetical unclassified, unknown PA4917 2.05 2.13 Hypothetical unclassified, unknown PA4918 1.26 Hypothetical unclassified, unknown PA4925 1.91 2.41 2.58 Hypothetical unclassified, unknown PA4926 1.49 Hypothetical unclassified, unknown PA4927 1.39 1.38 Hypothetical unclassified, unknown PA4928 2.09 Hypothetical unclassified, unknown PA4933 1.16 Hypothetical unclassified, unknown PA4939 1.22 1.24 Hypothetical unclassified, unknown PA4940 18.16 15.77 Hypothetical unclassified, unknown PA4948 1.37 1.50 1.95 Hypothetical unclassified, unknown PA4949 1.30 1.44 Hypothetical unclassified, unknown 38% similar to outer membrane usher Protein A faC [E. coli]. 70% similar to putative sugar nucleotide epimerase [E. coli] 49% similar to Paraquat-inducible Protein B [E. coli]. 48% similar to paraquat-inducible Protein A [E. coli]. 66% similar to putative reductase [E. coli] 53% similar to hypothetical Protein [Synechocystis sp] 42% similar to regulatory subunit of cAMP-dependent histone kinase 55% similar to putative Protein [Aquifex aeolicus] 45% similar to putative Phenzine biosynthesis Protein phzF [P. fluores 62% similar to hypothetical Protein [E. coli] 69% similar to hypothetical Protein YhbC [E. coli]. 56% similar to speE gene product (spermidine synthase) [E. coli] 52% similar to hypothetical Protein YdeD [E. coli] 48% similar to hypothetical Protein ZK632.3 IN Chromosome III [C.elegan 53% similar to a region of hypothetical Protein YqkA [B. subtilis]. 49% similar to FlaR Protein [Listeria monocytogenes]. 44% similar to Met(adenosyl) methyltransferase [S. erythraea]. 52% similar to hypothetical integral membrane Protein [T. pallidum]. 59% similar to putative regulator [E. coli]. 55% similar to hypothetical Protein [Methanococcus jannaschii]. 47% similar to carboxyphosphonoenolpyruvate phosphonomutase 64% similar to phosphate starvation-inducible Protein PsiF [E. coli]. 67% similar to hypothetical Protein YhjG [E. coli]. 44% similar to hydrogenase, cytochrome subunit [H. pylori J99]. 51% similar to hypothetical Protein [Synechocystis sp.]. 50% similar to putative transport Protein YggB [E. coli]. 53% similar to hypothetical Protein Rv2569c [M. tuberculosis]. 51 % similar to hypothetical Protein Rv2567 [M. tuberculosis]. N-terminus 78% similar to hypothetical Protein YgiR of [E. coli] 47% similar to putative histidyl-tRNA synthetase HisS [B. subtilis] 61% similar to hypothetical Protein YjeT [E. coli]. 70% similar to hypothetical Protein YjeE (E. coli] 60% similar to hypothetical Protein YjeF [E. coli] PA4950 1.20 1.31 Hypothetical unclassified, unknown PA4952 1.65 2.05 Hypothetical unclassified. unknown PA4955 1.16 Hypothetical unclassified, unknown PA4962 1.39 Hypothetical unclassified. unknown PA4963 1.47 Hypothetical unclassified unknown PA4972 1.21 Hypothetical unclassified. unknown PA4991 1.39 1.43 Hypothetical unclassified unknown PA4993 1.36 Hypothetical unclassified unknown PA4998 1.16 1.83 Hypothetical unclassified unknown PA4999 1.52 2.23 Hypothetical unclassified unknown PA5001 1.89 Hypothetical unclassified unknown PA5002 2.30 1.72 2.72 Hypothetical unclassified unknown PA5003 1.71 Hypothetical unclassified unknown PA5006 1.72 Hypothetical unclassified unknown PA5007 1.40 Hypothetical unclassified unknown PA5022 1.76 Hypothetical unclassified unknown PA5024 1.72 1.96 2.06 Hypothetical unclassified unknown PA5026 1.22 Hypothetical unclassified unknown PA5027 1.33 Hypothetical unclassified unknown PA5037 1.83 Hypothetical unclassified unknown PA5047 1.41 1.51 Hypothetical unclassified unknown PA5055 1.64 Hypothetical unclassified unknown PA5061 2.34 2.70 Hypothetical unclassified unknown PA5078 1.44 1.55 1.58 Hypothetical unclassified unknown PA5081 1.25 Hypothetical unclassified unknown PA5086 1.24 Hypothetical unclassified unknown PA5088 1.63 Hypothetical unclassified unknown PA5108 1.43 Hypothetical unclassified unknown PA5109 2.10 2.17 1.63 Hypothetical unclassified unknown PA5114 1.62 Hypothetical unclassified unknown PA5120 1.04 Hypothetical unclassified unknown PA5133 1.75 Hypothetical unclassified unknown PA5138 1.50 Hypothetical unclassified unknown PA5146 1.40 Hypothetical unclassified unknown PA5176 1.54 1.96 Hypothetical unclassified unknown PA5178 1.71 Hypothetical unclassified unknown PA5184 1.33 Hypothetical unclassified unknown PA5225 1.58 Hypothetical unclassified unknown PA5226 1.31 Hypothetical unclassified unknown PA5228 1.66 Hypothetical unclassified unknown PA5229 1.83 2.26 Hypothetical unclassified unknown PA5232 1.43 Hypothetical unclassified unknown PA5237 1.52 1.49 Hypothetical unclassified unknown PA5244 1.26 1.33 Hypothetical unclassified unknown PA5245 1.29 1.35 1.35 Hypothetical unclassified unknown PA5247 1.32 Hypothetical unclassified unknown PA5248 1.47 1.59 Hypothetical unclassified unknown PA5251 1.46 1.57 Hypothetical unclassified unknown 75% similar to hypothetical Protein YjeS [E. coli] 65% similar to conserved hypothetical Protein YjeQ [H. influenzae Rd] 68%> similar to hypothetical Protein Ybcl [E. coli] 66% similar to putative toluene tolerance Protein Ttg8 [P. putida]. 46% similar to hypothetical Protein Ttn [P. putida]. 45%> similar to putative transcription activator Mig-14 [S. typhimurium]. 100%> similar to putative heptose kinase W a p Q [P.aeruginosa]. 59% similar to hypothetical Protein AefA [E. coli]. 63% similar to YtnM [B. subtilis] 43% similar to hypothetical Protein [Methanococcus jannaschii]. 46% similar to unknown O R F of [Myxococcus xanthus] 75%) similar to phal Protein [P. oleovorans] 82%) similar to mdoG gene product of [E. coli] 47% similar to a hypothetical mutT-like Protein of [Streptomyces lividans] 54% similar to putative membrane Protein YibP [E. coli]. 66% similar to hypothetical yrfE gene product of [E. coli] 72% similar to a hypothetical Protein [E. coli] 46% similar to chorismate mutase [Erwinia herbicola]. 46% similar to hypothetical ygfB gene product of [E. coli] 55%) similar to putative ligase YgfA [E. coli]. 63% similar to a hypothetical Protein of [Synechocystis sp.] 78%) similar to putative membrane Protein Yhil [E. coli]. 87% similar to hypothetical yigC gene product of [E. coli] 55%> similar to hypothetical yohD gene product of [E. coli] 79% similar to sigma cross-reacting Protein 27A of [E. coli] 58% similar to hypothetical yail gene product of [E. coli] 47% similar to a region of putative transport Protein YggB [E. coli]. PA5257 4.04 3.90 Hypothetical, unclassified, unknown PA5269 1.85 Hypothetical, unclassified, unknown PA5279 1.19 Hypothetical, unclassified, unknown PA5285 1.22 Hypothetical, unclassified, unknown PA5286 4.63 Hypothetical, unclassified, unknown PA5289 1.91 Hypothetical, unclassified, unknown PA5305 1.26 Hypothetical, unclassified, unknown PA5335 1.38 Hypothetical, unclassified, unknown PA5392 1.17 Hypothetical, unclassified, unknown PA5395 1.48 1.52 1.64 Hypothetical, unclassified, unknown PA5396 1.27 Hypothetical, unclassified, unknown PA5406 1.33 1.49 1.46 Hypothetical, unclassified, unknown PA5414 1.37 1.40 Hypothetical, unclassified, unknown PA5441 1.41 1.79 Hypothetical, unclassified, unknown PA5446 2.01 Hypothetical, unclassified, unknown PA5462 1.71 1.74 3.42 Hypothetical, unclassified, unknown PA5463 1.93 1.98 2.60 Hypothetical, unclassified, unknown PA5464 1.45 Hypothetical, unclassified, unknown PA5465 1.44 Hypothetical, unclassified, unknown PA5469 1.40 Hypothetical, unclassified, unknown PA5471 2.01 Hypothetical, unclassified, unknown PA5480 1.43 Hypothetical, unclassified, unknown PA5481 1.54 Hypothetical, unclassified, unknown PA5485 1.29 1.33 1.58 Hypothetical, unclassified, unknown PA5486 1.56 Hypothetical, unclassified, unknown PA5487 1.33 Hypothetical, unclassified, unknown PA5488 1.29 Hypothetical, unclassified, unknown PA5494 1.27 Hypothetical, unclassified, unknown PA5515 1.66 Hypothetical, unclassified, unknown PA5526 1.77 3.44 Hypothetical, unclassified, unknown PA5527 1.62 Hypothetical, unclassified, unknown PA5528 1.49 Hypothetical, unclassified, unknown PA5532 1.23 Hypothetical, unclassified, unknown PA5533 1.81 Hypothetical, unclassified, unknown PA5537 1.76 Hypothetical, unclassified, unknown PA5539 1.47 Hypothetical, unclassified, unknown PA0345 1.29 Membrane proteins PA1669 2.16 Membrane proteins PA2286 1.80 Membrane proteins PA4224 pchG 1.36 1.39 Membrane proteins PA4370 icmP 1.28 Membrane proteins PA4586 2.94 4.42 4.48 Membrane proteins PA4757 1.44 Membrane proteins PA5264 1.39 Membrane proteins PA5478 1.24 1.41 Membrane proteins PA1085 flgJ 4.25 Motility & Attachment PA3805 pilF 1.51 Motility & Attachment PA4528 pilD 1.34 Motility & Attachment 48% similar to (hypothetical?) hemY gene product of [E. coli] 78% similar to O R F 240 of [P. fluorescens] 72% similar to hypothetical yjbQ gene product of [E. coli] 60% similar to C-terminal end of hypothetical yqiC gene product [E. coli] 54% similar to hypothetical ydbL gene product of [E. coli] 66% similar to hypothetical Protein YicC [H. influenzae] & [E. coli]. 55% similar to hypothetical Protein 1 (vnfA 5' region) [A.vinelandii]. 73% similar to hypothetical Protein [Streptomyces lividans]. 51% similar to hypothetical membrane dipeptidase [P. horikoshii]. 76% similar to hypothetical transmembrane Protein YgdQ [H. influenzae 44% similar to hypothetical Protein [E. coli) 56% similar to hypothetical Protein AmpD [E. coli). 64% similar to hypothetical Protein YhgN [E. coli]. 64% similar to a region of hypothetical Protein [Aquifex aeolicus] N-terminal end is 64% similar to hypothetical Protein YeiR [E. coli]. 41% similar to putative membrane Protein [Synechococcus PCC7942). 44% similar to IcmF Protein [Legionella pneumophila] 99% similar to PchG [P.aeruginosa] 100% similar to metalloProteinase [P.aeruginosa] 70% similar to E. coli yeaS hypothetical gene product. 55% similar to putative spore maturation Protein A [B. subtilis). 43% similar to flagellar basal body Protein FlgJ [S. typhimurium] 100% identical to PilF [P.aeruginosa] 99% similar to pilD [P.aeruginosa] O PA4550 fimU 1.55 1.27 Motility & Attachment PA5043 pilN 1.19 -1.23 Motility & Attachment PA5044 pilM 1.28 Motility & Attachment PA0148 1.40 Nucleotide biosynthesis and metabolism PA0336 ygdP 1.78 1.82 Nucleotide biosynthesis and metabolism PA0441 dht 2.04 2.36 1.92 Nucleotide biosynthesis and metabolism PA0444 1.45 Nucleotide biosynthesis and metabolism PA0590 apaH 1.39 Nucleotide biosynthesis and metabolism PA3050 pyrO 1.28 1.54 Nucleotide biosynthesis and metabolism PA3527 pyrC 1.23 Nucleotide biosynthesis and metabolism PA3654 pyrH 1.33 Nucleotide biosynthesis and metabolism PA3770 guaB 1.30 1.42 Nucleotide biosynthesis and metabolism PA4314 purU1 1.63 1.63 Nucleotide biosynthesis and metabolism PA4670 prs 1.75 Nucleotide biosynthesis and metabolism PA4758 carA 1.19 Nucleotide biosynthesis and metabolism PA4854 purH 1.46 1.61 Nucleotide biosynthesis and metabolism PA4855 purD 1.42 Nucleotide biosynthesis and metabolism PA5129 got 1.11 Nucleotide biosynthesis and metabolism PA5425 purK 1.66 Nucleotide biosynthesis and metabolism PA0677 2.03 1.19 Protein secretion/export apparatus PA0678 1.71 Protein secretion/export apparatus PA1692 1.66 Protein secretion/export apparatus PA1693 pscR 1.68 Protein secretion/export apparatus PA1696 pscO 3.40 2.71 3.12 Protein secretion/export apparatus PA1698 popN 1.50 Protein secretion/export apparatus PA1720 pscG 1.81 Protein secretion/export apparatus PA1724 pscK 1.45 Protein secretion/export apparatus PA3405 hasE 1.42 Protein secretion/export apparatus PA4144 1.98 2.03 Protein secretion/export apparatus PA4276 secE 1.54 1.75 Protein secretion/export apparatus PA4747 secG 1.54 Protein secretion/export apparatus PA5068 tatA 1.73 Protein secretion/export apparatus PA5069 tatB 1.70 2.02 Protein secretion/export apparatus PA0219 1.31 Putative enzymes PA0249 2.50 Putative enzymes PA0298 1.41 Putative enzymes PA0299 2.04 1.49 Putative enzymes PA0372 1.58 Putative enzymes PA0421 1.14 1.14 Putative enzymes PA0531 1.47 Putative enzymes PA0656 2.52 Putative enzymes PA0657 1.56 Putative enzymes PA0658 1.80 3.22 Putative enzymes PA0779 2.10 2.23 Putative enzymes PA0817 1.41 1.79 Putative enzymes PA0863 2.76 2.71 Putative enzymes PA0954 1.42 1.33 Putative enzymes PA1046 1.88 Putative enzymes Identical to pilN gene product of PA01 . Identical to pilM gene product of P A O L 46% similar to adenosine deaminase [E. coli] 81% similar to putative invasion Protein YgdP [E. coli]. 95% similar to D-hydantoinase [P. putida] 55% similar to DL-hydantoinase (N-carbamyl-L-amino acid amidohydrola 64% similar to E. coli apaH gene product. 71% similar to E. coli dehydroorotate dehydrogenase. 66% similar to pyrC gene product of [E. coli] 80% similar to PyrH gene product of [E. coli] 81%> similar to E. coli guaB gene product 67% similar to formyltetrahydrofolate deformylase [Aquifex aeolicus] 82% similar to phosphoribosylpyrophosphate synthetase of [E. coli]. 78% similar to phosphoribosylaminoimidazolecarboxamideformyltransfera 81% similar to E. coli purD gene product. 66% similar to grxC gene product of [E. coli]. 43% similar to general secretory pathway Protein J [B. pseudomallei]. 50% similar to general secretory pathway Protein H [B. pseudomallei]. 88% similar to Yop Proteins translocation Protein YscS [Y. pseudotuber 90% similar to Yop Proteins translocation Protein R homolog [Y. pestis] 57% similar to YscO translocation Protein [Yersinia pseudotuberculosis] 64% similar to YopN [Yersinia enterocolitica] 69% similar to yscG gene product [Yersinia enterocolitica] 86% similar to PscK [P.aeruginosa] 77% similar to metalloprotease transporter HasE [Serratia marcescens]. 49% similar to secretion Protein CyaE [Bordetella pertussis]. 65% similar to preProtein translocase S e c E [E. coli] 67% similar to S e c G gene product of E. coli 62% similar to TatA Protein [E. coli] 66% similar to ORF4 [Azotobacter chroococcum] 68% similar to aldehyde dehydrogenase AldH [E. coli]. 54% similar to hypothetical Protein YiaC [E. coli] 50% similar to putative glutamine synthetase YcjK [E. coli] 53% similar to beta-alanine-pyruvate transaminase [P. putida] 53%) similar to hypothetical zinc protease Y 4 W A [Rhizobium sp. NGR234] 43%) similar to monoamine oxidase B [Homo sapiens]. 58%) similar to hypothetical Protein [Synechocystis sp.]. 72% similar to histidine triad nucleotide-binding Protein (HINT) [O. cunicul 57% similar to putative cell division cycle Protein [Synechocystis sp.]. 56%o similar to hypothetical Protein Rv1544 [M. tuberculosis]. 61%) similar to mitochrondial ATP-dependentprotease [Homo sapiens] 71% similar to hypothetical Protein [Bordetella pertussis]. 50% similar to putative zinc-binding dehydrogenase [S.pombe]. 68% similar to putative acylphosphatase AcyP [E. coli] 48% similar to beta-agarase B (AgaB) [Vibrio sp.] PA1535 PA1565 1.64 PA1576 PA1654 PA1737 1.67 PA1828 PA1990 PA2124 PA2125 3.09 PA2263 1.52 PA2302 1.29 PA2333 2.35 PA2402 1.63 PA2499 PA2631 1.69 PA2891 PA2922 1.41 PA3001 2.41 PA3035 2.49 PA3368 1.68 PA3444 PA3534 PA3774 PA3798 PA3803 gcpE PA4079 PA4089 PA4217 2.40 PA4330 1.48 PA4401 2.49 PA4619 PA4621 PA4715 1.86 PA4786 PA4819 PA4899 2.62 PA4907 PA4943 PA4980 PA5000 1.37 PA5004 3.08 PA5005 PA5008 1.46 PA5048 1.42 PA5084 PA5312 PA5384 PA5386 1.44 Putative enzymes 1.44 Putative enzymes 1.59 1.70 . Putative enzymes 1.19 Putative enzymes Putative enzymes 1.30 Putative enzymes 1.28 1.88 Putative enzymes 1.53 Putative enzymes 1.99 Putative enzymes Putative enzymes 1.21 1.43 Putative enzymes 2.42 Putative enzymes 1.46 Putative enzymes 1.87 Putative enzymes 1.78 Putative enzymes 1.35 1.37 Putative enzymes 1.49 Putative enzymes 2.33 Putative enzymes Putative enzymes Putative enzymes 1.58 1.69 Putative enzymes 1.51 Putative enzymes 1.47 1.42 Putative enzymes 1.87 Putative enzymes 1.70 Putative enzymes 1.67 Putative enzymes 1.48 1.91 Putative enzymes 1.63 2.16 Putative enzymes Putative enzymes 2.18 2.07 Putative enzymes 1.97 2.26 Putative enzymes 1.31 1.46 Putative enzymes 2.08 2.45 Putative enzymes 1.34 1.39 Putative enzymes 1.53 Putative enzymes Putative enzymes 1.61 2.24 Putative enzymes 2.78 Putative enzymes 1.42 Putative enzymes 1.33 1.62 Putative enzymes 1.74 2.48 Putative enzymes 1.22 1.31 Putative enzymes 1.81 1.94 Putative enzymes 1.41 1.37 Putative enzymes 1.41 1.43 Putative enzymes 1.27 Putative enzymes 1.26 Putative enzymes 1.25 Putative enzymes 56% similar to long chain acyl-CoA dehydrogenase [Rattus norvegicus] 58% similar to P R O B A B L E O X I D O R E D U C T A S E OrdL [E. coli] 53% similar to 3-hydroxyisobutyrate dehydrogenase [P.aeruginosa] 64% similar to ORF42 of [Yersinia pestis] 70% similar to putative fatty oxidation Protein [Streptomyces coelicolor]. 59% similar to 7-alpha-hydroxysteroid dehydrogenase [Eubacterium sp.]. 51% similar to hypothetical Protein of [Synechocystis sp] 46-50% similar to many dehydrogenases 56% similar to glycine betaine aldehyde dehydrogenase [B. subtilis] 68% similar to yiaE gene product (putative dehydrogenase) [E. coli]. 50% similar to regions of tnicrocystin synthase McyA [M. aeruginosa]. 49% similar to phosphonate monoester hydrolase [B. caryophylli]. 54% similar to bacitracin synthetase 3 [B. licheniformis] 61% similar to hypothetical Protein YkoA [B. subtilis] 59% similar to hypothetical Protein YjcF [B. subtilis]. 60% similar to M. tuberculosis BccA 63% similar to hippurate hydrolase [Campylobacter jejuni]. 76% similar to putative glyceraldehyde-3-phosphate dehydrogenase 60% similar to G S T A Protein [Rhizobium leguminosarum]. 95% similar to putative FmnH2-dependent monooxygenase SsuD 45% similar to hypothetical Protein Rv0197[M. tuberculosis] 56% similar to hypothetical Protein AF0130[Archaeoglobus fulgidus]. 69% similar to putative aminotransferase YbdL [E. coli]. 85% similar to E. coli G c p E Protein -50% similar to several putative dehydrogenases from diverse organisms. 58% similar to 3-oxoacyl-[acyl-carrier-Protein] reductase [E. coli]. 46% similar to hypothetical Protein [Bordetella pertussis]. 51% similar to putative enoyl-CoA isomerase PaaG[E. coli]. 47% similar to hypothetical glutathione S-transferase YfcF [E. coli]. 53% similar to cytochrome c553 [Gluconobacter suboxydans]. 94% similar to putative aminotransferase [E. coli]. 66% similar to putative 3-oxoacyl-[acyl-carrier Protein] reductase FabG4 65% similar to SLL0501 hypothetical Protein [Synechocystis sp.] 77% similar to aldehyde dehydrogenase [P. putida] 70% similar to putative oxidoreductase YdfG [E. coli]. 73% similar to putative G T P a s e [E. coli] 51% similar to putative enzyme PaaG [E. coli]. 54% similar to mucus-inducible Protein MigA [P.aeruginosa]. 45-48% similar to several Proteins involved capsule or LPS biosynthesis. 50% similar to MODULATION Protein NolO [Rhizobium sp. NGR234] 100% similar to putative heptose kinase WapP [P.aeruginosa]. 53% similar to thermonuclease of [Staphylococcus intermedius] 55% similar to dadA gene product of [E. coli] 71% similar to aldH gene product of [E. coli] 53% similar to putative lipase Lipl [M. tuberculosis]. 56% similar to putative 3-hydroxyacyl-CoA dehydrogenase [A. fulgidus]. PA0616 2.91 PA0617 3.02 PA0618 2.13 PA0619 3.19 PA0620 2.66 PA0621 2.37 4.31 PA0622 2.42 PA0623 1.61 2.88 PA0624 2.66 PA0625 2.50 PA0626 1.81 PA0627 2.34 PA0628 1.80 PA0629 2.21 PA0630 1.59 PA0631 2.39 PA0632 3.39 PA0633 1.96 4.35 PA0634 3.16 PA0635 2.02 PA0636 2.14 PA0637 2.70 PA0638 PA0639 2.98 PA0640 3.12 PA0641 2.43 PA0642 2.66 PA0643 1.94 PA0644 4.03 PA0645 1.83 PA0646 3.13 PA0647 3.75 PA0648 PA0763 mucA 2.11 PA0985 4.54 PA1150 pys2 PA1900 1.78 PA3319 pIcN 1.53 PA3542 alg44 2.18 2.03 PA3866 5.32 13.89 PA4457 1.30 1.90 PA1294 md 2.65 1.91 PA2976 me 1.96 PA3528 mt PA3744 rimM 1.28 PA3861 ml PA4238 rpoA PA4544 riuD 1.44 13.04 Related to phage, transposon, or plasmid 13.86 Related to phage, transposon, or plasmid 9.97 Related to phage, transposon, or plasmid 14.72 Related to phage, transposon, or plasmid 12.10 Related to phage, transposon, or plasmid 16.68 Related to phage, transposon, or plasmid 10.25 Related to phage, transposon, or plasmid 12.84 Related to phage, transposon, or plasmid 12.78 Related to phage, transposon, or plasmid 14.10 Related to phage, transposon, or plasmid 9.75 Related to phage, transposon, or plasmid 10.75 Related to phage, transposon, or plasmid 9.95 Related to phage, transposon, or plasmid 8.70 Related to phage, transposon, or plasmid 10.58 Related to phage, transposon, or plasmid 16.20 Related to phage, transposon, or plasmid 21.20 Related to phage, transposon, or plasmid 16.37 Related to phage, transposon, or plasmid 15.82 Related to phage, transposon, or plasmid 15.89 Related to phage, transposon, or plasmid 9.99 Related to phage, transposon, or plasmid 11.02 Related to phage, transposon, or plasmid 8.45 Related to phage, transposon, or plasmid 15.87 Related to phage, transposon, or plasmid 16.36 Related to phage, transposon, or plasmid 9.75 Related to phage, transposon, or plasmid 12.88 Related to phage, transposon, or plasmid 8.93 Related to phage, transposon, or plasmid 18.01 Related to phage, transposon, or plasmid 15.84 Related to phage, transposon, or plasmid 13.43 Related to phage, transposon, or plasmid 23.03 Related to phage, transposon, or plasmid 8.16 Related to phage, transposon, or plasmid 2.16 Secreted Factors (toxins, enzymes, alginate) 18.06 Secreted Factors (toxins, enzymes, alginate) 5.35 Secreted Factors (toxins, enzymes, alginate) Secreted Factors (toxins, enzymes, alginate) Secreted Factors (toxins, enzymes, alginate) Secreted Factors (toxins, enzymes, alginate) 51.31 Secreted Factors (toxins, enzymes, alginate) 2.42 Secreted Factors (toxins, enzymes, alginate) Transcription, RNA processing and degradation Transcription, RNA processing and degradation 1.53 Transcription, RNA processing and degradation 1.85 Transcription, RNA processing and degradation 1.22 Transcription, RNA processing and degradation 1.32 Transcription, RNA processing and degradation 1.88 Transcription, RNA processing and degradation 50% similar to putative baseplate Protein [P.aeruginosa phage phi CTX] . 65% similar to M Protein [Bacteriophage 186]. 69% similar to baseplate assembly Protein J (GpJ) [Bacteriophage P2]. 77% similar to orf19 [P.aeruginosa phage phi CTX] . 80% similar to orf20 [P.aeruginosa phage phi CTX] . 55% similar to orf21 [P.aeruginosa phage phi CTX] 91 % similar to contractile tail sheath Protein (gpFI) [Bacteriophage PS17). 84% similar to contractile tail tube Protein [bacteriophage PS17]. 50%> similar to a region of orf26 [P.aeruginosa phage phi CTX] . 56% similar to on*8 [P.aeruginosa phage phi CTX]. 48% similar to essential tail Protein GpD [Bacteriophage P2]. 56% similar to hypothetical Protein HI1415 [Haemophilus influenzae Rd]. 50% similar to a region of putative phage H tail component [S. typhi 53% similar to gp17 [Bacteriophage N15] 60% similar to Gp18 [Bacteriophage N15). 56% similar to Gp19 [Bacteriophage N15]. 68% similar to Gp20 [Bacteriophage N15] 68% similar to a region of Protein Gp21 [Bacteriophage N15]. 100% similar to MucA [P. aeruginosa] 66%> similar to a region of colicin Protein [E. coli]. 100% identical to PYOCIN S2 (KILLER Protein) [P.aenjginosa] 92% similar to PhzB [P.aeruginosa] 96% similar to alginate synthesis-related Protein [P.aeruginosa] 77% similar (but 71% identical) to pyocin S3 from P. aeruginosa strain 74% similar to yrbH (putative isomerase)[E. coli] 55% similar to ribonuclease D [E.coli]. 59% similar to ribonuclease E [E.coli]. 79%) similar to RNase T [Vibrio parahaemolyticus] 62% similar to 16S rRNA P R O C E S S I N G Protein [E. coli] 63% similar to ATP-dependent RNA helicase RhIB [E. coli]. 84% similar to RNA polymerase alpha subunit [E. coli]. 73% similar to SfhB [E. coli] PA4742 truB 1.40 Transcription, RNA processing and degradation 67% similar to tRNA pseudouridine 5S synthase [E. coii]. PA4853 fis 1.85 1.77 Transcription, RNA processing and degradation 80% similar to Fis Protein [E. coli]. PA4951 om 1.45 Transcription, RNA processing and degradation 79% similar to oligoribonuclease Orn (E. coli] PA5239 rho 1.42 Transcription, RNA processing and degradation 92% similar to transcription termination factor Rho [E. coli]. PA0217 1.24 Transcriptional regulators 65% similar to malonate decarboxylase operon regulator MdcR [K. pneu PA0294 1.35 1.31 Transcriptional regulators 49% similar to putative transcriptional regulator [Streptomyces coelicolor]. PA0306 1.52 Transcriptional regulators 46% similar to transcriptional regulator PhbR [P. sp. 61-3] PA0376 rpoH 1.57 Transcriptional regulators 100% similar to heat-shock sigma factor [P.aeruginosa] PA0527 dnr 1.46 Transcriptional regulators 100% identical to Dnr Protein [P.aeruginosa]. PA0528 1.37 Transcriptional regulators 49% similar to positive regulator CynR [E. coli]. PA0610 prtN 1.96 6.83 Transcriptional regulators PA0611 prtR 1.25 3.02 Transcriptional regulators PA0762 algU 2.06 Transcriptional regulators 100% identical to AlgU Protein [P.aeruginosa]. PA0815 1.72 Transcriptional regulators 46% similar to putative transcriptional regulator YjiE [E. coli]. PA1309 1.45 Transcriptional regulators 52% similar to putative transcriptional regulator YwfK [B. subtilis]. PA1347 2.17 Transcriptional regulators 65% similar to a region of putative regulator FimZ [S. typhimurium]. PA1526 1.26 Transcriptional regulators 46% similar to hypothetical Protein [E. coli]. PA1570 1.53 Transcriptional regulators 54% similar to transcriptional regulator Protein RbcR [C.vinosum]. PA1619 2.61 3.03 Transcriptional regulators 45% similar to putative transcriptional regulator [Streptomyces coelicolor]. PA1630 1.34 1.52 Transcriptional regulators 58% similar to hypothetical Protein [Bordetella pertussis]. PA1653 3.50 2.89 3.05 Transcriptional regulators 61 % similar to PetP Protein [Rhodobacter capsulatus]. PA1884 1.17 Transcriptional regulators 51 % similar to putative transcriptional regulator Protein [B. sp.]. PA1912 1.29 Transcriptional regulators 71% similar to probable RNA polymerase sigma factor Feci [E. coli]. PA2273 2.81 Transcriptional regulators 77% similar to SoxR gene product of [E. coli] PA2281 2.28 2.07 Transcriptional regulators 64% similar to the C-terminal end of regulatory Protein PchR [Synechoc PA2447 1.30 Transcriptional regulators 51% similar to hypothetical Protein [Haemophilus influenzae]. PA2467 1.35 1.36 Transcriptional regulators 53% similar to PupR Protein [P. putida]. PA2489 2.78 Transcriptional regulators 46% similar to putative transcriptional regulator [Streptomyces coelicolor]. PA2931 1.63 Transcriptional regulators 63% similar to socA3 Protein [Myxococcus xanthus]. PA3006 psrA 1.69 Transcriptional regulators 90% similar to PsrA [P. putida]. PA3184 1.43 1.58 Transcriptional regulators PA3587 metR 1.77 Transcriptional regulators 63% similar to regulator MetR [E. coli]. PA3721 1.15 Transcriptional regulators 43% similar to putative transcriptional regulator [Streptomyces coelicolor]. PA3778 2.07 2.16 Transcriptional regulators 46% similar to regulator CbbRI [Rhodobacter capsulatus]. PA3899 1.42 1.68 Transcriptional regulators 75% similar to sigma factor Pupl [P. putida]. PA3900 1.41 Transcriptional regulators 61% similar to regulatory Protein PupR [P. putida]. PA4157 1.58 Transcriptional regulators 51% similar to transcriptional regulatory Protein OhbR [P.aeruginosa]. PA4227 pchR 1.70 1.59 Transcriptional regulators 100% similar to regulatory Protein PchR [P.aeruginosa] PA4315 mvaT 1.51 Transcriptional regulators 82% similar to heteromeric transcriptional activator MvaT P16 subunit PA4508 2.32 2.10 Transcriptional regulators 70% similar to transcription regulatory Protein PdhR [Ralstonia eutropha]. PA4581 rtcR 1.45 1.68 1.64 Transcriptional regulators 79% similar to regulator RtcR [E. coli]. PA4596 1.79 1.53 Transcriptional regulators 74% similar to nfxB gene product of [P.aeruginosa]. PA4769 1.84 2.15 Transcriptional regulators 66% similar to regulator for pyruvate dehydrogenase complex PdhR PA4778 2.30 Transcriptional regulators 65% similar to putative transcriptional regulator Ybbl [E. coli). PA4784 1.46 Transcriptional regulators 57% similar to putative transcriptional regulator [B. subtilis]. PA4806 1.23 Transcriptional regulators 46% similar to D M S O reductase regulatory Protein DorX [R.sphaeroides] PA4914 1.52 1.47 Transcriptional regulators 53% similar to positive regulator GcvA [E. coli]. PA5029 1.31 1.32 Transcriptional regulators 72% similar to putative transcriptional regulator YnfL [E. coli]. PA5085 1.91 1.97 Transcriptional regulators PA5105 hutC 1.72 Transcriptional regulators PA5274 mk 1.50 Transcriptional regulators PA5301 1.33 Transcriptional regulators PA5380 1.31 Transcriptional regulators PA5389 1.15 Transcriptional regulators PA5525 1.12 Transcriptional regulators PA0067 prtC 1.29 Translation, post-translational modification, degradation PA2612 serS 1.25 1.28 1.17 Translation, post-translational modification, degradation PA2617 aat 1.74 Translation, post-translational modification, degradation PA2739 pheT 1.33 Translation, post-translational modification, degradation PA2742 rpml 3.59 Translation, post-translational modification, degradation PA3007 lexA 1.60 2.55 3.58 Translation, post-translational modification, degradation PA3987 leuS 1.31 Translation, post-translational modification, degradation PA4241 rpsM 1.61 1.50 1.57 Translation, post-translational modification, degradation PA4245 rpmD 1.48 1.38 Translation, post-translational modification, degradation PA4246 rpsE 1.28 Translation, post-translational modification, degradation PA4254 rpsQ 2.03 1.89 Translation, post-translational modification, degradation PA4259 rpsS 1.41 1.40 Translation, post-translational modification, degradation PA4261 rplW 1.71 1.74 Translation, post-translational modification, degradation PA4264 rpsJ 1.67 1.50 1.48 Translation, post-translational modification, degradation PA4482 gate 1.35 Translation, post-translational modification, degradation PA4542 dpB 1.52 1.90 Translation, post-translational modification, degradation PA4568 rplU 1.37 Translation, post-translational modification, degradation PA4850 prmA 1.75 2.07 2.13 Translation, post-translational modification, degradation PA4932 rpll 1.36 Translation, post-translational modification, degradation PA4935 rpsF 1.21 Translation, post-translational modification, degradation PA4945 miaA 1.46 2.56 Translation, post-translational modification, degradation PA5014 glnE 2.35 3.27 Translation, post-translational modification, degradation PA5018 msrA 1.91 1.79 Translation, post-translational modification, degradation PA5051 argS 1.24 1.27 Translation, post-translational modification, degradation PA5080 3.10 3.08 2.78 . Translation, post-translational modification, degradation PA5134 1.07 Translation, post-translational modification, degradation PA5315 rpmG 2.24 Translation, post-translational modification, degradation PA5470 1.44 Translation, post-translational modification, degradation PA5569 rnpA 1.35 2.00 Translation, post-translational modification, degradation PA5570 rpmH 1.94 2.87 Translation, post-translational modification, degradation PA0073 1.18 Transport of small molecules PA0129 gabP 1.83 Transport of small molecules PA0136 2.07 Transport of small molecules PA0185 1.47 Transport of small molecules PA0206 1.94 Transport of small molecules PA0215 1.15 Transport of small molecules PA0282 cysT 1.33 Transport of small molecules PA0295 1.49 Transport of small molecules PA0300 potF2 1.43 Transport of small molecules PA0324 1.85 Transport of small molecules PA0352 1.88 Transport of small molecules 47% similar to regulatory Protein LysR [E. coli]. 94% similar to hutC gene product of [P. putida] 63% similar to rnk gene product of [E. coli] 62% similar to hypothetical Protein YcjC [E. coli]. 59% similar to ArgR regulatory Protein [P.aeruginosa]. 58% similar to ArgR regulatory Protein [P.aeruginosa]. 49% similar to nta operon transcriptional regulator [E. coli). 73%) similar to oligopeptidase A [E. coli]. 77% similar to seryl-tRNA synthetase [E. coli]. 64% similar to Leu/Phe-tRNA-Protein transferase [E. coli]. 69% similar to phenylalanyl-tRNA synthetase, beta subunit [E. coli]. 72% similar to ribosomal Protein L35 [E. coli] 77% similar to lexA Protein [E. coli]. 71% similar to leucine tRNA synthetase of [E.coli]. 86%) similar to 30S ribosomal subunit Protein S13 [E. coli] 72% similar to 50S ribosomal subunit Protein L30 [E. coli] 84% similar to 30S ribosomal subunit Protein S5 [E. coli] 87% similar to 30S ribosomal subunit Protein S17 [E. coli] 90% similar to 30S ribosomal subunit Protein S19 [E. coli] 70% similar to 50S ribosomal subunit Protein L23 [E. coli] 97% similar to 30S ribosomal subunit Protein S10 [E. coli] 56% similar to glutamyl-tRNA (Gin) amidotransferase subunit C [A.aeolicu 82%) similar to heat shock Protein CIpB [E. coli] 87%) similar to ribosomal Protein L21 of [E. coli] 71%> similar to methylase for 50S ribosomal subunit Protein L11 [E. coli] 78%i similar to rpll gene product of [E. coli] 76%) similar to ribosomal Protein S6 [E. coli]. 83% similar to tRNA 62% similar to glnE gene product of [E. coli]. 81 % similar to peptide methionine sulfoxide reductase [Synechocystis 65% similar to argS gene product of [E. coli] 72% similar to xap gene product of [Xanthomonas campestris] 68%) similar to ctpA gene product of [Bartonella bacilliformis] 95%) similar to rpmG gene product of [E. coli] 71%) similar to putative peptide chain release factor PrfH [E. coli] 87%) similar to rnpAof [P. putida]; 61% similar to RNase P [E. coli] 90%i similar to 50S ribosomal subunit Protein L34 [E. coli] 59% similar to putative ATP-binding component of a transport system 77% similar to gamma-aminobutyrate permease GabP [E. coli] 59% similar to putative ribose A B C transporter, ATP-binding Protein 89% similar to putative ABC-type transporter, membrane subunit AtsB 62%) similar to ATP-binding component of spermidine/putrescine transport 71%) similar to putative malonate transporter MadL [Malonomonas rubra]. 71%. similar to sulfate/thiosulfate transport Protein CysT [E. coli]. 57%) similar to putrescine transport Protein PotF [E. coli] 74%) similar to E. coli potF gene product. 68% similar to spermidine/putrescine transport system permease Protein 75% similar to putative transport Protein Y icE [E. coli] PA0427 oprM 2.51 Transport of smal molecules PA0458 1.21 1.34 Transport of smal molecules PA0604 1.98 Transport of smal molecules PA0860 2.35 2.42 2.36 Transport of smal molecules PA0913 mgtE 1.58 Transport of smal molecules PA1183 dctA 1.71 Transport of smal molecules PA1425 1.16 Transport of smal molecules PA1507 1.97 1.98 Transport of smal molecules PA1569 1.83 Transport of smal molecules PA1783 nasA 1.48 1.64 Transport of smal molecules PA1819 1.26 Transport of smal molecules PA2060 3.32 2.39 Transport of smal molecules PA2278 arsB 2.17 Transport of smal molecules PA2295 1.60 1.51 Transport of smal molecules PA2340 1.62 Transport of smal molecules PA2390 1.24 Transport of smal molecules PA2409 1.42 Transport of smal molecules PA2500 3.51 2.04 Transport of smal molecules PA2811 1.32 Transport of smal molecules PA2914 1.37 1.40 Transport of smal molecules PA3264 2.31 Transport of smal molecules PA3394 nosF 1.95 1.88 Transport of smal molecules PA3514 1.15 Transport of smal molecules PA3753 1.40 Transport of smal molecules PA3760 1.55 Transport of smal molecules PA3781 1.50 1.75 Transport of smal molecules PA3889 1.56 1.69 Transport of smal molecules PA3890 1.48 1.55 Transport of smal molecules PA3920 1.32 Transport of smal molecules PA4160 fepD 1.52 Transport of smal molecules PA4192 1.53 1.33 1.48 Transport of smal molecules PA4195 2.22 Transport of smal molecules PA4208 1.34 1.88 Transport of smal molecules PA4218 1.83 1.79 1.79 Transport of smal molecules PA4233 1.76 Transport o smal molecules PA4358 1.97 2.33 Transport of smal molecules PA4365 1.44 1.58 Transport of smal molecules PA4500 1.83 Transport of smal molecules PA4502 2.01 Transport of smal molecules PA4504 2.20 1.51 Transport of smal molecules PA4597 oprJ 1.47 Transport of smal molecules PA4616 1.65 Transport of smal molecules PA4628 lysP 1.61 1.71 Transport of smal molecules PA4688 hitB 1.41 1.48 Transport of smal molecules PA4706 1.57 1.82 Transport of smal molecules PA4710 phuR 1.15 Transport of smal molecules PA4719 1.95 1.87 Transport of smal molecules PA4765 omIA 1.46 1.63 Transport of smal molecules 62% similar to outer membrane Protein OprJ [P.aeruginosa] 67% similar to putative drug resistance translocase YieO [E. coli] 49% similar to putative mannopine-binding periplasmic Protein MotA 51% similar to putative transporter YwjA [B. subtilis]. 55% similar to MgtE [B. firmus] 88% similar to C4-dicarboxylate transport Protein DctA [S. typhimurium] 51% similar to lincomycin resistance Protein LmrC [S. lincolnensis] 46% similar to uracil permease [B. caldolyticus] 46% similar to glycerol-3-phosphate transport Protein GIpT [E. coli]. 66% similar to nitrate transporter NasA [B. subtilis] 56% similar to putative amino acid/amine transport Protein YjdE [E. coli]. 55% similar to oligopeptide A B C transporter AppC (permease) [B. subtilis 97% identical to arsB gene product of [P. aeruginosa] 64% similar to putative nitrate transporter [Methanococcus jannaschii]. 94% similar to transport system Protein MUG [P. fluorescens]. 62% similar to ATP-binding component of a transport system YbjZ 55% similar to zinc transport Protein ZnuB [E. coli] 62% similar to CynX [E. coli]. 76% similar to hypothetical Protein YadH [E. coli]. 56% similar to predicted ABC-type permease [P.aeruginosa]. 74% similar to B. subtilis yocS hypothetical gene product. 76% similar to P. stutzeri nosF gene product. 66% similar to putative transporter [B. subtilis] 99% similar to Ferripyochelin binding Protein [P.aeruginosa] 56% similar to Multiphosphoryltransfer Protein M T P [X. campestris] 57% similar to hypothetical Protein YiaN [Haemophilus influenzae]. 51% similar to B. subtilis ProX Protein 65% similar to B. subtilis ProW Protein. 65% similar to putative metal-transporting A T P a s e YvgX [B. subtilis]. 59% similar to ferric enterobactin transport Protein FepD [E. coli] 76% similar to membrane transport Protein GlnQ [B. stearothermophilus]. 50% similar to glutamine A B C transporter [B. subtilis] 56% similar to oprN [P.aeruginosa]. 41 % similar to beta-lactamase induction signal transducer A m p G [E. coli]. 69% similar to hypothetical Protein YajR [E. coli] 78% similar to feoB (ferrous iron transport Protein B) [E. coli] 62% similar to hypothetical Protein YggA [E. coli]. 71% similar to dipeptide transporter Protein DppA [E. coli]. 66% similar to dipeptide transporter Protein DppA [E. coli]. 81 % similar to dipeptide transport system permease Protein DppC [E. col 99% similar to oprJ gene product of [P.aeruginosa] 59% similar to periplasmic C4-dicarboxylate binding-Protein DctP 84% similar to lysine specific permease [E. coli] 64% similar to iron (lll)-transport system permease HitB [H. influenzae]. 57% similar to hemV Protein (ATPase component) [Y.enterocolitica] 49% similar to heme receptor HutA [Vibrio cholerae] 73% similar to putative membrane / transport Protein Y ieG [E. coli]. 57% similar to Small Protein A [E. coli] PA4770 lldP 1.25 Transport of small molecules 84% similar to Putatitve l-lactate permease YghK [E. coli] PA4837 2.60 Transport of small molecules 45% similar to ferrichrome iron receptor FhuA [E. agglomerans] PA4860 1.52 Transport of small molecules 59% similar to hypothetical Protein [Synechocystis sp.]. PA4862 1.48 Transport of small molecules 65% similar to amino acid transport ATP-binding Protein [Synechocystis PA4898 1.54 Transport of small molecules 56% similar to PhaK [P. putida] PA49