UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification of bacterial factors associated with the survival of Burkholderia cenocepacia in a murine… Chung, Jacqueline W. 2004

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

Item Metadata


831-ubc_2004-901661.pdf [ 10.31MB ]
JSON: 831-1.0091756.json
JSON-LD: 831-1.0091756-ld.json
RDF/XML (Pretty): 831-1.0091756-rdf.xml
RDF/JSON: 831-1.0091756-rdf.json
Turtle: 831-1.0091756-turtle.txt
N-Triples: 831-1.0091756-rdf-ntriples.txt
Original Record: 831-1.0091756-source.json
Full Text

Full Text

Identification of Bacterial Factors Associated with the Survival oi Burkholderia cenocepacia in a Murine Host by Jacqueline W. Chung B. Sc., The University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Pathology and Laboratory Medicine We accept this th&sjs as coafoijningAo the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2004 © Jacqueline W. Chung, 2004 Abstract The Burkholderia cepacia complex (BCC) is a family of Gram-negative bacteria of evolving importance as opportunistic pathogens, particularly in patients with cystic fibrosis (CF). BCC infections are difficult to treat and control due to the BCC's intrinsic resistance to antibiotics and ability to spread among patients. Moreover BCC colonization can develop into septicemia. Although recognized as a serious respiratory pathogen, little is known about the pathogenesis of the BCC. The purpose of these studies was to identify bacterial factors involved in BCC infection and colonization using animal models that show differential persistence and virulence among BCC strains in a murine host. Serial mouse passage of non-persistent BCC strain C1394 in a pulmonary infection model resulted in its adaptation to the murine host and produced a variant, C1394mp2 that persisted in the murine lung. The parent strain and variant were indistinguishable by genetic typing, but differed in colonial morphology and were compared to identify the bacterial determinants required for long-term infection in a murine host. The parent strain, C1394 had a matte colonial phenotype, made scant exopolysaccharide (EPS), and was lightly piliated. The variant, C1394mp2 had a shiny colonial phenotype, produced abundant EPS, and was heavily piliated. Matte to shiny colonial transformation was induced by growth at 42°C. Proteomic analysis of C1394 and C1394mp2 protein profiles at 37°C revealed increased flagellin production in C1394mp2 whereas C1394 had increased production of metabolizing enzymes of the tricarboxylic acid cycle. Differential expression of the stress-induced protein peroxiredoxin was observed in the proteomes of C1394 and C1394mp2 at 37°C and 42°C respectively. Further comparisons included in v i t r o assays examining host cell interactions with C1394 and C1394mp2. Though both isolates had poor nonopsonic association with primary human phagocytic cells, C1394mp2 had even less association than C1394. Binding assays with A549 epithelial cells also revealed lower association with C1394mp2 than C1394. The results in this thesis identified BCC surface determinants that were associated with a colonial morphology change, which could be induced under stress, and correlated with long-term infection in a murine host and decreased association with human host cells. Taken together, these results suggest ii putative bacterial factors that are necessary for BCC survival and chronic infections susceptible hosts. iii Table of Contents Abstract ii Table of contents iv List of Tables viii List of Figures ix List of Abbreviations xi Acknowledgements xiii CHAPTER 1 Introduction 1 1.1 Burkholderia cepacia complex (BCC) 1 1.1.1 Taxonomy and microbiology 1 1.1.2 The BCC genome 3 1.1.3 Agricultural role of the BCC 4 1.1.4 Clinical significance of the BCC 5 1.1.5 Emergence of the BCC as a CF pathogen 6 1.2 Gram-negative bacterial cell surface determinants 8 1.2.1 Surface appendages 9 Pili 9 Flagella 9 Type III secretion systems (TTSS) 11 1.2.2 Outer membrane (OM) 11 Lipopolysaccharide (LPS) 12 Outer membrane proteins (OMPs) 13 1.2.3 Exopolysaccharide (EPS) 13 1.2.4 Phase variation of cell surface determinants 15 1.3 Virulence Factors of the BCC 15 1.4 Stress regulation of virulence genes 19 1.5 BCC and host infection studies 20 1.5.1 BCC and host cell interactions 20 Adherence and host cell receptors for BCC 20 Invasion 20 Intracellular survival 22 1.5.2 BCC and mouse models 23 1.6 Thesis objective 24 iv CHAPTER 2 Materials and Methods 26 2.1 Strains, media, and growth conditions 26 2.2 DNA manipulation and genetic typing 27 2.2.1 Generation of the non-persistent transconjugant control, K56-2C 27 2.2.2 Isolation of BCC genomic DNA for molecular analysis 28 2.2.3 Polymerase chain reaction (PCR) 28 Random amplified polymorphic DNA (RAPD) analysis 28 Amplification of the B. cepacia epidemic strain marker (BCESM), cblA gene, and the fliC gene 29 2.2.4 Restriction fragment length polymorphisms (RFLP) analysis of PCR products 29 2.2.5 Pulse field gel electrophoresis (PFGE) 30 2.3 Phenotypic characterization 30 2.3.1 Biochemical analysis 30 2.3.2 Crude LPS extraction 30 2.3.3 Serum sensitivity assay 31 2.3.4 Determination of minimal inhibitory concentrations (MICs) 31 2.3.5 Bacterial biofilm assays 32 2.3.6 EPS extraction and purification 32 2.3.7 lH nuclear magnetic resonance (NMR), sugar, and methylation analyses of EPS 33 2.3.8 Transmission electron microscopy (TEM) 33 2.4 Protein studies 34 2.4.1 Inner and outer membrane proteins extraction 34 2.4.2 Whole cell protein extractions 34 2.4.3 2D gel electrophoresis 35 2.4.4 Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS), liquid chromatography (LC) - MS/MS analysis, and sequencing 36 2.4.5 Western blotting 36 2.5 Mice 37 2.5.1 Intraperitoneal (i.p.) infection model 37 2.5.2 Pulmonary infection model 38 Immunosuppression of BALB/c mice 38 Intranasal (i.n.) infection of BALB/c mice 38 2.6 In vitro studies 38 2.6.1 Isolation and culture of monocyte-derived macrophages (MDMs) 38 2.6.2 Isolation of polymorphonuclear cells (PMNs) 39 2.6.3 Assay for association of bacteria with phagocytic cells 40 2.6.4 A549 cell association assay 40 2.7 Statistical analysis 41 CHAPTER 3 BCC adaptation to the host: role of colonial morphology in B. cenocepacia persistence in the mouse 42 3.1 Isolation and selection for persistent B. cenocepacia strain variants 42 3.2 Genetic typing of persistent strain variant C1394mp2 49 3.3 Phenotypic characterization of C1394 and C1394mp2 49 3.3.1 Basic phenotypes, biochemistry, and microbiology 49 3.3.2 Differential colonial morphology and MICs 52 3.3.3 Static cultures and biofilm formation ,..56 3.3.4 Differential EPS production ' 56 3.3.5 Electron microscopy 61 3.4 Discussion 61 CHAPTER 4 Host cell interactions with nonpersistent B. cenocepacia strain C1394 and persistent variant C1394mp2 66 4.1 Association levels of C1394 and C1394mp2 with phagocytic cells 66 4.2 Binding levels of C1394 and C1394mp2 with A549 cells 68 4.3 Discussion 71 CHAPTER 5 Identification of candidate proteins involved in B. cenocepacia persistence in the mouse 77 5.1 Membrane protein profiles of B. cenocepacia strain C1394 and variant C1394mp2 77 5.2 Proteomic analysis of B. cenocepacia strain C1394 and variant C1394mp2 77 5.2.1 Differentially expressed proteins of C1394 and C1394mp2 grown at37°C 79 Identification of BCC type II flagellin protein with increased expression in C1394mp2 81 Analysis of the fliC gene of C1394 and C1394mp2 87 5.2.2 Differentially expressed proteins of C1394 and C1394mp2 grown at 42°C and 37°C 87 5.3 Discussion 90 CHAPTER 6 General discussion 96 References 102 Appendix 122 vii L i s t of Tables Table 1. Members of the Burkholderia cepacia complex 2 Table 2. Bacterial strains and plasmid 27 Table 3. Primer sequences 29 Table 4. Results of API and biochemical tests for B. cenocepacia 51 Table 5. MICs of antibiotics for C1394 and variant C1394mp2 55 Table 6. LC-tandem MS identification of differentially expressed proteins of C1394 and C1394mp2 83 Table 7. Evaluation of potential recipient B. multivorans strains for the JTC cosmid library 124 viii List of Figures Figure 1. Structure of the two EPS types produced by BCC 14 Figure 2. Intraperitoneal infection of C57B1/6 mice with B. multivorans strain JTC and B. cenocepacia strain K56-2 43 Figure 3. Intraperitoneal infection of C57B1/6 mice with B. cenocepacia strain K56-2, K56-2C and its derivative K56-2Ce 44 Figure 4. Intranasal infection of PBS-treated and CPA-treated BALB/c mice with B. multivorans strain C1576 and B. cenocepacia strain K56-2 46 Figure 5. Intranasal infection of CPA-treated BALB/c mice with B. cenocepacia isolate K56-2C (naive) and its derivative K56-2Ce (mouse passaged)....47 Figure 6. Intranasal infection of PBS-treated and CPA-treated BALB/c mice with B. cenocepacia strain C1394 and its derivative, C1394mp2 48 Figure 7. Spe I generated macrorestriction fragments of C1394 and C1394mp2 separated by PFGE 50 Figure 8. Crude LPS samples from C1394 and C1394mp2 53 Figure 9. Colonial morphology differences between C1394 and C1394mp2 grown on blood agar or on CRM 54 Figure 10. Pellicle formation and settling exhibited by C1394 and C1394mp2 57 Figure 11. Biofilm formation by C1394 and C1394mp2 in different growth media 58 Figure 12. 'H NMR spectrum of O-deactylated EPS of C1394mp2 60 Figure 13. Electron micrographs showing piliated bacteria after 24 hours of growth on LB agar 62 Figure 14. Nonopsonized association of C1394 and C1394mp2 with human MDMs at different MOIs 67 Figure 15. Nonopsonic and opsonic association of C1394 and C1394mp2 with human neutrophils at an MOI of 50:1 69 Figure 16. Effects of bacterial culture preparation on bacterial binding to A549 cells at an MOI of 100:1 70 ix Figure 17. Binding of C1394 and C1394mp2 to A549 cells after centrifugation 72 Figure 18. Outer and inner membrane protein profiles of C1394 and C1394mp2....78 Figure 19. 2D-PAGE of protein profiles of the proteins in C1394 and C1394mp2 grown in LB broth at 37°C and extracted at stationary phase 80 Figure 20. Protein spots C4, C5, and C7ab in silver stained 2D-PAGE of C1394 and C1394mp2 grown at 37°C 82 Figure 21. Protein spots M l , M2, and M3 in silver stained 2D-PAGE of C1394 and C1394mp2 grown at 37°C 84 Figure 22. Immunoblot analysis of flagellin protein in C1394 and C1394mp2 86 Figure 23. fliC PCR products of C1394 and C1394mp2 and PCR/RFLP patterns generated with the endonucleases Haelll, Mspl, and Pstl 88 Figure 24. Protein spot C l in silver stained 2D-PAGE of C1394 and C1394mp2 grown at 37°C 89 X List of Abbreviations 2D two-dimensional aa amino acid AHL acylhomoserine lactone AhpC alkyl hydroperoxide reductase subunit C ASF airway surface fluid ATP adenosine triphosphate BCC Burkholderia cepacia complex BCESM Burkholderia cepacia epidemic strain marker BEC buccal epithelial cells BLAST basic local alignment search tool C-terminal carboxyl-terminal CAA casamino acids Cbl cable CF cystic fibrosis CFTR cystic fibrosis transmembrane regulator CFU colony forming units CGD chronic granulomatous disease CK13 cytokeratin 13 CPA cyclophosphamide CR3 complement receptor 3 CRM Congo Red medium DMEM Dulbecco's modified Eagle's medium DMSO dimethylsulfoxide DNA deoxyribonucleic acid dNTP nucleotide triphosphates DTT dithiothreitol EDTA ethylene diamine tetraacetic acid E M electron microscopy EPS exopolysaccharide GHBSS Hank's Balanced Salt Solution containing 0.1%w/v gelatin Gb3 globotriosylceramide Gg 3 gangliotriosylceramide Gg 4 gangliotetraosylceramide GLC gas-liquid chromatography HBSS Hank's Balanced Salt Solution HSL homoserine lactone ID identification IEF isoelectric focusing IL interleukin IM inner membrane i.n. intranasal i.p. intraperitoneal xi IPG immobilized pH gradient LB Luria-Bertani LC liquid chromatography LDH lactate dehydrogenase LPS lipopolysaccharide MALDI-TOF matrix-assisted laser desorption ionization-time of flight M D M monocyte-derived macrophage M H Mueller-Hinton MIC minimal inhibitory concentration MOI multiplicity of infection MS mass spectrometry N-terminal amino-terminal NCBI National Center for Biotechnology Information NMR nucleic magnetic resonance OD optical density OF oxidation-fermentation OM outer membrane OMP outer membrane protein PAGE polyacrylamide gel electrophoresis PAMP pathogen-associated molecular pattern PBS phosphate buffered saline PCR polymerase chain reaction PFGE pulsed-field gel electrophoresis PHS pooled human serum pi isoelectric point PMN polymorphonuclear cell PMSF phenylmethylsulfonyl fluoride RAPD random amplified polymorphic DNA RFLP restriction fragment length polymorphism SBSM supplemented basal salts medium SDS sodium dodecyl sulphate SEM standard error of the mean SLB supplemented Luria-Bertani TCA tricarboxylic acid TEM transmission electron microscopy TIR Toll/IL-1 receptor TLR toll-like receptor TNF-a tumour necrosis factor alpha TSA tryptic soy agar TTSS type III secretion system Y E M yeast-extract mannitol xii Acknowledgments I would like to thank my supervisor David Speert for his guidance, patience, unwavering confidence in my abilities, and for taking a chance on me as a lab volunteer way back when. I would also like to thank my supervisory committee Brett Finlay, Colin Fyfe, and Niamh Kelly for their advice and input on the project. I thank the Canadian Cystic Fibrosis Foundation for financial support, and many thanks go to Dr. Julian Davies and his lab for their generosity and use of their 2D apparatus. I am grateful to Dr. Eleonora Altman, Dr. Terry Beveridge, and Dianne Moyles for their help and contribution to the project. A big thank you to those who have nudged me and my project along: Dr. Esh Mahenthiralingam and Dr. Dave Simpson for their teachings in molecular biology; Keith Halsey and Dan Doxsee for their help with the mouse experiments; Karen Chu for her assistance in the animal work; Dr. Barb Conway for technical advice, helpful discussions, and for helping me keep things in perspective; and Dr. Lucy Brooks for introducing me to the very sexy 2D technique and assuaging many anxiety attacks in and out of the lab. I would also like to thank the girls of the third floor of the institute for their friendship and company, especially during late night experiments, and members of the Speert lab for their assistance, advice, camaraderie, and supply of UK music! Lastly, cheers to my friends for being my personal cheerleading section and keeping my sanity in check, and a very special thanks to my family for their love and emotional support. xiii CHAPTER 1 Introduction 1.1 Burkholderia cepacia complex (BCC) The Burkholderia cepacia complex is a diverse family of Gram-negative bacteria that have recently emerged as opportunistic pathogens, particularly in individuals with cystic fibrosis (CF). BCC colonization in these individuals can develop into devastating clinical infection outcomes, which include fatal pneumonia with bacteremia. This is in distinct contrast to colonization with Pseudomonas aeruginosa, the most common CF pathogen, which is associated with slow progressive lung disease that is rarely invasive or systemic in nature. Although recognized as serious respiratory pathogens, certain BCC strains are also recognized as attractive candidates for bioremediation and biocontrol due to their degradative capabilities and production of antimicrobial agents (60, 129). However, with its virulence still poorly understood and the differences between clinical and environmental strains not yet established, the release of the BCC into the environment may present a health hazard for susceptible individuals. The growing significance of the BCC in both medical and agricultural microbiology necessitate the need to investigate the pathogenicity of the BCC and identify the virulence, factors that enable these organisms to cause severe infections in certain compromised hosts. 1.1.1 Taxonomy and microbiology Burkholderia cepacia was originally described as a phytopathogen responsible for soft rot in onions and classifed as a species of Pseudomonas in 1950 (18). Since identified as a serious human pathogen, the taxonomy of this bacterium has rapidly evolved. In 1992, P. cepacia was reclassified under the new genus Burkholderia, which belongs to the p-subdivision of the phylum Proteobacteria, in contrast to the y-subdivision containing the genus Pseudomonas (207). In the subsequent isolation of Burkholderia strains, several researchers noted a marked heterogeneity among presumed "2?. cepacia''' strains that presented problems in the accurate identification of the organism (36, 194). Based on a polyphasic taxonomic approach that included whole-cell protein and fatty acid analysis, DNA-DNA homology, DNA-rRNA homology, and phenotypic characteristics, B. cepacia was found to consist of at least five phenotypically similar but genetically distinct species that were denoted as genomovars (194). These genomovars 1 shared moderate levels of DNA-DNA hybridization (30-60%) and were collectively referred to as the B. cepacia complex (BCC) (192, 194). Analysis of the conserved housekeeping gene recA provided a means to differentiate genomovars I-V and further delineated genomovar III into four distinct clusters, III-A to III-D (105, 193). Further work has identified at least nine genomovars comprised in the BCC (104). With the exception of genomovar VI, the genomovars that have been clearly differentiated from all others in diagnostic laboratories have been assigned the following species designation: B. cepacia (formerly genomovar I and identified as the type strain for the species), B. multivorans (genomovar II), B. cenocepacia (genomovar III), B. stabilis (genomovar IV), B. vietnamiensis (genomovar V), B. ambifaria (genomovar VII), B. anthina (genomovar VIII), and B. pyrrocinia (genomovar IX) (Table 1) (36, 192, 193). In light of the BCC's growing significance in agricultural and medical microbiology and evolving taxonomic structure, a panel of representative strains of the first five BCC genomovars was published in 2000 and recently updated to assist researchers and clinicians in the analysis of BCC strains (107). Table 1. M e m b e r s of the Burkholderia cepacia complex B C C member Species designation Source Genomovar I Genomovar II B. cepacia B. multivorans Plant pathogen (onion); rhizosphere, soil , water, humans (non-C F and C F ) . Includes type strain Rhizosphere, humans (non-CF and C F ) Genomovar III B. cenocepacia Hospital environment, humans (non-CF and C F ) , rhizosphere, soil Hospital environment, humans (non-CF and C F ) Genomovar I V B. stabilis Genomovar V B. vietnamiensis Rhizosphere, soil , humans (non-CF and C F ) . Fixes nitrogen Genomovar V I - Humans (CF) Genomovar VI I Genomovar VIII B. ambifaria B. anthina Rhizosphere, soil , humans (CF) . Includes many biocontrol strains Rhizosphere, hospital environment, humans (CF) Genomovar I X B. pyrrocinia Soi l , human (CF) Adapted from Parke et al. 2001 and Vandamme et al. 2002 (129, 192) Phenotypically, BCC organisms are described as aerobic, non-spore-forming, non-fluorescent, motile bacteria of rod shape and averaging 1.9 pm in length. BCC strains are 2 weakly oxidase positive and catalase positive, although some catalase negative strains have been reported. Various non-fluorescent pigments may be produced, particularly by genomovar I strains. The optimal temperature for growth is 30-35°C, and growth at 42°C is frequently used to distinguish among BCC genomovars. BCC organisms can occupy many different environments but their natural habitats have been described as soil, water and the rhizosphere of crop plants (129). In general, genomovar I contains predominantly environmental isolates and the phytopathogens of the BCC (107). B. ambifaria (genomovar VII) also contains environmental specimens with several biocontrol strains considered for commercial use (36). However, all genomovars have strains of clinical origin (129, 192). The ability to infect both plant and animal tissue is reflective of its unique genome, which permits survival in such diverse environments. 1.1.2 The BCC genome The BCC genome is GC-rich, consisting of two to four chromosomes (replicons) with one or more plasmids, and has an overall size ranging from 5 to 9 Mb (98, 129). The genome also harbours an extensive array of insertion sequences (IS) which contribute to the genomic plasticity and phenotypic diversity of the BCC. IS are mobile DNA elements that can promote genetic rearrangements, recruit foreign genes, and cause insertional gene inactivation (191). In the BCC, IS elements were reported to promote deletions and inversions of plasmids, promote replicon fusions, increase expression of neighbouring genes, and activate transcription of lac genes on a broad-host range plasmid (10, 69, 98, 165, 205, 206). More importantly IS elements have had a significant role in the recruitment and expression of novel catabolic functions which include the degradation of complex compounds such as 2, 4, 5-tricholorophenoxyacetic acid (2, 4, 5-T), a potent herbicide used extensively in agriculture (42, 70). The association of foreign genes with IS elements is implicated in expanding the gene pool within strains and likely plays a role in the evolution of the BCC (56, 99, 167). In particular, the pathogenic evolution of the BCC from plant pathogen to human pathogen may be attributable to IS elements which can facilitate the movement and acquisition of clustered virulence genes or pathogenicity islands from other pathogens. 3 Studies have reported a high frequency of recombination in the environmental population of the BCC, and BCC bacteriophages with inter-genomovar host ranges have also been implicated in the transfer of genes in the BCC (96, 129, 204). A genome sequencing project of the highly transmissible CF strain B. cenocepacia (genomovar III) strain J2315 has recently been completed in the United Kingdom (http://www.sanger.ac.uk/Proiects/B cepacia/) and is currently undergoing annotative analysis. This project will provide greater insight into the complex organization of the BCC genome and reveal genes that contribute to the pathogenic nature and versatility of the BCC. 1.1.3. Agricultural role of the BCC With their impressive metabolic capacities, BCC organisms are attractive candidates among the environmentally friendly agents being sought for bioremediation purposes. BCC organisms are able to utilize constituents of crude oil as a sole carbon source, and can degrade chlorinated aromatic substrates found in complex pesticides and herbicides, some with carcinogenic potential (62, 78). Most notable is their ability to degrade 2,4,5-T, an herbicide used world-wide for weed control and a component in the highly potent "agent orange" (42). These attributes have driven the development of BCC strains for use in the bioremediation of soil and groundwater contaminated with organic pollutants. BCC organisms also have significant appeal as biocontrol agents as they are able to repress many soilborne plant pathogens related to economic losses and can improve germination and crop yields of maize, wheat, and rice (129). BCC produce several antimicrobial agents such as pyrrolnitrins, altercidins, cepalycins and bacteriocin-like agents inhibiting bacterial and fungal phytopathogens and suppressing plant diseases such as leaf and stem blight caused by the fungus Alternaria, root rot of legumes by Aphanomyces euteiches, Rhizoctonia solani stem rot of poinsettia, and tobacco wilt by P. solanacearum (4, 17, 24, 78, 90). Other studies have shown that boosting the BCC soil population after planting seeds and seedlings or foliar spraying can prevent disease from early-invading pathogens, and is similar to the protection provided by toxic and non-biodegradable fungicides (129, 173, 174). BCC biocontrol can also be applied to prevent postharvest diseases and fungal spoilage of fruits (129). 4 As an alternative to chemical agents for biocontrol and bioremediation, the BCC is attractive to several companies for routine agricultural use and is under consideration by North American government agencies examining licence applications for the use of BCC strains as microbial pesticides. However, the medical community has raised several concerns about the release of BCC strains into the environment, an event that may be a health hazard for susceptible patients and individuals. There is the potential risk of contaminating water supplies and foodstuffs with BCC strains, which can survive for long periods in water and nutritionally-deficient conditions (61). At present, there is little to clearly distinguish environmental BCC strains from clinical isolates and reports have revealed that the pathogenic potential of these organisms is not necessarily confined to their environmental niche. This was demonstrated by an environmental isolate causing foot lesions (swamp foot) in military personnel during jungle training and a clinical isolate that could macerate onion tissue (21,61). The genetic plasticity of the BCC genome is also a cause for concern for environmental isolates causing human infections de novo or through adaptation. Moreover the virulence traits that enable BCC to succeed as a phytopathogen may be relevant for human infections. For example, type III secretion systems, which deliver bacterial toxins into host cells, are recognized virulence determinants in both plant and animal pathogens. Consequently the medical community has pressed for a moratorium on the development of BCC strains for agricultural use until more is known about the pathogenic potential of environmental isolates and the bacterial factors that differentiate them from virulent clinical strains. 1.1.4 Clinical significance of the BCC Prior to 1980, sporadic cases of human infections caused by the BCC were predominantly limited to hospitalized patients and were attributable to contaminated disinfectant and anesthetic solutions (60). Upon analysis of the U.S. database of nosocomial infections from 1980 to 1985, a study by Jarvis et al. noted a significant rise in the incidence of BCC infections with the lower respiratory tract as the most frequently reported site of infection (59, 87). More disturbing was that these infections were often fatal. BCC tropism for the respiratory tract was also exhibited in patients with chronic 5 granulomatous disease (CGD) who are at high risk for BCC septicemia and pneumonia, which is the second leading cause of death in these individuals (166, 201). CGD is a rare congenital disease, affecting 1 in approximately 250,000 individuals and is characterized by recurrent, often life-threatening bacterial and fungal infections, and dysregulated granuloma formation. The basis for this clinical phenotype is the inability of phagocytic leukocytes (neutrophils, eosinophils, monocytes, and macrophages) to generate reactive oxygen radicals required for intracellular killing of phagocytosed microorganisms. Hence CGD neutrophils are reliant upon nonoxidative killing mechanisms (175). A study by Speert et al. revealed that BCC strains are resistant to nonoxidative killing by neutrophils, explaining their particular virulence in these hosts (175). A few cases of BCC infections in healthy individuals have been documented albeit very rarely (60). Generally, BCC infections appear to be confined to individuals with compromised immune status; additional reports include BCC association with community-acquired bacteremia and pneumonia, and multiple brain abscesses in immunocompetent individuals, recovery from HIV patients with acute bronchiectasis, and pneumonia in cancer patients (77, 97, 134, 136, 197). However, the BCC has gained the most notoreity as human pathogens in individuals with CF who are susceptible to bacterial infections in the lung. Interestingly, a predilection for compromised hosts is also shared with the BCC phytopathogens which do not usually cause disease in healthy onions but do infect "wounded" onions that have been damaged during harvesting (60, 173, 174). 1.1.5 Emergence of the BCC as a CF pathogen Much of what is known about the BCC as human pathogens comes from its association with severe pulmonary infections in CF patients. CF is an autosomal recessive disorder affecting approximately 1 in 2500 live births (carrier frequency of 1 in 25) and caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. CFTR functions as an ATP-driven pump that actively transports chloride ions across the apical membrane of epithelial cells. Approximately 70% of the CF population carry a deletion in a single codon for phenylalanine in one of the nucleotide-binding folds of the CFTR at position 508 (AF508). Defects in the CFTR results in 6 altered epithelial secretions and buildup of thick dehydrated mucus in male sex ducts, ducts of the pancreas, and the airways of the CF lung (59, 117). The increased viscosity of airway mucus impairs normal mucociliary clearance mechanisms, promoting persistent microbial infection by various pathogens. The CF airway surface fluid (ASF), in particular, has been the subject of much debate with two competing theories that have been proposed for the ASF in its contribution to microbial infections. The hypotonic theory suggests that defective chloride transport of the CFTR results in increased salt concentration in ASF lining CF airways which inactivates antimicrobial peptides (83, 200). The isotonic theory suggests that CFTR dysfunction is responsible for the increased absorbance of the surface fluid affecting the volume layer and contributing to the concentrated mucus that impairs mucociliary clearance (83, 200). Both theories can explain, in part, the deficienct nature in clearing pulmonary infections in CF, however recent evidence has favoured the isotonic theory, demonstrating that the ASF is similar in healthy control individuals and subjects with CF (57). Chronic bacterial infection of the lung and associated inflammation are the major causes of morbidity and mortality in patients with CF. P. aeruginosa is the most common CF pathogen and the mucoid variants of this bacterium are recognized as a poor prognostic indicator as they are seldom eradicated after colonization. However, in the early 1980s, BCC organisms were isolated with increasing frequency with prevalence rates as high as 40% in some CF clinics (59). Seminal reports documenting the increased BCC colonization described a syndrome of necrotizing pneumonia and bacteremia, culminating in rapid deterioration and death which was not observed with P. aeruginosa (86). This syndrome ("B. cepacia syndrome") affects approximately 20% of BCC-colonized patients while others sustain BCC infections for prolonged periods of time without adverse effects. The variability in clinical outcomes is thought to be due in part to heterogeneous strains representing different genomovars in the BCC, all of which have been isolated from CF patients. Further distinction between the BCC and other CF pathogens was made with evidence of patient-to-patient transmission in and out of hospitals, and led to cohorting of BCC-colonized individuals. Hastened by these events, epidemiologic research identified B. cenocepacia (genomovar III), particularly of the III-A cluster, among infections 7 associated with epidemic spread, high mortality rates and highest prevalence in CF patients (104, 177). In North America, B. multivorans (genomovar II) is secondary to B. cenocepacia as a CF pathogen (177). Transmission outbreaks caused by B. multivorans appear to be more region-specific with isolated cases in the UK and France (104, 177). Moreover, a study of the Vancouver, BC CF clinic reported the replacement of B. multivorans infections by virulent B. cenocepacia strains, further establishing the pathogenic status of the latter (110). The acquisition of these strains in CF patients appears to be mainly from other colonized patients, with a higher risk if hospitalized. However, with the advancements made in BCC selective media, more B. cenocepacia isolates have been recovered with greater frequency in agricultural soil and suggests the environment as a potential source of new clinical BCC isolates (9, 129). The reasons for the emergence of the BCC as a major pathogen are still not clear, though it is speculated that the use of aggressive antibiotic therapies led to the selection of these organisms. Treatment of BCC infections is especially difficult due to the innate multi-drug-resistance of BCC, which is accented by its ability to use penicillin G as a sole carbon source (60). Current therapies often require triple antibiotic combinations to control BCC infections (1). BCC infections are also associated with high rates of mortality following lung transplantation of CF patients, resulting in the denial of this procedure to BCC-infected CF patients in some centers (5, 44). The clinical significance of the BCC is evident from the implementation of severely stringent infection control policies. Despite advancements made in the taxonomy and epidemiological studies of these organisms, the pathogenesis and virulence of the BCC remains largely unexplained. 1.2 Gram-negative bacterial cell surface determinants Cell surface structures of Gram-negative bacteria are directly involved in the interaction with various environments and often have essential roles in bacterial virulence. The following includes key components in the Gram-negative cell envelope that may contribute to bacterial survival and adaptation to host niches. 8 1.2.1 Surface appendages Pili Pili and/or fimbriae, are filamentous structures composed of protein subunits or pilins that extend from the bacterial surface. Pili have a role in gene transfer by conjugation, as receptors for bacteriophage, and attachment to host mucosal and cellular surfaces, which is a critical step in successful bacterial colonization and subsequent disease (54, 94). Type I and IV pili represent two well-characterized model systems and are ubiquitous among important Gram-negative pathogens such as E. coli and P. aeruginosa respectively. In addition, type IV pili are responsible for a form of surface movement called twitching motility and have been implicated in the initial attachment and development of biofilm by P. aeruginosa (127). P pilus-mediated adherence of uropathogenic E. coli to its receptor stimulated transcription of several RNAs in the attached bacterium (208). Hence binding by pili not only establishes infection in a host but also appears to trigger a cascade of factors that enable the organism to survive in its new environment (94). Pilus-mediated attachment to respiratory epithelial cells has been suggested as an important step leading to chronic BCC colonization and sometimes invasion (117). The cable pilus is the only well-characterized adhesin of the BCC, however four other fimbrial structures have been described which include mesh (Msh), filamentous (Fil), spine (Spn), and spike (Spk) pili (58). Al l five types of BCC pili appear to be arranged peritrichously. Msh pili are the most widespread among BCC isolates, often co-expressed with other pili types and detected in both epidemic and non-epidemic CF isolates, as well as environmental strains (58). Distribution of the other BCC pilus types are more restricted where expression of Fil pili correlated with non-epidemic CF isolates, Spn with non-CF clinical isolates and Spk with environmental isolates (58). Al l four pilus types remain poorly characterized and their potential role in BCC pathogenesis is unknown. Flagella Flagella are proteinaceous structures arranged as polar or peritrichous extensions responsible for bacterial motility and chemotaxis, and are well conserved among diverse bacterial species. These complex organelles are thicker than pili and consist of a flagellin 9 filament, hook, and basal body that span the bacterial membrane with distal components secreted via a type III secretion system (TTSS) located at the base of the structure. Assembly of the flagella is highly complex, and flagella biosynthesis genes are classified by their temporal requirement and processed in a hierarchal order. Control of expression is regulated by positive and negative feedback loops in which transcription of each class is sensitive to the previous stage of assembly. Class I genes are the regulatory proteins that are first transcribed, and control expression of the entire regulon. Class II genes include most of the structural proteins of the hook-basal-body and TTSS while class III genes encode proteins of the filament, motor force generators, and chemosensory machinery (3). Bacterial motility has been associated with the virulence of pathogens such as Helicobacter pylori, P. aeruginosa, B. pseudomallei and Campylobacter jejuni (33, 50, 53, 120, 122). Flagellar motility enables bacteria to come in contact with the host cell or surfaces for biofilm formation and appears to enhance the invasiveness of some pathogens, facilitating penetration of host epithelial cell barriers. Flagella have also been shown to participate in the adhesion process for mucin-binding and the initial attachment that precedes host cell invasion or biofilm formation (7, 47, 85). Flagella are highly immunogenic in both animal and plant hosts. Flagellin protein is considered one of the conserved motifs regarded as pathogen-associated molecular patterns (PAMPs) that are recognized by the innate immune system of eukaryotic hosts. PAMPs are recognized by toll-like receptors (TLRs), a family of type I transmembrane proteins with a Toll/IL-1 receptor (TIR) signalling domain that initiate a signal transduction cascade promoting effective immunity (2, 72, 196). Flagellin is specifically recognized by TLR5 (72). Recently the role of flagella was described for the BCC in which mutants with impaired flagellar biogenesis were less invasive in cultured epithelial cells than the wild type (188). The flagellar genes that were identified among analysed mutants encoded homologs of the structural and regulatory components of flagella in E. coli and Salmonella spp. (188). Thus far, two types of BCC flagellin have been identified and sequenced. Type I and II of BCC flagellin are 55 and 45 kDa respectively. Both flagellin types have over 70% amino acid sequence identity with the flagellin of B. pseudomallei, a 10 closely related intracellular pathogen that causes meliodosis (65). Moreover the BCC flagellin genes exhibit significant sequence variability among BCC isolates and have been used as a genotyping tool for epidemiological and phylogenetic studies (190, 202). Type III secretion systems (TTSS) As previously mentioned, type III secretion systems (TTSS) are found in the flagellar apparatus to export and assemble flagella at the bacterial cell surface. However, this conserved secretory system is primarily involved in delivering bacterial virulence proteins directly to the cytosol of eukaryotic host cells. This delivery system has been associated with inhibition of phagocytosis and evasion of the host immune response by Yersinia spp. and Bordetella bronchiseptica, cytotoxicity by P. aeruginosa, and invasion by S. typhimurium and Shigella flexneri (80). TTSS are also essential for phytopathogens to cause disease in susceptible host plants and induce the hypersensitive response (HR) in resistant plants (80). Probes constructed from the TTSS genes from the plant pathogen Ralstonia solanacearum and B. pseudomallei, identified putative TTSS genes in B. cenocepacia strain J2315 (132). More recently a null mutant was generated in strain J2315 in the TTSS gene designated as bscN (187). This gene encodes an ATP-binding protein required to provide energy for protein secretion. Compared to the wild-type parent, the bscN mutant showed attenuated virulence in mice with reduced extrapulmonary spread to the spleen and less pathology. These data provide evidence of TTSS in the BCC with a role in establishing infection. 1.2.2 Outer membrane (OM) Gram-negative bacteria possess a cell envelope of two membranes separated by the periplasmic space and a peptidoglycan layer. The inner membrane is a phospholipid bilayer containing active transport systems whereas the outer membrane (OM) contains an inner leaflet of phospholipids, an outer leaflet of LPS, and OM proteins (OMP) (67, 101). 11 Lipopolysaccharide (LPS) Lipopolysaccharide (LPS) is a potent agent of inflammation, known for its immuno-stimulatory effects on mononuclear cells, granulocytes and B-lymphocytes, and a described PAMP recognized by TLR4 (2, 143). LPS-induced cytokine production can develop into an overwhelming inflammatory response that can lead to severe complications such as septic shock. The endotoxic activity of LPS is attributed to the hydrophobic glucosamine-based phospholipid component, lipid A. Lipid A anchors LPS to the outer leaflet of the OM and is covalently attached to a core oligosaccharide, which is linked to the O-antigenic side chain. The core oligosaccharide consists of an inner and outer core of which the inner core is proximal to lipid A and contains sugars heptose and 2-keto-3-deoxyoctonic acid (KDO). The outer core consists of hexoses and provides an attachment site for O polysaccharide (O-antigen) which has a variable number of repeating saccharide units contributing to its structural heterogeneity and determines serotype specificity. In addition, the O-antigen is responsible for biological functions such as resistance to serum-mediated killing, phagocytosis and cationic peptides (39, 41, 139). Bacteria that lack O-antigenic side chains are commonly referred to as rough due to their appearance in colonial morphology, while bacteria possessing this portion are referred to as smooth. P. aeruginosa isolates from chronically infected CF patients exhibit alginate biosynthesis and are often serum sensitive due to the absence of the O-antigen side chain (22, 66). Conversely, BCC isolates from CF patients can have rough or smooth LPS chemotypes which do not necessarily predict susceptibility to serum killing as some BCC isolates with smooth LPS are serum sensitive (22, 51). Rough LPS forms in BCC strains associated with bacteremia in CF suggests other mechanisms of serum resistance or contrasting LPS forms under in vitro conditions. Variation in LPS chemotype and antigenic cross reactivity was observed in the same BCC strain grown in different conditions (51). Hence, modifying the LPS phenotype may have a role in BCC adaptability in the host. 1 2 Outer Membrane Proteins (OMPs) Outer membrane proteins (OMPs) include integral membrane proteins and lipoproteins that make up approximately 50% of the OM mass (101). OMPs have P-barrel structures and are often arranged to form channels or pores through which small molecules can pass (13). Essential for the integrity and selective permeability of the membrane, OMPs are regulated by environmental signals and have roles in bacterial adaptive responses to nutrient deprivation, antimicrobial agents, and serum in the host. They can also serve as non-fimbrial adhesins, facilitating attachment to host cells or cell autoaggregation (8, 73, 91). Previous reports of BCC antibiotic resistance have implicated the impermeability of the OM as the primary mechanism of resistance (6, 20, 121). Small channel sizes in BCC pores were thought to play a role in decreasing diffusion of P-lactam antibiotics while a lack of cation-binding sites in the LPS provided BCC resistance against polycations such as aminoglycosides and polymyxin (20, 130). BCC resistance to chloramphenicol, trimethoprim, and ciprofloxacin was ascribed to an OM lipoprotein that was part of an antibiotic efflux pump with homology to the P. aeruginosa multiple antibotic resistance operon mexA-mexB-oprM (20). Another characterized BCC OMP is an iron-regulated OM receptor for the siderophore ornibactin. Inactivation of the gene encoding this receptor resulted in attenuated virulence in a murine model of chronic respiratory infection, indicating the importance of siderophore uptake and utilization in BCC pathogenesis (171). 1.2.3 Exopolysaccharide (EPS) Exopolysaccharide (EPS) is produced by several bacterial species and has been implicated as an important virulence factor for many pathogens. Described EPS functions including prevention of dessication, adherence, and resistance to host immunity (144). The EPS production of P. aeruginosa has been extensively studied due to its ramification in the infection course of CF patients. Increased production of the EPS alginate results in the mucoid phenotype of P. aeruginosa isolates associated with chronic pulmonary infection in CF (45, 59). However, evidence strongly suggests that the initial colonizers of the CF lung are non-mucoid P. aeruginosa strains that eventually convert to 13 mucoid form due to selective pressures in the CF lung (59). Alginate contributes to the extracellular matrix of biofilm, provides a barrier against certain antimicrobial agents and opsonic and nonopsonic phagocytosis, scavenges reactive oxygen intermediates, and has immunomodulatory effects that include suppression of lymphocyte functions (59). Recent reports of mucoid BCC CF isolates have raised interest in the pathogenic role for the EPS in BCC. Most BCC strains, including non-mucoid isolates, are capable of EPS production (26). BCC produce EPS that are biochemically distinct from alginate which is a linear copolymer of (1^4)-linked P-D-mannuronic acid and its C-5 epimer a-L-guluronic acid (45). ->3)-p-D-Glcp(l->3)-a-D-Galp(l-> X CH3 COOH PS I p-D-Galp( 1 ->2)-cc-D-Rhap( 1 ->4) (6<-1 )-p-D-Galp I I ->3)-a-D-GlcpA(l->3)-a-D-Manp(l->3)-p-D-Glcp-> I a-D-Galp(l->2) PS II Figure 1. Structure of the two EPS types produced by BCC. Adapted from Cerantola et al., 2000 (26). Two EPS types have been characterized in the BCC (Figure 1). EPS type PS I is made up of a disaccharide composed of 3-linked P-D-glucopyranosyl and 4,6-0-(l-carboxyethylidene)-a-D-galactopyranosyl residues (28). EPS type PS II is more complex, consisting of a heptasaccharide repeat unit of a linear poly (l—»3) trisaccharide with a mannosyl, a glucosyl and glucuronosyl residue (28). The glucuronic acid residue is substituted on C-2 and on C-3 with two single galactosyl residues and a disaccharide of galactosyl and rhamnosyl residue, respectively. The heptasaccharide complex appears to 14 be the most prevalent EPS type in clinical BCC isolates, suggesting the importance of this EPS in BCC infection (28). The disaccharide complex has a limited distribution in the BCC, however, concomitant production of both EPS types was reported in environmental strains (26). While the roles of these EPS types remain unknown, several studies are underway to investigate their pathogenic potential. 1.2.4 Phase variation of cell surface determinants Many bacterial pathogens employ phase variation, a strategy that promotes a heterogeneous bacterial population as a result of high frequency, reversible switching of phenotypic traits (48, 74). Phase variation usually involves surface structures such as pili, flagella, OMP, and capsule that facilitate colonization or evasion of host defense (48). As these structures are often targets of the host immune response, varying these components enables pathogens to generate diversity and infectious strain-derivatives to evade the immune system and permit reinfection (159). Phase variation is often mediated by genomic rearrangements that activate or repress gene transcription. For example, type I fimbrial variation of E. coli is mediated by site-specific DNA inversions while DNA recombination is associated with the variation of type IV pili of Neisseria gonorrhoeae (74). While phase variation is often associated with the unstable presence of cellular components, antigenic variation describes the ability of bacteria to express several different phenotypes of a component (159). The latter mechanism is also demonstrated in N. gonorrhoeae, which undergoes homologous recombination between two pilin loci to generate antigenic variants of pili (48, 63). Cell surface changes resulting from such alterations are often reflected in the colonial morphology. 1.3 Virulence Factors of the BCC BCC organisms possess a wide range of potential virulence determinants of which most have been isolated in relation to its pathogenesis in CF. Several extracellular products identified in BCC strains have been important virulence factors in other pathogens such as P. aeruginosa. Among the characterized BCC exoproducts are a purified protease that caused bronchopneumonia after intratracheal instillation in rat lungs, and a hemolysin of the highly transmissible B. cenocepacia strain J2315, that 15 induced apoptosis in cultured mouse macrophages and degranulation of human neutrophils (84, 115). This strain also produced a melanin-like pigment that scavenged superoxide radicals, providing protection against the respiratory burst of macrophages (210). In a study using rat alveolar macrophages, BCC lipase was found to significantly impair phagocytic activity and modify the number of pseudopodia and microvilli on treated macrophages (183). In addition, Melnikov et al., demonstrated BCC secretion of ATP-utilizing enzymes that had cytotoxic effects on phagocytic cells by modulating their effluxed ATP levels (116). A follow up study by Punj et al., revealed two additional secreted proteins, homologues of two redox proteins, that triggered apoptosis in macrophages and mast cells (137). While these studies show exoproducts that may allow BCC evasion of host defenses, the production of these factors is variable among CF isolates and their role in CF lung disease remains unclear. LPS, a major constituent of the outer membrane of the Gram-negative bacterial cell envelope, has been studied extensively in the BCC. The LPS of the BCC appears to be highly endotoxic and may contribute to the elevated inflammatory response associated with BCC infections. LPS preparations from clinical and environmental BCC isolates were able to induce high levels of the cytokine tumour necrosis factor alpha (TNF-a) from human blood and cultured monocytes (168, 209). LPS from B. cenocepacia strain J2315 was able to stimulate upregulation of proinflammatory cytokines interleukin-6 (IL-6) and IL-8 from whole blood, prime neutrophil respiratory burst responses, and increase surface expression of complement receptor 3 (CR3) on human neutrophils (81, 82). Since increased surface expression of CR3 is involved in neutrophil functions that include adhesion, phagocytosis, activation and transmigration, the LPS of BCC may have a role in neutrophil recruitment and primed respiratory burst that contributes to the inflammatory damage resulting in BCC bacteremia (81). Studies investigating BCC adherence to the respiratory epithelial surface have revealed the piliation of these organisms. Sajjan et al. demonstrated piliated CF isolates of the BCC that showed specific, saturable binding to CF and non-CF mucins (mucus glycoproteins) (153). Moreover the isolates exhibiting the highest average mucin binding appeared to correlate with severe illness, suggesting some clinical relevance of BCC mucin-binding (153): Further studies revealed that the mucin binding was mediated by a 16 22 kDa adhesin dispersed along the length of peritrichous pili of novel morphological structure (149). These cable (Cbl) pili were named for their unique appearance as giant intertwined fibers extending 2-4 um in length, and enhancing the adhesion of BCC isolates to host cells (58, 154). The major subunit pilin of the Cbl pili is a 17 kDa protein with homology to the CS1 family of fimbriae of enterotoxigenic Escherichia coli (ETEC). The cblA gene encoding this pilin protein has been sequenced from an isolate of the epidemic B. cenocepacia strain lineage ET12, which was associated with intercontinental transmission among CF patients in North America and Europe (154, 184). The prevalence of the Cbl pilus appears primarily restricted to clones of the ET12 lineage, however the cblA gene has been detected in a non-CF clinical isolate and an onion pathogenic isolate of BCC (140). The Burkholderia cepacia Epidemic Strain Marker (BCESM) is another genetic element used in epidemiological studies, and is considered as a marker of transmissibility due to its association with seven epidemic CF strains (108). The sequence analysis of this 1.4 kb marker revealed an open reading frame with significant homology to negative transcriptional regulators (108). Though not fully characterized, these putative gene products are speculated to have some role in regulating BCC virulence determinants relevant for the spread of BCC among patients with CF. Siderophores enable bacteria to acquire and compete for iron in the host. BCC organisms produce four different iron-binding siderophores: salicylic acid, ornibactins, pyochelin, and cepabactin (40). Salicylic acid and ornibactins are the major siderophores produced by CF isolates of the BCC. A study by Sokol et al. showed that ornibactin-deficient mutants were attenuated in virulence in chronic and acute murine models of respiratory infection (172). From these data it was concluded that the acquisition of iron is important for effective colonization and persistence in BCC lung infections. Further studies investigating altered siderophore production in a BCC mutant identified genes encoding a quorum sensing system. Quorum sensing is a mechanism demonstrated by several Gram-negative bacteria that regulate the expression of virulence factors by sensing the local population density through the production and response to N-acylhomoserine lactone (AHL) signaling molecules. The BCC genes cepR and cepl are homologs for luxR and luxl, which 17 respectively encode genes for the quorum-sensing transcriptional activator and AHL synthetase of Vibrio fisheri (100). In one study, inactivation of cepR or cepl resulted in hyperproduction of the siderophore ornibactin and a protease-negative phenotype, indicating that BCC quorum sensing had positive and negative control of extracellular virulence factors (172). Recent focus on the BCC quorum sensing system has identified the AHL molecules produced and an association with biofilm formation. Biofilms are surface-associated communities of bacteria enclosed in a self-produced polymeric matrix that can withstand host immune responses, and are often more resistant to antibiotics than planktonic bacteria (79). Hence, biofilms are often linked to many persistent and chronic bacterial infections. In particular, P. aeruginosa variants with increased production of the EPS alginate form a biofilm which is regulated by quorum sensing signals and associated with its persistence in the CF lung (170). Previous reports have described CF patients who experienced a more severe decline in health when co-colonized by P. aeruginosa and BCC isolates, suggesting a synergistic interspecies relationship (189). Infection with P. aeruginosa was shown to enhance adhesion by BCC to epithelial cell cultures while the addition of P. aeruginosa culture supernatants increased BCC production of siderophores, lipase, and protease (114, 147). Further studies investigating interspecies AHL communication show that BCC strains can respond to the AHL signals produced by P. aeruginosa although converse communication appears to be BCC strain-dependent (142). Cocolonization of the two species in vitro can also result in the formation of mixed biofilms, which may be more advantageous for growth or protection (142, 189). In a BCC-specific study, Conway et al. examined the correlation between AHL synthesis and biofilm formation by representative members of the BCC under various growth conditions (38). At least six AHL signal molecules have been reported for the BCC which include N-octanoly-L-homoserine lactone (C8-HSL), C4-, C6-, C10-, C12-HSLs, and N-3-oxohexanoyl-HSL (30C6-HSL); C8-HSL is produced via Cepl and appears to be the predominant AHL among BCC isolates (37, 38). In contrast to P. aerugionsa, Conway et al. revealed that AHL synthesis and biofilm formation were not always associated in BCC, suggesting that BCC quorum sensing may not regulate biofilm formation under all conditions (38). 18 However B. multivorans and B. cenocepacia group III-A produced the most abundant biofilms which, like P. aeruginosa, may correlate with their virulence in CF. 1.4 Stress regulation of virulence genes Bacteria are equipped to constantly monitor their surroundings and register changes to adapt to new conditions or hostile environments. There is substantial evidence correlating virulence gene expression with key environmental signals that bacteria encounter in the host. Many of these signals include temperature, pH, nutrients, presence or absence of specific ions, and oxidation, all of which can be perceived as environmental stresses for various pathogens and result in the expression of gene products necessary for survival. Hence several studies have investigated the stress response of bacteria to understand the regulation of virulence factors and mechanisms of pathogenesis as it has been shown that lab-generated mutants that cannot survive in extreme conditions are often attenuated in virulence. In particular, alginate biosynthesis genes in P. aeruginosa are affected by suboptimal nutrition, dehydration, osmolarity, and oxygen availability, all of which can be related to the CF lung environment (45). Extensive studies on the effects of heat shock and oxidative stress revealed the coordinate regulation of heat shock and conversion to the mucoid phenotype in P. aeruginosa (16, 113, 162, 163). The alternative sigma factor (a), Alg U (T), is the transcriptional activator of alginate biosynthesis genes and shares significant homology with the sigma factor, a E, from E. coli and Salmonella. Sigma factors are master regulators involved in global regulatory systems that control the expression of several genes (145). RNA polymerase bound with a sigma factor recognizes specific promoters and has increased efficiency in transcription initiation. Both sigma factors, Alg U (T) and aE, control extreme stress responses that enhance bacterial resistance to adverse conditions (16, 46). Mutations in Muc A, the anti-sigma factor controlling Alg U (T) activity, are the predominant cause of the mucoid phenotype conversion in P. aeruginosa (45). Virulence genes controlled at the level of sigma/anti-sigma systems implicate the global regulation of these genes among other regulons that are coordinately expressed to combat stress (123). 19 1.5 BCC and host infection studies 1.5.1 BCC and host cell interactions Adherence and host cell receptors for BCC In the search for preventive strategies against BCC infection, efforts have focused on the BCC adhesins and their respective host cell receptors facilitating BCC attachment to the cell surface. Krivan et al. reported BCC binding of an unclassified CF isolate to two glycolipids, gangliotriosylceramide (asialo-GM2 or Gg3) and gangliotetraosylceramide (asialo-GMi or Gg4), that contained the GalNAcpi-4Gal motif which was the proposed binding site for BCC; however, these studies did not identify the corresponding bacterial adhesins (92). In another study, Sylvester et al. used glycolipid extracts from human erythrocytes, human buccal epithelial cells (BEC), and Hep-2 laryngeal epithelial cells, and showed that galactolipids like globotriosylceramide (Gb3) were strong candidates as lipid receptors for BCC, particularly cable pilus-negative strains (185). Alternatively, BECs were also used to identify cytokeratin 13 (CK13) as the receptor for the 22 kDa adhesin on cable pilus (155). Cytokeratins are primarily cytoplasmic proteins that form intermediate filaments connecting eukaryotic cell organelles to the nucleus and plasma membrane. However, Sajjan et al. was able to demonstrate low concentrations of CK13 on the surface of human bronchial epithelial cells and proposed that CK13 availability for BCC binding was increased from tissue damage generated by chronic inflammation in the CF lung (155). This was further demonstrated using explants of CF epithelia, which were damaged from chronic infections, enriched in CK13, and more susceptible to cable pilus-positive BCC binding than non-CF tissues (152). Invasion The "cepacia syndrome" associated with some BCC infections highlight the ability of BCC to invade and gain access to deeper tissues. Burns et al. first described epithelial cell invasion by a clinical BCC isolate using the respiratory epithelial A549 cell line in a modified gentamicin protection assay (19). These results were confirmed with electron microscopy (EM) which showed the presence of intracellular bacteria in membrane-bound vacuoles. In addition Cytochalasin D, an inhibitor of microfilament formation in eukaryotic cells, was able to inhibit invasion, suggesting the involvement of 20 microfilaments in BCC uptake. Cytochalasin D also inhibits the cell entry of invasive pathogens such as Shigella flexneri, which possesses TTSS to reorganize the host cell actin cytoskeleton for bacterial-induced uptake (158). With the recent identification of a putative TTSS in BCC, it is possible that BCC utilizes similar invasion tactics. Other researchers have utilized A549 cells to determine the invasion frequencies among BCC members. These studies have shown B. multivorans and B. cenocepacia as the most invasive among the genomovars tested which correlate with their virulence in clinical reports (34, 89). However there did not appear to be any correlation between degrees of invasiveness and the presence or absence of Cbl pilus or BCESM (34, 89). By contrast, Sajjan et al. reported that B. cenocepacia strain BC7 [Cbl pilus positive, 22 kDa adhesin positive (cblA+/Adh+)], was able to invade CK13-enriched squamous cells differentiated from primary cultures (150). BCC isolates that were cblA+/Adh- or cblA-/Adh- showed minimal binding and did not invade. In addition, the invasive isolates caused cytotoxicity, stimulated IL-8 release by host cells, and migrated across cells by disrupting intermediate filament structures, increasing epithelial damage. Hence, these data provided evidence for the inflammation, tissue damage and septicemia associated with severe BCC infections and further demonstrated the importance of CK13 binding as the initial step in this cascade of events. The variability in the invasive phenotype among BCC isolates may be explained by in vitro evaluation using cell cultures that were differentiated, potentially affecting the accessibility to certain receptors (150). In general there appears to be very few in vitro systems that can faithfully demonstrate host cell interactions with all members of the BCC, as many observations appear to be strain-specific, even within the same genomovar. Although usage of the A549 cell line is widespread, the invasion frequencies that are obtained for BCC isolates are comparatively low as compared with values obtained for well-characterized invasive pathogens (34). Primary airway cell cultures have also shown differential invasion patterns among BCC members. Schwab et al. demonstrated that biofilm formation was associated with invasion and epithelial cell damage by the B. cenocepacia strain BC7 whereas a B. stablis strain did not form a biofilm and penetrated the epithelia via paracytosis without cell destruction (164). Conversely , a B. multivorans strain appeared to utilize both invasion 21 routes demonstrated by the other genomovar strains (164). These results may partially explain the differences in virulence among BCC members; some isolates may cause "cepacia syndrome", while others may invade and persist in an intracellular niche. Furthermore, the cell necrosis and invasion pattern shared by B. cenocepacia and B. multivorans correlated with associated clinical presentations. Intracellular survival Studies demonstrating BCC invasion have also shown evidence of intracellular survival, and provide some insight into the persistent infections caused by these organisms. Burns et al. described the ability of a BCC isolate to survive and replicate intracellularly, following A549 cell invasion (19). Subsequent studies with other BCC isolates have also demonstrated intracellular survival and replication in A549 cells as well as macrophages (88, 112, 148). Strains of B. vietnamensis (genomovar V) and genomovar VI survived in cultured murine macrophages for at least 5 days without multiplication (148). In particular, phagocytosis of the genomovar VI strain resulted in changes in macrophage morphology including cytoplasmic enlargement and vacuolization, with primed superoxide release and TNF-a production indicating macrophage activation. Hence certain BCC isolates are able to survive intracellularly and resist the consequences of macrophage activation. Alternatively, a separate study using cultured human macrophages demonstrated that the intracellular survival and replication of B. cenocepacia strain J2315 was associated with cytotoxicity and destruction of the infected monolayer (112). BCC is also able to survive in amoeba, suggesting a potential environmental reservoir for BCC and a vehicle for patient-to-patient transmission (95, 111). Intracellular survival and replication of BCC isolates was evident for at least 5 days in Acanthamoeba spp. with one study reporting extensive cytoplasmic vacuolization in amoeba after BCC infection (111). Taken together, these reports provide evidence demonstrating BCC as a facultative intracellular pathogen, however the specific mechanisms employed by BCC are still unclear, and are confounded by strain to strain differences. 22 1.5.2 BCC and mouse models Various experimental animal models have been modified or developed for the study of BCC pathogenesis and characterization of the host response. BCC infections have been examined in CF mouse models with comparisons made between wild-type and CFTR-deficient mice. In general, Cftr-/Cftr- mice are more susceptible to BCC infections, resulting in impaired bacterial clearance and severe lung pathology (43, 151). In addition Sajjan et al. demonstrated that a CF isolate of BCC induced more lung pathology than an environmental BCC isolate in Cftr-/Cftr- mice (151). Other murine models of pulmonary infection have relied on agar beads to entrap bacteria and ensure retention of bacteria in the lungs to promote colonization. Using this system, Starke et al. was able to achieve chronic BCC infection in which viable BCC could be recovered from murine lungs up to 21 days post-infection (181). Histology of infected murine lungs revealed inflammatory changes similar to those that were found in the lungs of CF patients, which included squamous metaplasia, fibrosis and bronchiectasis. In a different study, an agar bead model of BCC infection revealed a histopathological pattern that appeared to differ from that of P. aeruginosa under the same treatment (172). P. aeruginosa pulmonary infections facilitated by agar beads were characterized by multiple microabscesses, mucus plugging, and alveolar destruction (25). In contrast, Sokol et al. reported histological changes from BCC pulmonary infection that included a mixed cellular infiltrate of mononuclear phagocytes and lymphocytes, and increased lymphoid follicle size (172). While agar bead models have shown bacterial dissemination in mice infected with invasive BCC isolates, Speert and colleagues developed a murine model of systemic infection to examine invasive BCC infections reflective of "cepacia syndrome" and septicemia in CGD (34, 179). Following intraperitoneal inoculation, mice developed chronic splenic infection with certain BCC strains. In particular, a CGD isolate of B. multivorans persisted in the murine spleen for up to 56 days post-infection. However, no histopathology was observed in the lung, liver or spleen of infected animals. This same B. multivorans strain was later evaluated in a pulmonary model of infection developed by Chu et al. using mice that were rendered mildly leukopenic with cyclophosphamide (CPA) treatment (32). The immunosuppressive nature of the model sustained BCC 23 infections with a low infectious dose. Furthermore, this model was able to differentiate BCC genomovars based on animal infectivity. B. multivorans persisted in the murine lung for up to 16 days post-infection, causing minimal illness in the animals while B. cenocepacia was cleared by day 4, and caused variable degrees of illness. With a selection of BCC infection models to choose from, different aspects of BCC pathogenesis may be examined and allow for greater insight into the host factors that facilitate BCC persistence. 1.6 Thesis objective Though recent investigative efforts have rapidly advanced our knowledge of the taxonomy and clinical relevance of the BCC, its pathogenesis, particularly in CF, remains poorly understood. Several putative virulence factors have been identified for the BCC, however, their presence does not necessarily correlate with the severity of disease. Moreover, many of these determinants have been characterized in vitro and may not provide useful insights into their roles in vivo. Our laboratory has developed two animal models to study BCC infections and identify critical bacterial determinants necessary for infection and survival in a host. Both models demonstrated BCC genomovar differences in animal infectivity: B. multivorans can persist in a murine host while B. cenocepacia is rapidly cleared. The aim of this research project was to identify the bacterial factors enabling BCC persistence in the murine host to gain insight into BCC colonization and infection in susceptible hosts. The approach taken to achieve this goal is divided into two parts and explained in detail below: 1. Isolation of a persistent variant of a non-persistent B. cenocepacia strain from animal infection models. A. Generation of a persistent transconjugant of a non-persistent B. cenocepacia strain from the intraperitoneal infection model. Genetic complementation was employed to identify the genes that enabled B. multivorans strain JTC to persist in the murine host. A cosmid library of strain JTC was constructed then used to transform nonpersistent B. cenocepacia strain K56-2 to generate transconjugants. K56-2 transconjugants were pooled and sequentially passaged in the i.p. infection model to enrich for a persistant 24 transconjugant. Possible candidates were chosen after the third mouse passage and then analysed further to identify the genes that conferred enhanced survival. B. Isolation of a persistent variant of a non-persistent B. cenocepacia strain in the pulmonary infection model. This experiment was an alternative to the genetic complementation studies. A natural variant of non-persistent B. cenocepacia strain C1394 was selected and enriched in vivo in the pulmonary infection model and subsequently used for the aim of this thesis 2. Comparative analysis of the persistent variant and non-persistent B. cenocepacia strain. Phenotypic characterization for comparison of B. cenocepacia strain C1394 and its persistent derivative included Congo Red binding, heat shock induction, EPS production and analysis, EM, host cell interactions, and outer membrane and whole cell protein analysis by SDS-PAGE and 2D-PAGE, respectively. 25 CHAPTER 2 Materials and Methods 2.1 Strains, media, and growth conditions The bacterial strains used in this study are listed in Table 2. A detailed background of the two BCC isolates that are the focus of this thesis is provided here. B. cenocepacia (genomovar III-B) strain C1394 was an isolate recovered from an outbreak among CF patients in Manchester, England (107, 169). Strain variant C1394mp2 was derived by sequential passaging of C1394 through the pulmonary infection model described below. C1394 was recovered at day 16 from the lungs of a cyclophosphamide (CPA)-treated mouse and passaged twice more in CPA-treated mice. Strain variant C1394mp2 was recovered at day 4 of the second of these two additional passages. Bacterial strains were stored at -70°C in Mueller-Hinton (MH) broth or Luria-Bertani (LB) broth containing 8.0% (v/v) dimethylsulfoxide (DMSO) and were routinely cultured at 37°C on blood agar (PML Microbiologicals, Richmond, Canada) or LB agar. Liquid cultures were grown in LB or SOB broth, and 5 ml volumes or less were routinely grown with end-over-end rotation at 37°C (157). Static cultures were grown in 5 or 10 ml volumes inoculated 1:100 with overnight cultures. For stable maintenance of cosmid vector pScosBCl in BCC strain K56-2, cultures were grown in media containing 200 p.g/ml of trimethoprim (ICN Biomedical Inc.). Viable counts in lung homogenates were enumerated on tryptic soy agar (TSA) quadrant plates. To visualize cell surface differences, bacteria were grown on LB agar containing 0.01% (w/v) Congo Red (Congo Red medium [CRM]). For EPS purification, bacteria were grown on a modified version of yeast extract-mannitol (YEM) agar containing 0.05% (w/v) yeast extract, 0.4% (w/v) mannitol, and 1.5% (w/v) agar. For heat shock studies, bacteria were grown on LB agar or in LB broth at 42°C. For motility assays, LB swimming plates containing 0.3% (w/v) agar were stab inoculated with a single bacterial colony and incubated for 24 and 48 hours at 37°C. 26 Table 2. Bacterial strains and plasmid. Strains and plasmid used in this study and references. Strain Relevant characteristics Reference/source B. multivorans JTC C G D isolate, U S A (175) C1576 CF epidemic isolate, U K (107) B. cenocepacia K56-2 CF epidemic isolate, Canada (107) K56-2C K56-2 transformed with cosmid vector, pScosBCl This study K56-2Ce Enriched variant of K56-2C isolated from murine spleen, day 4 of This study second mouse passage C1394 CF epidemic isolate, U K (107, 169) C1394mp2 Variant of C1394 isolated from CPA mouse lung, day 4 of second This study mouse passage (mp2) P. aeruginosa PI CF isolate, serum sensitive (176) M2 Isolate from mouse intestine, serum resistant (182) pScosBCl Broad-host-range cosmid cloning vector based on pSuperCosl (172) (198); A p R T p R 2.2 DNA manipulation and genetic typing 2.2.1 Generation of the non-persistent transconjugant control, K56-2C The 8.8 kb broad-host-range cosmid vector, pScosBCl, was intended for the construction of a cosmid library of B. multivorans strain JTC (172). Cosmid pScosBCl is a derivative of the ColEl-based cloning vector pSuperCosl [Stratagene; (198)] with the kanamycin resistance gene and simian virus 40 origin of replication removed. The features of this vector are as follows: (i) a replication stablizing fragment facilitating the stable maintenance of ColEl-based replicons in many gram-negative bacteria, (ii) a trimethoprim gene cassette for selection in BCC strains, (iii) ampicillin resistance, and (iv) bacterial transformation by electroporation only. K56-2 was transformed, with pScosBCl containing no DNA insert to generate K56-2C, which served as a non-persistent control in the i.p. infection model. An overnight LB broth culture of K56-2 was used to inoculate 5 ml of SOB broth. After 2 hours of end-over-end tumbling at 37°C, the SOB culture was harvested and washed twice in sterile 10% (v/v) glycerol warmed to room temperature. Bacteria were resuspended in 500 pi of 10% glycerol and 100 pi aliquots were electroporated with ~1 pg pScosBCl DNA using a Gene Pulser (Pulse Controller II, Bio-Rad) set at 2.5 kV, 25 pFD and 400 ohms. Electroporated cells were transferred to microfuge tubes containing 900 pi of SOB broth and left to incubate with end-over-end rotation at 37°C for 2 to 3 hours. K56-2 transconjugants were 27 recovered after 2 days of incubation at 37°C on LB agar containing 200 ug/ml trimethoprim. 2.2.2 Isolation of BCC genomic DNA for molecular analysis Genomic DNA was isolated as previously described with some modifications (106). Overnight bacterial cultures grown in 2 ml LB broth were harvested and resuspended in 200 ul of TE10 buffer (10 mM Tris-Cl, 10 mM EDTA, pH 8.0). One hundred microlitres of this suspension was transferred to a 2 ml screw cap microcentrifuge tube containing approximately 500 ul of washed glass beads (0.1 mm diameter) and 500 ul of lysis buffer (50 mM Tris-Cl pH 8.0, 70 mM EDTA pH 8.0, 1% [w/v] SDS, 0.5 mg/ml pronase [Roche]). Bacteria were lysed by mechanical disruption with vortexing for 10 seconds, and pronase digestion was carried out at 37°C for 1 hour. After incubation, 200 ul of saturated ammonium acetate was added to the lysate. The contents in the tube were vortexed briefly then extracted with 600 ul of chloroform to remove protein, vortexed again and centrifuged at 14,000 x g for 5 minutes. Genomic DNA was precipitated by transferring the clear aqueous phase to 2.5 volumes of 100% ethanol. The DNA was recovered by centrifugation and washed with 70% ethanol, air dried, then dissolved in 100 ul of TE (10 mM Tris-Cl, pH 8.0; 1 mM EDTA, pH 8.0) containing 0.5 yig/rnl RNase A (Roche). DNA was quantitated by measuring its absorbance at 260nm (A.26o)-2.2.3 Polymerase chain reaction (PCR) All oligonucleotide primers used are listed in Table 3. PCR reactions were amplified in a Peltier Thermal Cycler (PTC-0200 DNA Engine; MJ Research, Waltham, MA). All amplifications were in 25 ul volumes. Random amplified polymorphic DNA (RAPD) analysis RAPD PCR mixtures were set up as previously described using oligonucleotide primers 270, 272, and 275 (106). Reaction contained 40 ng of genomic DNA template, 40 pmol oligonucleotide primers, 1 U of Taq polymerase (Invitrogen), 250 uM nucleotide triphosphates (dNTPs), and IX PCR buffer (10 mM Tris-Cl [pH 8.0], 50 mM KCL, 0.001% [w/v] gelatin, 3 mM MgCb). Each reaction was amplified as follows: (i) 4 28 cycles of 5 minutes each at 94°C, 36°C, and 72°C, (ii) 30 cycles of 1 minute at 94°C, 1 minute at 36°C, and 2 minutes at 72°C, and (iii) a final extension step at 72°C for 10 minutes. Amplification of the B. cepacia epidemic strain marker ( B C E S M ) , cblA gene, and the JliC gene Reaction mixes for BCESM and cblA gene amplifications contained 40 ng of genomic DNA template, 250 pM dNTPs, IX PCR buffer (Invitrogen), 1 U Taq polymerase (Invitrogen), 2 mM MgCfi, 10 pmol oligonucleotide primers. PCR cycle conditions consisted of 30 cycles of 1 minute at 94°C, 1 minute at 56°C, and 2 minutes at 72°C, followed by a final extension step at 72°C for 10 minutes. Flagellin (fliC) gene amplification with oligonucleotide primers BC4 and BCR12 was carried out as previously described (65)(Winstanley 1998). Each reaction contained 40 ng of genomic DNA template, 200 nM of each primer, 100 pM dNTPs, IX Taq Buffer (Invitrogen), 2.5 mM of MgCl 2, and 2.5 U of Taq polymerase (Invitrogen). PCR was performed for 30 cycles consisting of 1 minute at 94°C, 1 minute at 56°C, and 2 minutes at 72°C, followed by a final extension at 72°C for 10 minutes. Table 3. Primer sequences Primer Sequence (5' to 3') Reference R A P D P C R 208 A C G G C C G A C C (106) 270 T G C G C G C G G G 272 A G C G G G C C A A B C E S M B C E S M 1 C C A C G G A C G T G A C T A A C A (108) B C E S M 2 C G T C C A T C C G A A C A C G A T CblA sense C C A A A G G A C T A A C C C A (154) antisense A C G C G A T G T C C A T C A C A fliC B C 4 C T G G T C G C A C A G C A G A A C C T G A A C (65) B C R 1 2 A C A T G T T C G C G G T T T C C T G 2.2.4 Restriction fragment length polymorphism (RFLP) analysis of PCR products Flagellin gene PCR-amplified products (5 pi) were digested with restriction endonucleases Haelll and Mspl, or PstI as previously described (65, 190, 202). 29 Reactions were carried out in volumes of 20 ul under the conditions recommended by the supplier (New England Biolabs). 2.2.5 Pulsed-field gel electrophoresis (PFGE) PFGE was performed as previously described (195). Overnight bacterial cultures grown in 5 ml of LB broth were harvested, resuspended and adjusted to an OD (620nm) of between 0.9 and 1.0 in SE buffer (75 mM NaCl, 25 mM EDTA, pH 7.4). Equal volumes of bacterial suspension and molten 2% (w/v) low-melting-point agarose (Type 7; Sigma) were mixed together then poured into 70 ul disposable plug molds (Bio-Rad) and chilled at 4°C for 15 minutes. Plugs were ejected into 10 ml of PEN buffer (0.5 M EDTA, 1% [w/v] N-lauroyl sarcosine, pH 9.6) containing 1 mg of pronase (RocheVml. After overnight incubation with rocking at 37°C, plugs were washed with five changes (one per hour) of TE buffer (10 mM Tris-Cl pH 8.0, 1 mM EDTA pH 8.0). For macrorestriction digests, plug slices (2 mm) were incubated overnight with 10 U of Spe I in a 150 ul reaction mixture at 37°C, then loaded into the wells of a 1.2% (w/v) agarose gel made with IX TBE buffer (Tris-borate-EDTA). Electrophoresis was performed with a CHEF-DR II apparatus (Bio-Rad) at 5.0 V/cm for 40 hours, with pulse switches ramped from 20 to 75 seconds. Bacteriophage lambda concatemers were use as molecular size markers (Bio-Rad). 2.3 Phenotypic characterization 2.3.1 Biochemical analysis Identification of BCC strains were performed with the API Rapid NFT system (PML Microbiologicals, Richmond, BC) and with glucose, maltose, lactose, mannitol, and sucrose oxidation-fermentation (OF) reactions, followed by Moeller lysine decarboxylase test (Difco). API strips were incubated at 32°C for 48 hours. OF sugars were incubated for 3 days. 2.3.2 Crude LPS extraction. LPS was extracted by the method developed by Hitchcock and Brown (76). Overnight cultures of bacteria grown in 5 ml LB broth at 37°C were harvested and 30 washed with PBS, then adjusted to an ODgoo of 0.2. A portion of the bacterial suspension (1.5 ml) was centrifuged at 1000 x g for 2 minutes. The pellets were solubilized in 50 pi of Laemmli SDS-polyacrylamide gel electrophoresis (PAGE) buffer (62.5 mM Tris-HCl, pH 6.8, 2% [w/v] SDS, 10% [v/v] glycerol, 1% [v/v] 2-mercaptoethanol, 0.001% [w/v] bromophenol blue) and heated at 100°C for 10 minutes. To digest protein, 10 pi of Laemmli SDS-PAGE buffer containing 25 pg of of Proteinase K (Roche) was added to the boiled lysates and incubated at 60°C for 1 hour. Samples were electrophoresed in a 12%o (w/v) polyacrylamide gel then visualized using Pro-Q Emerald 300 Lipopolysaccharide Gel Stain Kit (Molecular Probes). 2.3.3 Serum sensitivity assay. The sensitivity of the strains to pooled human sera was determined as previously described (179). Overnight bacterial cultures were grown in 5 ml of LB broth, centrifuged, and resuspended in gHBSS. Bacteria were adjusted to a final concentration of 105 CFU in 1ml of gHBSS and incubated at 37°C in the presence of 10% (v/v) fresh frozen pooled human serum, collected from 5 healthy adult donors. One hundred microlitre samples were taken at 0, 2, and 3 hours, diluted serially in gHBSS, and quantitated on TSA plates after 48 hours of incubation at 37°C. Serum sensitivity was assessed on the following scale: resistant if there was no change in viability after 3 hours; intermediate if there was no change in log CFU at 2 hours but 1 logio drop in CFU after 3 hours; and sensitive if there was 1 logio drop in CFU at 2 hours. Assays used P. aeruginosa strains PI and M2 as serum sensitive and resistant controls respectively. 2.3.4 Determination of minimal inhibitory concentrations (MICs) The MICs for antibiotics were performed according to the standard microtitre broth dilution method as described in The National Committee for Clinical Laboratory Standards Manual. MICs were conducted in 96-well U-bottom microtitre plates (Becton Dickenson) using MH broth. Doubling dilutions provided an antibiotic concentraion range of 1 pg/ml to 1024 pg/ml. The last well, used as a control, did not contain antibiotic. The following antibiotics were used in the assay: ciprofloxacin, tetracycline, trimethoprim (ICN Biomedicals Inc); chloramphenicol (Sigma); piperacillin (Wyeth 31 Ayerst, pharmacy grade, lot #91125); tobramycin (Nebcin [Lily, pharmacy grade]); ceftazidime (Lily, pharmacy grade, lot #100242C). Bacteria were added at 1 x 105 CFU/ml to each well. Microtitre plates were incubated at 37°C and read after 24 and 48 hours. MICs were determined as the iowest concentration of antibiotic where no bacterial growth was observed (by eye). Additional tests were performed with LB agar containing the following other antibiotics: ampicillin (Ampicin, pharmacy grade, lot #A8W51A), amikacin (Amikin, pharmacy grade, lot #0J43094), and gentamicin (Sigma). 2.3.5 Bacterial biofilm assays Biofilm assays were performed as previously described (38). Bacteria grown on LB agar (24 hours at 37°C) were inoculated into media (100 ul/well) in 96-well polypropylene microtiter plate (Corning Incorporated Life Sciences) using a pin-inoculation device. Three types of media were tested for biofilm formation: supplemented LB (SLB) which contained 0.5% (w/v) Casamino Acids (CAA); supplemented basal salts medium (SBSM) containing 20mM citrate and 0.5% (w/v) CAA; and SBSM containing 20mM glucose and 0.5% (w/v) CAA. Microtiter plates were incubated at 37°C for 48 hours in a humidified plastic container. Plates were washed to remove planktonic bacteria and stained with 125 ul of 1% (w/v) crystal violet for 15 minutes. Wells were washed with water at least three times and 200 ul of 95% ethanol was added to release the stain. Biofilm staining was determined by measuring the absorbance of the resulting solution at 590 nm (A590). 2.3.6 EPS extraction and purification. EPS purification and analysis was performed by Dr. Eleonora Altman of the National Research Centre, Ottawa. Crude yield extraction studies were performed on EPS extracted from bacteria grown on Y E M agar at 37°C for 48 hours. The wet weight of cells was determined prior to EPS isolation. Bacteria were scraped off Y E M agar with aqueous phenol [2% (v/v) phenol in 0.9% (w/v) NaCl solution] and centrifuged for 30 minutes at 10,000 x g. The supernatant was dialysed against distilled water at 4°C for 2 days and lyophilized. Crude EPS was purified by gel filtration chromatography with Sephadex G-100 (Pharmacia Fine Chemicals, Uppsala) followed by ion-exchange 32 chromatography with DEAE-Sephacel (Pharmacia Fine Chemicals, Uppsala) equilibrated with 0.05 M Tris-HCl buffer (pH 7.2). The column was irrigated with 0.05 M Tris-HCl buffer (50 ml), and a gradient of 0 to 0.5 M NaCl in the same buffer was used to remove contaminating nucleic acids and LPS. 2.3.7 'il nuclear magnetic resonance (NMR), sugar, and methylation analyses of EPS. EPS samples (0.5 mg) were hydrolyzed with 2 M trifluoroacetic acid for 1 hour at 125°C. Released sugars were determined by gas-liquid chromatography (GLC) of their derived alditol acetates followed by GLC-mass spectrometry (MS) as previously described (146). Alternatively, EPS samples were subjected to methanolysis with refluxing methanolic 2.5% hydrogen chloride for 16 hours at 80°C. Samples were neutralized with silver carbonate, reduced with sodium borodeuteride in anhydrous methanol for 16 hours at 25°C, hydrolyzed with 2 M trifluoroacetic acid, and analyzed by GLC-MS. EPS was O deacetylated with 0.1 N NaOH at 37°C for 5 hours or 10 % (v/v) ammonium hydroxide at 37 °C for 18 hours. Samples (1 mg) were methylated according to the Hakomori procedure, and the methylated polysaccharides were hydrolyzed with 4 M trifluoroacetic acid for 1 hour at 125 °C (64). Methylation analysis was performed as previously described (146). NMR analysis was performed using a Varian INOVA 500 NMR spectrometer with a 5 mm probe. Measurements were made at 22°C and at a concentration of approximately 2 mg/ml in deuterium oxide (D2O), subsequent to several exchanges with D2O. Chemical shifts were reported relative to the methyl resonance of external acetone at 2.225 ppm. 2.3.8 Transmission Electron Microscopy (TEM). TEM was performed by Dr. Terry Beveridge in the Department of Microbiology at the University of Guelph. Bacteria were grown on LB agar at 37°C for 24 and 48 hours. Cells were gently resuspended in 1 drop of deionized H2O, and samples were placed on carbon- and Formvar-coated nickel grids for 30 seconds. Grids were floated on 1 drop of 33 1% (w/v) aqueous uranyl acetate, blotted dry, and then viewed with a Philips EM300 electron microscope at 60kV under standard operating conditions. 2.4 Protein Studies 2.4.1 Inner and outer membrane proteins extraction Membrane proteins were isolated by modification of the method of Hancock and Nikaido (67). Four hundred millilitre bacterial cultures were grown shaking in LB broth for 9 hours (approximately mid-log) at 37°C or 16 hours at 37°C and 42°C. Bacteria were centrifuged at 7000 x g for 15 minutes at 4°C and resuspended in 6 ml of cold 10 mM Tris-HCl (pH 8) containing 20% (w/v) sucrose. Suspensions were frozen at -20°C and thawed to assist cell breakage, treated with 50 |Lig/ml of DNase (Roche) and passed twice through a French pressure cell at 10,000 psi. Unbroken cells were removed by centrifugation at 1000 x g for 10 minutes at 4°C. Four millilitres of the supernatant (in 20% sucrose) was layered onto a two-step sucrose density gradient containing 4 ml of 70% (w/v) sucrose and 4 ml of 50% (w/v) sucrose. Samples were centrifuged for 39000 rpm (100000 x g) in an SW41 rotor overnight at 4°C without braking. The inner membrane fraction lay between the 20 and 50% sucrose layers while the outer membrane proteins were between the 50 and 70% sucrose layers. Each fraction was collected into separate tubes, diluted in 2 to 3 volumes of distilled water, centrifuged at 17000 x g for 1 hour, and resuspended in 100 ul of distilled water. Proteins were quantitated using the Bradford and Lowry assay. Approximately 15 ug of protein was mixed with Laemmli buffer, boiled, electrophoresed on SDS-12.5% PAGE gels then stained with 0.02% (w/v) Coomassie brilliant blue in 25% (v/v) ethanol and 10% (v/v) acetic acid. 2.4.2 Whole cell protein extractions Bacteria were grown in 100 ml of LB broth for 16 hours at 37°C or 42°C. Bacteria were also grown in one to eight tubes containing 5 ml of LB broth for 16 hours at 37°C, and pooled. Bacteria were centrifuged at 4500 x g at 4°C, washed once with cold PBS then resuspended in 1 ml of cold lysis buffer (5 mM EDTA and 1 mM Phenylmethylsulfonyl fluoride [PMSF; Sigma] in PBS). This suspension was added to 500 ul of washed glass beads (0.1 mm diameter) in a 2 ml tube; mechanical lysis was 34 achieved by using a mini bead-beater (Biospec Products, Bartlesville, Okla.) for three 1 minute pulses with 20 seconds intervals on ice to prevent overheating. Lysates were centrifuged at 14,000 x g for 3 minutes at 4°C, and the supernatant was collected. The pellet was resuspended in 1 ml of lysis buffer, disrupted by bead-beating for 1 minute, centrifuged, and the supernatants collected and pooled. Protein concentrations were determined using the BCA microassay kit (Pierce) and 150 pg aliquots were frozen down at -20°C. 2.4.3 2D gel electrophoresis All reagents, equipment for isoelectric focusing (IEF; Ettan IPGphor IEF unit) and 2D-PAGE (Ettan Dalt six electrophoresis unit) were purchased from Amersham Biosciences except where indicated. Whole cell protein extracts (150 pg) were incubated with 50 pi of DNase/RNase solution (1 mg DNase I and 1 mg RNase A [Roche]/ml) for 1 hour on ice. Proteins were concentrated by ultrafiltration at a molecular mass cut-off of 10000 Da. Concentrated proteins were resuspended in 450 pi of rehydration solution (8 M Urea, 2% [w/v] CHAPS, 0.5% [v/v] carrier ampholytes, 0.002%> [w/v] bromophenol blue, 0.3% [w/v] Dithiothreitol [DTT]) for 1 hour at 4°C and applied to 24 cm long immobilized pH 4-7 gradient (IPG) strips. The IPG strips were rehydrated overnight with the solubilized proteins (450 pi) at 20°C under mineral oil. Isoelectric focussing followed rehydration for a total of 99.7 kVh at 20°C. Focused strips were equilibrated in buffer (50 mM Tris-HCL, pH 8.8; 6 M urea; 30% [v/v] glycerol; 2% [w/v] SDS; 1% [w/v] DTT) for 30 minutes then placed atop second dimension high-tensile-strength slab gels containing 12.5%) acrylamide (Duracryl, Genomic Solutions). Gels were electrophoresed overnight with a broad-range molecular mass marker (Bio-Rad) at 1.5 mA per gel then stained with silver nitrate or stained with SYPRO Ruby red (Molecular probes). Protein spots were excised in a Laminar flow hood with methanol-cleaned instruments then stored in \% acetic acid. 35 2.4.4 Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS), liquid chromatography (LC) - MS/MS analysis, and sequencing Protein spots excised from 2D-PAGE gels were sent for processing and analysis at the Genome BC Proteomics Centre at the University of Victoria, BC. In-gel digestion was performed with 400 ng of modified porcine trypsin (Promega, #V5111) for 4 hours at 37°C. Peptides were extracted from the gel piece using 30 ul of 10% (v/v) Formic acid and spotted onto MALDI-TOF plates with a-cyano-4-hydroxycinnamic acid (CHCA) matrix using C18 ZipTips. The peptide mass data were submitted to MASCOT (http://www.matrixscience.com/cRi/index.pl?page=/search form select.html) for identification. Samples were further processed by LC-MS/MS if there were no confident search results. Samples were separated by one-dimension reversed-phase chromatography using Analyst software (Applied Biosystems) controlled Ultimate gradient pumps, SwitchOS II and FAMOS Autosampler (LC Packings/Dionex, The Netherlands) with a 75 um I.D. X 15 cm PepMap C18 3 um, 100A nanocolumn (LC Packings/Dionex). Peptides were eluted and ionized by electrospray into a PE Sciex Qstar Pulsar I Quadrupole TOF MS. The MS/MS data obtained were submitted to Pro ID (proprietary Applied Biosystems software) for bioinformatics analysis of public protein databases (NCBI) and identification. 2.4.5 Western blotting Whole cell and membrane proteins were separated by electrophoresis in a 12.5% (w/v) SDS-polyacrylamide gel and flagellin protein was detected by immunoblotting. The flagellin-specific rabbit polyclonal antiserum used was raised against purified B. pseudomallei flagellin and was a generous gift from D. Woods (University of Calgary). Separated proteins were electroblotted onto PVDF membrane (Bio-Rad) for one hour at 100 V. The membrane was rinsed with distilled water and with wash buffer (25 mM Tris, 137 mM NaCl, pH 7.5; 0.1% [v/v] Tween-20), then incubated in blocking buffer (wash buffer containing 5% [w/v] bovine serum albumin [BSA]) for one hour at room temperature. After incubation, the membrane was rinsed in wash buffer three times for five minutes then incubated overnight at 4°C with the primary flagellin-specific antiserum 36 at a dilution of 1:10,000 in blocking buffer. The membrane was rinsed again in wash buffer as above, and incubated for 1 hour at room temperature with secondary antibody (HRP-linked anti-rabbit immunoglobulin G; Cell Signaling) at a 1:2000 dilution in blocking buffer. Blots were detected and developed using the Phototope-horseradish peroxidase (HRP) Western Blot Detection System (Cell Signaling), following the manufacturer's instructions. The protein standards used were a prestained broad-range molecular mass marker (BioRad) and the biotinylated protein ladder provided by the detection system (Cell Signalling). 2.5 Mice Female C57B1/6 and BALB/c mice were purchased from Charles River Laboratories Canada (St. Constant, Quebec, Canada). All mice were housed and cared for in accordance with regulations of the University of British Columbia Animal Care Committee and The Canadian Council on Animal Care. Mice weighed between 18 and 22 g and were six to eight weeks old at the commencement of each experiment. In the pulmonary infection model, the general health of BALB/c mice was monitored daily using a three point system based on appearance, weight, and food and water consumption. Mice were euthanized if two of the three following observations were made: >10% weight loss, <3 g of food or water consumed, and general ill appearance (ruffled coats and huddled appearance). 2.5.1 Intraperitoneal (i.p.) infection model Bacteria, with the exception of the non-persistent transconjugant control, K56-2C and its derivative K56-2Ce, were cultured from frozen stocks for 24 to 48 hours on blood agar at 37°C. K56-2C and K56-2Ce required culturing in media containing 200 pg/ml trimethoprim. A single isolated colony was used to inoculate 5 ml of LB broth, which was incubated for 16 hours. Cultures were harvested by centrifugation and resuspended in 2 ml of Hank's balanced salt solution containing 0.1% (w/v) gelatin (gHBSS; Invitrogen Life Technologies). The optical density (OD, at 620 nm; OD620) of the culture was determined and the bacteria were diluted to 8 x 106 CFU/ml with gHBSS. C57B1/6 mice were challenged intraperitoneally with 0.5 ml of this bacterial suspension through a 37 25-gauge needle. At pre-selected time points, groups of three mice were sacrificed by cervical dislocation. After removal of the gall bladder, the spleen, liver, and lungs were excised, weighed, homogenized and serially diluted in gHBSS. Dilutions of homogenates were spotted on TSA quadrant plates and B. cepacia selective agar to enumerate viable bacteria (75). 2.5.2 Pulmonary infection model Immunosuppression of BALB/c mice Mice were anaesthetized with gaseous methoxyflurane (Janssen, Toronto, Canada) and received an i.p. injection of CPA (150 mg/kg of body weight; Bristol, Montreal, Canada). CPA was administered on days -1, 4, 9, and 14 of each experiment. Mice used as healthy controls received phosphate-buffered saline (PBS) instead of CPA at the same time points. Intranasal (i.n.) infection of BALB/c mice Bacteria were cultured and prepared as previously described [section; (32)]. After the OD620 of the culture was determined, the bacteria were diluted to 4 x 105 CFU/ml with gHBSS. Prior to i.n. challenge, mice were anaesthetized with an i.p. injection of ketamine hydrochloride (MTC Pharmaceuticals), 60 mg/kg of body weight. Bacteria were delivered drop-wise from a syringe fitted with a 25-gauge needle to alternate nares at an infectious dose of 1.6 x 104 CFU/40 pi. At pre-selected time points, groups of three mice were sacrificed by cervical dislocation. Lungs were excised, weighed, homogenized and serially diluted in gHBSS. Dilutions of lung homogenates were spotted on TSA quadrant plates and B. cepacia selective agar to enumerate viable bacteria. 2.6 In vitro Studies 2.6.1 Isolation and culture of monocyte-derived macrophages (MDMs) Human venous blood from healthy adult donors was collected in sodium heparin Vacutainer collection tubes (Becton Dickinson). Blood (100 ml) was mixed 1:1 with warm RIO (RPMI 1640 media supplemented with 10% [v/v] heat-killed fetal calf serum, 38 2mM L-glutamine, 1 nM sodium pyruvate; Invitrogen Life Technologies). Twenty millilitre aliquots were gently overlaid onto 10 ml of Ficoll-Paque Plus (Amersham Pharmacia Biotech) in 50 ml centrifuge tubes and centrifuged at 680 x g for 20 minutes at room temperature without braking. The upper interface containing mononuclear cells (T and B cells and monocytes) was removed using a transfer pipette and collected into a new tube. Cells were washed twice with warm PBS, resuspended in RIO and transferred to a sterile Teflon pot at a concentration of 2 x 106 cells/ml. Autologous serum was added at 10% (v/v) to R10 for the generation of monocyte-derived macrophages. Monocytes were cultured for five days with one third of the medium replaced on day 3 with fresh R10 containing 10% autologous serum. On day 5, MDMs were harvested as follows: cells were removed from the teflon pot with three washes using warm PBS; pooled washes were collected and centrifuged at 220 x g for 5 minutes. The cell pellet was washed twice more with PBS and resuspended in 1 ml of macrophage serum-free media supplemented with 2 mM L-glutamine and 1 nM sodium pyruvate. 2.6.2 Isolation of polymorphonuclear cells (PMNs) PMNs and monocytes (section 2.5.1.) were isolated from the same venous blood sample. After removing the mononuclear cell layer, the contents of the tube were diluted by one third volume with R10, gently mixed at 1:1 with 2% (w/v) dextran (Amersham Biosciences) in physiological saline, and left to sediment for 45 minutes. The upper phase was transferred to a new tube and centrifuged at 220 x g at 4°C. Remaining erythrocytes were lysed as follows: 6 ml of distilled water (Gibco, BRL, endotoxin-free) was added to cells on ice; after 30 seconds, 2 ml of PBS containing 2.5% (w/v) NaCl was added, followed by 6 ml of Krebbs-Ringer-Glucose balanced salt solution (KRG; 120 mM NaCl, 4.9 mM KC1, 2.5 mM MgS0 4, 1.7 mM KH 2 P0 4 , 9.3 mM Na 2HP0 4, 1.0 mM Glucose) without CaCl 2. PMNs were centrifuged at 400 x g for 7 minutes at 4°C and the lysis step was repeated if erythrocytes were still present. The pellet was then washed twice with KRG without CaCl 2 and resuspended in 1 ml of KRG containing 1 mM CaCl 2. 39 2.6.3 Assay for association of bacteria with phagocytic cells Assay conditions were performed in triplicate with the exception of cells treated with Cytochalasin D and non-infected controls, which were performed in duplicate. Assays were set up in polypropylene tubes containing 4 x 105 MDMs in 250 pi of gHBSS. To test for binding alone, Cytochalasin D (Sigma), an inhibitor of actin polymerization, was used to inhibit phagocytosis and was added to phagocytic cells at 10 pg/ml; both Cytochalasin D-treated and non-treated phagocytic cells were incubated at 37°C for 5 minutes prior to addition of bacteria. Overnight 5 ml LB broth cultures of bacteria were harvested, adjusted to an OD (600 nm) of 0.6, and diluted in gHBSS to the desired multiplicity of infection (MOI). For nonopsonic interactions, bacteria (50 pi) were added to phagocytic cells and the final volume in tubes was made up to 500 pi with gHBSS. For opsonic interactions, bacteria were mixed with 10% (v/v) heat-inactivated pooled human serum in a 500 pi volume with gHBSS. Tubes were incubated for 1 hour with end-over-end rotation at 37°C and the reaction was stopped by adding 1 ml of cold PBS. Cells were washed twice with 1 ml of cold PBS, fixed in 300 pi of 2%> formaldehyde, and cytocentrifuged onto glass slides. Slides were fixed in 100% methanol and stained with 5%> (v/v) Giemsa to visualize associated bacteria under light microscopy. The number of bacteria associated per cell was enumerated for at least 60 phagocytic cells on each slide. 2.6.4 A549 cell association assay The A549 cell line is a human alveolar epithelial carcinoma cell line and was obtained from the American Type Culture Collection (ATCC, Manassas, Va). Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen Life Technologies) supplemented with 10%> (v/v) heat inactivated fetal calf serum and 2 mM L-glutamine. Confluent A549 monolayers were trypsinized using 0.1%> (w/v) trypsin (Sigma) in PBS, seeded onto coverslips (round, 12mm diameter; Fisher Scientific) in 24-well plates at 5 x 104 cells/ml in DMEM supplemented medium, and incubated at 37°C in 5% CO2 overnight. Prior to the addition of bacteria, A549 cells were washed twice with warm HBSS then overlaid with 900 pi of HBSS. Bacteria were grown in LB broth at 37°C statically (48 hours) or with end-over-end rotation (16 hours). To test the effects of bacterial culture preparation, bacteria were either centrifuged and washed before OD 40 adjustment or were diluted in HBSS directly to an OD600 of 0.6, and then further diluted in HBSS to the desired MOL Bacteria (100 ul) were added to A549 cells, centrifuged at 220 x g for 5 minutes, and incubated at 37°C for 1 hour. Wells were washed four times with warm HBSS and 10% formaldehyde was added for 5-10 minutes. Coverslips were stained in 2.5% (v/v) Giemsa for 30 minutes, washed with distilled water, dried and mounted onto glass slides before examination by light microscopy. All conditions were performed in triplicate and the number of bacteria associated with each A549 cell was enumerated for at least 80 cells per coverslip. To test the effects of culture supernatant in A549 cell binding, supernatants from shaken cultures were passed through a 0.2 urn filter and 100 ul were added to cells prior to incubation as described above. Alternatively, cell free supernatant was added back at a ratio of 1:1 (v/v) to harvested bacteria from their respective cultures or to the other bacterial isolate being compared. These mixed suspensions were also added to cells prior to incubation. To test for any cytotoxic effects of the bacteria or its supernatant, the A549 culture supernatant was collected after infection for 1 hour, and the concentration of LDH-1 was quantified using a Cytotoxicity detection kit (Roche) according to the manufacturer's instructions. 2.7 Statistical analysis Data are expressed as mean + standard error of the mean (SEM). When applicable, the Student's t test for independent means was used to evaluate data. 41 C H A P T E R 3 B C C adaptation to the host: role of colonial morphology in B. cenocepacia persistence in the mouse 3.1 Isolation and selection for persistent B. cenocepacia strain variants B. multivorans and B. cenocepacia exhibited differential persistence in two mouse models of infection developed in this laboratory. In the i.p. infection model, B. multivorans strain JTC was able to persist in the spleens of C57B1/6 mice for up to 8 weeks; infection was sustained only in the spleen as bacterial titres were cleared from the lung and liver within a few days (179). Conversely, B. cenocepacia strains such as K56-2 were completely cleared from the murine host within one week. To identify genes that were integral to the persistent phenotype, the non-persistent strain K56-2 was complemented with a cosmid DNA library of persistent strain JTC. The JTC cosmid library was constructed with the cosmid vector pScosBCl, a pSuperCos 1 derivative that conferred trimethroprim resistance and could only be electroporated into BCC strains (172). This research plan involved pooling K56-2 transconjugants and evaluating their survival in the murine host. Since i.p. infections by JTC or K56-2 could be differentiated by day 3 post-infection (Figure 2), transconjugants were isolated at day 4 and passaged again in the same manner to enrich for the persistent phenotype. Strain K56-2 containing the pScosBCl cosmid vector (K56-2C) was a non-persistent control subjected to the same treatment as the transconjugants. Although these studies did not reveal genes involved in the survival of JTC, the sequential passaging of the control, K56-2C, appeared to enrich for the persistent phenotype. In its second mouse passage, K56-2C persisted in the murine spleen at high titres (104 CFU); this enriched K56-2C isolate was designated as K56-2Ce. It is unlikely that the cosmid vector itself conferred enhanced survival to K56-2 since K56-2C showed similar infection kinetics as the parent strain after the first passage, however that possibility cannot be excluded. Figure 3 shows the bacterial loads of K56-2, K56-2C, and its derivative K56-2Ce, cultured from the murine spleen at sequential mouse passages. In the first passage of K56-2 and K56-2C, mice cleared K56-2 by day 7 while less than five colonies of K56-2C were recovered from only one mouse at this timepoint. The second passage of K56-2 and K56-2C was performed with one of a 42 -i 1 1 1 1 r 0 1 2 3 4 5 Days Post-infection Figure 2. Intraperitoneal infection of C57B1/6 mice with B. multivorans strain JTC and B. cenocepacia strain K56-2. Bacterial counts in the spleen were assessed daily for five days. Day 0 is 4 hours post-infection. Data are the mean + SEM from three animals at each time point. 43 0 1 2 3 4 5 6 7 Days Post-infection Figure 3. Intraperitoneal infection of C57B1/6 mice with B. cenocepacia strain K56-2, K56-2C and its derivative K56-2Ce. Filled symbols represent the first animal passage of K56-2 (• ) and K56-2C ( • ) in the murine host . Open symbols represent the second passage of these isolates in the murine host. The dashed line represents the third mouse passage of the persistent derivative, K56-2Ce ( A ) . Bacterial counts in the spleen were determined at days 0 (4 hours post-infection), 3, and 7. Data are the mean + SEM from three animals at each time point. 44 few colonies recovered at day 3 from each isolate's first infection in a previous experiment. While infection with the secondary K56-2 still resulted in nonpersistence, the repeat passage of K56-2C resulted in the recovery of a persistent derivative similar to K56-2Ce, which maintained its persistence upon its third passage in the mouse. Al l colonies grew on BCSA, confirming the recovery of BCC strains and all derivatives of K56-2C maintained the cosmid vector. These results highlighted the adaptive nature of the BCC, which was later corroborated by studies using the subsequent pulmonary infection model, a more physiologically relevant model to study BCC infections. Since the infection kinetics of B. multivorans and B. cenocepacia in the murine spleen were comparable to that in the lung, the i.p. model was replaced by the pulmonary infection model to investigate BCC infection in the host (Figure 4). B. cenocepacia strains maintained the non-persistent phenotype in CPA-treated mice in which all infections were cleared by day 16, and in most cases, by day 4. In addition, i.n. challenge with most of the B. cenocepacia strains tested were associated with symptoms of systemic illness, whereas mice infected with B. multivorans appeared to be well and robust throughout the experiment. Although CPA-treated mice challenged with K56-2Ce did not rapidly clear the infection (Figure 5), the presence of the cosmid vector complicated the analysis of the persistent phenotype and was not pursued further. Another candidate to study BCC adaptation was found among seven genomovar III strains evaluated in the pulmonary infection model. Colonies of B. cenocepacia strain C1394 were recovered from one of three CPA-treated mice at numbers equivalent to the inoculating dose (104 CFU). The isolated clone was passaged twice in the same mouse model to evaluate its capacity to persist in the murine lung. Like K56-2C, higher bacterial titres of C1394 were recovered after each passage, demonstrating that the in vivo passage of a non-persistent strain promoted the selection of a persistent variant. A C1394 derivative recovered from the second mouse passage was chosen for further study and designated C1394mp2. Figure 6 shows the differences in infectivity observed between the parent strain and variant after challenge in the pulmonary infection model. A proportion (<17%) of the CPA-treated mice developed symptoms of systemic illness but were still capable of clearing C1394 from the lungs (typical of B. cenocepacia strains). The derivative, C1394mp2, persisted 45 C1576 PBS K56-2 PBS C1576 C P A K56-2 C P A 2 4 6 8 10 12 . 14 16 Days Post-infection 18 Figure 4. Intranasal infection of PBS-treated and CPA-treated BALB/c mice with B. multivorans strain C1576 and B. cenocepacia strain K56-2. Two mice were assessed for the deposition of bacteria in the lung at day 0 (3 hours post-infection). Bacterial counts in the lung were determined at days 4 and 16. Data are the mean + SEM from two animals at day 0 and three animals at days 4 and 16. 46 Figure 5. Intranasal infection of CPA-treated BALB/c mice with B. cenocepacia isolate K56-2C (naive) and its derivative K56-2Ce (mouse passaged). Two mice were assessed for the deposition of bacteria in the lung at Day 0 (3 hours post-infection). Bacterial counts in the lung were determined at days 1 and 4. Data are the mean + SEM from two animals at day 0 and days 1 and 4. 47 WD jo WO s = fc C1394 PBS C1394mp2 PBS C1394 CPA C1394mp2 CPA 2 4 6 8 10 12 Days Post-infection 14 16 Figure 6. Intranasal infection of PBS-treated and CPA-treated BALB/c mice with B. cenocepacia strain C1394 and its derivative, C1394mp2. Either one or two mice were evaluated for the deposition of bacteria in the lung at day 0 (3 hours post-infection). Bacterial counts in the lung were determined at days 4 and 16. Data are the mean + SEM from three animals at days 4 and 16. 48 for at least 16 days and fewer CPA mice appeared ill (<8%), similar to infections with B. multivorans. Thus it appeared that C1394mp2 had a similar in vivo phenotype to B. multivorans in the pulmonary infection model. All analyses that follow were performed onC1394andC1394mp2. 3.2. Genetic typing of persistent strain variant C1394mp2 To confirm the genetic background of C1394mp2, genetic typing was performed using RAPD PCR and PFGE analysis. Using oligonucleotide primer 270, which determines BCC RAPD types, C1394mp2 produced the same RAPD group 13 profile as the parent strain C1394 (106). Variances may be detected by other RAPD primers such as 272 (used to differentiate P. aeruginosa isolates), 275, and 208, however all three primers failed to distinguish C1394mp2 from C1394. PCR amplification of the BCESM marker was also performed as the BCESM region of the genome is unstable in genomovar III-B strains (104). However both C1394 and C1394mp2 produced the same 1.4 kb DNA band. PFGE separation of chromosomal and macrorestriction digested DNA did not detect genomic rearrangements or deletions (Figure 7). Hence C1394mp2 was genetically indistinguishable from C1394 at the macrogenomic level. 3.3. Phenotypic characterization of C1394 and C1394mp2 3.3.1. Basic phenotypes, biochemistry, and microbiology Sugar oxidation-fermentation profiles and oxidase tests were identical for both C1394 and C1394mp2, as were the API identification tests (Table 4). Although C1394mp2 required a slightly longer incubation time for growth on agar at 37°C, growth rates in LB broth were not significantly different. In addition, swimming motility was observed for both C1394 and C1394mp2 by light microscopy. In two separate sensitivity assays, both C1394 and C1394mp2 were considered resistant to 10% pooled human serum (PHS) when compared to P. aeruginosa strains PI and M2. After 3 hours of incubation with PHS, both isolates showed less than 1 logio drop in viability (C1394 [0.5 + 0.3], C1304mp2 [0.3 ± 0.2]) while PI was reduced by almost 2 logs and M2 grew in the presence of PHS. However, relative to each other, 4 9 F i g u r e 7. Spe I generated macrorestriction fragments of C1394 and C1394mp2 separated by PFGE. Gels were visualized after staining with ethidium bromide. A Lambda ladder (Bio-Rad) was run in the first lane. 50 Table 4. Results of A P I and biochemical tests for B. cenocepacia C1394 C1394mp2 Assimilation of: Glucose + + Arabinose + + Mannose + + Mannitol + + N-acetyl-glucosamine - -Maltose - -Gluconate + + Caprate + + Adipate - -Malate + + Citrate + weak + weak Phenyl-acetate + + Oxidation of: Glucose + + Maltose + + Lactose + + Adonitol + + Sucrose + + Lysine decarboxylation + + Nitrate reduction + + Glucose acidification - -Arginine dihydrolase - -Urea - -Esculin - -Gelatin - -PNPG* + + Oxidase + slow + slow p-nitrophenyl-B-D-galactopyranoside 51 C1394mp2 appeared to be slightly more serum resistant than C1394. Crude LPS was also extracted from each isolate to detect changes accounting for the slight difference in serum sensitivity. The analysis of LPS by PAGE showed that C1394 and C1394mp2 possessed rough LPS (Figure 8). 3.3.2. Differential colonial morphology and MICs Differences in colonial morphology were observed between C1394 and C1394mp2 (Figure 9). On blood agar, TSA and LB agar, the surface of C1394 colonies had a dry, matte appearance whereas the surface of C1394mp2 colonies was shiny (Figure 9A). These morphotypes were also displayed by K56-2C and K56-2Ce, in which the latter, persistent isolate appeared shiny. The distinction between the matte and shiny phenotypes was most apparent when bacteria were grown as a dense lawn or when grown on Congo Red medium (CRM). Congo Red has been shown to distinguish colonial variants of other bacterial species by differential uptake and binding of the dye to the cell surface (23, 133, 138). Strain C1394 absorbed the dye and produced red colonies whereas C1394mp2 colonies were pink indicating less absorption of Congo Red (Figure 9B). After 72 hours of growth, both C1394 and C1394mp2 colonies had absorbed the dye and were red (data not shown). The colonial morphotypes of C1394 and C1394mp2 were stable for at least five subcultures on LB agar incubated at 37°C. However, when grown at 42°C C1394 colonies displayed the shiny morphology typical of C1394mp2 colonies. In addition, growth at this temperature was much slower and required longer incubation time than at 37°C. When the heat-shocked, shiny C1394 was removed to 37°C, this resulted in reversion to the matte phenotype. In contrast, C1394mp2 colonies maintained their shiny morphotype when grown at either 37°C or 42°C. To test whether the shiny colonial morphology of C1394mp2 affected its resistance/susceptibility to different antibiotic families, MIC assays were performed. Compared to C1394, the variant was markedly more resistant to ceftazidime and piperacillin, and exhibited two- to four-fold increased MIC values for trimethoprim, chloramphenicol, and tetracycline (Table 5). Crude assessments were also performed with LB agar containing various concentrations of amikacin, ampicillin, gentamicin and 52 ON Figure 8. Crude LPS samples from C1394 and C1394mp2 (~1 pg), and purified smooth LPS from E. coli serotype 055:B5 were electrophoresed in a 12% acrylamide gel then stained with the Pro- Emerald 300 Lipopolysaccharide Gel Stain Kit. 53 Figure 9. Colonial morphology differences between C1394 (left panels) and C1394mp2 (right panels) grown on blood agar (A) or on CRM (B). Isolates were grown at 37°C for 48 hours. 54 Table 5. MICs of antibiotics for C1394 and variant C1394mp2. Isolates were incubated at 37°C for 24 hours in the presence of increasing concentrations of antibiotics as described (section 2.4.4). The lowest concentration of antibiotic that inhibited growth of the isolates was recorded. Results from one of three experiments are shown. Antibiotic C1394 (pg/ml) C1394mp2 (pg/ml) Ciprofloxacin 2 2 Piperacillin 2 32 Tobramycin 1024 1024 Trimethoprim 32 128 Chloramphenicol 8 32 Ceftazidime <1 16 Tetracycline 256 512 55 polymyxin which resulted in comparable susceptibility levels between C1394 and C1394mp2. Nonetheless, the overall elevated resistance displayed by C1394mp2 provided additional evidence of an altered cell surface as the multi-resistance of BCC strains is largely attributed to a highly impermeable and selective cell envelope (6, 121, 130). 3.3.3. Static cultures and biofilm formation Static cultures of C1394 and C1394mp2 were grown in 5 or 10 ml of LB broth in glass tubes. Growth was not apparent until 48 hours after inoculation. Incubation at 37°C was continued to five days at which time pellicle formation was observed in each culture (Figure 10). The pellicle produced by C1394 was thin, flakey and settled immediately when disturbed. The liquid medium underneath was resolved over time such that three distinct phases were seen: pellicle, clear liquid, and precipitate. Conversely, C1394mp2 produced a dense, flocculent pellicle that went into suspension as long strands when the tube was agitated. The liquid medium remained turbid throughout incubation although sedimented cells increased over time. When incubation was extended to 7 days, the cell aggregates settled to the bottom of C1394 culture tubes. In addition, the cell residue that remained on the glass surface at the air-liquid interface was analogous to the residual attached growth in biofilm cultures. In C1394mp2 cultures, the residue left behind was less substantial and could be washed away, suggesting a weaker surface attachment by C1394mp2. In light of these observations, preliminary biofilm assays were performed with C1394 and C1394mp2 using 96-well polypropylene microtiter plates. The assay utilized three different media as previous studies had shown these media to promote favourable conditions for BCC biofilm formation. In the rich SLB medium, C1394 had greater biofilm growth than C1394mp2, corresponding with the substantial attached growth observed in static cultures (Figure 11). Moreover, the biofilm-forming ability of C1394 was enhanced in the supplemented minimal media containing either citrate or glucose whereas C1394mp2 appeared to have impaired biofilm-forming ability under all growth conditions tested. 56 Figure 10. Pellicle formation and settling exhibited by C1394 and CT394mp2. Static liquid cultures of C1394 (left) and C1394mp2 (right) were incubated for five days at 37°C. 57 0.3 • C1394 • C1394mp2 SLB Citrate Glucose Biofilm growth media Figure 11. Biofilm formation by C1394 and C1394mp2 in different growth media. Isolates were grown in either SLB, SBSM-citrate (Citrate), or SBSM-glucose (Glucose) in 96-well polypropylene microtiter plates for 48 hours. Data represent the mean + SEM for n=8. 58 3.3.4. Differential EPS production To determine if EPS contributed to the differences in colonial morphology, C1394 and C1394mp2 were grown on Y E M agar. C1394mp2 produced abundant EPS, which surrounded and masked individual colonies, whereas C1394 produced minimal to moderate amounts of EPS. In addition, when C1394 was grown for 48hrs on Y E M agar containing 0.01% (w/v) Congo Red, it absorbed the dye whereas C1394mp2 did not. This suggested that an increased amount of EPS produced by C1394mp2 may have interfered with Congo Red binding. Crude EPS was extracted from C1394mp2 grown on Y E M agar and was significantly greater than the amount of EPS extracted from C1394 under the same conditions (10.88 mg of crude EPS per g [wet weight] of C1394 cells vs. 39.40 mg per g of C1394mp2 cells). Characterization of the EPS was performed by 'H-NMR and sugar and methylation analyses. The 'H-NMR spectrum of crude EPS from C1394mp2 showed the presence of at least three O-acetyl groups in the 2.1-2.2 ppm region (Figure 12) whereas EPS from C1394 was O-acetylated to a much lesser degree. Due to the extreme viscosity of the polysaccharide in water it was not subjected to further purification by ethanol precipitation. Instead, EPS was purified by gel filtration chromatography on Sephadex G-100 and ion-exchange chromatography on DEAE-Sephacel. Purified EPS eluted in the void volume of the Sephadex G-100 gel filtration column, indicating that it had a high molecular mass. Sugar analysis of the purified EPS from both parent and derivative revealed the presence of rhamnose, mannose, glucose and galactose in the ratio of 0.5 : 1.0 : 0.8 : 2.1. To confirm the presence of glucuronic acid, EPS was esterified with methanolic hydrogen chloride, followed by reduction of the resultant methyl ester of glucuronic acid with sodium borodeuteride in anhydrous methanol. Subsequent GLC-MS analysis of its hydrolysis products as alditol acetates showed rhamnose, mannose, glucose and galactose in the ratio of 0.6 : 1.0 : 1.0 : 2.0, where 30% of glucose was labeled with deuterium, confirming the presence of glucuronic acid. Methylation analysis was carried out without the carboxyl group reduction step and the substitution pattern of the glucuronic acid residue was not determined. However, the methylation analysis was consistent with the presence of 2-linked rhamnose, terminal galactose, 3-linked glucose and 3, 6-linked mannose as has been demonstrated for 59 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1 1 I I I I I I 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 ppm Figure 12. 'H NMR spectrum of O-deactylated EPS of C1394mp2. The spectrum was recorded in D 2 0 at 37°C. The inset shows the O-acetyl group region of the same polysaccharide prior to O-deacetylation. 60 62 distinguishable from the shiny colonies of C1394mp2. Initial studies using the i.p. infection model produced similar results with non-persistent B. cenocepacia strain K56-2 containing an empty cosmid vector (K56-2C). The persistent derivative, K56-2Ce, was also shiny compared to the matte phenotype of the parent, K56-2C. These observations suggest a potential role for colonial morphology in differentiating B. cenocepacia strains, particularly those that may cause persistent rather than transient infection. Congo Red dye was incorporated into solid media to confirm morphology variation between parent and derivative, as this has been used for several other diverse bacterial pathogens (23, 133, 138). C1394 colonies readily bound Congo Red dye whereas C1394mp2 colonies did not, indicating differences at the cell surface. Reduced dye absorption may be due to the loss or modification of a ligand-binding site, or secretion of a factor impeding Congo Red binding. The growth of C1394 and C1394mp2 on Y E M agar containing Congo Red suggested that the increased EPS produced by C1394mp2 may have prevented efficient dye binding. Recent reports of mucoid CF isolates of the BCC suggest that EPS production may be important in its pathogenesis (27, 141). Both C1394 and C1394mp2 produced EPS on Y E M agar, although the shiny colonial morphology of C1394mp2 correlated with greater EPS production. Characterization studies revealed that both parent and variant produced a polymer with glucosyl, rhamnosyl, galactosyl, mannosyl and glucuronosyl residues which have been shown to be present as a heptasaccharide repeat unit in the EPS of mucoid BCC clinical isolates (27, 30). Although the pathogenic role of this EPS has yet to be determined, it is possible that the BCC EPS provides a protective barrier that contributes to resistance to host immune effectors and delay penetration of antimicrobial agents. The latter part may be responsible for the increased MIC values obtained with C1394mp2. The elevated resistance to ceftazidime is particularly interesting, as it is one of a very few antibiotics effective in the clinical treatment of BCC infections. Alternatively, the larger amount of EPS produced by C1394mp2 may have prolonged its survival by masking bacterial surface antigens that are targets for phagocytosis or complement. The production of a 1, 2-linked glucan in the EPS of C1394 was also detected. The presence of periplasmic glucans has been described for many bacteria (15). Glucans have 63 been proposed to play a structural role in envelope organization as well as serve as information molecules sensed by certain proteins in the periplasmic compartment (15). While further investigation is required to verify the role of EPS in the enhanced survival of C1394mp2, the results suggest an association between abundant EPS production and persistence in the murine host. TEM revealed the increased presence of peritrichously-arranged pili on C1394mp2 cells, which may have altered its colonial morphology and promoted persistence in the murine lung since pili-mediated attachment is often a critical first step in colonization. The effect of increased piliation was demonstrated in a study by Kuehn et al. in which a heavily piliated B. cepacia strain bound more extensively to A549 cells (93). Functional studies of the only genetically characterized pili of the BCC, Cbl pili, further validate the role of these structures in BCC infection. Cable piliated strains bind to mucin and bound preferentially to CF knockout mouse nasal epithelia and human CF lung explants compared to non-CF controls (149, 152). The cbl A gene was not detected by PCR in either C1394 or C1394mp2, nor were cable pili observed by TEM. Because other fimbrial types of the BCC have not been fully characterized, the pili expressed by both C1394 and C1394mp2 could not be classified. However, the pili observed on C1394 and C1394mp2 most closely resembled the filamentous pili found in non-epidemic CF strains (58). In addition, the mesh-like pili detected on both isolates, resembled the BCC mesh pili that are coexpressed with either cable or filamentous pili as previously described (58). Hence it is possible that C1394 and C1394mp2 express at least two types of fimbriae. Moreover, the increased piliation of C1394mp2 may have conferred enhanced adherence to host cells and establishment of a persistent infection in the mouse model. The function of these pili may be clarified with adherence assays to host cells and is investigated in the following chapter. The aggregative behaviour of C1394 and C1394mp2 was another observation suggesting differences at the cell surface. Autoaggregation of some organisms has been attributed to pili and non-pilus adhesins such as OMPs (14, 73). The cable pilus has also been implicated in the clumping of B. cenocepacia strains (31). Statically grown cultures of C1394 and C1394mp2 showed differences in pellicle formation and attachment to inert surfaces. The pellicle formed by C1394 was a thin interface that easily sedimented when 64 disturbed, leaving attached growth at the glass surface. In contrast, the C1394mp2 pellicle was thick, went into suspension as long threads when the tube was gently shaken, and did not attach to the culture tube. These observations were supported by preliminary biofilm assays performed in microtitre plates that showed biofilm formation by the parent but not by the variant. C1394mp2 may be impaired in biofilm formation due to its inability to form a tight aggregate and attach to the substratum, which is required in the initial stages of biofilm development. In this regard, the increased piliation or EPS production in C1394mp2 may result in altered adhesive properties or may physically hinder attachment by other putative adhesins. Conversely, C1394mp2 may not express all the necessary adhesins required for surface attachment or autoaggregation. Bacterial surface components associated with adhesion and colonial morphology are often subject to phase variation (48, 74). Phase variation occurs at high frequencies (>10~5 per generation), and the instability and plasticity of the BCC genome may enhance recombination events and rearrangements associated with phase variation (74). Although the frequency of colonial morphology variation was not determined, we did not detect any genetic rearrangements between C1394 and C1394mp2. Furthermore, the colonies of both C1394 and C1394mp2 were uniform in their respective colonial morphotypes and were stable for at least five in vitro passages. A switch from matte to shiny morphotype was observed only when C1394 was incubated at 42°C. The temperature-induced shiny phenotype in C1394 was transient, suggesting that a selective mutational event in C1394mp2 may have stabilized the shiny phenotype and led to its enhanced survival in the host. Heat induction of the shiny phenotype may provide insight into the regulation of EPS production and piliation in C1394. This response may be part of a defense and adaptation mechanism activated by conditions encountered in the host. Whereas EPS and pili are putative factors associated with the shiny colonial morphology, they may not be the only factors involved in the persistence of C1394mp2 in the mouse model. 65 CHAPTER 4 Host cell interactions with nonpersistent B. cenocepacia strain C1394 and persistent variant C1394mp2 4.1 Association levels of C1394 and C1394mp2 with phagocytic cells To determine if resistance to phagocytic clearance played a role in the persistent phenotype of C1394mp2, the cellular uptake of C1394 and C1394mp2 was compared in a nonopsonic phagocytosis assay. Focus was kept on nonopsonic interactions since the first encounter between pulmonary phagocytic cells and infecting bacteria likely occurs in the absence of serum opsonins (109). Given that C1394 is readily eliminated within four days, the initial interactions with phagocytes may be a critical stage that explains the longevity of C1394mp2. Although C1394mp2 was isolated from a murine host, human cells were used to evaluate and relate the persistent phenotype within the context of a human host infection. After 5 days of culture, human monocyte-derived macrophages (MDMs) were infected with bacteria at an MOI of 50:1 and incubated for 1 hour. These culturing and assay conditions were favoured as previous results in this laboratory have shown that human MDMs prepared and used, as described in section 2.5, can ingest P. aeruginosa strains and zymosan particles with great efficiency (personal communication, Dr. Andrew Currie and Speert lab). Pretreatment of MDMs with Cytochalasin D inhibited phagocytosis and allowed quantitation of bound bacteria; the bacteria associated with untreated MDMs represented both bound and internalized bacteria. There was little evidence of bacterial binding or uptake for either isolate as the association levels with or without Cytochalasin D were extremely low and not significantly different (Figure 14). Nonetheless, MDM association by C1394 was almost two-fold greater than association by C1394mp2 at an MOI of 50:1 (Figure 14A). Higher MOIs were tested to increase bacteria-phagocyte interaction and to investigate whether the difference between C1394 and C1394mp2 would be amplified under these conditions. At MOIs of 100:1 and 200:1, the mean association with MDMs for both isolates were comparable and remained poor (<2 bacteria per cell), and there appeared to be no internalization of C1394 (Figure 14B & C). C1394 and C1394mp2 were also tested with human neutrophils in the phagocytic assay. While alveolar macrophages are present as the first line of defense against 66 = 2.5 o to n 1.5 0.5 1 , T — 2.5 I (1) o <u 2 a ro '£ o 1.5 o ro .a 1 6 c 0.5 c ro o S 0 C1394 C1394mp2 MOI 50:1 C1394 C1394mp2 MOI 100:1 - cytochalasin D I + cytochalasin D C1394 C1394mp2 MOI 200:1 Figure 14. Nonopsonized association of C1394 and CT394mp2 with human MDMs at different MOIs. Differences between MDMs treated with or without Cytochalasin D demonstrated bacterial internalization. The mean + SEM are shown for n=3 to 6 (2 experiments). 67 bacterial infection in the lung, the inflammatory response that follows results in the massive influx of neutrophils and serum factors to help clear the offending microbe (52, 178). In these studies, bacteria were opsonized and unopsonized to test their interaction with neutrophils at an MOI of 50:1. Heat inactivated PHS was used in the opsonized assay due to previous results that suggested a slight increase in susceptibility to PHS with C1394 (see section 3.3.1). Similar to the results from MDM infections at an MOI of 50:1, nonopsonized C1394 had a better association than C1394mp2 with phagocytic cells (Figure 15A). Opsonization dramatically enhanced the neutrophil interaction with bacteria but did so equally for both C1394 and C1394mp2 (Figure 15B). Comparisons with Cytochalasin D-treated neutrophils also revealed comparable levels of internalization for both isolates. Hence opsonization did not result in differential phagocytic interaction of C1394 and C1394mp2 but small differences were observed in nonopsonic associations at a low MOI. In comparison to C1394, unopsonized C1394mp2 association with phagocytes was exceptionally poor but may be an indication of how C1394mp2 avoids rapid clearance from the host. 4.2 Binding levels of C1394 and C1394mp2 with A549 cells To determine whether the enhanced survival of C1394mp2 was attributed to increased adherence to epithelial cells, binding assays were performed with a human lung pneumocyte cell line (A549). Since this assay did not distinguish internalized from surface-bound bacteria, binding refers to bacteria associated with A549 cells. The A549 cell line has morphological features characteristic of type II pneumocytes of the lung, and has been used extensively in studies examining the interactions of BCC with epithelial cells (19, 29, 31, 93). Initial binding experiments were performed without forcing bacteria-to-cell contact. The effects of bacterial culture preparation were tested, as certain bacterial adhesins can be sensitive to shear factors generated by shaking or centrifugation. Al l infections were performed at an MOI of 100:1 or higher to detect cell-associated bacteria. Bacteria diluted or harvested by centrifugation from shaken cultures displayed modest binding levels whereas bacteria diluted from statically grown cultures had a higher affinity for A549 cells (Figure 16). This observation was consistent with the notion that bacteria grown in standing cultures have more intact adhesins than bacteria 68 C1394 C1394mp2 Nonopsonic • - cytochalasin D • + cytochalasin D C1394 C1394mp2 Opsonic Figure 15. Nonopsonic (A) and opsonic (B) association of C1394 and C1394mp2 with human neutrophils at an MOI of 50:1. Differences between neutrophils treated with and without Cytochalasin D demonstrated bacterial internalization. The mean + SEM is shown for n=4 to 6 (2 experiments). 69 0.7 0.6 i 0.5 H 0.4 0.3 • C1394 • C1394mp2 Harvested Diluted Static Bacterial culture preparation Figure 16. Effects of bacterial culture preparation on bacterial binding to A549 cells at an MOI of 100:1. Shaken cultures were either harvested then resuspended in HBSS (Harvested) or diluted in HBSS (Diluted) or grown statically and then diluted in HBSS (Static). The mean + SEM are shown for n=5 to 6 (2 experiments). 70 from shaken cultures (125). Overall, no significant differences were observed in the binding levels of C1394 and C1394mp2. To facilitate bacteria-to-cell contact, bacteria grown in shaken cultures were centrifuged onto monolayers to improve their binding. Under these conditions, significant binding differences were observed with enhancement of association of C1394 but not of C1394mp2 (p<0.05). When bacteria were prepared from harvested cultures, C1394 association with A549 cells was approximately three times greater than that of C1394mp2 at an MOI of 100:1 and 250:1 (Figure 17). A slightly greater difference was obtained from cultures that were diluted at an MOI of 250:1 (Figure 17). These results indicate that with forced contact, the parent strain had a better binding efficiency with A549 epithelial cells despite C1394mp2 having an increased display of pili. Therefore the pili present on C1394mp2 may not have a significant role in adherence to A549 cells. The supernatant of each isolate was also tested to investigate whether C1394 secreted a factor that assisted in its binding to A549 cells. Cell-free supernatant from shaken cultures of C1394 was mixed with harvested C1394 or C1394mp2 cells and then added to A549 cells. The supernatant of C1394 did not improve the binding levels of C1394mp2 nor did it affect the binding of C1394, suggesting that an auxiliary adhesion factor did not facilitate binding by the parent (data not shown). Lactate dehydrogenase (LDH) assays were performed to detect any cytotoxic effects from infection with the isolates or the culture supernatants themselves on A549 cells since certain organisms produce products that injure or compromise the host cell to uncover sites for attachment. Results showed no differences in cell viability between infected cells and uninfected cells, indicating that neither C1394 and C1394mp2 nor their supernatants were cytotoxic to these cells. 4.3 Discussion The aim of these studies was to elucidate mechanisms of C1394mp2 persistence through host cell interactions. In vitro assays were performed with human phagocytes and epithelial cells to relate C1394mp2 persistence within a human infection scenario. In the pulmonary infection model, C1394mp2 had prolonged survival in the murine host whereas C1394 was eradicated immediately. This discrepancy could be due to resistance 71 Figure 17. Binding of C1394 and C1394mp2 to A549 cells after centrifugation. Bacteria were prepared from either shaken cultures that were harvested, resuspended, then added to A549 cells at an MOI of 100:1 or 250:1 (Harvested) or shaken cultures were diluted and added to A549 cells at an MOI of 250:1 (Diluted). The mean + SEM are shown for n=6 (2 experiments). Asterisk indicates significant binding differences between C1394 and C1394mp2 ( P<0.05). 72 of C1394mp2 to phagocytic clearance. From its phenotypic characterization, C1394mp2 appeared to have an altered cell surface in components such as EPS that may act to prevent efficient phagocytosis and establish persistence in the host. Since complement and antibody are present in low quantities in the endobronchial space, initial studies centred on nonopsonic interactions of the bacteria with phagocytic cells. Nonopsonic phagocytosis is probably most important in the protection against infection and disease progression prior to induction of an inflammatory response and may be the determining factor that explains C1394mp2 persistence (178). Assay conditions utilized end-over-end tumbling and small reaction volumes to promote bacterium-phagocyte contact while Cytochalasin D treatment provided a crude estimate of the ratio of bound-bacteria and internalized bacteria. However, even at elevated MOIs under these assay conditions, both C1394 and C1394mp2 showed weak association levels denoting poor interaction with primary human MDMs and neutrophils. Nonopsonic interactions at an MOI of 50:1 demonstrated that C1394mp2 had less association with phagocytes than C1394. Conversely, opsonic interactions with neutrophils at the same MOI failed to differentiate C1394 and C1394mp2, but resulted in the enhanced contact and internalization of both isolates. Based on these observations, a lower MOI in nonopsonic instead of opsonic conditions appears to be more discriminating in revealing differences between C1394 and C1394mp2. It is possible that the poor nonopsonic interaction between C1394mp2 and primary human phagocytic cells is a mechanism by which this organism can resist host defense. Evasion of phagocytic clearance may lie altogether within the weak affinity for phagocytic cells. Weak interaction with phagocytes was found among other clinical BCC strains tested, which also showed poor nonopsonic interaction with primary human MDMs (personal communication with Dr. Barb Conway and Speert lab). Since phagocytic activity could not be properly assessed from such low association indices, a cell line may have been more informative as others have provided evidence of BCC invasion and intracellular survival using the human monocytic cell line U937, and the murine macrophage cell line PU5-1.8. (112, 148). The assay and cells that were used in this study proved to be limited in demonstrating potential differences between C1394 and C1394mp2. Primary cells are subject to donor variability and a more conclusive result might be obtained from a greater 73 pool of donors. While it is plausible that the persistent phenotype of C1394mp2 is valid only within a murine host context, in vitro experiments using murine macrophage and other relevant cell lines are required to confirm this possibility and further elucidate C1394mp2 persistence. With the exception of the Cbl pilus, the ligands involved in BCC adherence to host cells and colonization are not well known. The cable pilus is the only one of five classified pili of the BCC that has an established role in mediating host cell adhesion. Neither C1394 nor C1394mp2 carried the cblA gene encoding the subunit of the cable pilus (section 3.3.5). TEM results showed another pili type on both isolates with increased expression on C1394mp2 and resembling the previously described BCC filamentous pili (58). Despite the heavier piliation on C1394mp2, this feature did not confer an enhanced ability to bind to A549 cells. In fact, the binding levels of C1394 were at least three times" higher than that of C1394mp2 after forced contact. Bacteria diluted from shaken cultures showed the greatest difference in binding and suggested the involvement of secreted components in C1394 adhesion, which may have been removed when cultures were harvested. However, the binding levels of C1394mp2 did not significantly improve after cell-free supernatants from C1394 was provided exogenously, and LDH assays did not detect cytotoxicity in A549 cells incubated with supernatant or with bacteria. Nevertheless the supernatants from both isolates were not analysed for the content of secreted products or for the presence of surface appendages released into the medium that could have enhanced or interfered with the binding process. The initial binding experiments without forced contact showed that static cultures had a better association than shaken cultures. However, subsequent binding experiments did not include static cultures largely due to the inconsistencies encountered while adjusting their OD during preparation. Future binding experiments with statically grown cultures of C1394 and C1394mp2 and analysis of their supernatants may further elucidate association differences as well as the growth conditions that might promote A549 cell binding. TEM micrographs also detected mesh-like pili which has previously been described on the cell surface of BCC strains (58). It is theoretically possible that the mesh pili mediate C1394 adherence to A549 cells, but the increased filamentous-like piliation and 74 EPS associated with C1394mp2 may cause steric hindrance to the epithelial surface, thereby reducing the binding efficiency of this variant. Evidence of another adhesin system in the BCC was reported by Sylvester et al., showing adherence to epithelial cells via membrane lipids, particularly galactolipids (185). This was demonstrated by piliated and nonpiliated BCC isolates in which the latter bound more tightly to globotriosylceramide (Gb3), suggesting another adhesin that was obscured by the presence of pili. Likewise, the heavier piliation observed on C1349mp2 might physically obstruct attachment to host cells by a similar non-pilus adhesin. Adherence to membrane lipids may provide more conclusive evidence, as would experiments that test inhibition of binding with varying concentrations of purified pili or EPS and antibodies raised against the two pilus types. A549 cells have been frequently used to investigate BCC invasion of epithelial cells. A study by Cieri et al. revealed that invasion among BCC strains, including C1394, was relatively low compared to other well-characterized invasive pathogens such as Salmonella and Yersinia (34). Because the binding assay could not clearly distinguish internalized from surface-bound bacteria, future studies investigating the internalization of these isolates may further distinguish C1394 from C1394mp2. Additional work is also required with other cell types, particularly differentiated cells which may be more representative of the respiratory tract. Nonetheless studies with A549 cells showed reduced binding by C1394mp2 and correspond with histological findings from the pulmonary infection model. Immunohistology of murine lungs infected with other persistent BCC strains demonstrated bacteria within the airways but only loosely associated with type II pneumocytes (personal communication, Karen Chu). Moreover, C1394mp2 displayed reduced attachment to inert surfaces when grown in standing culture (see section 3.3.3). Hence C1394mp2 may not require tight adherence to epithelial cells to establish infection and C1394mp2 survival may be attributed more towards its poor affinity for phagocytic cells and the protection provided by increased EPS production. In this regard, an organism that avoids close interaction with host cells may not elicit an effective immune response that leads to its elimination. Studies by Chu et al. reported degrees of systemic illness accompanied by infections with B. cenocepacia strains in the pulmonary infection model, possibly reflecting the induction of an 75 aggressive immune response leading towards clearance (32). In contrast, persistent strains of B. multivorans caused minimal illness. These reported in vivo phenotypes are consistent with C1394 and C1394mp2 respectively. An interesting study would be the analysis of the cytokine response of both phagocytes and epithelial cells infected with C1394 and C1394mp2 as differences could reveal how these isolates modulate the immune response responsible for their outcome in the host. 76 CHAPTER 5 Identification of candidate proteins involved in B. cenocepacia persistence in the mouse 5.1 Membrane protein profiles of B. cenocepacia strain C1394 and variant C1394mp2 The results thus far indicate that the major changes associated with C1394 adaptation in the murine host are chiefly at the bacterial cell surface. Protein profiles of the inner and outer membranes of C1394 and C1394mp2 were analyzed to identify the structural differences responsible for the matte and shiny colonial morphologies. Crude membrane extracts were separated by 12.5% PAGE and are shown in Figure 18. Proteins were extracted from log (-9 hours) and stationary (16 hours) phase cultures grown shaking at 37°C or from shaking cultures grown at 42°C for 16 hours. A ~45 kDa band was the most significant difference between the two isolates and was more prominent in the IM protein extracts of C1394mp2, particularly from stationary phase cultures. When C1394mp2 was grown at 42°C, the intensity of this band was reduced to the log phase amounts observed at 37°C. Hence, at 42°C, C1394 and C1394mp2 appeared to have equivalent amounts of this protein whereas at 37°C, the -45 kDa protein was more apparent in the variant. This particular protein may be upregulated in C1394 at 42°C and correlate with the heat induction of the shiny morphology in C1394, as demonstrated in Chapter 3. The -45 kDa band was excised, sequenced, and identified by LC-tandem MS, yielding the following amino acid (aa) sequence: INSAADDAAGLAISTR. This sequence matched with the B. cepacia flagellin protein (accession no. gi4210944). 5.2 Proteomic analysis of B. cenocepacia strain C1394 and variant C1394mp2 Since this thesis was undertaken, progress has been made in the genetic manipulation of BCC strains to study molecular determinants of virulence. However, the success rate of mutagenesis and cloning is still highly strain dependent, and the innate antibiotic resistance of BCC strains continues to limit which cloning tools can be utilized. Furthermore, the stability of such manipulations can vary, as the genome of BCC strains is plastic and prone to rearrangements (98). Due to previous complications encountered in the genetic complementation of K56-2 (Chapter 3), a proteomic approach was taken to 77 37°C 42°C B 7.5 37°C 7.5 Figure 18. Outer and inner membrane protein profiles of C1394 and C1394mp2 electrophoresed on SDS - 12.5% PAGE gel and stained with Coomassie brilliant blue. IM profiles are shown in A; OM profiles are shown in B. OM profiles of isolates grown at 42°C were identical to the OM profiles presented and are, therefore, not shown. Each lane contained 15 pg of crude protein preparations. Abbreviations: L, molecular mass marker; C, C1394; M , C1394mp2. Arrows denote the ~45 kDa band difference detected in the IM profiles. 78 more fully comprehend the differences between C1394 and C1394mp2. Proteomics provides a comparative means to study the protein expression of parental strains and mutants, detect differentially expressed protein products, and thus expose the genes that are affected. Proteomic analysis has also been used to test strains under various conditions to monitor responses to specific stimuli such as stress. The sensitivity and analytical power of this technique is largely due to its resolving capacity, which separates proteins by two dimensions. Isoelectric focussing (IEF) in the first dimension separates proteins according to their isoelectric points (pi) and SDS-PAGE in the second dimension further separates these proteins by their molecular weights. As each spot on the resulting two-dimensional (2D) gel theoretically corresponds to a single protein species of the sample, thousands of proteins can be separated (128). Thus, by comparing the whole cell protein profiles of C1394 and C1394mp2, differentially expressed proteins can be detected and identified to provide insight into the persistent phenotype of C1394mp2. Apart from reproducibility, the remaining caveat to this approach was the incomplete annotation of the genome of B. cenocepacia strain J2315 at the time of analysis (http://www.sanger.ac.uk/Proiects/Bcepacia/). Putative identifications of analyzed proteins were obtained from nonredundant protein databases, using the online BLAST search engine at the National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih.gov/blast/). 5.2.1 Differentially expressed proteins of C1394 and C1394mp2 grown at 37°C A large proportion of the proteomic studies were focused on proteins expressed in stationary phase from cultures grown at 37°C. The rationale behind this was to mimic the growth conditions of bacterial cultures prior to their preparation for animal infection (see section A preliminary IEF run with a pH gradient of 3-10 provided an overview of the proteins expressed by C1394 and C1394mp2, and showed that the majority of the proteins had acidic pis. Hence a pH gradient of 4-7 was used for separation in the first dimension. Figure 19 shows representative silver stained gels of whole cell protein profiles of C1394 and C1394mp2. All protein differences between C1394 and C1394mp2 were detected in at least three separate protein extractions of both isolates, and 2D-PAGE was performed at least twice with each extraction. The significant 79 L_J .:, 9 o as oo O a oo g - f l CQ c o 5 o £ -fl o 1/5 Bb g • - s Cd ON 1 ctS •fl > -a 5 3 15 ^ <u eg c <D u o a ~ eg C J H fl u u fl .2 '3 & o OT +_> S a 0) on U § J < 8 O u _ f l G o CJ J3 ffl g .S e s "3 2 43 fl OH 5 " w ** . « U r H a j t . O N K! OT U CJ 1 "2 u f• • .fl O H OT cu CD C S B fl >, -cd o l s s 60 T3 CL, O cd CQ •g cu OT cd . & ts is «-O r f l OH 0 0 fl fl 2 o £ "8 w CO cj 13 a CM '+3 O OT W -a < ^ cx I Q CN co s cu cd a a y S J O o •— r--cd > OH &H CJ ^ cj ir> <u CN £ T M Q o c m O w fl OT .2 .1 td +-> cd CJ CH OT >. x> T3 CJ OT >, Id fl cd CJ 3 X3 CJ CJ CJ +3 CU T3 _cu 2 '3 fl T3 O (H CL. CU 5 Vg I T ) O S I T ) < S 80 differences between the two isolates (i.e. loss or gain of spots) are described in the following sections, but some quantitative differences are reported here. An enlargement of specific spot differences are shown in Figure 20. Spots C4, C5, and C7ab were present in both isolates with increased levels in C1394 (Figure 20). Their characteristics, aa sequence, and identification from LC-tandem MS analysis are listed in Table 6. The aa sequence from spot C5 was matched to the probable succinyl-CoA synthetase alpha chain protein of Ralstonia solanacearum whereas spots C4 and C7ab were identified as proteins of B. fungorum. Ralstonia, like Burkholderia, is a diverse group of phytopathogens and classified under the same taxonomic branch of beta-Proteobacteria (156). B. fungorum is predominantly an environmental organism of the Burkholderia sp. and is closely associated with the white-rot fungus, Phanerochaete chryosporium (35). Spot C7ab appeared as two separate spots on 2D-PAGE, but was excised together and generated one short peptide in the MS analysis. This sequence matched with the pyruvate/2-oxoglutarate dehydrogenase complex, dihydrolipoamide acyltransferase (E2) component, and related enzymes of B. fungorum which has 70% sequence identity to the dihydrolipoamide succinyltransferase of R. eutropha. Spot C4 was identified as the malate/lactate dehydrogenase protein of B. fungorum. Al l three proteins that were putatively identified for spots C4, C5, and C7ab are enzymes involved with energy metabolism, particularly within the tricarboxylic acid (TCA) cycle. The tentative identification of these proteins suggests that C1394 may have an increased metabolism relative to C1394mp2 even though they had similar growth rates (chapter 3). Identification of BCC type II flagellin protein with increased expression in C1394mp2 One of the significant differences observed in the proteomes of C1394 and C1394mp2 was the presence of three spots (Ml , M2, and M3) in C1394mp2 grown at 37°C, with M l as the most prominent spot of the three (Figure 21). Spot M l was also present in C1394 but was barely detectable. All three spots yielded sequence that matched BCC flagellin (Table 6). The existence of multiple 2D-PAGE spots of the same gene product may indicate posttranslational protein modifications, although, spot M3 was identified as the BCC type II flagellin despite containing the same peptide recovered from 81 A B 6.0 5.6 6.8 6.4 pH pH Figure 20. Protein spots C4, C5, and C7ab in silver stained 2D-PAGE of C1394 (A and B) and C1394mp2 (C and D) grown at 37°C. Spots C4, C5, and C7ab represent proteins that were upregulated in C1394 which were detected by SYPRO Ruby red staining. 82 I If CS E \ai .a (Ni <D CN a. 5 a. O o r » r o .3 O o u r o tn r o ft — D U .5 O o •a r-~ <L> r O t. r o OH ~ .S O o T3 f-U CO CD O >- r o a, — D O U .S°r-.2 CN 3 & 60 g *-> ST Cl, ov X 2 O O o o • ! * U o o r -s= ^ • A IS CO D , 5> E U B O CD c s I r o .13 o «J c CD 60 O UH T3 >^ JS CD T 3 a 5 IN 2 «s 3 a < o U • •S 1 o 3 oo JS cS js o cu cn r— 'eb CD C3 i> cB 2 c« <L> 3 60 2 o i i •P o >> C3 -b -a •SrS 3 s i s 1.U vr i3 O . -2 "5b S 0 -a x >> 1 3 £ x" CD S S <~o O g CN Of O r o P bo r o O C oo _ ^ CN 60 p . T .5 u ct\ — a © 60 CN E CQ 3 2? 8 CO «3 cj Os a o CN 60 cS 53 I T ) M 8 1 CL r o O S" ON p . O ? j , <D 1 O o o r o o p o r o p o fN VO vd o ov T i -r o o o T t > X a < < < < < < J z a i4 lu Q CH > H OH ci cn PS < s cn J a < z > a CM —1 r/3 ^< ^ > a < o > a z > a a a, —i oo ^< < H J H J > a oo < o> a > o z > a <y OH H J 00 z < < > 00 a < <y o > z o . & 00 U T t u vn U •S U c C3 w <e H U ft 83 A . O i t • O B 4 .9 4 .3 pH Figure 21. Protein spots M l , M2 and M3 in silver stained 2D-PAGE of C1394 (A) and C1394mp2 (B) grown at 37°C. Spot M l was more apparent in C1394mp2 than in C1394. Spots M2 and M3 were detected only in C1394mp2. 84 spots M l and M2 (145). While two flagellin types have been described for the BCC, the unclassified BCC flagellin identified for spots M l and M2 is closer in sequence to the type II flagellin (83% identity) than type I (72% identity) (65). These results support the identification and increased expression of the ~45 kDa protein detected in the IM protein profiles of C1394mp2 (section 5.1). In addition, these observations support the described abundancy of flagellar fragments in the TEM micrographs of C1394mp2 (section 3.3.5). 1 Previous studies have identified the BCC fliC gene encoding the type II flagellin, which reportedly has a molecular mass of 45 kDa (65). This protein is also reported to have 77.5%o identity to the aa sequence of the flagellin protein of B. pseudomallei (65). To verify the differential expression of the BCC flagellin protein in C1394 and C1394mp2, rabbit polyclonal antisera raised against the B. pseudomallei flagellin was used to probe Western blots with separated protein extracts from both isolates (Figure 22). Immunoblots detected a 45 kDa band that was more concentrated in the whole cell and OM protein extracts of C1394mp2 grown at 37°C, confirming the increased expression of this protein. However, within the whole cell protein extracts of C1394mp2, other bands below the 45 kDa protein were unmistakenly highlighted with the antisera. It is possible that these bands correspond with the extra 2D-PAGE spots identified as flagellin. Further tests were performed on swimming motility agar. C1394mp2 appeared more motile as the diameter of the zone of motility was twice that of C1394 [C1394 = 9 mm; C1394mp2 = 18 mm]. Hence the increased expression of flagellin contributed to the increased swimming motility of C1394mp2. Immunoblots also detected equal.amounts of flagellin protein in the OM extracts of C1394 and C1394mp2 grown at 42°C. Compared to flagellin levels at 37°C, flagellin expression at 42°C is suppressed in C1394mp2 and increased in C1394. The latter is supported by the equivalent flagellin levels at 42°C detected in the immunoblots, which surpass the amount detected in C1394 at 37°C. Given the heat induction of the shiny phenotype in C1394, it is tempting to speculate that flagellin expression is one of the contributing surface components affected in this response. 85 37°C W OM 45 k D a Figure 22. Immunoblot analysis of flagellin protein in C1394 and C1394mp2. Lanes contained 15 pg of protein which were separated on SDS-12% PAGE gels and immunoblotted with anti-flagellin antibody as described in Materials and Methods. L denotes the biotinylated molecular marker. Lanes alternate between the C1394 and C1394mp2. C1394 is in lanes 1, 3, and 5; C1394mp2 is in lanes 2, 4, and 6. Lanes 1 to 4 are extracted from stationary phase cultures grown at 37°C and lanes 5 and 6 are extracted from 16 hour cultures grown at 42°C. Lanes contain: whole cell lysates (W), and OM proteins (OM). Position of the 45 kDa flagellin protein band is indicated by an arrow. 86 Analysis of the fliC gene of C1394 and C1394mp2 The type II fliC gene of the BCC exhibits considerable heterogeneity among strains (65). Variation among strains has been detected using restriction fragment length polymorphism (RFLP) analysis of amplified fliC genes (190, 202). PCR/RFLP analysis was employed to determine whether C1394 and C1394mp2 carried the same fliC gene. Primers from the N-terminal and carboxyl (C)-terminal sequence of the flagellin protein were used to amplify the fliC gene from both isolates and the PCR product was digested with Haelll, Mspl and Pstl restriction endonucleases (190, 202). Figure 23 shows a -1.0 kb band amplified from both C1394 and C1394mp2, corresponding with previous reports of the type II flagellin PCR product (65). The RFLP analysis also showed identical restriction patterns for C1394 and C1394mp2, indicating that, by these methods, there was no variation in the fliC genes of C1394 and C1394mp2 (Figure 23). 5.2.2 Differentially expressed proteins of C1394 and C1394mp2 grown at 42°C and 37°C The heat induction of the shiny phenotype in C1394 prompted the investigation of protein expression of both isolates at 42°C. Proteins were extracted at 16, 18, and 20 hours, but the cellular mass at each time point was less than extracts from 37°C cultures, and subsequently affected the number of proteins resolved by 2D-PAGE. In these protein profiles, the flagellin protein was not evident in either isolate, and the only significant difference was protein spot, C l , that was detected in C1394mp2 only (Figure 24). At 37°C this spot was detected in C1394 yet was not present in C1394mp2 at this temperature. The analysis of this protein spot yielded the aa sequences shown in Table 6 and was identified as the peroxiredoxin protein of B. fungorum in the analysis by LC-tandem MS. A BLASTP search with this protein produced matches with the peroxiredoxin protein, alkyl hydroperoxidase reductase subunit C (AhpC), with 80% aa sequence identity to the putative AhpC of R. solanacearum. The AhpC protein is a highly conserved protein among many organisms, and has a particular role in the stress response to organic peroxides. The expression of AhpC is regulated by OxyR, a transcriptional factor that activates several genes whose products have roles in eliminating reactive oxygen compounds during oxidative stress (160). While both C1394 87 PCR Haelll Mspl Pstl L 1 2 3 4 5 6 7 8 L m IBB mm mis M i n i m lIWi (MM! ill • 4V Figure 23. /7/C PCR products of C1394 and C1394mp2 and PCR/RPLP patterns generated with the endonucleases Haelll, Mspl and Pstl. PCR products and digests were separated on a 2% agarose gel. A 100 bp ladder was used as indicated (L). Lanes alternate between C1394 (odd lane no.) and C1394mp2 (even lane no.). Lane 1 and 2 are the fliC PCR products; lanes 3 and 4 are Haelll digests; lanes 5 and 6 are Mspl digests; lanes 7 and 8 are Pstl digests. 88 B Figure 24. Protein spot C l in silver stained 2D-PAGE of C1394 (A) and C1394mp2 (B) from 16 hour cultures grown at 37°C. At 37°C spot C l was present in C1394 (A) and absent in C1394mp2 (B). The reverse pattern was detected at 42°C in which spot C l was present only in C1394mp2 and absent in C1394. 89 and C1394mp2 putatively express an AhpC peroxiredoxin protein, the regulation of this protein appears to be different and implicates OxyR as well as the recognition of different stress stimuli as discerning factors between these two isolates. 5.3 Discussion To identify the factors involved in the persistence and shiny colonial morphology of C1394mp2, the protein profiles of C1394 and C1394mp2 were compared for differentially expressed proteins. While mutagenesis is predominantly used to analyse putative bacterial determinants of infection, a proteomic approach bypassed the complications associated with genetic manipulation of BCC strains and indirectly identified the genes involved. Moreover, this approach can reveal expression levels and post-translational modifications of proteins that cannot be predicted from the gene sequence alone (11). Flagella are major surface components that may have contributed to the different colonial morphologies of C1394 and C1394mp2. Both ID and 2D SDS-PAGE detected the enhanced presence of BCC flagellin in C1394mp2. This observation corroborated with TEM results that demonstrated more flagella among C1394mp2 cells. In addition, immunoblots and soft agar confirmed the augmented expression of flagellin and its function in swimming motility respectively. Previous studies have described two major BCC flagellin types based on the molecular mass (45 and 55 kDa) (65). The majority of clinical BCC strains possess the type II flagellin (45 kDa) which, by RFLP analysis, has considerable variation in the BCC (65, 202). To test whether any polymorphisms exist in the fliC genes of C1394 and C1394mp2, PCR/RFLP analysis was performed with previously described primers and restriction endonucleases (190, 202). From these results it appeared that C1394 and C1394mp2 possessed identical type II fliC genes and suggested that differences were due to regulation of gene expression rather than to changes in the FliC primary aa sequence. The biogenesis of flagella is a highly complex process involving more than 40 structural and regulatory genes arranged in three classes (188). Flagellar genes are expressed in a hierarchical order such that the transcription of each class is not activated until the preceding class of genes has been expressed (3, 54). Various environmental stimuli (temperature and cAMP levels) regulate the transcription 90 of class I genes, which encode the regulators that control expression of the entire flagella regulon, and activate the class II genes encoding the structural proteins that are assembled 28 F into the hook-basal-body. A member of the a family of alternative sigma factors, o~ , is among the class II genes that are expressed and required to activate class III genes, which include the flagellin structural gene and genes encoding the chemosensory machinery. Sequencing analysis by Hales et al. revealed a putative aF-type promoter upstream of the transcriptional start site of the type II BCC fliC gene, and identical to the promoter upstream of the B. pseudomallei flagellin gene (65). While the C1394 and C1394mp2 fliC genes appeared identical, the same might not be true for their aF-type promoter sequences. Moreover, Pseudomonas and Vibrio spp. appear to have other sigma factors in addition to a that can enhance the transcription of the flagellin gene (55). Sequencing the regions upstream of the fliC genes may provide information about the potential promoter differences that may give rise to the flagellin expression of C1394 and C1394mp2. Alternatively, the anti-a28 factor, FlgM, which inhibits the activity of CT28, may be altered between the two isolates and may also account for differences in flagellin expression. A recent study by Tomich et al. described the role of the BCC flagella in host cell invasion (188). Their results demonstrated that functional flagella and motility were required for full invasiveness in A549 cells and determined that the flagella did not function as adhesins in direct binding to A549 cells. These findings described a possible mechanism by which BCC strains breached the epithelial barrier and ultimately caused systemic infections, as observed in "5. cepacia syndrome" (188). Previous work by Cieri et al. showed a modest level of A549 cell invasion by C1394, however, no data was available for C1394mp2 (34). Moreover, C1394mp2 was not assessed for its ability to cause systemic infections in the pulmonary infection model. Future experiments evaluating the invasiveness of both isolates may reveal the advantages of overexpressing flagella for survival in the host, particularly if this enables C1394mp2 to survive in an intracellular niche. Alternatively, the reduced A549 binding of C1394mp2 may support the claim that flagella do not function as adhesins. In this respect, the enhanced expression of flagella might even interfere with stable attachment to host cells. This may also explain, in part, why C1394mp2 cannot form a biofilm. Evidence from E. coli and 91 P. aeruginosa studies show that expression of flagella is decreased in biofilm-associated cells while EPS production is increased (199). If the regulation of C1394mp2 flagella expression is affected such that the bacteria cannot stably attach to the substratum, then it may explain why C1394mp2 cells are unable to aggregate and form microcolonies that mature into biofilm-forming cells. Flagella have a role in enhancing the pathogenicity of certain organisms, either by promoting adherence to host tissues or by activating host inflammatory signaling pathways (49). Flagellin is considered to be highly immunogenic, and is among the listed pathogen-associated molecular patterns (PAMPs) recognized by toll-like receptors (TLR) on host cells that mediate the production of cytokines for an effective immune response (72). Thus the enhanced expression of flagellin in C1394mp2 is somewhat in conflict with evidence of flagellin as a pro-inflammatory molecule whose expression is repressed by some pathogens to prevent host recognition (203). To evade detection by the immune system, antigenic and phase variation of flagella has been observed in Salmonella spp. and Campylobacter spp., which possess two flagellin types that can undergo recombination events (203). However C1394 and C1394mp2 appeared to express the same flagellin, and Hales et al. concluded from their own studies that, like P. aeruginosa, BCC strains possess the genes required for the production of only one flagellin type (65). Antigenic variation may also be generated by posttranslational modifications such as glycosylation, which may explain the discrepancy between the predicted molecular weight of the BCC type II flagellin (38.7 kDa) and its estimated weight (45 kDa) observed in SDS-PAGE. Due to the excess production of flagellin in C1394mp2, it is possible that the two additional 2D-PAGE spots (M2 and M3) and the extra bands highlighted in the immunoblots represent progressively modified version of flagellin. Two of the three flagellin spots detected by 2D-PAGE (Ml and M2) have the same MW yet possess different pis which may represent different modifications of the C1394mp2 flagellin that coexist together in the flagellar filament. The mix of these modified flagellin proteins may alter the biochemical properties of the flagella expressed by C1394mp2, and consequently change the immune response against C1394mp2. Such modifications of flagellin could be determined by chemical deglycosylation to detect the existence of glycosyl moieties and additional MS analysis. 9 2 Another differentially expressed protein of C1394 and C1394mp2 was unveiled at different growth temperatures. A peroxiredoxin protein with high similarity to the C subunit of alkyl hydroperoxidase reductase (AhpC) was observed in the C1394 proteome only at 37°C and in the C1394mp2 proteome only at 42°C. Peroxiredoxins such as AhpC are among a number of antioxidant enzymes expressed by bacteria under oxidative stress, particularly in the presence of peroxides (118). In Bacillus subtilis, AhpC was induced not only under oxidative stress but also under heat or salt stress or glucose starvation (12). Likewise, AhpC was also expressed in Myxococcus xanthus when subjected to heat stress (126). These findings indicate a role for AhpC under several stress conditions. While neither C1394 nor C1394mp2 were directly exposed to reactive oxygen compounds prior to protein extraction, a peroxiredoxin protein was detected in C1394 and C1394mp2 during stationary phase (nutrient starvation and oxidation) and heat stress respectively, suggesting divergent responses to various stress. Possibly peroxiredoxin protein expression in C1394mp2 is not required under the same conditions that induce the same protein in C1394 due to compensatory pathways that may include EPS production for resistance or upregulation of other oxidative protection enzymes that can provide equivalent protection, as observed in Xanthomonas strains (119). In several organisms, OxyR is the central regulator that controls the expression of AhpC and other hydrogen peroxide-inducible genes such as katG (catalase) and dps (DNA binding protein) (118). Extensive studies on the OxyR of E. coli have shown this regulatory protein to also be a repressor of the phase variable OMP, Antigen 43 (Ag 43), a self-recognizing surface protein responsible for the autoaggregation and frizzy colonial morphology of E. coli (68, 160, 161). The role of OxyR in the aggregative properties of E. coli may provide some insight into the aggregative behaviors of C1394 and C1394mp2. OxyR mutants of B. pseudomallei, a close relative of the BCC, also demonstrated altered patterns in autoaggregation and biofilm formation (103). These collective findings imply a potential role for OxyR in the expression of surface components affecting the colonial morphology of C1394mp2. Within this context it would be interesting to investigate whether the type II flagellin expression at 37°C and 42°C was directly affected by the regulation of the peroxiredoxin as well. Indeed, the 93 presence of the peroxiredoxin appeared to correlate with a comparatively reduced expression of flagellin for C1394 and C1394mp2 at 37°C and 42°C respectively. A study by Taylor et al. demonstrated that the AhpC expressed by S. typhimurium during infection of BALB/c mice was not essential for virulence but elicited an inflammatory response (186). Interestingly the inactivation of AhpC in Mycobacteria tuberculosis H37Rv also had no effect in the ability of the organism to establish an infection in mice (180). The work by Springer et al. demonstrated that while AhpC expression was silenced in virulent M. tuberculosis under oxidative and normal growth conditions, it was upregulated in static culture suggesting more of a physiological rather than pathogenic role. Taken all together, the results of these studies may apply to the differential expression of the putative AhpC in C1394mp2 and C1394, and the rapid clearance of the latter in CPA-treated mice. The expression of AhpC by C1394 may have provided a target for the immune system that led to elimination from the host. On the other hand, C1394mp2 did not express AhpC, which did not diminish its capacity to survive in the host and thus, remained "undetected" by the immune system. Hence a feasible explanation for C1394 adaptation to the host was the selection for variants that suppressed expression of immunogenic proteins like AhpC as that may have provided an advantage within the host. The remaining proteins detected by 2D-PAGE were metabolizing enzymes involved in the TCA cycle, which appeared to be upregulated in C1394. From phenotypic characterization, C1394 and C1394mp2 had similar growth rates in liquid cultures although C1394mp2 took a slightly longer incubation time for growth on agar at 37°C. The apparent increase in succinyl-CoA synthetase, dihydrolipoamide succinyltransferase-related protein, and malate/lactate dehydrogenase was detected by quantitative staining with SYPRO Ruby red, although the exact abundance ratio of these proteins in C1394 and C1394mp2 was not determined. The apparent upregulation of these energy-related proteins in C1394 may have a more physiological rather than pathogenic relevance but suggests a higher metabolism in C1394. However, it is possible that organisms that are more metabolically active are a target for the immune system of the host. The dihydrolipoamide succinyltransferases from Brucella melitensis and Coxiella burnetii 94 were previously reported as immunogenic proteins that were detected in the sera of infected animals and patients respectively (124, 211). While only five differentially expressed proteins were analysed and reported in this study they were not the only differences observed within the proteomes of C1394 and C1394mp2. It is also worth noting that protein expression in vitro may not exactly represent what is expressed in the host. Although some of the identified proteins may not have a principal role in the persistence of C1394mp2, differences may point towards the concomitant regulation of genes that are directly involved. In particular, regulation of stress-induced proteins can coincide with the expression of virulence determinants that are also controlled by the same global regulators and sigma factors. Further 2D-PAGE analysis of proteomes from various stress and growth conditions, complemented with the monitoring of EPS production, piliation and flagellin expression, will provide a more comprehensive story of the persistence of C1394mp2. 95 CHAPTER 6 General discussion In this thesis, the differential persistence of two B. cenocepacia isolates in a murine model of infection was investigated to understand the pathogenesis of the BCC. This laboratory has developed two murine models to evaluate BCC strains in systemic and pulmonary infections. Animal models were used to determine the relative virulence of BCC isolates and the bacterial factors required for infectivity. In these models, B. multivorans persists as a chronic infection in the murine host, while B. cenocepacia is rapidly cleared. Initial animal studies revealed the capacity for B. cenocepacia strains to adapt to the murine host after sequential passaging, resulting in a persistent phenotype similar to that of B. multivorans. The adaptive nature of BCC strains has been described in association with their ability to survive in many different environments, and is attributed to the plasticity of their genome (98). Studying the adaptation of these organisms could uncover putative virulence determinants that enable a phytopathogen to cause infections in susceptible human hosts. Through in vivo selection, a persistent derivative of B. cenocepacia strain C1394 was obtained in the pulmonary infection model and the comparative analyses between the non-persistent parent C1394 and its adapted derivative C1394mp2 were performed. The intent of these studies was to identify putative virulence factors that enabled C1394mp2 to colonize and persist in a susceptible host. Genetic analyses by PCR typing and PFGE confirmed that C1394 and C1394mp2 shared the same strain background, with no evidence of major genetic rearrangements accounting for the persistent phenotype of C1394mp2. A colonial morphology change from matte to shiny differentiated C1394 from C1394mp2 respectively, and indicated changes at the cell surface level. This conclusion was confirmed by reduced Congo red binding, increased MIC values and reduced autoaggregation by C1394mp2. The shiny colonial morphology of C1394mp2 correlated with increased EPS production, piliation, and flagella expression. These surface features have been described at length as virulence factors for other pathogens, affecting their colonial morphology and contributing to their colonization and persistence in the host. Therefore, colonial morphology may assist in differentiating B. cenocepacia strains, particularly those that cause persistent rather than transient infection. Supporting this diagnostic role was a recent report correlating the 96 presence of small-colony variants (SCV) of the BCC to the fatal outcome of CF patients that underwent lung transplantation (71). The SCVs were recovered after patients had developed fatal systemic infections post transplantation, but were clonally identical to wildtype BCC strains that colonized the patients prior to lung transplantation. Hence, discrimination of BCC morphotypes may be an important prognosticator of the clinical outcome in CF patients. There are a growing number of reports describing BCC CF isolates that display a mucoid phenotype, similar to that of persistent CF isolates of P. aeruginosa. Since the alginate produced by mucoid P. aeruginosa isolates is considered to be an important virulence factor and hallmark of persistent infections and poor clinical prognosis, the role of EPS for BCC strains may have some significance in their pathogenesis. The conversion of C1394 from a matte to shiny colony appearance has some parallels to P. aeruginosa where non-mucoid strains are cleared until they convert to mucoid and cause chronic infections. Interestingly, the EPS produced by both C1394 and C1394mp2, with elevated levels in the latter, has the same EPS composition associated with mucoid clinical BCC isolates (27, 30). To date, a correlation between this EPS and animal infectivity has not been previously described. It is possible that the increased production of this EPS may contribute to the long-term survival of BCC in the host and provide a protective barrier against host immune effectors. The increased O acetylation detected in the EPS of C1394mp2 may also interfere with the rapid clearance of this variant from the murine host, as O acetylation has been implicated in enhanced resistance to phagocytosis in P. aeruginosa (59, 135). Further experiments to examine the role of EPS will require identification of the genes encoding EPS production. This task should be feasible with the recent report of a putative gene cluster for capsular polysaccharide synthesis in BCC strain J2315 (131). The in vitro data presented in this thesis helps to clarify the roles of the putative surface components involved in C1394mp2 persistence. However these studies revealed that the nonopsonic interaction between B. cenocepacia and primary human phagocytes was exceptionally poor compared to other organisms that have been tested in this system. This observation has been verified by others in this laboratory testing other B. cenocepacia isolates, and may, therefore, denote another characteristic of this genomovar 97 (personal communication, Dr. Barb Conway and Dr. Andrew Currie). While these results may also suggest that C1394mp2 persistence is murine host-specific, the intent of these in vitro studies was to evaluate C1394mp2 within a human infection context. Currently there are very few in vitro systems that enable the study of BCC interactions with host cells, and very little has been described about the phagocytosis of BCC strains or possible ligand-receptor interaction with phagocytes. Using human MDMs and neutrophils, both C1394 and C1394mp2 exhibited low association, affecting the assessment of phagocytic uptake of either isolate. Although differences were not statistically significant between the two isolates, C1394mp2 had consistently lower nonopsonic association with human phagocytes. It is possible that C1394mp2 avoids clearance by maintaining a lower affinity for phagocytic cells. Further analysis may require longer incubation times to confirm these differences in the nonopsonic uptake of these isolates. Other cell lines and types (murine and human) should be used to validate these results or reveal if these low association levels are confined to a particular cell type. Evaluating the intracellular survival of C1394 and C1394mp2 may also highlight differences between the two isolates and the mechanism by which C1394mp2 persists. Since other researchers have demonstrated the intracellular survival of BCC strains using different conditions and cell lines from the ones applied in these studies, more information may be obtained from those methods (112, 148). A549 cells have been used frequently in studies investigating BCC adherence, invasion, and intracellular survival in epithelial cells (19, 34, 88, 188). Based on these published works, the A549 cell line was selected to determine whether the increased piliation on C1394mp2 enhanced adhesion to epithelial cells, and subsequently promoted its persistence. Contrary to past reports that correlated heavily piliated BCC strains to enhanced A549 cell binding, C1394mp2 had reduced binding levels compared to C1394 after forced contact. The two pili types that were detected in C1394 and C1394mp2 closely resembled filamentous and mesh pili of the BCC, which have been previously described but not functionally characterized (58). Therefore it is possible that the differentially expressed filamentous pili do not have a specific function in adhesion to A549 cells or its adhesiveness is altered in C1394mp2. Alternatively another adhesin may be involved that is not expressed or is masked on the surface of C1394mp2. From 98 the data acquired, it is difficult to conclude whether the increased piliation on C1394mp2 occludes A549 binding by another adhesin or whether the increased flagella or EPS production interferes with the binding activities of either the filamentous or mesh pili present on C1394mp2. A better understanding would be gained by constructing isogenic mutants for these surface elements. While successful mutagenesis of BCC strains is still relatively strain-dependent, progress in this area has advanced rapidly in recent years. The sequencing of the B. cenocepacia strain J2315 genome should facilitate the analysis of genes encoding these surface structures and lead to the construction of isogenic mutants in pili, flagella or EPS. The evaluation of these mutants through in vivo and in vitro models should help resolve the roles of these surface determinants in BCC infection and survival in a host. The in vivo phenotype of C1394mp2 in the pulmonary infection model was similar to that of B. multivorans; the number of CPA-treated mice infected with C1394mp2 that appeared symptomatically ill was less than those infected with C1394. Chu et al. speculated that infection by B. cenocepacia strains in CPA-treated mice provoked an aggressive immune response, manifesting in systemic illness and rapid clearance of this genomovar from the murine lung (32). In contrast, B. multivorans caused minimal illness in CPA-treated mice and persisted. The similarities between C1394mp2 and B. multivorans infections are intriguing as the persistence of these organisms may result from their lack of inflammatory potential and failure to evoke an effective immune response. Interestingly, on-going studies in this laboratory showed that many of the persistent B. multivorans strains evaluated in the pulmonary infection model also produced copious amounts of EPS on Y E M agar but were not mucoid on LB or blood agar. While the EPS type from these strains has not been analysed, it is plausible that the ability to produce EPS may enhance the survival of both B. multivorans and C1394mp2 by masking bacterial surface antigens that are targets for the immune system. Alternatively, the decreased association and adherence to host cells exhibited by C1394mp2 may be a means to prevent the induction of an immune response. Also supporting this theory is the decreased expression of possibly immunogenic proteins in C1394mp2. In other pathogens, alkyl hydroperoxidase reductase (AhpC) and dihydrolipoamide succinyltransferase illicit an immune response in animals, and similar 99 proteins were either absent or reduced in cultures of C1394mp2. Since the symptoms associated with C1394 and C1394mp2 infections in mice were analogous to acute and chronic BCC infections respectively, analysis of the host response may reveal how clonal isolates cause such divergent infection outcomes. A comparison of the cytokine responses of phagocytes and epithelial cells induced by C1394 and C1394mp2 pili, flagellin and EPS would provide insight into the type of immune response provoked by each isolate. Protein studies detected the increased production of flagellin by C1394mp2, agreeing with TEM results that showed more flagella with C1394mp2 cells. From PCR and RFLP analysis, C1394 and C1394mp2 carried identical fliC genes encoding the flagellin protein, suggesting that increased flagellin production was due to a change in the regulation of this protein. Proteomic analysis also demonstrated possible posttranslational modifications of the flagellin in C1394mp2, which could have altered the biochemical and immunogenic properties of this protein. Though flagella has been established as an adhesin for some pathogens, the upregulation of this structure did not appear to promote C1394mp2 adhesion to A549 cells. Tomich et al. also demonstrated that BCC flagella did not function as adhesins but were important for the BCC invasion of A549 cells (188). Although C1394 and C1394mp2 were not compared for A549 cell invasion, it is possible that the increased flagellin expression in C1394mp2 may have enhanced its ability to enter and survive in an intracellular niche in the murine host, resulting in the persistent phenotype of this variant. Another significant result from proteomic studies was the differential expression of a peroxiredoxin protein with high similarity to AhpC. This protein is among several antioxidant proteins regulated by OxyR and expressed under oxidative stress. However, AhpC has also been expressed by organisms under other stress conditions, which may indicate its importance in adaptation and response to several environments (12, 126). In addition, stress-induced proteins can be expressed concomitantly with virulence factors in other organisms, indicating a shared regulatory network (45, 123). The differential expression of the putative AhpC protein in C1394 and C1394mp2 may elucidate the regulation of putative BCC virulence factors, including the pili, flagellin and EPS described in this investigation. Furthermore, the heat induction of the shiny phenotype 100 also suggests an association between the stress response and cell surface changes in C1394mp2. Hence, these surface components may be essential for BCC survival in adverse conditions, including infection in a host. However, it is unclear if one of these surface components contributes more to the persistence of C1394mp2 or whether they all collectively work together. Interestingly, all three surface components were increased in C1394mp2, which may suggest some overlap in the regulatory network of these surface features. It is possible that the BCC pili, flagellin and EPS are under global regulation, mediated by alternate sigma factors that recognize stress signals and control several regulons, including virulence genes. Examining the regulation of pili, flagella, and EPS, particularly under different stress conditions, would provide a fuller understanding of how C1394 evolved into C1394mp2, and thus, provide insight into BCC persistence in a host. This thesis describes putative factors that may have been involved in the adaptation and persistence of B. cenocepacia in a murine host. The majority of these factors appeared to be cell surface determinants, which have been described as virulence factors for other pathogens. The identification of non-cable pili, flagella, and EPS in the conversion from a nonpersistent to persistent phenotype suggests their relevance in the long-term survival of BCC in a host. Moreover these findings coincide with recent reports describing the role of these surface components in mechanisms of BCC virulence, further validating their importance. Hence, the presence of these surface determinants may serve as markers of chronic respiratory infections and highlights a potential role for the colonial morphology of BCC in differentiating clinical strains. While results were limited in clarifying the mechanisms of BCC persistence, they suggested an advantage in maintaining an inconspicuous existence in the host that included masking of surface antigens, weak association with host cells, and modification or reduction of immunogenic proteins. Such tactics would prevent the illicitation of an effective immune response and prolong survival within the host. Combined with the sequenced genome of B. cenocepacia strain J2315, proteomic analysis between non-persistent and persistent BCC isolates will undoubtedly identify and reveal more proteins that are involved in the pathogenesis of the BCC. 101 References 1. Aaron, S. D., W. Ferris, D. A. Henry, D. P. Speert, and N. E. Macdonald. 2000. Multiple combination bactericidal antibiotic testing for patients with cystic fibrosis infected with Burkholderia cepacia. Am J Respir Crit Care Med 161:1206-12. 2. Aderem, A., and R. J. Ulevitch. 2000. Toll-like receptors in the induction of the innate immune response. Nature 406:782-7. 3. Aldridge, P., and K. T. Hughes. 2002. Regulation of flagellar assembly. Curr Opin Microbiol 5:160-5. 4. Aoki, M., K. Uehara, K. Tsuji, K. Ono, and M. Iijima. 1993. Large-scale culture and preservation methods of Pseudomonas cepacia B5 for biological control against bacterial wilt disease. Biosci Biotechnol Biochem 57:668-669. 5. Aris, R. M., J. C. Routh, J. J. LiPuma, D. G. Heath, and P. H. Gilligan. 2001. Lung transplantation for cystic fibrosis patients with Burkholderia cepacia complex. Survival linked to genomovar type. Am J Respir Crit Care Med 164:2102-6. 6. Aronoff, S. C. 1988. Outer membrane permeability in Pseudomonas cepacia: diminished porin content in a beta-lactam-resistant mutant and in resistant cystic fibrosis isolates. Antimicrob Agents Chemother 32:1636-9. 7. Arora, S. K., B. W. Ritchings, E. C. Almira, S. Lory, and R. Ramphal. 1998. The Pseudomonas aeruginosa flagellar cap protein, FliD, is responsible for mucin adhesion. Infect Immun 66:1000-7. 8. Azghani, A. O., S. Idell, M. Bains, and R. E. Hancock. 2002. Pseudomonas aeruginosa outer membrane protein F is an adhesin in bacterial binding to lung epithelial cells in culture. Microb Pathog 33:109-14. 9. Balandreau, J., V. Viallard, B. Cournoyer, T. Coenye, S. Laevens, and P. Vandamme. 2001. Burkholderia cepacia genomovar III Is a common plant-associated bacterium. Appl Environ Microbiol 67:982-5. 10. Barsomian, G., and T. G. Lessie. 1986. Replicon fusions promoted by insertion sequences on Pseudomonas cepacia plasmid pTGL6. Mol Gen Genet 204:273-80. 11. Berkelman, T., and T. Stenstedt. 1998. 2-D Electrophoresis: using immobilized pH gradients, principles & methods. Amersham Pharmacia Biotech. 102 12. Bernhardt, J., U. Volker, A. Volker, H. Antelmann, R. Schmid, H. Mach, and M. Hecker. 1997. Specific and general stress proteins in Bacillus subtilis--a two-dimensional protein electrophoresis study. Microbiology 143:999-1017. 13. Bernstein, H. D. 2000. The biogenesis and assembly of bacterial membrane proteins. Curr Opin Microbiol 3:203-9. 14. Bieber, D., S. W. Ramer, C. Y. Wu, W. J. Murray, T. Tobe, R. Fernandez, and G. K. Schoolnik. 1998. Type IV pili, transient bacterial aggregates, and virulence of enteropathogenic Escherichia coli. Science 280:2114-8. 15. Bohin, J. P. 2000. Osmoregulated periplasmic glucans in Proteobacteria. FEMS Microbiol Lett 186:11-9. 16. Boucher, J. C , M. J. Schurr, H. Yu, D. W. Rowen, and V. Deretic. 1997. Pseudomonas aeruginosa in cystic fibrosis: role of mucC in the regulation of alginate production and stress sensitivity. Microbiology 143 ( Pt ll):3473-80. 17. Bowers, J., and J. Parke. 1993. Epidemiology of Pythium damping-off and Aphanomyces root rot of peas after seed treatment with bacterial agents for biological control. Phytopathology 83:1466-73. 18. Burkholder, W. H. 1950. Sour skin: a bacterial rot of onion bulbs. Phytopathology 40:115-117. 19. Burns, J. L., M. Jonas, E. Y. Chi, D. K. Clark, A. Berger, and A. Griffith. 1996. Invasion of respiratory epithelial cells by Burkholderia (Pseudomonas) cepacia. Infect Immun 64:4054-9. 20. Burns, J. L., C. D. Wadsworth, J. J. Barry, and C. P. Goodall. 1996. Nucleotide sequence analysis of a gene from Burkholderia (Pseudomonas) cepacia encoding an outer membrane lipoprotein involved in multiple antibiotic resistance. Antimicrob Agents Chemother 40:307-13. 21. Butler, S. L., C. J. Doherty, J. E. Hughes, J. W. Nelson, and J. R. Govan. 1995. Burkholderia cepacia and cystic fibrosis: do natural environments present a potential hazard? J Clin Microbiol 33:1001-4. 22. Butler, S. L., J. W. Nelson, I. R. Poxton, and J. R. Govan. 1994. Serum . sensitivity of Burkholderia (Pseudomonas) cepacia isolates from patients with cystic fibrosis. FEMS Immunol Med Microbiol 8:285-92. 23. Cangelosi, G. A., C. O. Palermo, J. P. Laurent, A. M. Hamlin, and W. H. Brabant. 1999. Colony morphotypes on Congo red agar segregate along species and drug susceptibility lines in the Mycobacterium avium-intracellulare complex. Microbiology 145 ( Pt 6): 1317-24. 103 24. Cartwright, D. K., W. S. Chilton, and D. M. Benson. 1995. Pyrrolnitrin and phenazine production by Pseudomonas cepacia, strain 5.5B, a biocontrol agent of Rhizoctonia solani. Applied Microbiological Biotechnology 43:211-216. 25. Cash, H. A., D. E. Woods, B. McCullough, W. G. Johanson, Jr., and J. A. Bass. 1979. A rat model of chronic respiratory infection with Pseudomonas aeruginosa. Am Rev Respir Dis 119:453-9. 26. Cerantola, S., J. Bounery, C. Segonds, N. Marty, and H. Montrozier. 2000. Exopolysaccharide production by mucoid and non-mucoid strains of Burkholderia cepacia. FEMS Microbiol Lett 185:243-6. 27. Cerantola, S., A. Lemassu-Jacquier, and H. Montrozier. 1999. Structural elucidation of a novel exopolysaccharide produced by a mucoid clinical isolate of Burkholderia cepacia. Characterization of a trisubstituted glucuronic acid residue in a heptasaccharide repeating unit. Eur J Biochem 260:373-83. 28. Cerantola, S., and H. Montrozier. 2001. Production in vitro, on different solid culture media, of two distinct exopolysaccharides by a mucoid clinical strain of Burkholderia cepacia. FEMS Microbiol Lett 202:129-33. . * 29. Cervin, M. A., D. A. Simpson, A. L. Smith, and S. Lory. 1994. Differences in eucaryotic cell binding of Pseudomonas. Microb Pathog 17:291-9. 30. Cescutti, P., M. Bosco, F. Picotti, G. Impallomeni, J. H. Leitao, J. A. Richau, and I. Sa-Correia. 2000. Structural study of the exopolysaccharide produced by a clinical isolate of Burkholderia cepacia. Biochem Biophys Res Commun 273:1088-94. 31. Chiu, C. H., S. Wong, R. E. Hancock, and D. P. Speert. 2001. Adherence of Burkholderia cepacia to respiratory tract epithelial cells and inhibition with dextrans. Microbiology 147:2651-8. 32. Chu, K. K., D. J. Davidson, T. K. Halsey, J. W. Chung, and D. P. Speert. 2002. Differential persistence among genomovars of the Burkholderia cepacia complex in a murine model of pulmonary infection. Infect Immun 70:2715-20. 33. Chua, K. L., Y. Y. Chan, and Y. H. Gan. 2003. Flagella are virulence determinants of Burkholderia pseudomallei. Infect Immun 71:1622-9. 34. Cieri, M. V., N. Mayer-Hamblett, A. Griffith, and J. L. Burns. 2002. Correlation between an in vitro invasion assay and a murine model of Burkholderia cepacia lung infection. Infect Immun 70:1081-6. 104 35. Coenye, T., S. Laevens, A. Willems, M. Ohlen, W. Hannant, J. R. Govan, M. Gillis, E. Falsen, and P. Vandamme. 2001. Burkholderia fungorum sp. nov. and Burkholderia caledonica sp. nov., two new species isolated from the environment, animals and human clinical samples. Int J Syst Evol Microbiol 51:1099-107. 36. Coenye, T., P. Vandamme, J. R. Govan, and J. J. LiPuma. 2001. Taxonomy and identification of the Burkholderia cepacia complex. J Clin Microbiol 39:3427-36. 37. Conway, B. A., and E. P. Greenberg. 2002. Quorum-sensing signals and quorum-sensing genes in Burkholderia vietnamiensis. J Bacteriol 184:1187-91. 38. Conway, B. A., V. Venu, and D. P. Speert. 2002. Biofilm formation and acyl homoserine lactone production in the Burkholderia cepacia complex. J Bacteriol 184:5678-85. 39. Daniels, C , C. Griffiths, B. Cowles, and J. S. Lam. 2002. Pseudomonas aeruginosa O-antigen chain length is determined before ligation to lipid A core. Environ Microbiol 4:883-97. 40. Darling, P., M. Chan, A. D. Cox, and P. A. Sokol. 1998. Siderophore production by cystic fibrosis isolates of Burkholderia cepacia. Infect Immun 66:874-7. 41. Darveau, R. P., and R. E. Hancock. 1983. Procedure for isolation of bacterial lipopolysaccharides from both smooth and rough Pseudomonas aeruginosa and Salmonella typhimurium strains. J Bacteriol 155:831-8. 42. Daubaras, D. L., C. E. Danganan, A. Hiibner, R. W. Ye, W. Hendrickson, and A. M. Chakrabarty. 1996. Biodegradation of 2,4,5-trichlorophenoxyacetic acid by Burkholderia cepacia strain AC 1100: evolutionary insight. Gene 179:1-8. 43. Davidson, D. J., J. R. Dorin, G. McLachlan, V. Ranaldi, D. Lamb, C. Doherty, J. Govan, and D. J. Porteous. 1995. Lung disease in the cystic fibrosis mouse exposed to bacterial pathogens. Nat Genet 9:351-7. 44. De Soyza, A., A. McDowell, L. Archer, J. H. Dark, S. J. Elborn, E. Mahenthiralingam, K. Gould, and P. A. Corris. 2001. Burkholderia cepacia complex genomovars and pulmonary transplantation outcomes in patients with cystic fibrosis. Lancet 358:1780-1. 45. Deretic, V., M. J. Schurr, J. C. Boucher, and D. W. Martin. 1994. Conversion of Pseudomonas aeruginosa to mucoidy in cystic fibrosis: environmental stress and regulation of bacterial virulence by alternative sigma factors. J Bacteriol 176:2773-80. 105 46. Deretic, V., M. J. Schurr, and H. Yu. 1995. Pseudomonas aeruginosa, mucoidy and the chronic infection phenotype in cystic fibrosis. Trends Microbiol 3:351-6. 47. Deziel, E., Y. Comeau, and R. Villemur. 2001. Initiation of biofilm formation by Pseudomonas aeruginosa 57RP correlates with emergence of hyperpiliated and highly adherent phenotypic variants deficient in swimming, swarming, and twitching motilities. J Bacteriol 183:1195-204. 48. DiRita, V. J., and J. J. Mekalanos. 1989. Genetic regulation of bacterial virulence. Annu Rev Genet 23:455-82. 49. Donnelly, M. A., and T. S. Steiner. 2002. Two nonadjacent regions in enteroaggregative Escherichia coli flagellin are required for activation of toll-like receptor 5. J Biol Chem 277:40456-61. 50. Eaton, K. A., D. R. Morgan, and S. Krakowka. 1992. Motility as a factor in the colonisation of gnotobiotic piglets by Helicobacter pylori. J Med Microbiol 37:123-7. 51. Evans, E., I. R. Poxton, and J. R. Govan. 1999. Lipopolysaccharide chemotypes of Burkholderia cepacia. J Med Microbiol 48:825-32. 52. Everett, K. D., S. Barghouthi, and D. P. Speert. 1996. In vitro culture of murine peritoneal and alveolar macrophages modulates phagocytosis of Pseudomonas aeruginosa and glucose transport. J Leukoc Biol 59:539-44. 53. Feldman, M., R. Bryan, S. Rajan, L. Scheffler, S. Brunnert, H. Tang, and A. Prince. 1998. Role of flagella in pathogenesis of Pseudomonas aeruginosa pulmonary infection. Infect Immun 66:43-51. 54. Fernandez, L. A., and J. Berenguer. 2000. Secretion and assembly of regular surface structures in Gram-negative bacteria. FEMS Microbiol Rev 24:21-44. 55. Frisk, A., J. Jyot, S. K. Arora, and R. Ramphal. 2002. Identification and functional characterization of flgM, a gene encoding the anti-sigma 28 factor in Pseudomonas aeruginosa. J Bacteriol 184:1514-21. 56. Galas, D. J., and M. Chandler. 1989. Bacterial insertion sequences, p. 109-162. In D. E. Berg and M . M . Howe (ed.), Mobile DNA. American Society for Microbiology, Washington, D.C. 57. Gibson, R. L., J. L. Burns, and B. W. Ramsey. 2003. Pathophysiology and management of pulmonary infections in cystic fibrosis. Am J Respir Crit Care Med 168:918-51. 106 58. Goldstein, R., L. Sun, R. Z. Jiang, U. Sajjan, J. F. Forstner, and C. Campanelli. 1995. Structurally variant classes of pilus appendage fibers coexpressed from Burkholderia (Pseudomonas) cepacia. J Bacteriol 177:1039-52. 59. Govan, J. R., and V. Deretic. 1996. Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev 60:539-74. 60. Govan, J. R., J. E. Hughes, and P. Vandamme. 1996. Burkholderia cepacia: medical, taxonomic and ecological issues. J Med Microbiol 45:395-407. 61. Govan, J. R., and P. Vandamme. 1998. Agricultural and medical microbiology: a time for bridging gaps. Microbiology 144 ( Pt 9):2373-5. 62. Grifoll, M., S. A. Selifonov, C. V. Gatlin, and P. J. Chapman. 1995. Actions of a versatile fluorene-degrading bacterial isolate on polycyclic aromatic compounds. Appl Environ Microbiol 61:3711-23. 63. Hagblom, P., E. Segal, E. Billyard, and M. So. 1985. Intragenic recombination leads to pilus antigenic variation in Neisseria gonorrhoeae. Nature 315:156-8. 64. Hakomori, S. 1964. Rapid permethylation of glycolipids and polysaccharides by methylsulfinyl carbanion in dimethyl sulfoxide. J Biochem (Tokyo) 55:205-208. 65. Hales, B. A., J. A. Morgan, C. A. Hart, and C. Winstanley. 1998. Variation in flagellin genes and proteins of Burkholderia cepacia. J Bacteriol 180:1110-8. 66. Hancock, R. E., L. M. Mutharia, L. Chan, R. P. Darveau, D. P. Speert, and G. B. Pier. 1983. Pseudomonas aeruginosa isolates from patients with cystic fibrosis: a class of serum-sensitive, nontypable strains deficient in lipopolysaccharide O side chains. Infect Immun 42:170-7. 67. Hancock, R. E., and H. Nikaido. 1978. Outer membranes of gram-negative bacteria. XIX. Isolation from Pseudomonas aeruginosa PAOl and use in reconstitution and definition of the permeability barrier. J Bacteriol 136:381-90. 68. Hasman, H., M. A. Schembri, and P. Klemm. 2000. Antigen 43 and type 1 fimbriae determine colony morphology of Escherichia coli K-12. J Bacteriol 182:1089-95. 69. Haugland, R. A., U. M. Sangodkar, and A. M. Chakrabarty. 1990. Repeated sequences including RSI 100 from Pseudomonas cepacia ACI 100 function as IS elements. Mol Gen Genet 220:222-8. 107 70. Haugland, R. A., D. J. Schlemm, R. P. Lyons, 3rd, P. R. Sferra, and A. M . Chakrabarty. 1990. Degradation of the chlorinated phenoxyacetate herbicides 2,4-dichlorophenoxyacetic acid and 2,4,5-trichlorophenoxyacetic acid by pure and mixed bacterial cultures. Appl Environ Microbiol 56:1357-62. 71. Haussler, S., C. Lehmann, C. Breselge, M . Rohde, M . Classen, B. Tummler, P. Vandamme, and I. Steinmetz. 2003. Fatal Outcome of Lung Transplantation in Cystic Fibrosis Patients due to Small-Colony Variants of the Burkholderia cepacia Complex. Eur J Clin Microbiol Infect Dis 22:249-53. 72. Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Y i , D. R. Goodlett, J. K. Eng, S. Akira, D. M . Underhill, and A. Aderem. 2001. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410:1099-103. 73. Henderson, I. R., M . Meehan, and P. Owen. 1997. Antigen 43, a phase-variable bipartite outer membrane protein, determines colony morphology and autoaggregation in Escherichia coli K-12. FEMS Microbiol Lett 149:115-20. 74. Henderson, I. R., P. Owen, and J. P. Nataro. 1999. Molecular switches-the ON and OFF of bacterial phase variation. Mol Microbiol 33:919-32. 75. Henry, D. A., M . E. Campbell, J. J. LiPuma, and D. P. Speert. 1997. Identification of Burkholderia cepacia isolates from patients with cystic fibrosis and use of a simple new selective medium. J Clin Microbiol 35:614-9. 76. Hitchcock, P. J., and T. M . Brown. 1983. Morphological heterogeneity among Salmonella lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J Bacteriol 154:269-77. 77. Hobson, R., I. Gould, and J. Govan. 1995. Burkholderia (Pseudomonas) cepacia as a cause of brain abscesses secondary to chronic suppurative otitis media. Eur J Clin Microbiol Infect Dis 14:908-11. 78. Holmes, A., J. Govan, and R. Goldstein. 1998. Agricultural use of Burkholderia (Pseudomonas) cepacia: a threat to human health? Emerg Infect Dis 4:221-7. 79. Huber, B., K. Riedel, M . Hentzer, A. Heydorn, A. Gotschlich, M . Givskov, S. Molin, and L. Eberl. 2001. The cep quorum-sensing system of Burkholderia cepacia H i l l controls biofilm formation and swarming motility. Microbiology 147:2517-28. 80. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 62:379-433. 108 81. Hughes, J. E., J. Stewart, G. R. Barclay, and J. R. Govan. 1997. Priming of neutrophil respiratory burst activity by lipopolysaccharide from Burkholderia cepacia. Infect Immun 65:4281-7. 82. Hutchison, M. L., E. C. Bonell, I. R. Poxton, and J. R. Govan. 2000. Endotoxic activity of lipopolysaccharides isolated from emergent potential cystic fibrosis pathogens. FEMS Immunol Med Microbiol 27:73-7. 83. Hutchison, M. L., and J. R. Govan. 1999. Pathogenicity of microbes associated with cystic fibrosis. Microbes Infect 1:1005-14. 84. Hutchison, M. L., I. R. Poxton, and J. R. Govan. 1998. Burkholderia cepacia produces a hemolysin that is capable of inducing apoptosis and degranulation of mammalian phagocytes. Infect Immun 66:2033-9. 85. Inglis, T. J., T. Robertson, D. E. Woods, N. Dutton, and B. J. Chang. 2003. Flagellum-mediated adhesion by Burkholderia pseudomallei precedes invasion of Acanthamoeba astronyxis. Infect Immun 71:2280-2. 86. Isles, A., I. Maclusky, M. Corey, R. Gold, C. Prober, P. Fleming, and H. Levison. 1984. Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J Pediatr 104:206-10. 87. Jarvis, W. R., D. Olson, O. Tablan, and W. J. Martone. 1987. The epidemiology of nosocomial Pseudomonas cepacia infections: endemic infections. Eur J Epidemiol 3:233-6. 88. Keig, P. M., E. Ingham, and K. G. Kerr. 2001. Invasion of human type II pneumocytes by Burkholderia cepacia. Microb Pathog 30:167-70. 89. Keig, P. M., E. Ingham, P. A. Vandamme, and K. G. Kerr. 2002. Differential invasion of respiratory epithelial cells by members of the Burkholderia cepacia complex. Clin Microbiol Infect 8:47-9. 90. King, E., and J. Parke. 1996. Population density of the biocontrol agent Burkholderia cepacia AMMDR1 on four pea cultivars. Soil Biology and Biochemistry 28:308-312. 91. Kjaergaard, K., M. A. Schembri, H. Hasman, and P. Klemm. 2000. Antigen 43 from Escherichia coli induces inter- and intraspecies cell aggregation and changes in colony morphology oi Pseudomonas fluorescens. J Bacteriol 182:4789-96. 109 92. Krivan, H. C , V. Ginsburg, and D. D. Roberts. 1988. Pseudomonas aeruginosa and Pseudomonas cepacia isolated from cystic fibrosis patients bind specifically to gangliotetraosylceramide (asialo GM1) and gangliotriaosylceramide (asialo GM2). Arch Biochem Biophys 260:493-6. 93. Kuehn, M., K. Lent, J. Haas, J. Hagenzieker, M. Cervin, and A. L. Smith. 1992. Fimbriation of Pseudomonas cepacia. Infect Immun 60:2002-7. 94. Kuehn, M. J. 1997. Establishing communication via gram-negative bacterial pili. Trends Microbiol 5:130-2. 95. Landers, P., K. G. Kerr, T. J. Rowbotham, J. L. Tipper, P. M. Keig, E. Ingham, and M. Denton. 2000. Survival and growth of Burkholderia cepacia within the free-living amoeba Acanthamoeba polyphaga. Eur J Clin Microbiol Infect Dis 19:121-3. 96. Langley, R., D. T. Kenna, P. Vandamme, R. Ure, and J. R. Govan. 2003. Lysogeny and bacteriophage host range within the Burkholderia cepacia complex. J Med Microbiol 52:483-490. 97. Ledson, M. J., M. J. Gallagher, and M. J. Walshaw. 1998. Chronic Burkholderia cepacia bronchiectasis in a non-cystic fibrosis individual. Thorax 53:430-2. 98. Lessie, T. G., W. Hendrickson, B. D. Manning, and R. Devereux. 1996. Genomic complexity and plasticity of Burkholderia cepacia. FEMS Microbiol Lett 144:117-28. 99. Lessie, T. G., M. S. Wood, A. Byrne, and A. Ferrante. 1990. Transposable gene-activating elements in Pseudomonas cepacia, p. 279-291. In S. Silver, A. M . Chakrabartry, B. Iglewski, and S. Kaplan (ed.), Pseudomonas: Biotransformations, Pathogenesis and Evolving Biotechnology. American Society for Microbiology, Washington, D.C. 100. Lewenza, S., B. Conway, E. P. Greenberg, and P. A. Sokol. 1999. Quorum sensing in Burkholderia cepacia: identification of the LuxRI homologs CepRI. J Bacteriol 181:748-56. 101. Lin, J., S. Huang, and Q. Zhang. 2002. Outer membrane proteins: key players for bacterial adaptation in host niches. Microbes Infect 4:325-31. 102. Linker, A., L. R. Evans, and G. Impallomeni. 2001. The structure of a polysaccharide from infectious strains of Burkholderia cepacia. Carbohydr Res 335:45-54. 110 103. Loprasert, S., R. Sallabhan, W. Whangsuk, and S. Mongkolsuk. 2002. The Burkholderia pseudomallei oxyR gene: expression analysis and mutant characterization. Gene 296:161-9. 104. Mahenthiralingam, E., A. Baldwin, and P. Vandamme. 2002. Burkholderia cepacia complex infection in patients with cystic fibrosis. J Med Microbiol 51:533-8. 105. Mahenthiralingam, E., J. Bischof, S. K. Byrne, C. Radomski, J. E. Davies, Y. Av-Gay, and P. Vandamme. 2000. DNA-Based diagnostic approaches for identification of Burkholderia cepacia complex, Burkholderia vietnamiensis, Burkholderia multivorans, Burkholderia stabilis, and Burkholderia cepacia genomovars I and III. J Clin Microbiol 38:3165-73. 106. Mahenthiralingam, E., M. E. Campbell, D. A. Henry, and D. P. Speert. 1996. Epidemiology of Burkholderia cepacia infection in patients with cystic fibrosis: analysis by randomly amplified polymorphic DNA fingerprinting. J Clin Microbiol 34:2914-20. 107. Mahenthiralingam, E., T. Coenye, J. W. Chung, D. P. Speert, J. R. Govan, P. Taylor, and P. Vandamme. 2000. Diagnostically and experimentally useful panel of strains from the Burkholderia cepacia complex. J Clin Microbiol 38:910-3. 108. Mahenthiralingam, E., D. A. Simpson, and D. P. Speert. 1997. Identification and characterization of a novel DNA marker associated with epidemic Burkholderia cepacia strains recovered from patients with cystic fibrosis. J Clin Microbiol 35:808-16. 109. Mahenthiralingam, E., and D. P. Speert. 1995. Nonopsonic phagocytosis of Pseudomonas aeruginosa by macrophages and polymorphonuclear leukocytes requires the presence of the bacterial flagellum. Infect Immun 63:4519-23. 110. Mahenthiralingam, E., P. Vandamme, M. E. Campbell, D. A. Henry, A. M. Gravelle, L. T. Wong, A. G. Davidson, P. G. Wilcox, B. Nakielna, and D. P. Speert. 2001. Infection with Burkholderia cepacia complex genomovars in patients with cystic fibrosis: virulent transmissible strains of genomovar III can replace Burkholderia multivorans. Clin Infect Dis 33:1469-75. 111. Marolda, C. L., B. Hauroder, M. A. John, R. Michel, and M. A. Valvano. 1999. Intracellular survival and saprophytic growth of isolates from the Burkholderia cepacia complex in free-living amoebae. Microbiology 145 ( Pt 7):1509-17. 112. Martin, D. W., and C. D. Mohr. 2000. Invasion and intracellular survival of Burkholderia cepacia. Infect Immun 68:24-9. i l l 113. Mathee, K., O. Ciofu, C. Sternberg, P. W. Lindum, J. I. Campbell, P. Jensen, A. H. Johnsen, M. Givskov, D. E. Ohman, S. Molin, N. Hoiby, and A. Kharazmi. 1999. Mucoid conversion of Pseudomonas aeruginosa by hydrogen peroxide: a mechanism for virulence activation in the cystic fibrosis lung. Microbiology 145 (Pt 6): 1349-57. 114. McKenney, D., K. E. Brown, and D. G. Allison. 1995. Influence of Pseudomonas aeruginosa exoproducts on virulence factor production in Burkholderia cepacia: evidence of interspecies communication. J Bacteriol 177:6989-92. 115. McKevitt, A. I., S. Bajaksouzian, J. D. Klinger, and D. E. Woods. 1989. Purification and characterization of an extracellular protease from Pseudomonas cepacia. Infect Immun 57:771-8. 116. Melnikov, A., O. Zaborina, N. Dhiman, B. S. Prabhakar, A. M. Chakrabarty, and W. Hendrickson. 2000. Clinical and environmental isolates of Burkholderia cepacia exhibit differential cytotoxicity towards macrophages and mast cells. Mol Microbiol 36:1481-93. 117. Mohr, C. D., M. Tomich, and C. A. Herfst. 2001. Cellular aspects of Burkholderia cepacia infection. Microbes Infect 3:425-35. 118. Mongkolsuk, S., and J. D. Helmann. 2002. Regulation of inducible peroxide stress responses. Mol Microbiol 45:9-15. 119. Mongkolsuk, S., W. Whangsuk, P. Vattanaviboon, S. Loprasert, and M. Fuangthong. 2000. A Xanthomonas alkyl hydroperoxide reductase subunit C (ahpC) mutant showed an altered peroxide stress response and complex regulation of the compensatory response of peroxide detoxification enzymes. J Bacteriol 182:6845-9. 120. Montie, T. C , D. Doyle-Huntzinger, R. C. Craven, and I. A. Holder. 1982. Loss of virulence associated with absence of flagellum in an isogenic mutant of Pseudomonas aeruginosa in the burned-mouse model. Infect Immun 38:1296-8. 121. Moore, R. A., and R. E. Hancock. 1986. Involvement of outer membrane of Pseudomonas cepacia in aminoglycoside and polymyxin resistance. Antimicrob Agents Chemother 30:923-6. 122. Morooka, T., A. Umeda, and K. Amako. 1985. Motility as an intestinal colonization factor for Campylobacter jejuni. J Gen Microbiol 131 ( Pt 8): 1973-80. 112 123. Morris, J. G. 1993. Bacterial shock responses. Endeavour 17:2-6. 124. Nguyen, S. V., H. To, T. Yamaguchi, H. Fukushi, and K. Hirai. 1999. Characterization of the Coxiella burnetti sucB gene encoding an immunogenic dihydrolipoamide succinyltransferase. Microbiol Immunol 43:743-9. 125. Old, D. C., and J. P. Duguid. 1970. Selective outgrowth of fimbriate bacteria in static liquid medium. J Bacteriol 103:447-56. 126. Otani, M., J. Tabata, T. Ueki, K. Sano, and S. Inouye. 2001. Heat-shock-induced proteins from Myxococcus xanthus. J Bacteriol 183:6282-7. 127. O'Toole, G. A., and R. Kolter. 1998. Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol Microbiol 30:295-304. 128. Palzkill, T. 2002. Proteomics. Kluwer Academic Publishers, Boston. 129. Parke, J. L., and D. Gurian-Sherman. 2001. Diversity of the Burkholderia cepacia complex and implications for risk assessment of biological control strains. Annu Rev Phytopathol 39:225-58. 130. Parr, T. R., Jr., R. A. Moore, L. V. Moore, and R. E. Hancock. 1987. Role of porins in intrinsic antibiotic resistance of Pseudomonas cepacia. Antimicrob Agents Chemother 31:121-3. 131. Parsons, Y. N., R. Banasko, M. G. Detsika, K. Duangsonk, L. Rainbow, C. A. Hart, and C. Winstanley. 2003. Suppression-subtractive hybridisation reveals variations in gene distribution amongst the Burkholderia cepacia complex, including the presence in some strains of a genomic island containing putative polysaccharide production genes. Arch Microbiol 179:214-23. 132. Parsons, Y. N., K. J. Glendinning, V. Thornton, B. A. Hales, C. A. Hart, and C. Winstanley. 2001. A putative type III secretion gene cluster is widely distributed in the Burkholderia cepacia complex but absent from genomovar I. FEMS Microbiol Lett 203:103-8. 133. Payne, S. M., and Finkelstein, R.A. 1977. Detection and differentiation of iron-responsive avirulent mutants on Congo red agar. Infect Immun 18:94-98. 134. Pegues, D. A., L. A. Carson, R. L. Anderson, M. J. Norgard, T. A. Argent, W. R. Jarvis, and C. H. Woernle. 1993. Outbreak of Pseudomonas cepacia bacteremia in oncology patients. Clin Infect Dis 16:407-11. 1 1 3 135. Pier, G. B., F. Coleman, M. Grout, M. Franklin, and D . E. Ohman. 2001. Role of alginate O acetylation in resistance of mucoid Pseudomonas aeruginosa to opsonic phagocytosis. Infect Immun 69:1895-901. 136. Pujul, M., C. X., J. Carratala, and F. Guidiol. 1992. Community acquired bacteremic Pseudomonas cepacia pneumonia in an immunocompetent host. Clinical Infectious Diseases 15:887-888. 137. Punj, V., R. Sharma, O. Zaborina, and A. M. Chakrabarty. 2003. Energy-generating enzymes of Burkholderia cepacia and their interactions with macrophages. J Bacteriol 185:3167-78. 138. Qadri, F., S. A. Hossain, I. Ciznar, K. Haider, A. Ljungh, T. Wadstrom, and D . A. Sack. 1988. Congo red binding and salt aggregation as indicators of virulence in Shigella species. J Clin Microbiol 26:1343-8. 139. Raetz, C. R., and C. Whitfield. 2002. Lipopolysaccharide endotoxins. Annu Rev Biochem 71:635-700. 140. Richardson, J., D . E. Stead, and R. H. Coutts. 2001. Incidence of the cblA major subunit pilin gene amongst Burkholderia species. FEMS Microbiol Lett 196:61-6. 141. Richau, J. A., J. H. Leitao, M. Correia, L. Lito, M. J. Salgado, C. Barreto, P. Cescutti, and I . Sa-Correia. 2000. Molecular typing and exopolysaccharide biosynthesis of Burkholderia cepacia isolates from a Portuguese cystic fibrosis center. J Clin Microbiol 38:1651-5. 142. Riedel, K., M. Hentzer, O. Geisenberger, B. Huber, A. Steidle, H. Wu, N. Heiby, M. Givskov, S. Molin, and L. Eberl. 2001. N-acylhomoserine-lactone-mediated communication between Pseudomonas aeruginosa and Burkholderia cepacia in mixed biofilms. Microbiology 147:3249-62. 143. Rietschel, E. T., T. Kirikae, F. U. Schade, A. J. Ulmer, O. Hoist, H. Brade, G. Schmidt, U. Mamat, H. D . Grimmecke, S. Kusumoto, and et al. 1993. The chemical structure of bacterial endotoxin in relation to bioactivity. Immunobiology 187:169-90. 144. Roberts, I . S. 1996. The biochemistry and genetics of capsular polysaccharide production in bacteria. Annu Rev Microbiol 50:285-315. 145. Rosen, R., and E. Z. Ron. 2002. Proteome analysis in the study of the bacterial heat-shock response. Mass Spectrom Rev 21:244-65. 114 146. Sadovskaya, I., J. R. Brisson, E. Altman, and L. M. Mutharia. 1996. Structural studies of the lipopolysaccharide O-antigen and capsular polysaccharide of Vibrio anguillarum serotype 0:2. Carbohydr Res 283:111-27. 147. Saiman, L., G. Cacalano, and A. Prince. 1990. Pseudomonas cepacia adherence to respiratory epithelial cells is enhanced by Pseudomonas aeruginosa. Infect Immun 58:2578-84. 148. Saini, L. S., S. B. Galsworthy, M. A. John, and M. A. Valvano. 1999. Intracellular survival of Burkholderia cepacia complex isolates in the presence of macrophage cell activation. Microbiology 145 ( Pt 12):3465-75. 149. Sajjan, S. U., and J. F. Forstner. 1992. Identification of the mucin-binding adhesin of Pseudomonas cepacia isolated from patients with cystic fibrosis. Infect Immun 60:1434-40. 150. Sajjan, U., C. Ackerley, and J. Forstner. 2002. Interaction of cblA/adhesin-positive Burkholderia cepacia with squamous epithelium. Cell Microbiol 4:73-86. 151. Sajjan, U., G. Thanassoulis, V. Cherapanov, A. Lu, C. Sjolin, B. Steer, Y. J. Wu, O. D. Rotstein, G. Kent, C. McKerlie, J. Forstner, and G. P. Downey. 2001. Enhanced susceptibility to pulmonary infection with Burkholderia cepacia in Cftr(-/-) mice. Infect Immun 69:5138-50. 152. Sajjan, U., Y. Wu, G. Kent, and J. Forstner. 2000. Preferential adherence of cable-piliated Burkholderia cepacia to respiratory epithelia of CF knockout mice and human cystic fibrosis lung explants. J Med Microbiol 49:875-85. 153. Sajjan, U. S., M. Corey, M. A. Karmali, and J. F. Forstner. 1992. Binding of Pseudomonas cepacia to normal human intestinal mucin and respiratory mucin from patients with cystic fibrosis. J Clin Invest 89:648-56. 154. Sajjan, U. S., L. Sun, R. Goldstein, and J. F. Forstner. 1995. Cable (cbl) type II pili of cystic fibrosis-associated Burkholderia (Pseudomonas) cepacia: nucleotide sequence of the cblA major subunit pilin gene and novel morphology of the assembled appendage fibers. J Bacteriol 177:1030-8. 155. Sajjan, U. S., F. A. Sylvester, and J. F. Forstner. 2000. Cable-piliated Burkholderia cepacia binds to cytokeratin 13 of epithelial cells. Infect Immun 68:1787-95. 115 156. Salanoubat, M., S. Genin, F. Artiguenave, J. Gouzy, S. Mangenot, M. Arlat, A. Billault, P. Brottier, J. C. Camus, L. Cattolico, M. Chandler, N. Choisne, C. Claudel-Renard, S. Cunnac, N. Demange, C. Gaspin, M. Lavie, A. Moisan, C. Robert, W. Saurin, T. Schiex, P. Siguier, P. Thebault, M. Whalen, P. Wincker, M. Levy, J. Weissenbach, and C. A. Boucher. 2002. Genome sequence of the plant pathogen Ralstonia solanacearum. Nature 415:497-502. 157. Sambrook, J., E. F. Fritsch, and T. Maniatus. 1989. Molecular Cloning: a laboratory manual, 2 ed. Cold Spring Harbor Laboratory Press. 158. Sansonetti, P. J. 1992. Molecular and cellular biology of Shigella flexneri invasiveness: from cell assay systems to shigellosis. Curr Top Microbiol Immunol 180:1-19. 159. Saunders, J. R. 1990. Genetic mechanisms for modulating virulence determinants on the bacterial surface. Sci Prog 74:279-90. 160. Schembri, M. A., L. Hjerrild, M. Gjermansen, and P. Klemm. 2003. Differential expression of the Escherichia coli autoaggregation factor antigen 43. J Bacteriol 185:2236-42. 161. Schembri, M. A., and P. Klemm. 2001. Coordinate gene regulation by fimbriae-induced signal transduction. Embo J 20:3074-81. 162. Schurr, M. J., and V. Deretic. 1997. Microbial pathogenesis in cystic fibrosis: co-ordinate regulation of heat-shock response and conversion to mucoidy in Pseudomonas aeruginosa. Mol Microbiol 24:411-20. 163. Schurr, M. J., H. Yu, J. C. Boucher, N. S. Hibler, and V. Deretic. 1995. Multiple promoters and induction by heat shock of the gene encoding the alternative sigma factor AlgU (sigma E) which controls mucoidy in cystic fibrosis isolates of Pseudomonas aeruginosa. J Bacteriol 177:5670-9. 164. Schwab, U., M. Leigh, C. Ribeiro, J. Yankaskas, K. Burns, P. Gilligan, P. Sokol, and R. Boucher. 2002. Patterns of epithelial cell invasion by different species of the Burkholderia cepacia complex in well-differentiated human airway epithelia. Infect Immun 70:4547-55. 165. Scordilis, G. E., H. Ree, and T. G. Lessie. 1987. Identification of transposable elements which activate gene expression in Pseudomonas cepacia. J Bacteriol 169:8-13. 166. Segal, B. H., T. L. Leto, J. I. Gallin, H. L. Malech, and S. M. Holland. 2000. Genetic, biochemical, and clinical features of Chronic Granulomatous Disease. Medicine (Baltimore) 79:170-200. 116 167. Shapiro, J . A. 1995. The discovery and significance of mobile genetic elements, p. \-\6.In D. J. Sherratt (ed.), Mobile Genetic Elements. Oxford University Press, New York. 168. Shaw, D., I . R . Poxton, and J . R . Govan. 1995. Biological activity of Burkholderia (Pseudomonas) cepacia lipopolysaccharide. FEMS Immunol Med Microbiol 11:99-106. 169. Simpson, I . N., J . Finlay, D. J . Winstanley, N. Dewhurst, J . W. Nelson, S. L. Butler, and J . R . Govan. 1994. Multi-resistance isolates possessing characteristics of both Burkholderia (Pseudomonas) cepacia and Burkholderia gladioli from patients with cystic fibrosis. J Antimicrob Chemother 34:353-61. 170. Singh, P. K., A. L. Schaefer, M. R . Parsek, T. O. Moninger, M. J . Welsh, and E. P. Greenberg. 2000. Quorum-sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms. Nature 407:762-4. 171. Sokol, P. A., P. Darling, S. Lewenza, C. R . Corbett, and C. D. Kooi. 2000. Identification of a siderophore receptor required for ferric ornibactin uptake in Burkholderia cepacia. Infect Immun 68:6554-60. 172. Sokol, P. A., P. Darling, D. E. Woods, E. Mahenthiralingam, and C. Kooi. 1999. Role of ornibactin biosynthesis in the virulence of Burkholderia cepacia: characterization of pvdA, the gene encoding L-ornithine N(5)-oxygenase. Infect Immun 67:4443-55. 173. Speert, D. 2001. Understanding Burkholderia cepacia: epidemiology, genomovars, and virulence. Infect Med 18:49-56. 174. Speert, D. P. 2002. Advances in Burkholderia cepacia complex. Paediatr Respir Rev 3:230-5. 175. Speert, D. P., M. Bond, R . C. Woodman, and J . T. Curnutte. 1994. Infection with Pseudomonas cepacia in chronic granulomatous disease: role of nonoxidative killing by neutrophils in host defense. J Infect Dis 170:1524-31. 176. Speert, D. P., F. Eftekhar, and M. L. Puterman. 1984. Nonopsonic phagocytosis of strains of Pseudomonas aeruginosa from cystic fibrosis patients. Infect Immun 43:1006-11. 177. Speert, D. P., D. Henry, P. Vandamme, M. Corey, and E. Mahenthiralingam. 2002. Epidemiology of Burkholderia cepacia complex in patients with cystic fibrosis, Canada. Emerg Infect Dis 8:181-7. 117 178. Speert, D. P., and D. A. Simpson. 1999. Phagocytosis of Pseudomonas aeruginosa, p. 159-172, Advances in Cell and Molecular Biology of Membranes and Organelles, vol. 6. JAI Press Inc. 179. Speert, D. P., B. Steen, K. Halsey, and E. Kwan. 1999. A murine model for infection with Burkholderia cepacia with sustained persistence in the spleen. Infect Immun 67:4027-32. 180. Springer, B., S. Master, P. Sander, T. Zahrt, M. McFalone, J. Song, K. G. Papavinasasundaram, M. J. Colston, E. Boettger, and V. Deretic. 2001. Silencing of oxidative stress response in Mycobacterium tuberculosis: expression patterns of ahpC in virulent and avirulent strains and effect of ahpC inactivation. Infect Immun 69:5967-73. 181. Starke, J. R., M. S. Edwards, C. Langston, and C. J. Baker. 1987. A mouse model of chronic pulmonary infection with Pseudomonas aeruginosa and Pseudomonas cepacia. Pediatr Res 22:698-702. 182. Stieritz, D. D., and I. A. Holder. 1975. Experimental studies of the pathogenesis of infections due to Pseudomonas aeruginosa: description of a burned mouse model. J Infect Dis 131:688-91. 183. Straus, D. C , M. K. Lonon, and J. C. Hutson. 1992. Inhibition of rat alveolar macrophage phagocytic function by a Pseudomonas cepacia lipase. J Med Microbiol 37:335-40. 184. Sun, L., R. Z. Jiang, S. Steinbach, A. Holmes, C. Campanelli, J. Forstner, U. Sajjan, Y. Tan, M. Riley, and R. Goldstein. 1995. The emergence of a highly transmissible lineage of cbl+ Pseudomonas (Burkholderia) cepacia causing CF centre epidemics in North America and Britain. Nat Med 1:661-6. 185. Sylvester, F. A., U. S. Sajjan, and J. F. Forstner. 1996. Burkholderia (basonym Pseudomonas) cepacia binding to lipid receptors. Infect Immun 64:1420-5. 186. Taylor, P. D., C. J. Inchley, and M. P. Gallagher. 1998. The Salmonella typhimurium AhpC polypeptide is not essential for virulence in BALB/c mice but is recognized as an antigen during infection. Infect Immun 66:3208-17. 187. Tomich, M., A. Griffith, C. A. Herfst, J. L. Burns, and C. D. Mohr. 2003. Attenuated virulence of a Burkholderia cepacia type III secretion mutant in a murine model of infection. Infect Immun 71:1405-15. 188. Tomich, M., C. A. Herfst, J. W. Golden, and C. D. Mohr. 2002. Role of flagella in host cell invasion by Burkholderia cepacia. Infect Immun 70:1799-806. 118 189. Tomlin, K . L., O . P. Coll, and H. Ceri. 2001. Interspecies biofilms of Pseudomonas aeruginosa and Burkholderia cepacia. Can J Microbiol 47:949-54. 190. Tungpradabkul, S., S. Wajanarogana, S. Tunpiboonsak, and S. Panyim. 1999. PCR-RFLP analysis of the flagellin sequences for identification of Burkholderia pseudomallei and Burkholderia cepacia from clinical isolates. Mol Cell Probes 13:99-105. 191. Tyler, S. D., K . R. Rozee, and W. M. Johnson. 1996. Identification of IS1356, a new insertion sequence, and its association with IS402 in epidemic strains of Burkholderia cepacia infecting cystic fibrosis patients. J Clin Microbiol 34:1610-6. 192. Vandamme, P., D. Henry, T. Coenye, S. Nzula, M. Vancanneyt, J. J. LiPuma, D. P. Speert, J. R. Govan, and E. Mahenthiralingam. 2002. Burkholderia anthina sp. nov. and Burkholderia pyrrocinia, two additional Burkholderia cepacia complex bacteria, may confound results of new molecular diagnostic tools. FEMS Immunol Med Microbiol 33:143-9. 193. Vandamme, P., B . Holmes, T. Coenye, J. Goris, E. Mahenthiralingam, J. J. LiPuma, and J. R. Govan. 2003. Burkholderia cenocepacia sp. nov.-a new twist to an old story. Res Microbiol 154:91-6. 194. Vandamme, P., B . Holmes, M. Vancanneyt, T. Coenye, B . Hoste, R. Coopman, H. Revets, S. Lauwers, M. Gillis, K . Kersters, and J. R. Govan. 1997. Occurrence of multiple genomovars of Burkholderia cepacia in cystic fibrosis patients and proposal of Burkholderia multivorans sp. nov. Int J Syst Bacteriol 47:1188-200. 195. Vandamme, P., E. Mahenthiralingam, B . Holmes, T. Coenye, B . Hoste, P. De Vos, D. Henry, and D. P. Speert. 2000. Identification and population structure of Burkholderia stabilis sp. nov. (formerly Burkholderia cepacia genomovar IV). J Clin Microbiol 38:1042-7. 196. Vasselon, T., and P. A. Detmers. 2002. Toll receptors: a central element in innate immune responses. Infect Immun 70:1033-41. 197. Verghese, A., M. al-Samman, D. Nabhan, A. D. Naylor, and M. Rivera. 1994. Bacterial bronchitis and bronchiectasis in human immunodeficiency virus infection. Arch Intern Med 154:2086-91. 198. Wahl, G. M., K . A. Lewis, J. C. Ruiz, B . Rothenberg, J. Zhao, and G. A. Evans. 1987. Cosmid vectors for rapid genomic walking, restriction mapping, and gene transfer. Proc Natl Acad Sci U S A 84:2160-4. 1 1 9 199. Watnick, P., and R. Kolter. 2000. Biofilm, city of microbes. J Bacteriol 182:2675-9. 200. Wine, J. J. 1999. The genesis of cystic fibrosis lung disease. J Clin Invest 103:309-12. 201. Winkelstein, J. A., M. C. Marino, R. B. Johnston, Jr., J. Boyle, J. Curnutte, J. I. Gallin, H. L. Malech, S. M. Holland, H. Ochs, P. Quie, R. H. Buckley, C. B. Foster, S. J. Chanock, and H. Dickler. 2000. Chronic granulomatous disease. Report on a national registry of 368 patients. Medicine (Baltimore) 79:155-69. 202. Winstanley, C , M. G. Detsika, K. J. Glendinning, Y. N. Parsons, and C. A. Hart. 2001. Flagellin gene PCR-RFLP analysis of a panel of strains from the Burkholderia cepacia complex. J Med Microbiol 50:728-31. 203. Winstanley, C , and J. A. Morgan. 1997. The bacterial flagellin gene as a biomarker for detection, population genetics and epidemiological analysis. Microbiology 143 ( Pt 10):3071-84. 204. Wise, M. G., L. J. Shimkets, and J. V. McArthur. 1995. Genetic structure of a lotic population of Burkolderia (Pseudomonas) cepacia. Appl Environ Microbiol 61:1791-8. 205. Wood, M. S., A. Byrne, and T. G. Lessie. 1991. IS406 and IS407, two gene-activating insertion sequences for Pseudomonas cepacia. Gene 105:101-5. 206. Wood, M. S., C. Lory, and T. G. Lessie. 1990. Activation of the lac genes of Tn951 by insertion sequences from Pseudomonas cepacia. J Bacteriol 172:1719-24. 207. Yabuuchi, E., Y. Kosako, H. Oyaizu, I. Yano, H. Hotta, Y. Hashimoto, T. Ezaki, and M. Arakawa. 1992. Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the type species Burkholderia cepacia (Palleroni and Holmes 1981) comb, nov. Microbiol Immunol 36:1251-75. 208. Zhang, J. P., and S. Normark. 1996. Induction of gene expression in Escherichia coli after pilus-mediated adherence. Science 273:1234-6. 209. Zughaier, S. M., H. C. Ryley, and S. K. Jackson. 1999. Lipopolysaccharide (LPS) from Burkholderia cepacia is more active than LPS from Pseudomonas aeruginosa and Stenotrophomonas maltophilia in stimulating tumor necrosis factor alpha from human monocytes. Infect Immun 67:1505-7. 120 210. Zughaier, S. M . , H . C. Ryley, and S. K. Jackson. 1999. A melanin pigment purified from an epidemic strain of Burkholderia cepacia attenuates monocyte respiratory burst activity by scavenging superoxide anion. Infect Immun 67:908-13. 211. Zygmunt, M . S., M . A. Diaz, A. P. Teixeira-Gomes, and A. Cloeckaert. 2001. Cloning, nucleotide sequence, and expression of the Brucella melitensis sucB gene coding for an immunogenic dihydrolipoamide succinyltransferase homologous protein. Infect Immun 69:6537-40. 121 APPENDIX 7.1 Genetic Complementation ofB. multivorans strain J T C B. multivorans strain JTC persists in the spleen for up to 8 weeks in a murine model of intraperitoneal (i.p.) infection. Genetic complementation with a cosmid library of strain JTC into a non-persistent strain was used to identify the critical bacterial determinants for BCC infection and persistence. Unlike JTC, B. cenocepacia strain K56-2 is eliminated from the spleen within 7 days after i.p. infection. Strain K56-2 was electroporated and transformed with the cosmid library of JTC, which generated transconjugants that were evaluated for any enhanced survival in the murine spleen. Transconjugants recovered from the spleen (at CFU comparable to strain JTC) were passaged at least twice to enrich for persistence. This would allow the isolation of genes that played a role in BCC persistence and survival in the mouse. 7.1.1 Cosmid L i b r a r y of strain J T C A cosmid library of strain JTC was generated with the cosmid vector, pScosBCl, which contained both ampicillin and trimethoprim resistance genes for selection. Genomic DNA of strain JTC was partially digested with restriction endonuclease Sau3A and ligated to 5awHI-cut pScosBCl, then packaged into recombinant lambda phage (Gigapack III Gold Packaging extract; Stratagene), and used to transfect E. coli strain MRI-XL Blue (Stratagene) to propagate the library. After transfection, E. coli were grown on LB agar containing ampicillin (100 |ig/ml). Transfection titres from several attempts averaged at 105CFU/ml of undiluted packaged library. Cosmids were extracted from at least twenty randomly selected colonies, digested with BamWl, and resolved on a 1% (w/v) agarose to analyze for random distribution of the library and DNA insert size. No siblings were detected among these clones and the average genomic insert size was 30 kb. The cosmid library was extracted and purified using a plasmid miniprep kit (Qiagen) according to the manufacturer's instructions. 122 7.1.2 Selection of a recipient BCC host for the JTC cosmid library B. multivorans strains from clinical and environmental sources were evaluated for the following as potential host strains: i) high transformation frequency with electroporation to take up the JTC cosmid library; ii) serum resistance to select for strains that were similar to strain JTC, which is also serum resistant; and iii) non-persistence in the i.p. mouse model (Table 7). Selection of serum resistant strains allowed for the isolation of other integral genes that were required for systemic survival, apart from those that were directly involved with serum resistance. Five B. multivorans strains showed resistance to pooled human serum, and attained moderate to high transformation frequencies when electroporated with pScosBCl. However all five strains demonstrated delayed clearance, with bacterial titres recovered at 104 CFU in the spleens of C57B1/6 mice on Day 14 after i.p. challenge. This observation suggested a general persistent phenotype among B. multivorans strains in the i.p. infection model. Therefore the search for a non-persistent host strain was extended to other genomovars of the BCC, specifically B. cenocepacia. It is important to note that at this period of time, the distinction between B. multivorans and B. cenocepacia were not yet clearly defined and B. cenocepacia was still referred as genomovar III. Hence, it was not known if the genetic backgrounds between these two genomovars would be too divergent for cross species gene expression. All B. cenocepacia strains that were tested in the i.p. infection model were cleared from the murine spleen by Day 14. Previous work by Lewenza et al. had established B. cenocepacia strain K56-2 as one of the few BCC strains that were amenable to genetic manipulation (100). Hence strain K56-2 was chosen among the non-persistent B. cenocepacia strains as the recipient BCC strain for the JTC cosmid library. 123 Table 7. Evaluation of potential recipient B. multivorans strains for the J T C cosmid library Strain Strain origin Transformation Serum Persistence in frequency resistant/sensitive the mouse per ug of D N A (R/S) (to Day 14) C4861 Clinical, B.C. 106 R Yes C7906 Clinical, B.C. 105 R Yes CEP208 Environmental, Edinburgh 102 R Yes CEP691 Clinical, Denmark 106 R Yes CEP694 Clinical, Denmark >103 S N/D CEP695 Clinical, Denmark >103 S N/D CEP698 Clinical, Denmark >103 R N/D CEP781 CF epidemic, Glasgow >103 R Yes Bp23 Environmental, Vancouver <103 R N/D Bp 102 Environmental, Vancouver N T R N/D NT, no transformants N/D,not done K56-2 transconjugants and enrichment of persistent clones Strain K56-2 was electroporated with 1 pg of the JTC cosmid library and grown on LB agar containing trimethoprim for the selection of transconjugants. Random K56-2 transconjugants were selected and verified on BCSA. Transconjugants that grew on selective LB agar were pooled together, transferred to fresh LB broth containing trimethoprim, and incubated for 4 hours at 37°C. The expanded transconjugant pool was harvested, resuspended in gHBSS, and prepared for i.p. infection as previously described (section 2.5.1). Four mice were challenged with transconjugants, sacrificed at days 0 and 3, and spleens were removed, homogenized and plated onto LB agar with trimethoprim to recover transconjugants. Persistent K56-2 transconjugants recovered at day 3 were pooled and frozen at -70°C in LB broth with 8% DMSO in cryovials. Enrichment of persistent transconjugants involved transferring day 3 isolates from cryovials to fresh LB broth with trimethoprim, incubating bacteria for 12 hours, and preparing transconjugants for i.p. infection as described above. Persistent transconjugants were subjected to two enrichment passages in the i.p. model. Cosmids from enriched transconjugants were extracted, and digested with EcoRl to resolve vector from DNA insert. Results showed that one particular clone was predominantly enriched from several transconjugant pools that were prepared for mouse infection. The cosmid from this clone contained DNA insert, however, it did not confer enhanced survival to new K56-2 transconjugants that were generated with this cosmid. Control infections and enrichment passages were performed with the control transconjugant K56-2C, which contained an empty cosmid vector as previously described (section 3.1). The persistence of K56-2C suggested that 124 the cosmid vector pScosBCl itself may have contributed to the persistence of K56-2 or that K56-2 itself was capable of adapting to the murine host during this enrichment process. Furthermore, the persistent derivative of K56-2C exhibited a shiny colonial morphology, which was in contrast to the matte colonial morphology of naive K56-2C. The matte colonial morphology was also observed among K56-2 transconjugants prior to infection whereas the shiny morphology was detected among enriched transconjugants. K56-2 transconjugant pool library Since one particular clone dominated the enriched K56-2 transconjugant pools in the i.p. model, a transconjugant library was produced to obtain even representation of different clones within the population. Since the average BCC genome size was calculated as 8-9Mb and cosmids contained an average DNA insert size of 30 kb, 300 transconjugants would represent the entire genome. To allow for sufficient coverage of the BCC genome, 700 colonies were isolated in 96-well microtitre plates, producing a transconjugant library of seven microtitre plates. The library was stored at -70°C in LB broth containing 8.0% (v/v) DMSO and cultured at 37°C in new microtitre plates with each well containing 150 ul of LB broth containing trimethoprim. Transconjugants from each microtitre plate were pooled as a population (i.e. total of seven pools) and prepared for i.p. infection in the mice. This involved taking 75 ul from each well after 48 hours incubation, pooling together the volumes from one microtitre plate, then harvesting this mixed culture by centrifugation, and resuspending the pellet in 2 ml of gHBSS. Transconjugant pools were adjusted to an inoculum of 8 x 106 CFU/ml, and mice were infected as previously described (section 2.5.1). Two mice were challenged with each pool, sacrificed at days 0 and 3, and spleens were removed, homogenized and plated onto LB agar with trimethoprim to recover transconjugants. Cosmids were extracted from persistent transconjugants recovered at day 3. Although different clones were recovered from the library pools, the extracted cosmids did not confer enhanced survival in the mouse when analysed further. 125 7.1.4 Analysis of K56-2C adaptation to the murine host To determine the role of pScosBCl in K56-2C persistence, the enriched isolate, K56-2Ce was cured of its vector by several in vitro passages on LB agar without trimethoprim. Sensitivity to trimethoprim and cosmid extractions demonstrated that cured K56-2Ce no longer contained the pScosBCl vector. In addition cured K56-2Ce maintained the shiny colonial morphology. C57B1/6 mice were challenged intraperitoneally with the cured K56-2Ce isolate and results showed that this isolate remained in the spleen at bacterial titres that were intermediate between persistent K56-2Ce and naive K56-2C. This suggested that K56-2 itself, was capable of adapting to the murine host, however, it is possible that pScosBCl may have potentiated K56-2 adaptation or its enhanced survival. 7.2 Discussion Complementation of B. cenocepacia strain K56-2 with the cosmid library of B. multivorans strain JTC showed that the cosmid vector pScosBCl might have influenced the adaptation of K56-2 to the murine host. Alternatively, expression of B. multivorans genes may have been limited in a B. cenocepacia host. It is also possible that the tranformation frequency of K56-2 was not sufficiently high enough to get ample representation of the JTC library. In addition, transconjugants pools may have had uneven representation of the JTC library with one particular clone that dominated over underrepresented or slow growing transconjugants in the population. These studies first demonstrated B. cenocepacia adaptation to the murine host and revealed observations with colonial morphology and persistence that were further analyzed in subsequent studies highlighted in this thesis. Moreover, these results indicated that B. cenocepacia possessed genes that were required for enhanced survival in the murine host. 126 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items