PATHOGENIC MECHANISMS OF CAMPYLOBACTER JEJUNI: CHARACTERIZATION AND IDENTIFICATION OF THE ROLE OF CADF IN CAMPYLOBACTER MEDIATED ENTERITIS

Transcription

1 PATHOGENIC MECHANISMS OF CAMPYLOBACTER JEJUNI: CHARACTERIZATION AND IDENTIFICATION OF THE ROLE OF CADF IN CAMPYLOBACTER MEDIATED ENTERITIS By MARSHALL RENO MONTEVILLE A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY (Microbiology) WASHINGTON STATE UNIVERSITY School of Molecular Biosciences May 2003

2 ii To the Faculty of Washington State University: The members of the Committee appointed to examine the dissertation of Marshall Reno Monteville find it satisfactory and recommend that it be accepted. Chair

3 iii ACKNOWLEDGMENTS I would like to thank my committee: Drs. Michael E. Konkel, Michael L. Kahn, Philip F. Mixter, and Thomas E. Besser for their assistance in providing me with guidance throughout my graduate studies. A special thanks goes to Dr. Michael Konkel for his dedication in providing the necessary direction and continuous support in seeing me through my doctoral studies in a timely manner. Furthermore I thank the United States Navy for affording me the opportunity to attend graduate school while on active duty.

4 iv PATHOGENIC MECHANISMS OF CAMPYLOBACTER JEJUNI: CHARACTERIZATION AND IDENTIFICATION OF THE ROLE OF CADF IN CAMPYLOBACTER MEDIATED ENTERITIS Chair: Michael E. Konkel Abstract by Marshall Reno Monteville, Ph.D. Washington State University May 2003 Campylobacter jejuni, a Gram-negative, spiral shaped bacterium is currently recognized as the leading cause of gastroenteritis in humans worldwide. Despite this prevalence, little is known regarding the mechanism by which C. jejuni causes disease. Researchers have proposed a multifactorial infection process whereby motility, chemotaxis, host cell-translocation, host celladherence, host cell-invasion, and toxin production function in a coordinated manner resulting in disease. The primary goal of the research described herein was to characterize the role of adherence in establishment of a successful infection in a human host using in vitro model systems. More specifically, work focussed on characterizing the role of the C. jejuni adhesin CadF, Campylobacter adhesion to fibronectin, in C. jejuni-mediated enteritis. The CadF adhesin is a 37 kda outer membrane protein which facilitates the adherence of C. jejuni to the host extracellular matrix component fibronectin. The results generated revealed that the CadF adhesin functions to localize C. jejuni to specific receptors on host-cells thereby maximizing invasion. Furthermore, the CadF adhesin enables the bacterium to adhere to fibronectin localized at the basolateral surface of hostcells following translocation via a paracellular route. This interaction prevents the organism from translocation completely across an intact epithelial barrier and therefore allows the bacterium to subsequently invade host cells. I hypothesize that C. jejuni association with fibronectin leads to integrin (α 5 β 1 ) occupancy and clustering. This hypothesis is supported by data which revealed that

5 v host-cell signalling events, such as phosphorylation of paxillin, were dependent upon the CadF adhesin. Following the binding of C. jejuni to fibronectin, data further indicate that C. jejuni is taken up by host cells via cooperative interaction between both microfilaments and microtubules. Additional studies using the C. jejuni CadF amino acid sequence and synthesized peptides with overlapping regions revealed that the CadF adhesin possesses a single fibronectin binding domain. These results provide a more detailed model of C. jejuni-mediated enteritis and support the hypothesis that the CadF adhesin contributes to the virulence of C. jejuni by maximizing the organism s adherence to appropriate host-cell receptors. In turn, this binding increases the organism s invasive potential.

6 vi TABLE OF CONTENTS ACKNOWLEDGMENTS...iii ABSTRACT...iv TABLE OF CONTENTS...vi LIST OF TABLES...xiv LIST OF FIGURES...xvi DEDICATION...xix Chapter 1 INTRODUCTION...1 Background...1 Work completed in this dissertation...4 REFERENCES...8 Chapter 2 The pathogenesis of Campylobacter jejuni-mediated enteritis...11 ABSTRACT...12 INTRODUCTION...13 MOTILITY AND FLAGELLA...15 CHEMOTAXIS...17 TRANSLOCATION...19 ADHESINS AND THE ROLE OF ADHERENCE...21 INVASION...27 SECRETION...29

7 vii CYTOLETHAL DISTENDING TOXIN...31 LIPOPOLYSACCHARIDE AND CAPSULAR POLYSACCHARIDE...34 IRON ACQUISITION...36 INTRACELLULAR SURVIVAL IN MONONUCLEAR PHAGOCYTES...39 ROLE OF OXIDATIVE RADICALS IN PHAGOLYSOSOME SURVIVAL...40 A MODEL OF C. jejuni PATHOGENESIS...41 CONCLUDING COMMENTS...44 ACKNOWLEDGEMENTS...45 REFERENCES...46 Chapter 3 In vitro and in vivo models used to study Campylobacter jejuni-virulence properties...62 ABSTRACT...63 INTRODUCTION...64 INTERACTIONS OF C. jejuni WITH NON-PROFESSIONAL PHAGOCYTIC CELLS...67 Adherence...67 Invasion...70 Translocation...74 INTERACTIONS OF C. jejuni WITH PROFESSIONAL PHAGOCYTIC CELLS...76 Engulfment...77 Survival...78 Role of Phagolysosome Processing in Macrophage Survival...79 IN VIVO COLONIZATION MODELS...80 IN VIVO INFECTION MODELS...81 Small Animal Models...82

8 viii Large Animal Models...83 MOLECULAR APPROACHES USED IN CAMPYLOBACTER RESEARCH...84 Methods Used to Generate Defined C. jejuni Mutants...85 Methods Used to Generate Random C. jejuni Mutants (Transposons)...88 NOVEL MOLECULAR APPROACHES...90 SUMMARY...92 REFERENCES...93 Chapter 4 Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface ABSTRACT INTRODUCTION MATERIALS AND METHODS Bacterial isolates and growth condition Tissue culture T84 non-polarized adherence and internalization assays T84 polarized translocation, adherence and internalization assays Competitive binding assays Statistical analysis RESULTS Translocation across T84 polarized cells is independent of Cia protein secretion Translocation results from saturation of host-cell receptors Invasion of T84 cells occurs primarily via the basolateral surface DISCUSSION...125

9 ix ACKNOWLEDGEMENTS REFERENCES Chapter 5 Maximal adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer membrane protein and microfilament reorganization ABSTRACT INTRODUCTION MATERIALS AND METHODS Bacterial isolates and growth condition Binding of C. jejuni to immobilized ECM Gel electrophoresis and immunoblot analysis Bactericidal concentration of gentamicin INT 407 binding and internalization assays Competitive inhibition assays Inhibitor studies Reversibility of cytochalasin D Preparation of the polyclonal antisera Confocal microscopy examination of C. jejuni infected cells Immunoprecipitation Other analytical procedures RESULTS CadF promotes the binding of C. jejuni to Fn...154

10 x CadF is required for the maximal binding and internalization of C. jejuni to INT 407 cells C. jejuni internalization involves actin cytoskeletal reorganization The effect of cytochalasin D on C. jejuni internalization is reversible C. jejuni internalization is sensitive to microtubule inhibitors Maximal C. jejuni entry is accompanied by the tyrosine phosphorylation of paxillin DISCUSSION ACKNOWLEDGEMENTS REFERENCES Chapter 6 Identification of a fibronectin-binding domain within the Campylobacter jejuni CadF adhesin ABSTRACT INTRODUCTION MATERIALS AND METHODS Synthesis of CadF peptides Peptide-fibronectin binding assay Peptide inhibition assay Expression of cadf in Escherichia coli Purification of recombinant CadF RESULTS The FRLS residues within CadF are essential for Fn-binding activity The binding of rcadf to Fn is specific DISCUSSION...197

11 xi ACKNOWLEDGEMENTS REFERENCES Chapter 7 The bile salt sodium deoxycholate alters the kinetics of Campylobacter jejuni invasion ABSTRACT INTRODUCTION MATERIALS AND METHODS Bacterial isolates and growth condition Preparation of secreted proteins One and two-dimensional gel electrophoresis Immunoblot analysis Isolation of a C. jejuni flagellar export apparatus mutant Phenotypic analysis of the C. jejuni flhb mutant Tissue culture Binding and internalization assays Other analytical procedures RESULTS Deoxycholate stimulates the synthesis of the Cia proteins The profile of secreted proteins does not include flagellar proteins Cia protein secretion is dependent upon a functional flagellar export apparatus Deoxycholate enhances C. jejuni internalization of T84 cells Pre-culturing C. jejuni with deoxycholate alters the kinetics of bacterial uptake DISCUSSION...224

12 xii ACKNOWLEDGEMENTS REFERENCES Chapter 8 Secretion of virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES Bacterial isolates and growth conditions Isolation of C. jejuni flagellar mutants Phenotypic analysis of the C. jejuni flagellar mutants Complementation analysis Analysis of the C. jejuni secreted proteins Examination of the interactions of C. jejuni with INT 407 cells Other analytical procedures RESULTS Generation of C. jejuni flagellar mutants Secretion of the Cia proteins requires the intact flagellar apparatus Complementation analysis restores Cia protein secretion Maximal C. jejuni invasion requires Cia protein secretion and motility Cia protein export is independent of σ DISCUSSION ACKNOWLEDGEMENTS...263

13 xiii REFERENCES Chapter 9 Conclusions and future directions...281

14 xiv LIST OF TABLES Chapter 4 Table 1. Cia proteins are required for maximal invasion of T84 non-polarized cells by C. jejuni. Table 2. The ability of C. jejuni to translocate across T84 polarized cells is independent of invasive potential. Table 3. C. jejuni that become cell-associated remain associated with T84 polarized cells. Table 4. The CadF protein is required for maximal T84 cell-association regardless of MOI. Table 5. The functional role of CadF as an adhesin is shared among the C. jejuni clinical isolates and F Table 6. C. jejuni association with fibronectin maximizes bacterial adherence and invasion of T84 non-polarized cells. Table 7. C. jejuni association with fibronectin localized to the basolateral surface of T84 polarized cells maximizes bacterial adherence and invasion. Chapter 5 Table 1. Competitive inhibition of C. jejuni binding to Fn with antibodies reactive against either Fn or CadF and upon the addition of exogenous Fn. Table 2. The functional role of CadF as an adhesin is conserved in C. jejuni clinical isolate Table 3. Effect of microfilament inhibitors on C. jejuni binding and internalization. Table 4. Effect of mycalolide B microfilament inhibitor on C. jejuni binding and internalization. Table 5. Reversibility of cytochalasin D on C. jejuni binding and internalization. Table 6. Effect of microtubule inhibitors on C. jejuni binding and internalization. Table 7. Combined effects of microtubule and microfilament inhibitors on C. jejuni binding and internalization.

15 xv Chapter 6: Table 1. Peptides (30-mers) used in this study. Chapter 7: Table 1. Pre-culturing C. jejuni on deoxycholate-supplemented MH agar plates versus non-supplemented MH agar plates results in a two-fold increase in protein secretion. Table 2. Pre-culturing C. jejuni on deoxycholate-supplemented MH agar plates overcomes the inhibitory affect of chloramphenicol on T84 cell invasion. Chapter 8: Table 1. Phenotypes displayed by the C. jejuni wild-type isolates and isogenic mutants.

16 xvi LIST OF FIGURES Chapter 2 Figure 1. C. jejuni infection cycle. Figure 2. Diagram depicting C. jejuni-virulence determinants and their potential role in the development of C. jejuni-mediated enteritis. Figure 3. Hematoxylin and eosin stained sections of the small intestines of E. coli and C. jejuniinoculated piglets. Chapter 3 Figure 1. Diagram showing the adherence and invasion assay. Figure 2. Diagram showing cross-section of a polarized cell monolayer. Figure 3. Transmission electron micrograph of J774A.1 macrophage infected with C. jejuni M129. Chapter 4 Figure 1. Translocation kinetics of C. jejuni F38011 and the isogenic cadf mutant across polarized T84 cells at various MOIs. Chapter 5 Figure 1. Representative gel and immunoblot showing the detection of the CadF protein in the whole-cell extracts of C. jejuni F38011, , and an F38011 cadf mutant. Figure 2. Staining of microfilaments in INT 407 cells infected with C. jejuni, S. typhimurium, and C. freundii infected INT 407 cells. Figure 3. Staining of microtubules in INT 407 cells infected with C. jejuni, S. typhimurium, and C. freundii infected INT 407 cells.

17 xvii Figure 4. Binding of C. jejuni to INT 407 cells induces phosphorylation of the 68 kda focal adhesion-associated protein paxillin. Chapter 6 Figure 1. Fn binding to CadF peptides (30-mers). Figure 2. Fn binding to CadF peptides (16-mers). Figure 3. Competitive inhibition assay using peptides to block Fn binding to rcadf. Chapter 7: Figure 1. Culturing C. jejuni on deoxycholate-supplemented MH agar plates increases the amount of the Cia proteins in the supernatant fluids. Figure 2. Culturing C. jejuni on a deoxycholate-supplemented MH agar plate increases the amount of the Cia proteins in the supernatant fluids two-fold over that of C. jejuni cultured on a non-supplemented MH agar plate. Figure 3. The profile of secreted proteins does not include FlaA or FlaB. Figure 4. FlhB, a component of the flagellar export apparatus, is required for Cia protein secretion. Figure 5. Pre-culturing C. jejuni on deoxycholate-supplemented agar plates alters the kinetics of bacterial uptake. Chapter 8: Figure 1. Assessment of C. jejuni motility on Mueller-Hinton medium supplemented with 0.4% Select Agar. Figure 2. Immunoblot analysis with an anti-flagellin antibody. Figure 3. Autoradiographs showing the secretion profiles of the C. jejuni flagellar mutants. Figure 4. Autoradiograph showing the secreted protein profiles of a C. jejuni filament mutant with a recombinant plasmid harboring either flaa or flab.

18 Figure 5. Autoradiograph showing the secretin profile of the C. jejuni flia mutant. xviii

19 xix DEDICATION This dissertation is dedicated to my wife Kay and my children Joshua and Samantha for their continued support throughout my studies at Washington State University.

20 1 Chapter 1 INTRODUCTION Background C. jejuni is currently recognized as the leading cause of bacterial-induced gastroenteritis in developed nations worldwide (Tauxe, 1992; Altekruse et al., 1999). Nevertheless, our current understanding of how this pathogen causes disease in a human host remains less well defined than other Gram-negative organisms. There is a consensus among investigators that adherence (De Melo & Pechère, 1990; Fauchère et al., 1989; Jin et al., 2001; Kelle et al., 1998; Kervella et al., 1993; Konkel et al., 1997; McSweegan & Walker, 1986; Moser & Schröder, 1995; Moser et al., 1997; Moser et al., 1992; Pei & Blaser, 1993; Pei et al., 1998; Schröder & Moser, 1997), translocation (Brás and Ketley, 1999; Everest et al., 1992; Harvey et al., 1999; Konkel et al., 1992), invasion (Everest et al., 1992; Konkel et al., 1992; Konkel et al., 1993; Russell et al., 1993), and toxin production (Pickett and Whitehouse, 1999; Wassenaar, 1997) play a role in C. jejunimediated enteritis. As such, research presented herein was aimed at further elucidating the role of adherence and translocation in the infectious process, as well as determining the stage in which each of these virulence attributes is essential. Furthermore, an attempt was made to address the impact of translocation and adherence on C. jejuni invasion of human intestinal epithelial cells. Bacterial adherence to host cells is mediated by surface exposed molecules termed adhesins. To date, the best characterized C. jejuni adhesins are CadF, JlpA, and PEB1 (Konkel et al., 1997; Jin et al., 2001; Pei et al., 1998). The CadF (Campylobacter adhesion to Fn) adhesin is a 37 kda outer membrane protein that mediates C. jejuni binding to the extracellular matrix component fibronectin (Konkel et al., 1997). The cadf gene has thus far been found to be conserved among C. jejuni and

21 2 C. coli isolates (Konkel et al., 1999). In vivo studies have suggested that the CadF adhesin is required for the colonization of chickens by C. jejuni (Ziprin et al., 1999). However, the role of the CadF adhesin in a human host remains to be elucidated. The architecture of the intestinal epithelium dictates that the extracellular matrix component fibronectin is localized to the basolateral surface (Vuori, 1998). Therefore, C. jejuni would only have the capacity to associate with this hostcell receptor after traversing an epithelial barrier and entering the lamina propria. Bacterial translocation is the movement of bacteria across an intact polarized epithelial barrier. Polarized cells are characterized by defined apical and basolateral cell surfaces. Adjacent cells are separated by tight junctions, which limit the passage of solutes through the paracellular spaces (Madara, 1998). Translocation is considered an important virulence attribute for certain pathogens because it permits access to underlying tissues and may allow for the organism s dissemination throughout a host. Previous work has revealed that C. jejuni can translocate a Caco-2 polarized cell monolayer without a concomitant loss in transepithelial resistance (Everest et al., 1992; Harvey et al., 1999; Konkel et al., 1992), indicating that C. jejuni can translocate across a cell monolayer whose integrity remains intact. However, the ability of C. jejuni to translocate does not appear to quantitatively correlate with an organism s invasive capacity (Harvey et al., 1999). Nevertheless, Everest et. al. noted that 86% of Campylobacter clinical isolates from individuals with colitis translocated across Caco-2 polarized cells in vitro as compared to 46% of isolates from individuals with noninflammatory disease (Everest et al., 1992). These data suggest that while C. jejuni translocation is independent of invasion, translocation of the pathogen results in more pronounced symptomology associated with C. jejuni-mediated enteritis. Translocation of C. jejuni, independent of invasion, would suggest that C. jejuni are capable of traversing an intact monolayer via a paracellular (between cells) versus a transcellular (through cells) route of passage. Work by others has revealed that cellular tight junctions can reseal following bacterial penetration (Takeuchi, 1967), thus providing a basis for C. jejuni to utilize a paracellular route of passage without the long-term

22 3 disruption of the integrity of the cell monolayer. Based on our current model of C. jejuni pathogenesis, and that CadF appears to be a conserved virulence determinant, I hypothesize that CadF functions to promote interaction with the host cell extracellular matrix component fibronectin following translocation of an intact epithelial barrier. Bacterial invasion involves the participation of host cells via rearrangement of the cytoskeleton (e.g., microfilaments and microtubules). Published work indicates a relationship between the paracellular fibronectin matrix and the cytoskeleton (Gumbiner, 1996; Miyamoto et al., 1998; van der Flier & Sonnenberg, 2001). More specifically, fibronectin binds to α 5 β 1 integrin receptors at sites termed focal adhesions (Hynes, 1992; Miyamoto et al., 1998), which in turn serve as the active site for actin nucleation. Integrin occupancy and clustering results in a tyrosine phosphorylation of proteins associated with focal adhesions and subsequently induces host-cell signalling events facilitating actin rearrangement and bacterial uptake (Miyamoto et al., 1998; Tachibana et al., 1995). I hypothesize that C. jejuni association to fibronectin via CadF leads to integrin clustering and hostcell signalling resulting in bacterial induced uptake. The specific interaction between the CadF adhesin and fibronectin is yet to be defined. To continue the characterization of the C. jejuni CadF adhesin, experiments were performed to identify the CadF fibronectin binding domain/s. Using peptides (30-mers) with overlapping regions corresponding to the amino sequence of CadF, data suggests that a single domain within CadF facilitates binding of the adhesin to human fibronectin. Of interest, previous studies directed at characterizing the Mycobacterium avium FAP-A (fibronectin attachment protein-a) fibronectin binding domain identified a single region, consisting of 12 amino acids (FAP-A ), which was involved in adherence to fibronectin (Zhao et. al. 1999). Furthermore, the authors concluded that four FAP-A amino acids (RWFV) within the 12 amino acid region were essential in binding fibronectin. Based on my preliminary data, I hypothesize that the C. jejuni CadF adhesin harbors a single fibronectin

23 binding domain in which amino acids (FRLS) are essential in fibronectin binding activity. 4 Work completed in this dissertation The work presented in this dissertation is directed at further understanding the role of the CadF adhesin in C. jejuni-mediated enteritis. The manuscripts presented herein provide the reader with a current model of infection beginning with translocation of the organism across the intestinal epithelium, adherence of C. jejuni to fibronectin localized to the basolateral surface of host cells, and subsequent invasion of the host epithelium through exploitation of host-signalling processes. Chapter Two, The pathogenesis of Campylobacter jejuni-mediated enteritis, is a review article describing C. jejuni virulence determinants and their role in pathogenesis. My personal contribution to this article was in generating Figure 3 as well as writing sections describing Chemotaxis, Translocation, Adhesins and the Role of Adherence, Cytolethal Distending Toxin and Lipopolysaccharide and Capsular Polysaccharide. Dr. Vanessa Rivera-Amill contributed in generating Figure 1 as well as writing the section on Secretion. Dr. Lynn Joens, University of Arizona, contributed by providing the images used for Figure 2 and writing the sections on Intracellular Survival in Mononuclear Phagocytes and Role of Oxidative radicals in Phagolysosome Survival. Dr. Michael E. Konkel contributed in writing the remaining sections and was the major contributor to editing each section as well as writing the final manuscript. Chapter Three, In vitro and in vivo models used to study Campylobacter jejuni-virulence properties is a book chapter dedicated to describing current experimental systems used by researchers to study C. jejuni pathogenesis. My contribution was in generating Figures 1 and 2. I also assisted primarily in the writing initial drafts of the sections describing Interactions of C. jejuni with non-professional phagocytic cells, In vivo infection models, and Molecular approaches used in

24 5 Campylobacter research. I also contributed in editing the section on Interactions of C. jejuni with professional phagocytic cells. Dr. Lynn Joens contributed in writing the initial draft of the section describing Interactions of C. jejuni with professional phagocytic cells and In vivo infection models and provided the images used in Figures 3 and 4. Dr. John D. Klena contributed by editing and formatting the entire book chapter. Dr. Michael E. Konkel was the major contributor to writing the final drafts of each section as well as the Introduction and Summary. Chapter Four, Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface describes the role of the CadF adhesin following translocation of the organism across the intestinal epithelium. Data indicate that C. jejuni traverse an intact T84 polarized intestinal monolayer via a paracellular route thereby adhering to the fibronectin localized to the basolateral cell surface. C. jejuni association with fibronectin at the basolateral surface of host cells was found to facilitate maximal uptake of C. jejuni by intestinal epithelial cells. My contribution to this manuscript was in performing all experiments as well as generating each of the tables and figures. The majority experiments were designed by Dr. Michael E. Konkel with my technical assistance in experimental approach. I assisted Dr. Michael E. Konkel in writing and formatting the manuscript. Chapter Five, Maximal adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer membrane protein and microfilament reorganization focusses on determining the role of the C. jejuni CadF adhesin in the exploitation of the host cytoskeleton thereby inducing bacterial uptake. Data suggest that CadF promotes C. jejuni-host cell interactions which in turn stimulate signalling pathways associated with focal adhesions. Subsequently, C. jejuni is taken up via a cooperative interaction involving both microtubules and microfilaments. My contribution to this manuscript was in performing the experiments used to generate Tables 1 through 7 as well as Figures 1 through 3. Julie Yoon performed all work involving paxillin and

25 6 generated Figure 4. The majority of experiments were designed by Dr. Michael E. Konkel with my technical assistance in experimental approach. Dr. Michael E. Konkel was instrumental in performing the confocal microscopy for Figures 2 and 3. Dr. John D. Klena contributed by performing the gentamicin-sensitivity assay, generation of the Cj1477c (putative hydrolase gene) suicide vector, and PCR analysis to confirm plasmid carriage using gene specific primers targeting virb11 and teto. Joey Mickelson assisted in generating the Cj1477c knockout. I also assisted Dr. Michael E. Konkel in writing and formatting the manuscript. Chapter Six, Identification of a fibronectin-binding domain within the Campylobacter jejuni CadF adhesin describes research performed to determine the specific amino acids within the CadF adhesin which facilitate adherence to fibronectin. Data suggest that amino acids are essential in the binding of fibronectin by the C. jejuni CadF adhesin. My contribution to this preliminary manuscript was in performing all experiments used to generate all figures and tables. Dr. John D. Klena developed the vectors which were used in the expression of cadf in E. coli via an arabinose inducible promoter. Joey Mickelson assisted in the purification of the recombinant CadF protein. Experimental approach was directed by Dr. Michael E. Konkel with my technical assistance in optimizing ELISA procedures. I also assisted Dr. Michael E. Konkel in writing and formatting the manuscript. Chapter Seven, The bile salt deoxycholate alters the kinetics of Campylobacter jejuni invasion represents a collaborative effort by members of the Konkel lab. Research described in this chapter is directed at characterizing the Cia (Campylobacter invasion antigens) proteins from the C. jejuni clinical isolate and their role in promoting bacterial-host cell interactions. Data suggest the secreted Cia proteins from C. jejuni strain play a role in promoting bacterial uptake by human intestinal epithelial cells. Furthermore, the bile salt sodium deoxycholate serves as a stimulatory signal for Cia synthesis. These data are in agreement with that previously described in

26 7 regards to the C. jejuni clinical isolate F Nevertheless, this research provides novel insight describing the affects sodium deoxycholate has on the synthesis of Cia proteins and subsequently on bacterial-host cell interactions. For the first time it is shown that by pre-synthesizing Cia proteins the kinetics of invasion are significantly altered such that C. jejuni readily invades human intestinal epithelial cells upon contact with host cells. In addition, experiments were performed to quantify the concentration of Cia proteins secreted per bacterium in the presence, or absence, of presynthesis. This manuscript introduces the possibility that C. jejuni utilizes the Type III flagellar apparatus to secrete the Cia proteins. My contribution to this research was in generating the data for Figure 5 as well as Tables 1 and 2. I also assisted in editing the manuscript and formatting Tables and Figures. The majority of experiments were designed by Dr. Michael E. Konkel with my technical assistance in experimental approach relating to my specific contributions. The entire manuscript was written by Dr. Michael E. Konkel with editing assistance from Dr. John D. Klena and myself. Chapter Eight, Secretion of virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus represents a long-term collaborative effort by members of the Konkel lab. This chapter describes the role the flagellar apparatus plays in secretion of Cia proteins. Data indicate that an intact basal body and hook is not sufficient for secretion of the Cia proteins. In fact, an intact flagellar structure (containing the basal body, hook and at least a partial filament) is required for export of Cia proteins via the flagellum. My personal contribution to this manuscript was in performing all adherence and invasion assays used in Table 1. I also assisted in preparing Figure 1 with data generated by myself and others in the lab. Dr. Michael E. Konkel wrote the entire manuscript with editing assistance from Dr. John D. Klena and myself.

27 8 REFERENCES Altekruse, S. F., N. J. Stern, P. I. Fields, and D. L. Swerdlow. (1999). Campylobacter jejuni - - an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Brás, A. M., and J. M. Ketley Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: De Melo, M. A. and Pechère, J.-C. (1990). Identification of Campylobacter jejuni surface proteins that bind to eucaryotic cells in vitro. Infect. Immun. 58: Everest, P. H., H. Goossens, J. P. Butzler, D. Lloyd, S. Knutton, J. M. Ketley, and P. H. Williams. (1992). Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J. Med. Microbiol. 37: Fauchère, J.-L., Kervella, M., Rosenau, A., Mohanna, K. and Véron, M. (1989). Adhesion to HeLa cells of Campylobacter jejuni and C. coli outer membrane components. Res. Microbiol. 140: Gumbiner, B. M. (1996). Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 84: Harvey, P., T. Battle, and S. Leach. (1999). Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Hynes, R. O. (1992). Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69: Jin, S., Joe, A., Lynett, J., Hani, E. K., Sherman, P. and Chan, V. L. (2001). JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Kelle, K., Pagès, J.-M. and Bolla, J.-M. (1998). A putative adhesin gene cloned from Campylobacter jejuni. Res. Microbiol. 149: Kervella, M., Pagès, J.-M., Pei, Z., Grollier, G., Blaser, M. J. and Fauchère, J.-L. (1993). Isolation and characterization of two Campylobacter glycine-extracted proteins that bind to HeLa cell membranes. Infect. Immun. 61: Konkel, M. E., and W. Cieplak, Jr. (1992). Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M. E., Garvis, S. G., Tipton, S. L., Anderson, D. E., Jr. and Cieplak, W., Jr. (1997). Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24:

28 Konkel, M. E., S. A. Gray, B. J. Kim, S. G. Garvis, and J. Yoon. (1999). Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadf virulence gene and its product. J. Clin. Microbiol. 37: Konkel, M. E., D. J. Mead, and W. Cieplak, Jr. (1993). Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Madara, J. L. (1998). Regulation of the movement of solutes across tight junctions. Annu. Rev. Physiol. 60: McSweegan, E. and Walker, R. I. (1986). Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infect. Immun. 53: Miyamoto, S., Katz, B.-Z., LaFrenie, R. M. and Yamada, K. M. (1998). Fibronectin and integrins in cell adhesion, signaling, and morphogenesis. Ann N.Y. Acad Sci 857, Moser, I. and Schröder, W. (1995). Binding of outer membrane preparations of Campylobacter jejuni to INT 457 cell membranes and extracellular matrix proteins. Med. Microbiol. Immunol. 184: Moser, I., Schröder, W. and Salnikow, J. (1997). Campylobacter jejuni major outer membrane protein and a 59-kDa protein are involved in binding to fibronectin and INT 407 cell membranes. FEMS Mirobiol. Lett. 157: Moser, I., Schröder, and Hellmann, E. (1992). In vitro binding of Campylobacter jejuni/coli outer membrane preparations to INT 407 cell membranes. Med. Microbiol. Immunol. 180: Pei, Z. and Blaser, M. J. (1993). PEB1, the major cell-binding factor of Campylobacter jejuni, is a homolog of the binding component in Gram-negative nutrient transport systems. J. Biol. Chem. 268: Pei, Z., Burucoa, C., Grignon, B., Baqar, S., Huang, X.-Z., Kopecko, D. J., Bourgeois, A. L., Fauchere, J.-L. and Blaser, M. J. (1998). Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Pickett, C. L., and C. A. Whitehouse. (1999). The cytolethal distending toxin family. Trends in Microbiol. 7: Russell, R. G., M. O'Donnoghue, D. C. Blake, Jr., J. Zulty, and L. J. DeTolla. (1993). Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Schröder, W. and Moser, I. (1997). Primary structure analysis and adhesion studies on the major outer membrane protein of Campylobacter jejuni. FEMS Microbiol. Lett. 150: Tachibana, K., Sato, T., D'Avirro, N. and Morimoto, C. (1995). Direct association of pp125fak with paxillin, the focal adhesion-targeting mechanism of pp125fak. J. Exp. Med. 182:

29 10 Takeuchi, A Electron microscopic studies of experimental Salmonella infection. Penetration into the intestinal epithelium by Salmonella typhimurium. Am. J. Pathol. 50: Tauxe, R.V. (1992). Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Nachamkin, I., Blaser, M.J. and Tompkins, L. (eds.), Campylobacter jejuni:current status and future trends. American Society for Microbiology, Washington, D.C., pp van der Flier, A. and Sonnenberg, A. (2001). Function and interactions of integrins. Cell Tissue Res. 305: Vuori, K. (1998). Integrin signaling: tyrosine phosphorylation events in focal adhesions. J. Membrane Biol. 165: Wassenaar, T. M. (1997). Toxin production by Campylobacter spp. Clin. Microbiol. Rev. 10: Zhao, W., Schorey J.S., Groger R., Allen P.M., Brown E.J., Ratliff T.L. (1999). Characterization of the fibronectin binding motif for a unique mycobacterial fibronectin attachment protein, FAP. J. Biol. Chem. 274(8): Ziprin, R. L., Young, C. R., Hume, M. E. and Konkel, M. E. (1999). The absence of cecal colonization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectinbinding protein. Avian Dis. 43:586-9.

30 11 Chapter 2 The pathogenesis of Campylobacter jejuni-mediated enteritis Michael E. Konkel,a* Marshall R. Monteville,a Vanessa Rivera-Amill,a and Lynn A. Joensb School of Molecular Biosciences, Washington State University, Pullman, Washington 99164;a Departments of Veterinary Science and Microbiology, University of Arizona, Tucson, AZ 85721b Published in Current Issues in Intestinal Microbiology (2001) 2(2): 55-71

31 12 ABSTRACT Campylobacter jejuni, a Gram-negative spiral shaped bacterium, is a frequent cause of gastrointestinal food-borne illness in humans throughout the world. Illness with C. jejuni ranges from mild to severe diarrheal disease. This articles focuses on Campylobacter virulence determinants and their potential role in the development of C. jejuni-mediated enteritis. A model is presented that diagrams the interactions of C. jejuni with the intestinal epithelium. Additional work to identify and characterize C. jejuni virulence determinants is certain to provide novel insights into the diversity of strategies employed by bacterial pathogens to cause disease.

32 13 INTRODUCTION The genus Campylobacter contains 14 species of which C. jejuni, C. coli and C. fetus are the most frequently isolated from humans. Campylobacter species are Gram-negative rods that have curved or spiral morphology and are motile by means of unipolar or bipolar flagella. They grow best in microaerophilic atmosphere at temperatures ranging from 37 to 42 C. The genome of Campylobacter jejuni is roughly 1.6 to 1.7 Mbp with a GC ratio of approximately 30% (Owen and Leaper, 1981). Extrachromosomal elements including plasmids and bacteriophages have also been detected in Campylobacter sp. (Bradbury et al., 1983; Bacon et al., 2000). The illness associated with Campylobacter infections ranges from mild to severe diarrheal disease, with stool specimens often containing blood and leukocytes. Other symptoms include fever, nausea and abdominal pain. The incubation period of Campylobacter enteritis ranges from two to seven days and the illness is often self-limiting. In severe cases, patients are treated with erythromycin. Experimental C. jejuni infections in humans have revealed that a low dose of organisms ( ) is sufficient to cause diarrhea and that the numbers of individuals who incur the disease increases with higher doses (Black et al., 1988). The most notable complication of C. jejuni infections is the development of Guillain-Barré syndrome (GBS), an acute demyelinating polyneuropathy. Development of GBS follows gastrointestinal disease and is characterized by flaccid paralysis. O- side chain serotyping studies have revealed that certain Campylobacter serotypes are linked to GBS. Most GBS cases in the United States occur after an infection with serotype O:19. However, other C. jejuni serotypes have also been identified in association with GBS including O:1, O:2, O:2/44, O:4 complex, O:5, O:10, O:16, O:23, O:37, O:41, O:44 and O:64 (Reviewed in (Nachamkin et al., 1998); (Prendergast et al., 1998)). The ecologic cycle of C. jejuni involves water, animals and food (Fig. 1). Infection with C. jejuni is

33 14 most frequently acquired from the consumption and handling of chicken. Infections also occur from drinking unpasteurized milk and contaminated water. The majority of C. jejuni infections are sporadic in nature (Friedman et al., 2000). Despite this, C. jejuni is the leading cause of bacterial gastroenteritis in the United States with an estimated 2.4 million cases per year (Allos and Blaser, 1995; Altekruse et al., 1999; Allos, 2001). The incidence of C. jejuni infections in the United States is higher during the late summer and early fall with few cases occurring throughout the year (Allos and Blaser, 1995). C. jejuni affects all age groups but infants and young adults have the highest reported rates of infection (Allos and Blaser, 1995). In this article, we focus on C. jejuni virulence determinants and their potential role in the development of C. jejuni-mediated enteritis. In the past decade, there has been an explosion of new information as researchers have begun to study the molecular basis of C. jejuni pathogenesis. While a clear picture of C. jejuni virulence determinants and their role in disease has yet to emerge, research in this area is intensifying with the development of new tools to genetically manipulate the organism (Golden et al., 2000) and with the availability of the genome sequence (Parkhill et al., 2000). Perhaps one of the most interesting questions that researchers are just beginning to address is whether C. jejuni isolates possess a repertoire of virulence genes that can be expressed discordantly or are comprised of a mosaic of virulence-associated genes. Surely a picture is likely to evolve that involves both of these possibilities. In this article, we attempt to present a global view of C. jejuni pathogenesis. In most instances, the contribution of a given gene to the organism s virulence is tenuous due to the lack of availability of a simple and inexpensive animal model to researchers. We also present a working model that diagrams the interactions of C. jejuni with the intestinal epithelium. While our current working model is based on speculation, we hope that it will serve to stimulate additional discussion and research in this area. Finally, many fine review articles have been published that focus on only one of the topics discussed below in much greater detail (Ketley, 1997; Wassenaar, 1997; Wooldridge and Ketley, 1997; Nachamkin et al., 1998; Altekruse

34 et al., 1999; Pickett and Whitehouse, 1999; Allos, 2001). We encourage you to revisit the articles as we only briefly discuss some of this organism s virulence attributes due to space constraints. 15 MOTILITY AND FLAGELLA The role of motility in C. jejuni colonization and subsequent disease production has been intensely studied. Research in this area has been greatly aided by availability of simple and inexpensive models to study C. jejuni colonization and the ease of isolating C. jejuni nonmotile mutants by several methods including treatment of the bacteria with chemical mutagens. The flagellum of C. jejuni is composed of a basal body, hook, and filament. In C. jejuni and Campylobacter coli, the flagellar filament is comprised of two proteins termed FlaA and FlaB. Both C. jejuni flagellin proteins are synthesized concomitantly, but flaa, which is regulated by σ28 (Nuijten et al., 1991), is expressed at much greater levels than flab, which is regulated by σ54 (Hendrixson et al., 2001). In contrast to C. jejuni flaa+flab- mutants in which full-length filaments are produced, C. jejuni flaaflab+ mutants produce filaments that are truncated (Guerry et al., 1991; Wassenaar et al., 1991). In Campylobacter organisms, motility correlates with the synthesis of the FlaA protein. However, C. jejuni flaa-flab+ mutants have been isolated that form full-length flagella but are less motile than the C. jejuni wild-type isolate (Wassenaar et al., 1994). In 1985, Morooka et al. (Morooka et al., 1985) treated C. jejuni with N-methyl-N -nitro-nnitrosoguanidine and methyl methane sulphonate and isolated bacteria with motility defects as judged by the hanging drop method. Each of the C. jejuni nonmotile isolates tested, of which one was flagellated and two were nonflagellated, were unable to colonize suckling mice and were cleared from the intestinal tract 2 d post-challenge. The investigators hypothesized that motility was required for C. jejuni to swim through the viscous mucus (Morooka et al., 1985). Subsequently, Newell (Newell, 1986) reported that a C. jejuni nonflagellated isolate termed SF-2 poorly colonized

35 16 mice and was cleared from the intestinal tract 7 d post-challenge. In the same study, Newell (Newell, 1986) found that a C. jejuni nonmotile, flagellated isolate termed SF-1 colonized the intestinal tract of mice as successfully as the wild-type strain. It was latter established that the SF-1 isolate was indeed motile, although not to the same extent as that of the C. jejuni wild-type isolate, thus clarifying the discrepancy between their findings and those published by Morooka et al. (Morooka et al., 1985; Wassenaar et al., 1993). Wassenaar et al. (Wassenaar et al., 1993) used genetically defined C. jejuni flaa and flab mutants to determine the importance of each of the flagellin proteins in the colonization of 1-d-old chicks. The investigators found that neither flagellin expression nor motility were essential for colonization in this model, but that a C. jejuni flaa+flabmutant colonized chicks at a level 1000-fold greater than the wild-type isolate. Based on their results, Wassenaar et al. (Wassenaar et al., 1993) concluded that bacteria expressing the flaa+ gene promotes maximal colonization. Consistent with these findings, others have reported that motility is important in promoting the colonization of animals by C. jejuni (Pavlovskis et al., 1991; Nachamkin et al., 1993). In the early 1990s, several studies were undertaken to further dissect the importance of motility versus the actual flagellum in the interaction of C. jejuni with cultured epithelial cells (Wassenaar et al., 1991; Grant et al., 1993; Yao et al., 1994). Motility, and the expression of the flaa+ gene was found necessary for the maximal invasion of eukaryotic cells and for the translocation of polarized cell monolayers by C. jejuni (Wassenaar et al., 1991; Grant et al., 1993). However, differences were noted in the invasive potential of the C. jejuni flaa- flab+ and C. jejuni flaa- flab- isolates, with the former being more invasive (Wassenaar et al., 1991; Grant et al., 1993). Given the differences observed in invasive potential of the C. jejuni mutants, Grant et al. (Grant et al., 1993) concluded that the flagellar structure played a role in the internalization process of C. jejuni that was independent of motility.

36 17 Of interest is the finding that C. jejuni flagella undergo phase variation (Caldwell et al., 1985). In a study in which human volunteers were challenged with a mixture of a C. jejuni motile isolate and a non-motile phase-variant of the same isolate, only the motile phase variant was recovered from stool samples (Black et al., 1988). Consistent with this study and the importance of motility and flagella in bacteria-host cell interactions, Wassenaar et al. (Wassenaar et al., 1994) isolated a motile variant of a C. jejuni flaa-flab+ mutant after performing an invasion assay. As discussed above, flaa is normally expressed at greater levels than flab. Given this fact, Wassenaar et al. (Wassenaar et al., 1994) hypothesized that the co-cultivation of C. jejuni with intestinal cells leads to the expression of flab and the shutdown of flaa transcription. Collectively, previous work indicates that motility contributes significantly to the colonization of animals by C. jejuni and subsequently in the development of disease in susceptible hosts. CHEMOTAXIS Chemotaxis is the movement of an organism towards or away from a chemical stimulus. To determine whether a particular chemical acts as an attractant or repellent, investigators have commonly used a plate assay in which a bacteria are mixed in a PBS-solution containing 0.35 to 0.4% agar (Hugdahl et al., 1988). After the addition of the test chemicals to the plates by either hard-agar plugs (HAP) or filter discs, the plates can be incubated under a variety of environmental conditions and chemotactic response assessed. A zone of turbidity in an agar plate reflects an organism s migration toward the substrate, and is indicative of a positive chemoattractant response. Hazeleger et al. (Hazeleger et al., 1998) examined the chemotactic behavior of C. jejuni to a variety of chemical stimuli. In their study, C. jejuni was found to exhibit a positive chemotactic response to the carbohydrate L-fucose, the amino acids L-aspartate, L-cysteine, L-glutamate, and L-serine, and the organic acids pyruvate, succinate, fumarate, citrate, malate, and α-ketoglutarate (Hugdahl et al., 1988). The investigators also found that mucin, a glycoprotein of high molecular weight that

37 18 contains L-fucose as its terminal sugar, acts as chemoattractant for C. jejuni. The contribution of chemotaxis in colonization has been noted to be important for other pathogenic bacteria including V. cholerae, S. typhimurium, and E. coli (Allweiss et al., 1977). Several reports also suggest that chemotaxis is an important C. jejuni virulence determinant. In 1992, Takata et al. (Takata et al., 1992) found that mice were colonized when challenged orally with 110 CFU of a C. jejuni wild-type isolate, but not when challenged with as many as 5 x 107 CFU of twoindependently isolated C. jejuni chemotaxis mutants. The C. jejuni nonchemotactic mutants used in the study were isolated using the HAP assay, and judged to possess flagella of the same length and width as that of the wild-type isolate. Yao et al. (Yao et al., 1997) explored the in vitro and in vivo role of chemotaxis using a set of defined C. jejuni mutants. Here, a C. jejuni chey null mutant was generated, and found to display a nonchemotactic but motile phenotype. This C. jejuni mutant exhibited a threefold increase in adherence and invasion of INT 407 cells when compared to the wild-type isolate, but was unable to colonize mice or cause symptoms in infected ferrets. A possible explanation for these findings is that the motility of a chey mutant is altered such that organisms makes longer runs, resulting in increased host cell contacts that promote irreversible cell adherence and invasion. However, the increase in the lengths of the runs in vivo, without chemotaxis providing appropriate directionality towards mucus, could lead to the organism s expulsion from the host by fluid flow and peristaltic activity. In the same study, Yao et al. (Yao et al., 1997) performed similar experiments with a C. jejuni isolate containing two copies of chey. Of interest was the finding that the C. jejuni chey diploid isolate displayed remarkably different behavior than the chey null mutant. While the C. jejuni chey diploid isolate demonstrated chemotactic behavior as expected, it also exhibited a decrease in its in vitro adherence and invasion capabilities and colonized mice. Similar to the chey null mutant, the chey diploid isolate was unable to cause disease in the ferret model. In the in vivo experiments, it is possible that the organism migrated towards the mucus within the crypts, but was unable to produce runs of sufficient lengths

38 19 to penetrate the mucus due to its viscosity. Thus, the chemotactic response of C. jejuni appears important in directing the organism to specific sites in the host s intestinal tract. TRANSLOCATION Investigators have utilized a unique cell culture system to assess the ability of pathogens to translocate across cell barriers. Briefly, the cells are cultured on a permeable membrane, and fresh media is added to the apical and basolateral chambers to promote cell growth and differentiation. Cell differentiation results in the establishment of distinct apical and basolateral cell surfaces with their own set of surface markers. The apical cell surfaces are further characterized by welldeveloped microvilli and brush borders. Monolayer integrity can be monitored throughout the course of the assay by measuring transepithelial electrical resistance (TER), and a decrease, or loss of TER, indicates disruption of the tight junctions. The HT29, Caco-2, and T84 human adenocarcinoma cell lines have been most commonly used to examine the ability of enteric pathogens to translocate across a cell barrier. This capability is considered an important virulence attribute for some pathogens as it permits them access to underlying tissues and could promote their dissemination throughout the host. Nevertheless, the degree to which a pathogen translocates across a cell barrier and the organism s fate beyond the local environment differs greatly among pathogens. For example, S. typhi rapidly translocates across a polarized monolayer, causing cellular destruction and extrusion that leads to a complete loss of monolayer integrity. In contrast, the translocation of S. typhimurium across polarized cells causes minimal damage to the cell monolayer early in the process (Kops et al., 1996). Presumably these data are reflective of in vivo disease presentations where an infection with S. typhi is commonly septic in nature and an infection with S. typhimurium is generally localized to the intestinal mucosa.

39 20 In 1992, Everest et al. (Everest et al., 1992) noted that 86% of Campylobacter isolates from individuals with colitis were able to translocate across polarized Caco-2 cells versus 48% of strains isolated from individual with noninflammatory disease. The translocation of C. jejuni across polarized Caco-2 cell monolayers was determined by plating serial dilutions of the basolateral chamber media on agar plates. Interestingly, six C. jejuni isolates characterized as being noninvasive were able to translocate across the polarized monolayers. To further define the interactions of C. jejuni with polarized cells, Harvey et al. (Harvey et al., 1999) compared the ability of 4 clinical C. jejuni isolates to translocate across polarized Caco-2 cells with their ability to invade both polarized and non-polarized cells. The authors found that an organism s invasiveness does not quantitatively correlate with its ability to translocate across a cell monolayer. While the investigators also detected fluctuations in the measurable TER with the different C. jejuni isolates over the course of the assay (6 hr), monolayer integrity was maintained and final TER values comparable to starting baseline values. Maintenance of monolayer integrity, at least over a relatively short period of time (8 h), has also been reported by others (Konkel et al., 1992c; Brás and Ketley, 1999). Noteworthy is that Bras et al. (Brás and Ketley, 1999) detected a loss in TER of Caco-2 cells inoculated with C. jejuni after 24 h, indicating an eventual disruption of cellular tight junctions. The investigators proposed that the loss in monolayer integrity was either a result of long-term effects of translocation and/or invasion or the accumulation of a bacterial toxin(s). The aforementioned studies suggest that the genes that encode the products responsible for invasion in C. jejuni are distinct from those that confer translocation ability. The mechanism by which C. jejuni translocates across polarized cells is presently unclear but could be accomplished by either a transcellular (through a cell) or a paracellular (between cells) route. Evidence supporting the transcellular route of passage is the presence of intracellular bacteria and the fact that C. jejuni-cellular translocation is reduced at 20 C (Konkel et al., 1992c). Temperatures of C preferentially inhibit eukaryotic endocytic and phagocytic processes (Silverstein et al.,

40 ). Evidence supporting the paracellular route of passage is that C. jejuni can be recovered from the basolateral chamber as early as 15 min post-inoculation of a polarized cell monolayer (Konkel et al., 1992c). In addition, the invasiveness of C. jejuni isolates does not quantitatively correlate with translocation efficiency (Harvey et al., 1999). It is possible that C. jejuni organisms utilize the paracellular route of passage based on work indicating that cellular tight junctions can reseal following bacterial penetration (Takeuchi, 1967). Previous studies have also demonstrated that tight junctions temporally relax to allow regulated passage of both solutes and neutrophils (Madara, 1998). The consensus among investigators is that C. jejuni initially colonizes the jejunum and ileum, and then the colon (Allos and Blaser, 1995; Skirrow and Blaser, 2000). However, histological examination of C. jejuni-infected humans and animals has revealed pathology primarily in the colon (Black et al., 1988; Babakhani et al., 1993; Russell et al., 1993). Advantages for C. jejuni reaching the underlying tissue and submucosa include access to a different set of cellular molecules that serve as receptors and the fact that the organisms are no longer subject to the peristaltic action of the intestine. If cellular translocation is associated with the development of C. jejuni-mediated enteritis, then several mechanisms of translocation may occur depending on the target site. In the intestinal tract, access to the submucosa could be achieved via translocation of the intestinal epithelia by either the transcellular or paracellular routes discussed above. Alternatively, C. jejuni may gain access to the submucosa via uptake by M cells (Walker et al., 1988). Not known is whether C. jejuni can translocate across the cells in the colon. Noteworthy is that the incidence of C. jejuni septicemia is low (0.4% cases) (Allos and Blaser, 1995), suggesting that C. jejuni organisms are less well equipped to survive and proliferate following dissemination from the intestine. ADHESINS AND THE ROLE OF ADHERENCE Adhesins are surface-exposed molecules that facilitate a pathogen s attachment to host cell receptor

41 22 molecules. In vitro adherence assays have been used extensively to characterize the interactions of C. jejuni with host cells and to attempt to identify the bacterial proteins that mediate binding. Although C. jejuni are capable of binding to cells of human (INT 407, HEp-2, HeLa, and 293) and non-human origin (Vero, CHO-K1, and MDCK) with equal efficiency, the binding of C. jejuni to INT 407 (Henle, a human intestinal epithelial cell line) and Caco-2 (a human colonic cell line) cells has been most extensively studied as these cells are thought to be more reflective of those that C. jejuni encounters in vivo. To date, the molecules proposed to act as adhesins have been found to be synthesized constitutively by C. jejuni. This fact is consistent with early studies, in which metabolically inactive (heat-killed and sodium azide-killed) C. jejuni were found to bind to cultured cells at levels equivalent to metabolically active organisms (Konkel and Cieplak, 1992). In addition, treatment of C. jejuni with chloramphenicol, a specific inhibitor of bacterial protein synthesis, has no effect on adherence (Konkel and Cieplak, 1992). Prior to the identification of C. jejuni adhesins, the importance of C. jejuni binding to host target cells was questionable. Lee et al. (Lee et al., 1986) observed C. jejuni specifically associated with the intestinal mucus-blanket and mucus-filled crypts of BALB/c mice. This association involved highly motile organisms with no apparent adhesion to epithelial cells of the gut mucosa. However, mucus association was only studied over the course of several days due to experimental difficulties in maintaining depletion of normal surface-associated bacteria. In addition, the relevance of the model is debatable because C. jejuni-infected mice do not develop disease. The investigators hypothesized that the lack of pathology in the mouse model was a result of the host cells lacking the appropriate receptors for bacterial products. The interaction of C. jejuni with mucin was later investigated by Szymanski et al. (Szymanski et al., 1995) using non-polarized Caco-2 cells and carboxymethylcellulose. The addition of carboxymethylcellulose to the cells was used to mimic an in vivo mucus layer. The investigators observed increases in the binding and entry of C. jejuni to the carboxymethylcellulose-treated Caco-2 cells. While it is unclear whether the increase in binding

42 23 was a result of C. jejuni specifically binding to host cell receptors or to components of the carboxymethylcellulose, an increase in C. jejuni entry into the Caco-2 cells was also noted. The investigators also found that the increased viscosity imparted by the carboxymethylcellulose resulted in longer runs by C. jejuni. The longer runs were proposed to result in an increase in the frequency of contacts between C. jejuni and host cells, thus leading to increases in host cell adherence and invasion. Based on these observations, the association of C. jejuni with the mucus in the crypts was proposed to be essential for cell invasion. De Melo and Pechère (De Melo and Pechère, 1990) identified four outer membrane proteins (omps) with apparent molecular masses of 28, 32, 36 and 42 kda that may play a role in mediating C. jejuni binding to host cells using a ligand-binding assay. By screening a C. jejuni genomic - λgt11 library with a hyperimmune antibody raised against the 28 kda protein, Pei and Blaser (Pei and Blaser, 1993) cloned a gene encoding a protein with a calculated molecular mass of 28,181 Da. More recent evidence suggests that the 28 kda protein, termed PEB1, mediates the binding of C. jejuni to epithelial cells (Pei et al., 1998). PEB1 is homologous with membrane proteins from other Gram-negative bacteria that function in amino acid transport. We have cloned and partially characterized a 37 kda omp from C. jejuni, termed CadF, that mediates the binding of C. jejuni to fibronectin (Fn) (Konkel et al., 1997). CadF is conserved among all C. jejuni and C. coli isolates tested to date (Konkel et al., 1999a). Whether the 36 kda protein identified by De Melo and Pechère (De Melo and Pechère, 1990) is the CadF protein is not known. The role of the C. jejuni 32 and 42 kda omps in adherence remains to be elucidated. Jin et al. (Jin et al., 2001) identified a 42.3 kda lipoprotein (JlpA = jejuni lipoprotein A) that mediates the binding of C. jejuni to HEp-2 cells. A mutation in the jlpa gene resulted in an 18 to 19.4% reduction in adherence when compared to the C. jejuni wild-type isolate, but had no effect on C. jejuni invasion. In addition, pretreatment of Hep-2 cells with recombinant JlpA reduced the binding of C. jejuni to the cells in a dose-dependent fashion.

43 24 Other molecules proposed to act as adhesins include the flagellum, lipopolysaccharide (McSweegan and Walker, 1986; Moser et al., 1992), the major outer membrane protein (MOMP, also called OmpE) (Moser et al., 1997; Schröder and Moser, 1997), and P95 (Kelle et al., 1998). These molecules are listed as putative adhesins because their adhesive properties are less well characterized. While McSweegan and Walker (McSweegan and Walker, 1986) and Moser et al. (Moser et al., 1992) both reported that purified flagellin is capable of binding to host cells and INT 407 cell membrane fractions, Wassenaar et al. (Wassenaar et al., 1991) found that the addition of purified flagellin did not competitively inhibit the binding of C. jejuni to cultured cells. Thus, the role of the flagellum as an adhesin remains ill-defined. Noteworthy is that the examination of C. jejuni-infected INT 407 cells by scanning electron microscopy has shown the flagella in contact with host cells (Konkel et al., 1992a). In 1986, McSweegan and Walker (McSweegan and Walker, 1986) proposed that the binding of C. jejuni to INT 407 cells was mediated by LPS. This proposal was based on observations that radioactively labeled LPS bound to INT 407 cells and that pretreatment of INT 407 cells with LPS concentrations of 250 µg per well and greater reduced the binding of C. jejuni to cultured cells. Alterations in LPS have been shown to affect binding, and in certain instances internalization, of a number of enteric bacteria including E. coli (Bradley et al., 1991), S. typhi (Mroczenski-Wildey et al., 1989), and N. gonorrhoeae (Schwan et al., 1995) to host cells. These investigators hypothesized that LPS structural changes could affect the pathogen s binding potential by altering the organism s surface charge, masking the specific adhesins, and changing the integrity of the outer membrane. Because the extraction and labeling protocol used by McSweegan and Walker (McSweegan and Walker, 1986) is likely to have resulted in the use of material that is rich in both LPS and capsular polysaccharide, their data with respect to the role of C. jejuni LPS as an adhesive molecule is difficult to interpret. In 1997, Schroder et al. (Schröder and Moser, 1997) proposed that the MOMP of C. jejuni serves as an adhesin because it was found to bind to INT 407 cell membranes. In their study, the MOMP was prepared from crude outer membrane preparations using sarcosyl extraction and further purified using SDS-polyacrylamide

44 25 gel electrophoresis (PAGE) or native gel electrophoresis. In contrast to the MOMP that was purified by SDS-PAGE, the MOMP purified by native gel electrophoresis was found to bind to INT 407 membranes as determined by enzyme-linked immunosorbant assays. The authors concluded that in addition to its role as a porin, the MOMP may also serve as an adhesin. However, the specificity of the MOMP binding to the INT 407 cell membranes was not determined. The P95 protein was identified by screening C. jejuni clinical isolates with a degenerative oligonucleotide probe by Southern hybridization analysis (Kelle et al., 1998). The sequence of the probe was based on the nucleotide sequences of adhesins identified in other Gram-negative bacteria (Kelle et al., 1998). A hybridizing band was detected in six of thirteen C. jejuni clinical isolates as judged by Southern blot analysis. After cloning and sequencing the hybridizing C. jejuni chromosomal DNA fragment, an ORF was subsequently identified that was capable of encoding a polypeptide 869 amino acids with an M r of 95 kda. The investigators reported that the deduced amino acid sequence of the C. jejuni P95 protein shared similarity with adhesins found in other Gram-negative organisms including Bordetella pertussis and Haemophilus influenzae. Not determined was whether the P95 gene was expressed in the C. jejuni isolates examined and the phenotype of a C. jejuni P95 mutant. The role of each of the C. jejuni molecules discussed above in adherence requires additional study. Adhesive pili or fimbriae have been identified in enteric pathogens including E. coli, Salmonella and Vibrio cholerae (Hultgren et al., 1991; Hultgren et al., 1996; Low et al., 1996). In 1996, Doig et al. (Doig et al., 1996) reported that both C. jejuni and C. coli produced environmentally regulated peritrichous pilus-like appendages. The investigators also reported that a mutation in a gene termed pspa (pilus-synthesis protease) resulted in an isolate that was incapable of synthesizing pili when cultured in the presence of the bile salt deoxycholate as evidenced by examination of the bacteria by transmission electron microscopy. While in vitro assays demonstrated that pili played no role in promoting the organism s adherence or invasion of epithelial cells, in vivo studies revealed that the

45 26 C. jejuni pspa mutant exhibited reduced pathology in ferrets when compared to animals infected with the C. jejuni wild-type isolate. Subsequent work by Gaynor et al. (Gaynor et al., 2001) found that the pilus-like appendages in these Campylobacter isolates were bacteria-independent artifacts induced by the culture conditions. Whether Campylobacter organisms produce fimbriae that assist in colonization remains uncertain. Also not known is the role of the pspa gene from C. jejuni. Though several adhesins have been characterized, little is known regarding the host cell surface receptors to which the C. jejuni adhesins bind. CadF, a 37kDa outer membrane protein, binds to Fn. Fn is a component of the extracellular matrix (ECM). The host cell receptor molecules that JlpA and PEB1 bind remain to be elucidated. In addition to broadening of host range, the advantage of possessing multiple adhesins is that these molecules could act individually or in concert or at different stages of the infection. Certain adhesins might be involved initially in promoting the adherence of C. jejuni to the apical surface of intestinal epithelial cells or M cells, while other adhesins could be involved in promoting the organism s binding to molecules or receptors found on basolateral cell surfaces following translocation. Additional characterization of these interactions may provide a basis to help determine the sequential steps in Campylobacter pathogenesis. In summary, several reports suggest that the adhesins synthesized by C. jejuni are important in colonization. A correlation has been observed between the clinical symptoms of C. jejuni-infected individuals and the degree to which C. jejuni isolates adhered to cultured cells. Fauchere et al. (Fauchere et al., 1986) found that C. jejuni strains isolated from patients with fever and diarrhea adhered to cultured cells at a greater efficiency than those strains isolated from asymptomatic individuals. The role of the adhesins in enabling successful colonization is supported by the experimental inoculation of animals with C. jejuni mutants. For example, the C. jejuni peb1a null mutant exhibits a reduction in the duration of mouse intestinal colonization when compared to the

46 27 C. jejuni wild-type isolate (Pei et al., 1998). In addition, a C. jejuni cadf mutant lacks the ability to colonize the cecum of newly hatched leghorn chickens (Ziprin et al., 1999). However, studies are lacking to demonstrate that the colonization-impaired phenotype displayed by the peb1 and cadf mutants is due to the specific mutation introduced. In the case of the CadF protein, we have not been able to construct a Campylobacter shuttle vector harboring the cadf gene because expression of the gene from its endogenous promoter appears toxic in a heterologous host such as E. coli. INVASION The ability of C. jejuni to enter, survive, and replicate in mammalian cells has been studied extensively using tissue culture models (Newell et al., 1985; De Melo et al., 1989; Konkel and Joens, 1989; Konkel et al., 1990; Wassenaar et al., 1991; Everest et al., 1992; Konkel et al., 1992a; Konkel et al., 1992b; Grant et al., 1993; Oelschlaeger et al., 1993; Russell and Blake, 1994; Yao et al., 1994; Doig et al., 1996; Pei et al., 1998; Konkel et al., 1999b; Rivera-Amill et al., 2001). Typically such model systems involve determination of the number of intracellular C. jejuni by assaying bacterial protection from an antibiotic, such as gentamicin, that does not penetrate eukaryotic cell membranes (Hale et al., 1979). The relative ability of C. jejuni to invade cultured cells appears to be strain-dependent (Newell et al., 1985; Konkel and Joens, 1989; Everest et al., 1992). Newell et al. (Newell et al., 1985) found that environmental isolates were much less invasive for HeLa cells than clinical isolates as determined by immunofluorescence and electron microscopy examination of C. jejuni-infected cells. Everest et al. (Everest et al., 1992) observed a statistically significant difference in the level of invasion between C. jejuni strains isolated from individuals with colitis versus those isolated from individuals with noninflammatory diarrhea. Also of interest is that the ability of C. jejuni to invade cells has been noted to decrease after extensive in vitro passage (Konkel et al., 1990). The percent of the inoculum internalized for C. jejuni , a strain isolated from a milk-borne outbreak of diarrheal illness, has been reported to range between 0.8 to 1.8% (Yao et al., 1994; Doig et al., 1996; Yao et al., 1997). Biswas et al. (Biswas et al., 2000) and

47 28 Hu and Kopecko (Hu and Kopecko, 1999) found that C. jejuni invasion is optimal when mammalian cells are inoculated at low MOIs. However, Biswas et al. (Biswas et al., 2000) also noted that the maximal number of internalized bacteria occurs at higher MOIs. Once internalized, C. jejuni organisms can survive for extended periods of time within epithelial cells and ultimately induce a cytotoxic response (Konkel et al., 1992b). Parasite-directed endocytosis is a process in which a microorganism synthesizes the proteins required to promote their internalization by host cells. In addition to this basic requirement, investigators also discovered that the maximal uptake into cells occurs with metabolically active Haemophilus influenzae (St. Geme III and Falkow, 1990), Neisseria gonorrhoeae (Richardson and Sadoff, 1988), Rickettsia prowazekii (Walker and Winkler, 1978), Salmonella typhimurium (Finlay et al., 1989; Lee and Falkow, 1990), and Shigella flexneri (Hale and Bonventre, 1979; Headley and Payne, 1990). In the early 1990s, the internalization of C. jejuni was found to be significantly reduced in the presence of chloramphenicol, a specific inhibitor of bacterial protein synthesis (Konkel and Cieplak, 1992). This finding, coupled with the fact that metabolically inactive (heat-killed and sodium azide-killed) C. jejuni are not internalized, suggested that C. jejuni synthesize entry-promoting proteins (Konkel and Cieplak, 1992). One and two-dimensional electrophoretic analyses of metabolically labeled C. jejuni cultured in the presence and absence of epithelial cells revealed that a number of proteins were synthesized exclusively, or preferentially, in the presence of epithelial cells while others were selectively repressed (Konkel and Cieplak, 1992; Konkel et al., 1993). In support of these findings, Panigrahi et al. (Panigrahi et al., 1992) demonstrated that C. jejuni synthesized a number of proteins during growth in rabbit ileal loops that were not synthesized under standard laboratory conditions. Two of the newly synthesized proteins, with apparent molecular masses of 84 and 47 kda, were detectable using convalescent sera from C. jejuni-infected individuals. Additional work revealed that the de novo proteins synthesized by C. jejuni upon co-cultivation with INT 407 cells were unique from those proteins induced by thermal

48 stress of C. jejuni (Konkel et al., 1998). These findings suggest a coordinated response, whereby C. jejuni expresses certain genes after encountering the epithelial cell microenvironment. 29 The challenge of Macaca mulatta primates with C. jejuni has provided the most convincing experimental evidence that the primary mechanism of colonic damage and diarrheal disease is related to the organism s ability to invade colonic epithelial cells (Russell et al., 1993). Examination of colonic biopsy specimens from C. jejuni-infected primates revealed organisms, in association with dense concentrations of microfilaments, penetrating epithelial cells. C. jejuni were also observed within membrane bound vacuoles and within the cytoplasm of damaged cells. The investigators concluded that early mucosal damage, occurring prior to any inflammatory response, resulted from C. jejuni penetrating the colonic epithelial cells. Infection of a number of animals (i.e.: newborn piglets (Babakhani et al., 1993), chicken embryos (Field et al., 1986b), newly-hatched chicks (Welkos, 1984), hamsters (Humphrey et al., 1985), and gamma-irradiated mice (Sosula et al., 1988)) has supported the hypothesis that the ability of C. jejuni to cause illness is related to its ability to invade the epithelial cells lining the intestinal tract. SECRETION Campylobacter jejuni secretes a set of proteins termed the Campylobacter invasion antigens (Cia proteins). The M r of the Cia proteins range from 12.8 to 108 kda (Konkel et al., 1999b). In the laboratory, the synthesis of the Cia proteins can be induced by culturing C. jejuni on medium supplemented with bile salts whereas both Cia protein synthesis and secretion are induced by culturing C. jejuni with eukaryotic cells or in serum-supplemented medium (Rivera-Amill et al., 2001). Thus, it appears that C. jejuni Cia protein synthesis and secretion are induced in the laboratory when the organism is cultured using conditions that mimic the in vivo environment. To date, only one secreted protein termed CiaB has been identified (Konkel et al., 1999b). The C.

49 30 jejuni ciab gene encodes a protein of 610 aa with a calculated molecular mass of 73,154 Da. While confocal microscopic examination of C. jejuni-infected cells suggests that CiaB is translocated into the cytoplasm of the host cells, the specific function of CiaB is not known. Perplexing is that C. jejuni ciab null mutants are deficient in the secretion of the total pool of Cia proteins. One possible explanation for this observation is that the synthesis of a 3 -truncated version of CiaB may obstruct the secretory apparatus. Data suggest that the Cia proteins promote C. jejuni uptake by host cells. C. jejuni ciab null mutants bind to cultured epithelial cells at levels equal to or slightly greater than C. jejuni wild-type isolates, but exhibit a reduction in INT 407 cell invasion when compared to the wild-type isolates. In addition, preculturing C. jejuni wild-type isolates on plates supplemented with the bile salt deoxycholate retards the inhibitory effect of chloramphenicol on C. jejuni invasion as judged by the gentamicin-protection assay (Rivera-Amill et al., 2001). This finding supports the notion that the Cia proteins promote the organism s uptake because the synthesis of the Cia proteins is induced by deoxycholate prior to incubating the bacteria in chloramphenicol containing media. Infection of newborne piglets with C. jejuni Cia secretion-competent and secretion-deficient isolates has revealed that the secreted Cia proteins contribute to the pathology of C. jejuni-mediated enteritis. Piglets infected with the C. jejuni wild-type and complemented ciab isolate developed diarrhea 24 h post-infection, whereas diarrhea was not observed in piglets infected with the C. jejuni ciab mutant until 3 days post-infection. More severe histological lesions were also observed in piglets infected with the C. jejuni complemented ciab isolate when compared to the C. jejuni ciab mutant. Additional studies are necessary to determine the roles of the Cia proteins. While some of the Cia proteins may serve as functional components of the secretory apparatus, others likely directly interact with host cell molecules. Preliminary data has been generated in our laboratory suggesting that the Cia proteins are secreted via the flagellar type III secretion apparatus. A precedent for protein secretion via the flagellar apparatus does exist in that Yersinia secrete proteins termed Fops for flagellar outer proteins from

50 31 this apparatus (Young et al., 1999). While recent data indicates that the Fops contribute to the pathogenesis of Yersinia (Schmiel et al., 1998), the precise role of these proteins in infection has yet to be defined. If the Cia proteins are indeed secreted via the flagellar apparatus in C. jejuni, it would indicate that motility and virulence are linked in a novel fashion in this organism as Yersinia are nonmotile when cultured at 37 C. CYTOLETHAL DISTENDING TOXIN A number of Campylobacter strains, including C. jejuni, C. coli, C. lari, C. fetus, and C. upsaliensis, produce cytolethal distending toxin (CDT) (Johnson and Lior, 1988; Mooney et al., 2001). The production CDT by Campylobacter isolates was first reported by Johnson and Lior in 1988 (Johnson and Lior, 1988). The investigators reported that 41% of the 718 isolates examined produced CDT. Typically, the production of CDT by Campylobacter isolates is assessed by the addition of serially-diluted bacterial whole cell lysates (sonicates) to actively proliferating cells. In susceptible cells, toxin activity is evident from cell distension characterized by both elongation and swelling to nearly 5 times its normal size. Enlargement of nuclei is also common in the distended cells. Ultimately, CDT-treated cells die or disintegrate. Cell lines found to be susceptible to CDT include CHO, Vero, HeLa, HEp-2, Caco-2, COS-1, REF52, and INT 407 cells (Johnson and Lior, 1988; Whitehouse et al., 1998; Pickett and Whitehouse, 1999; Lara-Tejero and Galán, 2000). The sensitivity of different cell lines to CDT is variable, which may be due to differences in their surface receptors (Pickett and Whitehouse, 1999). Pickett et al. (Pickett et al., 1996) was the first to clone the cdt toxin genes. The CDT toxin was found to be encoded by three adjacent genes termed cdta, cdtb, and cdtc (Pickett et al., 1996). The cdta, cdtb, and cdtc genes encode proteins of approximately 30, 29 and 21 kda, respectively. Not known is whether the expression of the toxin genes is environmentally regulated. Moreover, no differences have been detected in CDT production by C. jejuni in response to modified

51 32 environmental conditions including alterations in iron, growth phase, and growth temperature (Pickett, 2000). With respect to its biological properties, CDT is both heat-labile and trypsinsensitive (Johnson and Lior, 1988). All three components of the toxin, which are associated with the bacterial outer membrane, are required for toxin delivery and activity (Hickey et al., 2000; Lara- Tejero and Galán, 2001). Microinjection studies have revealed that CdtB is the active subunit of the toxin (Lara-Tejero and Galán, 2000). The structure the mature holotoxin is not known, and attempts to purify the holotoxin from membrane fractions has proved inherently difficult due to association of the toxin with outer membrane components (Pickett, 2000). However, Lara-Tejero and Galán (Lara-Tejero and Galán, 2001) recently reported that the purified toxin components can be reconstituted by mixing the recombinant CdtA, CdtB, and CdtC proteins. The reconstituted toxin exhibits biological activity as evidenced by cytoplasmic distension and cell cycle arrest (see below). The investigators concluded that CdtA and CdtC most likely comprise the heterodimeric B subunit of the toxin, and are required for CdtB delivery into a cell. Based on their work, Lara-Tejero and Galán (Lara-Tejero and Galán, 2001) propose that CDT is an AB 2 heterodimeric toxin. CDT causes progressive cell distention by causing cells to irreversibly arrest in the G 2 /M transition phase of the cell cycle (Whitehouse et al., 1998; Lara-Tejero and Galán, 2000). CDT prevents dephosphorylation of CDC2. CDC2 is the catalytic subunit of the cyclin-dependent kinase and must be activated (dephosphorylated) for cells to enter mitosis. Thus, CDT prevents CDC2 dephosphorylation, which in turn causes cells to arrest in the G 2 phase. How this occurs is not yet understood. CDT may cause the G 2 block by directing the cell into a DNA damage/incomplete replication checkpoint pathway (Pickett and Whitehouse, 1999). Lara-Tejero and Galán (Lara- Tejero and Galán, 2000) reported that the deduced amino acid sequence of CdtB shares similarity with members of type I deoxyribonuclease protein family and speculated that DNA damage could

52 33 occur during the S phase of the cell cycle. In a recent report, Mooney et al. (Mooney et al., 2001) found that extracts prepared from CDT-producing isolates of C. upsaliensis induced cells to undergo programmed cell death (apoptosis) as judged by TUNEL and flow cytometric analyses. However, it is not yet known whether CDT from C. upsaliensis, or C. jejuni, is in itself capable of inducing apoptosis as purified toxin or cdt mutants were not included in their study (Mooney et al., 2001). The effects of CDT on cultured cells are profound, but little is known regarding the functional role of the toxin in bacterial pathogenesis. To begin to address the contribution of CDT in C. jejuni pathogenesis, Purdy et al. (Purdy et al., 2000) intragastrically challenged severe combined immunodeficient (SCID) mice with 109 cfu of a C. jejuni wild-type isolate and an isogenic cdtb mutant. Blood, liver and spleen samples were acquired at 2, 6, and 24 h post-challenge to assay for the presence of invasive Campylobacter organisms. A total of 30 mice were infected and 5 mice per C. jejuni isolate sacrificed at each timepoint. Wild-type bacteria were readily present in 8 of 15 samples (3 spleens, 4 livers, and 1 blood) at 2 h. However, the C. jejuni cdtb mutant was recovered in only 4 of 15 samples (1 spleen and 3 livers). Later timepoints showed approximately identical results for both the C. jejuni wild-type and cdtb mutant. Colonization levels were also monitored over the course of the assay and found to be identical between the two isolates. Based on these data, the authors suggested a possible role for CDT in invasion (Purdy et al., 2000). The investigators also noted that the sonicates of a cdtb mutant still exhibited some cytopathic effects on HeLa cells and suggested that there may be a second active toxin. Nevertheless, CDT appeared to be the principal toxin that is active in C. jejuni sonicates. Future work involving the characterization of CDT and its role in bacterial invasion or alternative functions in Campylobacter pathogenesis will surely prove interesting.

53 34 LIPOPOLYSACCHARIDE AND CAPSULAR POLYSACCHARIDE Lipopolysaccharide (LPS) is a major component of the outer membrane in Gram-negative bacteria. LPS has three distinct structural components: lipid A, which serves as the membrane anchor; a core composed of heterogeneous glycoses; and the somatic O antigen (O Ag) composed of a repeating unit of one or more glycosyl residues attached covalently to the core. LPS molecules without O- side chains are referred to as lipooligosaccharides. Early reports indicated that the LPS of C. jejuni is similar to that of Haemophilus and Neisseria spp. More specifically, C. jejuni LPS was characterized as being of low molecular weight (M r ) and lacking detectable amounts of O polysaccharide chains (Logan and Trust, 1984). Interestingly, the investigators remarked that it was unusual for the low M r LPS of C. jejuni to confer such an extensive number of serotypes as identified by the Penner serotyping scheme. The Penner serotyping scheme, based on the presence of soluble heat-stable (HS) antigens that were presumed to be LPS in nature, was introduced in 1980 (Penner and Hennessy, 1980). In 1987, Preston and Penner (Preston and Penner, 1987) reported that approximately one-third of C. jejuni isolates produced a high M r LPS, characteristic of LPS molecules with O side chains, that could be observed by immunoblotting with serotyping antisera. An impressive amount of work has been done to determine the structures of LPS molecules from various C. jejuni isolates. Three different lipid A backbones have been identified in serostrain HS:2. Approximately 73% of the LPS molecules in this strain have a disaccharide backbone composed of diaminoglucose and D-glucosamine, 15% contain a backbone with two diaminoglucose residues, and the remaining 12% contain a backbone consisting of two D-glucosamines (Moran et al., 1991; Moran and Penner, 1999). All three backbones are acylated and phosphorylated in a similar manner (Moran et al., 1991; Moran, 1997; Moran and Penner, 1999). Identical backbones are present in other C. jejuni serostrains, but the molar ratios of the disaccharide units varies (Moran, 1997). The

54 35 C. jejuni core oligosaccharide contains two distinct regions. The inner core is invariably comprised of a trisaccharide of 2-keto-3-deoxy-octulosonic acid (Kdo) and two heptoses (Aspinall et al., 1993). A second common feature is that the heptose adjacent to Kdo is substituted by D-glucose (beta 1-4 linkage) (Moran and Penner, 1999). Variation has been observed in the core region of the C. jejuni HS:1 and HS:2 reference strains, where the second heptose is substituted by glucose. Another variation observed among strains is that the heptose adjacent to Kdo is substituted at position 6 with either a phosphate or a phosphoethanolamine. In contrast to the conserved nature of the inner core, the outer core is more variable consisting of two or three hexoses that are substituted laterally or terminally with sialic acid or quinovosamine residues. Structural variations within this region were proposed to provide the diversity seen among serostrains containing only low M r components (Aspinall et al., 1992; Aspinall et al., 1993; Moran and Penner, 1999). Karlyshev et al. (Karlyshev et al., 2000) recently reported that all C. jejuni isolates, including those isolates previously thought to contain only low M r polysaccharide, produce high M r polysaccharide. As previously proposed by Chart et al. (Chart et al., 1996), this high M r component was found to be biochemically similar to group II capsular polysaccharide and not O-antigen (Karlyshev et al., 2000). The investigators identified kps-like genes showing significant sequence similarity and overall organization to capsular polysaccharide genes of E. coli. Mutagenesis of the kps-like genes in C. jejuni resulted in production loss of capsular material and attributed serotypic determinants. The high M r LPS material was also demonstrated by immunoblotting to be susceptible to phospholipase treatment presumably due to cleavage of a phospholipid moiety, instead of lipid A, from the polysaccharide. Removal of the lipid moiety hinders the polysaccharide from migrating in SDS-polyacrylamide gels (Tsai and Frasch, 1982). The investigators concluded that the basis of Penner serotyping is the capsular material and not the LOS or O-antigen. Bacon et al. (Bacon et al., 2001) reached similar conclusions, but noted the presence of a possible second glycan structure that

55 36 could be visualized in a kpsm mutant. These data are intriguing and provide the foundation for future studies to investigate the importance of capsular polysaccharide in the development of GBS and in evasion of the pathogen from the host immune response. Of additional interest will be dissecting the roles of C. jejuni LPS and the capsular polysaccharide in the pathogenesis of C. jejuni. IRON ACQUISITION In the host, free iron is complexed with transferrin and lactoferrin at binding constants of approximately 1020 (Ratledge and Dover, 2000), making it a limiting nutrient for bacterial growth. Therefore, most pathogenic bacteria have developed mechanisms to scavenge iron that enables them to successfully colonize and survive within a host. Siderophore mediated iron uptake (1022 to 1050 dissociation constant) is a common method used by bacterial pathogens to acquire iron in the extracellular environment (Drechsel and Winkelmann, 1997). Iron loaded siderophores are bound via specific outer membrane receptors and are transported to the cytosol. The synthesis of siderophores and corresponding uptake systems is often iron repressed. Field et al. (Field et al., 1986a) found that only 7 of 26 C. jejuni isolates produced siderophores when grown in iron depleted media. However, each of three C. jejuni isolates tested were also found to utilize exogenously supplied enterochelin and ferrichrome siderophores to satisfy their iron requirement (Field et al., 1986a). Genes encoding both the enterochelin (ceubcde) and ferrichome (fhuabd) uptake systems have been characterized in Campylobacter isolates (Richardson and Park, 1995; Galindo et al., 2001). The cfra gene of C. jejuni encodes a protein that shares similarity with the siderophore receptor BfrA from Bordetella bronchiseptica (Guerry et al., 1997). The siderophore that binds to the CfrA putative receptor is not known. The cfra gene was identified in 27 of 33 isolates as judged by Southern hybridization analysis (Guerry et al., 1997). A second iron uptake system, the fhuabd

56 37 operon, was recently found within a C. jejuni putative pathogenicity island. A C. jejuni fhua putative mutant failed to grow in iron depleted media, was readily killed following internalization in cultured epithelial cells, and elicited increased sensitivity to peroxide killing when compared to the wild-type isolate (L.A. Joens, unpublished). Interestingly, fhuabd was found in only 6 of 11 isolates tested (Galindo et al., 2001). Hence, cfra and fhuabd are not uniformly present in C. jejuni isolates and it is not known whether C. jejuni contain only one or both of these systems. Noteworthy is the absence of an outer membrane receptor in the ceubcde operon and a cytoplasmic ATPase in the fhuabd operon. It is possible that individual components of various iron-uptake systems may complement each other. Analysis of the C. jejuni NCTC genome has revealed the presence of additional iron acquisition systems including a hemin uptake operon consisting of the four proteins ChuABCD (Van Vliet et al., 1998). The NCTC genome also contains genes that encode for: a putative siderophore receptor; a periplasmic binding protein dependent system (Parkhill et al., 2000); and a ferrous uptake protein that shares similarity with the FeoB from E. coli (Van Vliet and Ketley, 2001). The importance of the FeoB protein in the pathogenesis of C. jejuni awaits investigation, however, the synthesis of FeoB affects the ability of H. pylori to colonize the stomach of mice (Velayudhan et al., 2000). Transport of siderophores and certain host-iron sequestering molecules across the outer membrane requires the activity of the TonB system (Ratledge and Dover, 2000). The system is encoded by the tonb-exbb-exbd genes. TonB couples the proton motive force to actively drive the movement of molecules across the outer membrane and into the periplasm (Larsen et al., 1999). Analysis of the C. jejuni genome has revealed three sets of genes whose deduced amino acid sequences share similarity with TonB, ExbB, and ExbD. Two sets of these genes are located in the proximity of a putative siderophore receptor and cfra (Parkhill et al., 2000). While most Enterobacteriaceae,

57 38 including E. coli, only possess one copy of the TonB system, Vibrio cholerae encodes two complete systems (Seliger et al., 2001). Studies are needed to determine whether each of the three sets of genes identified in the C. jejuni genome encode for a functional system, and whether there are particular environmental conditions under which each system operates. Of additional interest will be to determine whether any of the three putative systems displays specificity for a specific siderophore/heme. Fur is a repressor of iron regulated genes which uses ferrous ion (Fe2+) as a cofactor. In a C. jejuni fur mutant, the transcription of various iron uptake genes including cfra became derepressed under high iron conditions (Van Vliet et al., 1998). However, oxidative stress response genes such as ahpc and kata maintained iron regulation. Another Fur homolog, PerR, was shown to repress the activity of oxidative stress genes in the presence of iron (Van Vliet et al., 1999). PerR had no effect on the transcription of other iron-regulated genes. Hence, C. jejuni contains two fur homologs that repress the activity of separate iron-regulated genes. Ferritin, which stores iron intracellularly and prevents oxidative damage, has been purified from C. jejuni (Wai et al., 1995). Ferritin, encoded by cft, prevents oxidative damage by lowering the intracellular concentration of iron which may react to form various oxygen radicals (Wai et al., 1996). A C. jejuni cft mutant not only grows poorly in iron depleted media but is also more sensitive to peroxide killing than a wild-type isolate (Wai et al., 1996). Thus, the presence of the iron storage protein ferritin not only protects against oxidative stress but might also allow the organism to weather a varied range of iron concentrations. In summary, C. jejuni has a number of iron-uptake systems for both ferric (Fe3+) and ferrous ions (Fe2+). The role of these systems in C. jejuni pathogenesis is not yet clear, however, the redundancy

58 39 in function suggests that none will be singularly required for virulence. INTRACELLULAR SURVIVAL IN MONONUCLEAR PHAGOCYTES In the piglet model, C. jejuni penetrates and proliferates within the intestinal epithelium of the ileum and colon. Cell damage occurs with degeneration of the superficial epithelium leading to the shortening of the villi and the production of an exudate in the lumen of the intestine. In some cases, as in human infections (Blaser et al., 1980), there is deeper tissue involvement resulting in hemorrhagic necrosis in the lamina propria, the formation of crypt abscesses and the influx of inflammatory cell exudate. It is during this phase of the disease that C. jejuni encounters phagocytic cells of the lamina propria (Duffy et al., 1980; Blaser et al., 1983) and the blood. Subsequently, the ability of C. jejuni to survive phagocytosis could exacerbate the disease by enabling the organism to be disseminated in a host. Although C. jejuni is readily taken up by monocytes and macrophages in vitro (Kiehlbauch et al., 1985; Myszewski and Stern, 1991; Wassenaar et al., 1997; Day et al., 2000), contradictory data have been generated regarding whether this pathogen is capable of surviving in these cells. Myszewski and Stern (Myszewski and Stern, 1991) examined the ability of both a C. jejuni high passage clinical isolate and a C. jejuni chicken isolate to resist killing by macrophages. They found that both C. jejuni isolates were killed by peritoneal macrophages, which were harvested from chickens, within a 6 hr incubation period. A greater number of C. jejuni organisms were phagocytized when serum reactive against the corresponding isolate was included in the assay, however, the bacteria were killed within the 6 hr incubation period regardless of whether the antiserum was included or omitted from the assay. Similar findings were described by Wassenaar et al. (Wassenaar et al., 1997) in examining the survival of 16 isolates of C. jejuni internalized by activated human peripheral monocytes. Survival assays conducted for 72 h demonstrated the killing of C. jejuni by the majority of the donors monocytes within 24 to 48 h. However, approximately 10% of the monocytes

59 40 demonstrated normal uptake of C. jejuni but failed to kill the bacterium. The authors concluded that C. jejuni infected individuals are prone to develop a bacteremia if their monocytes fail to kill the organism. Contrary to the above work, other investigators have demonstrated intracellular survival of C. jejuni for at least 72 h after internalization by mononuclear phagocytes (Kiehlbauch et al., 1985; Day et al., 2000). Kiehlbauch et al. (Kiehlbauch et al., 1985) examined the survival of a C. jejuni clinical isolate using the J774G8 BALB/c mouse macrophage cell line, BALB/c macrophages, and human monocytes using acridine orange as a vital stain. The researchers were able to recover C. jejuni from all three cell types over a 6 d period. Day et al. (Day et al., 2000) reported similar findings in that a clinical isolate of C. jejuni (M129) was able to survive following phagocytosis by porcine peritoneal macrophages, murine peritoneal macrophages and J774A.1 cells. However, there was a noticeable reduction in the number of C. jejuni recovered from the three phagocytic cell types at 72 h post-inoculation; the greatest survival of C. jejuni was noted with the porcine peritoneal macrophages. It is our belief that the differences noted between laboratories with respect to the ability of C. jejuni to survive within phagocytes reflects the use of phagocyte cells of different origins and various bacterial isolates. ROLE OF OXIDATIVE RADICALS IN PHAGOLYSOSOME SURVIVAL Although intracellular existence provides bacteria an unoccupied niche and shelter from immune surveillance, internalized bacteria must be able to survive a variety of reactive oxygen species, especially in the phagolysosome of professional phagocytes. Bacterial factors such as superoxide dismutase and catalase, which inactivate these products, allow invasive bacteria to persist in host cells and tissues. Experiments have been conducted to examine the effect of oxygen radicals on the survival of Salmonella typhimurium (DeGroote et al., 1997). It was found that a sodc mutant of S. typhimurium was more susceptible to killing by superoxide and nitric oxide than the wild-type isolate. Moreover, greater numbers of the S. typhimurium sodc mutant were recovered when the respiratory burst inhibitor acetovanillone or nitric oxide synthase inhibitor NG-L-monomethyl

60 41 arginine was added to the culture medium. To address the role of superoxide dismutase in C. jejuni survival in macrophages, assays were performed with a C. jejuni sodb mutant and the J774A.1 murine macrophage-like cell line. No difference was noted in the survival of the C. jejuni sodb mutant in J774A.1 cells when compared to the C. jejuni wild-type isolate (L.A. Joens, unpublished data). However, a C. jejuni kata mutant was not recovered from J774A.1 cells 24 h post-inoculation (Day et al., 2000); the kata gene from C. jejuni encodes the enzyme catalase. Also noteworthy is that the C. jejuni kata mutant was recovered when the respiratory burst or production of nitric oxide was inhibited. This finding demonstrates that C. jejuni possesses certain virulence attributes that enable it to survive intracellularly within mononuclear phagocytes. Additional work is required to determine whether C. jejuni is able to alter its intracellular trafficking and the precise role of macrophages in the development of campylobacteriosis. A MODEL OF C. jejuni PATHOGENESIS We conclude this review by presenting a diagram that represents our current perspective of C. jejunivirulence determinants and their potential role in the development of the histological lesions observed in C. jejuni-infected individuals. Our model is based on articles discussed above and examination of biopsy specimens from piglets infected with C. jejuni clinical isolates. We believe that the infection of piglets with C. jejuni represents a relevant model for C. jejuni-mediated enteritis as these animals are anatomically similar to humans. Moreover, C. jejuni-infected piglets develop the clinical symptoms (e.g., bloody diarrhea) and histopathological lesions (e.g., epithelial cell degeneration, exudation of fibrin and inflammatory cells in both the small intestine and colon) similar to that of C. jejuni-infected humans. C. jejuni is proposed to initially colonize the jejunum and ileum, and then the colon, of humans (Allos and Blaser, 1995; Skirrow and Blaser, 2000). However, the precise in vivo target site of C. jejuni is not known because autopsy and surgical material is rare. Motility and chemotaxis likely

61 42 play critical roles in disease. Upon passage into the small intestine and migration of bacteria toward the mucus-filled crypts, we propose that C. jejuni engage in an adaptive response to the intestinal microenvironment where they synthesize a novel set of proteins that promotes their subsequent interaction with host target cells. In vitro data supports the notion that C. jejuni has the ability to migrate across the enterocytes via a paracellular or transcellular route (Everest et al., 1992; Konkel et al., 1992c; Grant et al., 1993; Brás and Ketley, 1999; Harvey et al., 1999). However, not known is the significance of either route of translocation versus the role of M cells in the organism s ability to breach the intestinal barrier. We speculate that adherence plays an early role in the infectious process and that C. jejuni binds specifically to host cell receptors. If studies continue to support the necessity of adhesins in establishing disease, it will prove interesting to determine whether C. jejuni has a predilection for receptors on the apical or basolateral surfaces of the host cells. Following the intimate binding of C. jejuni to host cells, in vivo evidence indicates that C. jejuni is internalized by a host cell. Not clear is the contribution of invasion, versus Cia protein secretion, to the severity of Campylobacter-mediated enteritis. One of the hallmarks of infection in C. jejuni-inoculated piglets is villous atrophy (Babakhani et al., 1993). More specifically, C. jejuni appears to destroy the cells at the tips of the villi that are fully differentiated rather than the undifferentiated cells in the crypt. We propose that the necrosis of the villi is primarily caused by one or more bacterial toxins. We speculate that CDT contributes to villous atrophy by targeting the actively proliferating cells within the crypt. Thus, replacement of differentiated cells at the tips of the villi could be retarded by inhibiting crypt cell hyperplasia. How CDT is delivered to the cells within the crypt is not known. We would be remiss without mentioning that C. jejuni infection is accompanied by an intense inflammatory response that no doubt results from the heightened production of cellular cytokines. While this aspect of C. jejuni infection warrants a review article all to itself, space constraints have not permitted us to discuss this area. We hypothesize that the inflammatory response is responsible for intensifying the symptoms exhibited by C. jejuni-infected individuals, but is not responsible for causing the villous atrophy observed early in infection. We base this statement on the apparent lack

62 43 of significant numbers of neutrophils and other inflammatory cells at tissue damaged sites. However, detailed studies are required to more closely examine the presence of inflammatory cell infiltrates at sites of C. jejuni infection.

63 44 CONCLUDING COMMENTS A more accurate and comprehensive understanding of C. jejuni-mediated enteritis will emerge as researchers functionally characterize putative virulence genes and discover virulence attributes that are unique among particular C. jejuni isolates. There is little doubt that additional work will unveil that C. jejuni organisms have a repertoire of unique virulence strategies. Moreover, elucidation of C. jejuni virulence determinants and the stages at which they contribute in infection will yield new insights into the diverse mechanisms by which bacteria cause disease.

64 45 ACKNOWLEDGMENTS We thank Brian Raphael for assistance in preparation of this manuscript. Work in MEK s laboratory is supported by a grant from the NIH (DK58911) and USDA National Research Initiative Competitive Grants Program ( ). Work in LAJ s laboratory is supported by USDA-NRICGP grants ( and ).

65 46 REFERENCES Allos, B.M Campylobacter jejuni infections: Update on emerging issues and trends. Clin. Infect. Dis. 32: Allos, B.M. and Blaser, M.J Campylobacter jejuni and the expanding spectrum of related infections. Clin. Infect. Dis. 20: Allweiss, B., Dostal, J., Carey, K.E., Edwards, T.F. and Freter, R The role of chemotaxis in the ecology of bacterial pathogens of mucosal surfaces. Nature 266: Altekruse, S.F., Stern, N.J., Fields, P.I. and Swerdlow, D.L Campylobacter jejuni -- an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Aspinall, G.O., McDonald, A.G., Raju, T.S., Pang, H., Kurjanczyk, L.A., Penner, J.L. and Moran, A.P Chemical structure of the core region of Campylobacter jejuni serotype 0:2 lipopolysaccharide. Eur. J. Biochem. 213: Aspinall, G.O., McDonald, A.G., Raju, T.S., Pang, H., Mills, S.D., Kurjanczyk, L.A. and Penner, J.L Serological diversity and chemical structures of Campylobacter jejuni low-molecular weight lipopolysaccharides. J. Bacteriol. 174: Babakhani, F.K., Bradley, G.A. and Joens, L.A Newborn piglet model for campylobacteriosis. Infect. Immun. 61: Bacon, D.J., Alm, R.A., Burr, D.H., Hu, L., Kopecko, D.J., Ewing, C.P., Trust, T.J. and Guerry, P Involvement of a plasmid in virulence of Campylobacter jejuni Infect. Immun. 68: Bacon, D.J., Szymanski, C.M., Burr, D.H., Silver, R.P., Alm, R.A. and Guerry, P A phasevariable capsule is involved in virulence of Campylobacter jejuni Mol. Microbiol. 40: Biswas, D., Itoh, K. and Sasakawa, C Uptake pathways of clinical and healthy animal isolates of Campylobacter jejuni into INT-407 cells. FEMS Immunol. Med. Microbiol. 29: Black, R.E., Levine, M.M., Clements, M.L., Hughes, T.P. and Blaser, M.J Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157: Blaser, M.J., Hardesty, H.L., Powers, B. and Wang, W.L Survival of Campylobacter fetus subsp. jejuni in biological milieus. J. Clin. Microbiol. 11: Blaser, M.J., Wells, J.G., Feldman, R.A., Pollard, R.A. and Allen, J.R Campylobacter enteritis in the United States. A multicenter study. Ann. Intern. Med. 98: Bradbury, W.C., Marko, M.A., Hennessy, J.N. and Penner, J.L Occurrence of plasmid DNA in serologically defined strains of Campylobacter jejuni and Campylobacter coli. Infect. Immun. 40:

66 47 Bradley, D.E., Anderson, A.N. and Perry, M.B Differences between the LPS cores in adherent and non-adherent strains of enteropathogenic Escherichia coli FEMS Microbiol. Lett. 64: Brás, A.M. and Ketley, J.M Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol Lett 179: Brás, A.M. and Ketley, J.M Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: Caldwell, M.B., Guerry, P., Lee, E.C., Burans, J.P. and Walker, R.I Reversible expression of flagella in Campylobacter jejuni. Infect. Immun. 50: Chart, H., Frost, J.A., Oza, A., Thwaites, R., Gillanders, S. and Rowe, B Heat-stable serotyping antigens expressed by strains of Campylobacter jejuni are probably capsular and not long-chain polysaccharides. J. Appl. Bacteriol. 81: Day, W.A., Sajecki, J.L., Pitts, T.M. and Joens, L.A Role of catalase in Campylobacter jejuni intracellular survival. Infect. Immun. 68: De Melo, M.A., Gabbiani, G. and Pechère, J.-C Cellular events and intracellular survival of Campylobacter jejuni during infection of HEp-2 cells. Infect. Immun. 57: De Melo, M.A. and Pechère, J.-C Identification of Campylobacter jejuni surface proteins that bind to eucaryotic cells in vitro. Infect. Immun. 58: DeGroote, M.A., Ochsner, U.A., Shiloh, M.U., Nathan, C., McCord, J.M., Dinaure, M.C., Libby, S.J., Vasquez-Torres, A., Xu, Y. and Fang, F.C Periplasmic superoxide dismutase protects Salmonella from products of phagocyte NADPH-oxidase and nitric oxidase synthase. Proc. Natl. Acad. Sci. USA. 94: Doig, P., Yao, R., Burr, D.H., Guerry, P. and Trust, T.J An environmentally regulated piluslike appendage involved in Campylobacter pathogenesis. Mol. Microbiol. 20: Drechsel, H. and Winkelmann, G.: Iron chelation and siderophores. In: Winkelmann, G. and Carrano, C.J. (Eds.), Transition metals in microbial metabolism. Harwood Acad, Amsterdam, Netherlands, 1997, pp Duffy, M.C., Benson, J.B. and Rubin, S.J Mucosal invasion in Campylobacter enteritis. Amer. J. Clin. Path. 73: Everest, P.H., Goossens, H., Butzler, J.P., Lloyd, D., Knutton, S., Ketley, J.M. and Williams, P.H Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J Med Microbiol 37: Fauchere, J.L., Rosenau, A., Veron, M., Moyen, E.N., Richard, S. and Pfister, A Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infect. Immun. 54:

67 Field, L.H., Headley, V.L., Payne, S.M. and Berry, L.J. 1986a. Influence of iron on growth, morphology, outer membrane protein composition, and synthesis of siderophores in Campylobacter jejuni. Infect. Immun. 54: Field, L.H., Headley, V.L., Underwood, J.L., Payne, S.M. and Berry, L.J. 1986b. The chicken embryo as a model for Campylobacter invasion: comparative virulence of human isolates of Campylobacter jejuni and Campylobacter coli. Infect. Immun. 54: Finlay, B.B., Heffron, F. and Falkow, S Epithelial cell surfaces induce Salmonella proteins required for bacterial adherence and invasion. Science 243: Friedman, C.R., Neimann, J., Wegener, H.C. and Tauxe, R.V.: Epidemiology of Campylobacter jejuni infections in the United States and other industrialzed nations. In: Nachamkin, I. and Blaser, M.J. (Eds.), Campylobacter. American Society for Microbiology, Washington, D.C., 2000, pp Galindo, M.A., Day, W.A., Raphael, B.H. and Joens, L.A Cloning and characterization of a Campylobacter jejuni iron-uptake operon. Curr. Microbiol. 42: Gaynor, E.C., Ghori, N. and Falkow, S Bile-induced 'pili' in Campylobacter jejuni are bacteria-independent artifacts of the culture medium. Mol. Microbiol. 39: Grant, C.C.R., Konkel, M.E., Cieplak, W., Jr. and Tompkins, L.S Role of flagella in adherence, internalization, and translocation of Campylobacter jejuni in nonpolarized and polarized epithelial cell cultures. Infect. Immun. 61: Guerry, P., Alm, R.A., Power, M.E., Logan, S.M. and Trust, T.J Role of two flagellin genes in Campylobacter motility. J. Bacteriol. 173: Guerry, P., Perez-Casal, J., Yao, R., McVeigh, A. and Trust, T.J A genetic locus involved in iron utilization unique to some Campylobacter strains. J. Bacteriol. 179: Hale, T.L. and Bonventre, P.F Shigella infection of Henle intestinal epithelial cells: role of the bacterium. Infect. Immun. 24: Hale, T.L., Morris, R.E. and Bonventre, P.F Shigella infection of Henle intestinal epithelial cells: role of the host cell. Infect. Immun. 24: Harvey, P., Battle, T. and Leach, S Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Hazeleger, W.C., Wouters, J.A., Rombouts, F.M. and Abee, T Physiological activity of Campylobacter jejuni far below the minimal growth temperature. Appl. Environ. Microbiol. 64: Headley, V.L. and Payne, S.M Differential protein expression by Shigella flexneri in intracellular and extracellular environments. Proc. Natl. Acad. Sci. USA 87: Hendrixson, D.R., Akerley, B.J. and DiRita, V.J Transposon mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol. Microbiol. 40:

68 Hickey, T.E., McVeigh, A.L., Scott, D.A., Michielutti, R.E., Bixby, A., Carroll, S.A., Bourgeois, A.L. and Guerry, P Campylobacter jejuni cytolethal distending toxin mediates release of interleukin-8 from intestinal epithelial cells. Infect. Immun. 68: Hu, L. and Kopecko, D.J Campylobacter jejuni associates with microtubules and dynein during invasion of human intestinal cells. Infect. Immun. 67: Hugdahl, M.B., Beery, J.T. and Doyle, M.P Chemotactic behavior of Campylobacter jejuni. Infect. Immun. 56: Hultgren, S.J., Jones, C.H. and Normark, S.: Bacterial adhesins and their assembly. In: Neidhardt, F.C. (Ed.), Escherichia coli and Salmonella. American Society for Microbiology, Washington, D.C., 1996, pp Hultgren, S.J., Normark, S. and Abraham, S.N Chaperone-assisted assembly and molecular architecture of adhesive pili. Annu. Rev. Microbiol. 45: Humphrey, C.D., Montag, D.M. and Pittman, F.E Experimental infection of hamsters with Campylobacter jejuni. J. Infect. Dis. 151: Jin, S., Joe, A., Lynett, J., Hani, E.K., Sherman, P. and Chan, V.L JlpA, a novel surfaceexposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Johnson, W.M. and Lior, H A new heat-labile cytolethal distending toxin (CLDT) produced by Campylobacter spp. Microb. Pathog. 4: Karlyshev, A.V., Linton, D., Gregson, N.A., Lastovica, A.J. and Wren, B.W Genetic and biochemical evidence of a Campylobacter jejuni capsular polysaccharide that accounts for Penner serotype specificity. Mol. Microbiol. 35: Kelle, K., Pages, J.M. and Bolla, J.M A putative adhesin gene cloned from Campylobacter jejuni. Res. Microbiol. 149: Ketley, J.M Pathogenesis of enteric infection by Campylobacter. Microbiol. 143: Kiehlbauch, J.A., Albach, R.A., Baum, L.L. and Chang, K.-P Phagocytosis of Campylobacter jejuni and its intracellular survival in mononuclear phagocytes. Infect. Immun. 48: Konkel, M.E., Babakhani, F. and Joens, L.A Invasion-related antigens of Campylobacter jejuni. J. Infect. Dis. 162: Konkel, M.E. and Cieplak, W., Jr Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M.E., Corwin, M.D., Joens, L.A. and Cieplak, W., Jr. 1992a. Factors that influence the interaction of Campylobacter jejuni with cultured mammalian cells. J. Med. Microbiol. 37:

69 Konkel, M.E., Garvis, S.G., Tipton, S.L., Anderson, D.E., Jr. and Cieplak, W., Jr Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M.E., Gray, S.A., Kim, B.J., Garvis, S.G. and Yoon, J. 1999a. Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadf virulence gene and its product. J. Clin. Microbiol. 37: Konkel, M.E., Hayes, S.F., Joens, L.A. and Cieplak, W., Jr. 1992b. Characteristics of the internalization and intracellular survival of Campylobacter jejuni in human epithelial cell cultures. Microbial. Pathog. 13: Konkel, M.E. and Joens, L.A Adhesion to and invasion of HEp-2 cells by Campylobacter spp. Infect. Immun. 57: Konkel, M.E., Kim, B.J., Klena, J.D., Young, C.R. and Ziprin, R Characterization of the thermal stress response of Campylobacter jejuni. Infect. Immun. 66: Konkel, M.E., Kim, B.J., Rivera-Amill, V. and Garvis, S.G. 1999b. Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32: Konkel, M.E., Mead, D.J. and Cieplak, W., Jr Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Konkel, M.E., Mead, D.J., Hayes, S.F. and Cieplak, W., Jr. 1992c. Translocation of Campylobacter jejuni across human polarized epithelial cell monolayer cultures. J. Infect. Dis. 166: Kops, S.K., Lowe, D.K., Bement, W.M. and West, A.B Migration of Salmonella typhi through intestinal epithelial monolayers: an in vitro study. Microbiol Immunol 40: Lara-Tejero, M. and Galán, J.E A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 290: Lara-Tejero, M. and Galán, J.E CdtA, CdtB, and CdtC form a tripartate complex that is required for cytolethal distending activity. Infect. Immun. 69: Larsen, R.A., Thomas, M.G., Postle, K Protonmotive force, ExbB and ligand-bound FepA drive conformational changes in TonB. Mol. Microbiol. 31: Lee, A., O'Rouke, J.L., Barrington, P.J. and Trust, T.J Mucus colonization as a determinant of pathogenicity in intestinal infection by Campylobacter jejuni: a mouse cecal model. Infect. Immun. 51: Lee, C.A. and Falkow, S The ability of Salmonella to enter mammalian cells is affected by bacterial growth state. Proc. Natl. Acad. Sci., USA 87: Logan, S.M. and Trust, T.J Structural and antigenic heterogeneity of lipopolysaccharides of Campylobacter jejuni and Campylobacter coli. Infect. Immun. 45:

70 Low, D., Braaten, B. and Van der Woude, M.: Fimbriae. In: Neidhardt, F.C. (Ed.), Escherichia coli and Salmonella. American Society for Microbiology, Washington, D.C., 1996, pp Madara, J.L Regulation of the movement of solutes across tight junctions. Annu. Rev. Physiol. 60: McSweegan, E. and Walker, R.I Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infect. Immun. 53: Mooney, A., Clyne, M., Curran, T., Doherty, D., Kilmartin, B. and Bourke, B Campylobacter upsaliensis excerts a cytolethal distending toxin effect on HeLa cells and T lymphocytes. Microbiol. 147: Moran, A.P Structure and conserved characteristics of Campylobacter jejuni lipopolysaccharides. J. Infect. Dis. 176 (Suppl 2): S Moran, A.P. and Penner, J.L Serotyping of Campylobacter jejuni based on heat-stable antigens: relevance, molecular basis and implications in pathogenesis. J. Appl. Microbiol. 86: Moran, A.P., Zahringer, U., Seydel, U., Scholz, D., Stutz, P. and Rietschel, E.T Structural analysis of the lipid A component of Campylobacter jejuni CCUG (serotype O:2) lipopolysaccharide. Description of a lipid A containing a hybrid backbone of 2-amino-2-deoxy-Dglucose and 2,3-diamino-2,3-dideoxy-D-glucose. Eur. J. Biochem. 198: Morooka, T., Umeda, A. and Amako, K Motility as an intestinal colonization factor for Campylobacter jejuni. J. Gen. Microbiol. 131: Moser, I., Schröder, W.F.K.J. and Hellmann, E In vitro binding of Campylobacter jejuni/coli outer membrane preparations to INT 407 cell membranes. Med. Micrbiol. Immunol. 180: Moser, I., Schroeder, W. and Salnikow, J Campylobacter jejuni major outer membrane protein and a 59-kDa protein are involved in binding to fibronectin and INT 407 cell membranes. FEMS Mirobiol. Lett. 157: Mroczenski-Wildey, M.J., Di Fabio, J.L. and Cabello, F.C Invasion and lysis of HeLa cell monolayers by Salmonella typhi: role of lipopolysaccharide. Microb. Pathog. 6: Myszewski, M.A. and Stern, N.J Phagocytosis and intracellular killing of Campylobacter jejuni by elicited chicken peritoneal macrophages. Avian Dis. 35: (?). Nachamkin, I., Allos, B.M. and Ho, T Campylobacter species and Guillian-Barré syndrome. Clin. Microbiol. Rev. 11: Nachamkin, I., Yang, X.H. and Stern, N.J Role of Campylobacter jejuni flagella as colonization factors for three-day-old chicks: analysis with flagellar mutants. Appl. Environ. Microbiol. 59: Newell, D.G Monoclonal antibodies directed against the flagella of Campylobacter jejuni: production, characterization and lack of effect on the colonization of infant mice. J. Hyg. (London) 96:

71 52 Newell, D.G., McBride, H., Saunders, F., Dehele, Y. and Pearson, A.D The virulence of clinical and environmental isolates of Campylobacter jejuni. J. Hyg. (London) 94: Nuijten, P.J., van der Zeijst, B.A. and Newell, D.G Localization of immunogenic regions on the flagellin proteins of Campylobacter jejuni Infect. Immun. 59: Oelschlaeger, T.A., Guerry, P. and Kopecko, D.J Unusual microtubule-dependent endocytosis mechanisms triggered by Campylobacter jejuni and Citrobacter freundii. Proc. Natl. Acad. Sci. USA 90: Owen, R.J. and Leaper, S Base composition, size and nucleotide sequence similarities of genome deoxyribonucleic acids from species of the genus Campylobacter. FEMS Microbiol. Lett. 12: Panigrahi, P., Losonky, G., DeTolla, L.J. and Morris, J.G Human immune response to Campylobacter jejuni proteins expressed in vitro. Infect. Immun. 60: Parkhill, J., Wren, B.W., Mungall, K., Ketley, J.M., Churcher, C., Basham, D., Chillingworth, T., Davies, R.M., Feltwell, T., Holroyd, S., Jagels, K., Karlshev, A.V., Moule, S., Pallen, M.J., Penn, C.W., Quall, M.A., Rajandrean, M.A., Rutherford, K.M., Van Vliet, A.H.M., Whitehead, S. and Barrell, B.G The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403: Pavlovskis, O.R., Rollins, D.M., Haberberger, R.L. Jr., Green, A.E., Habash, L., Strocko, S. and Walker, R.I Significance of flagella in colonization resistance of rabbits immunized with Campylobacter spp. Infect. Immun. 59: Pei, Z. and Blaser, M.J PEB1, the major cell-binding factor of Campylobacter jejuni, is a homolog of the binding component in Gram-negative nutrient transport systems. J. Biol. Chem. 268: Pei, Z., Burucoa, C., Grignon, B., Baqar, S., Huang, X.-Z., Kopecko, D.J., Bourgeois, A.L., Fauchere, J.-L. and Blaser, M.J Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Penner, J.L. and Hennessy, J.N Passive hemagglutination technique for serotyping Campylobacter fetus subsp. jejuni on the basis of soluble heat-stable antigens. J. Clin. Microbiol. 12: Pickett, C.L.: Campylobacter toxins and their role in pathogenesis. In: Nachamkin, I. and Blaser, M.J. (Eds.), Campylobacter. American Society for Microbiology, Washington, D.C., 2000, pp Pickett, C.L., Pesci, E.C., Cottle, D.L., Russell, G., Erdem, A.N. and Zeytin, H Prevalence of cytolethal distending toxin production in Campylobacter jejuni and relatedness of Campylobacter sp. cdtb genes. Infect. Immun. 64: Pickett, C.L. and Whitehouse, C.A The cytolethal distending toxin family. Trends in Microbiol. 7:

72 53 Prendergast, M.M., Lastovica, A.J. and Moran, A.P Lipopolysaccharides from Campylobacter jejuni O:41 strains associated with Guillain-Barré syndrome exhibit mimicry of GM 1 ganglioside. Infect. Immun. 66: Preston, M.A. and Penner, J.L Structural and antigenic properties of lipopolysaccharides from serotype reference strains of Campylobacter jejuni. Infect. Immun. 55: Purdy, D., Buswell, C.M., Hodgson, A.E., McAlpine, K., Henderson, I. and Leach, S.A Characterisation of cytolethal distending toxin (CDT) mutants of Campylobacter jejuni. J. Med. Microbiol. 49: Ratledge, C. and Dover, L.G Iron metabolism in pathogenic bacteria. Annu. Rev. Microbiol. 54: Richardson, P.T. and Park, S.F Enterochelin acquisition in Campylobacter coli: characterization of components of a binding-protein-dependent transport system. Microbiol. 141: Richardson, W.P. and Sadoff, J.C Induced engulfment of Neisseria gonorrhoeae by tissue culture cells. Infect. Immun. 56: Rivera-Amill, V., Kim, B.J., Seshu, J. and Konkel, M.E Secretion of the virulence associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J. Infect. Dis. 183: Russell, R.G. and Blake, D.C., Jr Cell association and invasion of Caco-2 cells by Campylobacter jejuni. Infect. Immun. 62: Russell, R.G., O'Donnoghue, M., Blake, D.C., Jr., Zulty, J. and DeTolla, L.J Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Schmiel, D.H., Wagar, E., Karamanou, L., Weeks, D. and Miller, V.L Phospholipase A of Yersinia enterocolitica contributes to pathogenesis in a mouse model. Infect. Immun. 66: Schröder, W. and Moser, I Primary structure analysis and adhesion studies on the major outer membrane protein of Campylobacter jejuni. FEMS Microbiol. Lett. 150: Schwan, E.T., Robertson, B.D., Brade, H. and van Putten, J.P Gonococcal rfaf mutants express Rd 2 chemotype LPS and do not enter epithelial host cells. Mol. Microbiol. 15: Seliger, S.S., Mey, A.R., Valle, A.-M., and Payne, S.M The two TonB systems of Vibrio cholerae: redundant and specific functions. Mol. Microbiol. 39: Silverstein, S.C., Steinman, R.M. and Cohn, Z.A Endocytosis. Annu. Rev. Biochem. 46: Skirrow, M.B. and Blaser, M.J.: Clinical aspects of Campylobacter infection. In: Nachamkin, I. and Blaser, M.J. (Eds.), Campylobacter, 2nd Edition. ASM Press, Washington, DC, 2000, pp Sosula, L., Nicholls, E.M. and Skeen, M Ultrastructure of Campylobacter jejuni in gammairradiated mouse jejunum. Am. J. Pathol. 131:

73 54 St. Geme III, J.W. and Falkow, S Haemophilus influenzae adheres to and enters cultured human epithelial cells. Infect. Immun. 58: Szymanski, C.M., King, M., Haardt, M. and Armstrong, G.D Campylobacter jejuni motility and invasion of Caco-2 cells. Infect. Immun. 63: Takata, T., Fujimoto, S. and Amako, K Isolation of nonchemotactic mutants of Campylobacter jejuni and their colonization of the mouse intestinal tract. Infect. Immun. 60: Takeuchi, A Electron microscopic studies of experimental Salmonella infection. Penetration into the intestinal epithelium by Salmonella typhimurium. Am J Pathol 50: Tsai, C.-M. and Frasch, C.E A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 119: Van Vliet, A.H., Baillon, M.L., Penn, C.W. and Ketley, J.M Campylobacter jejuni contains two fur homologs: characterization of iron-responsive regulation of peroxide stress defense genes by the PerR repressor. J. Bacteriol. 181: Van Vliet, A.H. and Ketley, J.M Pathogenesis of enteric Campylobacter infection. J. Appl. Microbiol. 90: 45S-56S. Van Vliet, A.H., Wooldridge, K.G. and Ketley, J.M Iron-responsive gene regulation in a Campylobacter jejuni fur mutant. J. Bacteriol. 180: Velayudhan, J., Hughes, N.J., McColm, A.A., Bagshaw, J., Clayton, C.L., Andrews, S.C. and Kelly, D.J Iron acquisition and virulence in Helicobacter pylori: a major role for FeoB, a highaffinity ferrous iron transporter. Mol. Microbiol. 37: Wai, S.N., Nakayama, K., Umene, K., Moriya, T. and Amako, K Construction of a ferritindeficient mutant of Campylobacter jejuni: contribution of ferritin to iron storage and protection against oxidative stress. Mol Microbiol 20: Wai, S.N., Takata, T., Takade, A., Hamasaki, N. and Amako, K Purification and characterization of ferritin from Campylobacter jejuni. Arch. Microbiol. 164: 1-6. Walker, R.I., Schmauder-Chock, E.A., Parker, J.L. and Burr, D Selective association and transport of Campylobacter jejuni through M cells of rabbit Peyer's patches. Can. J. Microbiol. 34: Walker, T.S. and Winkler, H.H Penetration of cultured mouse fibroblasts (L cells) by Rickettsia prowazeki. Infect. Immun. 22: Wassenaar, T.M Toxin production by Campylobacter spp. Clin. Microbiol. Rev. 10: Wassenaar, T.M., Bleumink-Pluym, N.M., Newell, D.G., Nuijten, P.J.M. and van der Zeijst, B.A.M Differential flagellin expression in a flaa flab+ mutant of Campylobacter jejuni. Infect. Immun. 62:

74 Wassenaar, T.M., Bleumink-Pluym, N.M. and van der Zeijst, B.A Inactivation of Campylobacter jejuni flagellin genes by homologous recombination demonstrates that flaa but not flab is required for invasion. EMBO J. 10: Wassenaar, T.M., Engelskirchen, M., Park, S. and Lastovica, A Differential uptake and killing potential of Campylobacter jejuni by human peripheral monocytes/macrophages. Med. Micrbiol. Immunol. 186: Wassenaar, T.M., van der Zeijst, B.A., Ayling, R. and Newell, D.G Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol. 139: Welkos, S.L Experimental gastroenteritis in newly-hatched chicks infected with Campylobacter jejuni. J. Med. Microbiol. 18: Whitehouse, C.A., Balbo, P.B., Pesci, E.C., Cottle, D.L., Mirabito, P.M. and Pickett, C.L Campylobacter jejuni cytolethal distending toxin causes a G 2 -phase cell cycle block. Infect. Immun. 66: Wooldridge, K.G. and Ketley, J.M Campylobacter-host cell interactions. Trends Microbiol. 5: Yao, R., Burr, D.H., Doig, P., Trust, T.J., Niu, H. and Guerry, P Isolation of motile and nonmotile insertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukaryotic cells. Mol. Microbiol. 14: Yao, R., Burr, D.H. and Guerry, P CheY-mediated modulation of Campylobacter jejuni virulence. Mol. Microbiol. 23: Young, G.M., Schmiel, D.H. and Miller, V.L A new pathway for the secretion of virulence factors by bacteria: The flagellar export apparatus functions as a protein-secretion system. Proc. Natl. Acad. Sci. USA 96: Ziprin, R.L., Young, C.R., Hume, M.E. and Konkel, M.E The absence of cecal colonization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectin-binding protein. Avian Dis. 43:

75 56 Figure. 1. C. jejuni infections are commonly acquired by handling and consuming undercooked chicken, and drinking unpasteurized milk and polluted water. Human illness with C. jejuni ranges from mild to severe diarrheal disease, the latter of which is characterized by the presence of blood and leukocytes in stool specimens.

76 57 Figure. 1. Processing and food handling Livestock and wildlife Fecal runoff Environment Transmission Diarrheal disease Contamination of surface water Human illness Ingestion by a susceptible host Passage through the stomach into the intestine Inflammatory response Colonization Cell invasion and toxin production

77 58 Figure 2. Diagram depicting C. jejuni-virulence determinants and their potential role in the development of C. jejuni-mediated enteritis. The top panel illustrates the interactions of C. jejuni with enterocytes and sites where certain virulence determinants might contribute in disease. The bottom panel illustrates the gross morphological changes that occur in the intestinal tract of an infected host. Not highlighted is the potential interaction of C. jejuni with M cells or professional phagocytic cells, as well as the contribution of the inflammatory response in infection.

78 59 Figure 2. Stomach Jejunum/Ileum Submucosa M cell Translocation Adherence Invasion & Toxin Production Motility, Chemotaxis, & Novel Synthetic Response Submucosa C. jejuni Small Intestine Large Intestine Mucus-Filled Crypt Villi Necrosis Mucosa Mucosa Submucosa Arterioles Venules Lymphatic Vessels Submucosa Arterioles Venules Lymphatic Vessels

79 60 Figure 3. Hematoxylin and eosin stained sections of the small intestines of E. coli and C. jejuniinoculated piglets. Note the presence of villi blunting and exudate (panel B) resulting from tissue necrosis. Panels: A, E. coli inoculated piglet; B, C. jejuni M129 inoculated piglet. Bar = 100 µm.

80 61 Figure 3. A villus crypt mucosa (m) submucosa (sm) B necrosis exudate m sm

81 62 Chapter 3 In vitro and in vivo models used to study Campylobacter jejuni-virulence properties Michael E. Konkel,a Marshall R. Monteville,a John D. Klena,a and Lynn A. Joensb School of Molecular Biosciences, Washington State University, Pullman, Washington 99164;a Departments of Veterinary Science and Microbiology, University of Arizona, Tucson, AZ 85721b Published in Current Topics in Food Safety in Animal Agriculture, Isaacson, R.E. and Torrence, M.E. (Eds.). In press.

82 63 ABSTRACT Campylobacter jejuni, a Gram-negative spiral shaped bacterium, is the leading cause of bacterial gastroenteritis in the world. Illness with C. jejuni ranges from mild to severe diarrheal disease. This article focuses on in vitro and in vivo models used to study some of the pathogenic properties of C. jejuni. We discuss the in vitro models currently being used to examine the interactions of C. jejuni with non-professional and professional phagocytic cells, and the in vivo models currently being used to examine C. jejuni colonization and infection. Also discussed are the methods being used to genetically manipulate the organism.

83 64 INTRODUCTION Members of the genus Campylobacter are Gram-negative curved rods that range in size from 0.2 to 0.8 mm in width and 0.5 to 5 mm in length. They are nonsporeforming, nonsaccharolytic, motile bacteria that possess unipolar or bipolar flagella. These bacteria grow optimally at temperatures between 37 and 42 C under microaerophilic conditions. When stressed, these bacteria typically become spherical in shape (Boucher et al., 1994; Federighi et al., 1998). The genome of C. jejuni is approximately 1.6 to 1.7 Mbp, with a mol (G + C) content of approximately 30 percent (Chang and Taylor, 1990; Karlyshev et al., 1998; Parkhill et al., 2000). To better understand the metabolic capacity and virulence properties of C. jejuni, the genome sequence of strain NCTC was determined (Parkhill et al., 2000). Protein coding regions account for 94.3 percent of the NCTC genome. Although members of Campylobacter spp. were initially recognized to cause disease in sheep and cattle, C. jejuni was not recognized as a human pathogen until much later (Skirrow, 1977). The first reported isolation of C. jejuni from diarrheal stools of humans was in 1972, with researchers using a filtration technique designed for veterinary diagnostics (Dekeyser et al., 1972). Subsequently, a selective medium for Campylobacter isolation from diarrheal stools of animals and humans was published in 1979 (Butzler and Skirrow, 1979). Now Campylobacter spp. have emerged as the leading cause of human gastroenteritis in developed countries (Tauxe, 1992; Altekruse et al., 1999). For example, in 1980, campylobacteriosis became a notifiable disease in New Zealand. By the year 2000, New Zealand had one of the highest reported incidence rates (greater than 230 per 100,000) for campylobacteriosis in the developed world (Anonymous, 2001; Savill et al., 2001). In the New Zealand context, this increase in reported cases does not appear to be caused by increased reporting or changes in methodologies for the detection of the pathogen (McNicholas et al., 1995). In other countries, the sharp rise in the number of reported cases of C. jejuni infections can be attributed to increased awareness of the disease by laboratory personnel and physicians, an

84 65 increase in attempts to isolate the microorganism, better isolation methods, and improved reporting (Skirrow, 1991; Cowden, 1992; Kapperud and Aasen, 1992). In the United States, the annual incidence of C. jejuni infection has been shown to be five to six per 100,000 persons (Tauxe et al., 1988). However, it has been estimated that 2.4 to 4 million cases of human campylobacteriosis occur every year in the United States (Tauxe, 1992), suggesting that this disease is significantly underreported in the United States (Blaser, 1997). All ages are affected by C. jejuni infection, although in developed countries, a greater number of infections occur in children less than four years of age and in young adults (Allos and Blaser, 1995). Infection caused by C. jejuni is more common in developing countries, but because of highlevel exposure within the first five years of life and the development of protective immunity, symptomatic infections among adults are much less common (Blaser and Reller, 1981). A slightly higher reporting rate exists among males than females (Skirrow, 1987). In developing countries, campylobacteriosis is usually a mild diarrheal disease with watery stools and little dysentery (Glass et al., 1983). In contrast, the presentation of disease in developed countries is more severe, characterized by bloody stools, abdominal pain, and fever (Blaser et al., 1979; Karmali and Fleming, 1979; Svedhem and Kaijser, 1980; Blaser et al., 1983). These differences probably reflect the repeated exposure of individuals in developing countries and may also reflect differences in the virulence among C. jejuni strains. Campylobacter infections can be fatal, with a case fatality rate for Campylobacter infection of 0.05 per 1,000 infections (Allos, 2001). However, the case fatality rate increases in immunocompromised individuals; this is especially true of individuals infected by the human immunodeficiency virus (Altekruse et al., 1998; Altekruse et al., 1999). Although most cases of campylobacteriosis are thought to be sporadic, some outbreaks have been associated mainly with drinking improperly treated water and unpasteurized milk (Vogt et al., 1982; Potter et al., 1983; Humphrey and Hart, 1988; Pebody et al., 1997). These outbreaks frequently occur in the spring and fall (Altekruse et al., 1998). In contrast, sporadic infections occur

85 66 most commonly as a result of handling or consumption of raw or undercooked meats, especially poultry (Hopkins et al., 1984; Harris et al., 1986; Deming et al., 1987; Adak et al., 1995; Eberhart- Philips et al., 1997). In temperate climates, sporadic infections peak in the early summer months (Brieseman, 1990; Skirrow, 1991; Tauxe, 1992; Allos and Blaser, 1995). The infective dose varies depending on the nature of the contaminated food, but as few as five hundred bacteria can cause disease (Robinson, 1981; Black et al., 1988; Black et al., 1992). Although Campylobacter-mediated enteritis can begin as early as eighteen hours after exposure, the average onset of disease is 3.2 days. The duration of clinical disease, diarrhea with abdominal pain, is usually three to four days. Following recovery, relapses of diarrhea and cramps have been reported (Blaser et al., 1979; Drake et al., 1981). There have also been reports of continued shedding of C. jejuni in the feces of individuals for weeks to months beyond the resolution of symptoms (Karmali and Fleming, 1979; Kapperud and Aasen, 1992). Extraintestinal infections associated with C. jejuni disease are rarely reported (Blaser et al., 1986; Allos, 2001). Although most people fully recover from campylobacteriosis, serious post-infectious sequelae can occur. The most notable post-infectious processes are Guillain-Barré (GBS) and Miller-Fisher (MFS) Syndromes, autoimmune-mediated disorders of the peripheral nervous system (Salloway et al., 1996; Nachamkin et al., 1998). These disorders are associated with host antibodies generated against the bacterial surface antigens; these antibodies may alternatively react with the host myelin sheaths of the sensory nerve fibers. This demyelination results in a flaccid paralysis; in some cases this can be fatal, but most affected individuals either completely recover or experience chronic disease (Mishu et al., 1993). Treatment of campylobacteriosis is not required for most individuals because the disease is self-limiting. For individuals who seek treatment, the macrolide erythromycin and fluoroquinolones are effective drug treatments for C. jejuni infections. However, emergence of antibiotic resistance has been reported in developed and developing countries (reviewed in Engberg et al., 2001). Erythromycin resistance appears to be chromosomally mediated, involving changes in the

86 67 peptidyltransferase binding site (23S rrna domain V) (Weisblum, 1995; Jensen and Aarestrup, 2001). Quinolone and fluoroquinolone resistance is mainly caused by alterations located in DNA gyrase (gyra) or, less commonly, topoisomerase IV (parc) (Gibreel et al., 1998). Use of antimicrobial agents such as quinolones and fluoroquinolones in animal feed appears to directly influence the emergence of resistant isolates in humans (Smith et al., 1999; van den Bogaard and Stobberingh, 2000; Engberg et al., 2001). Azithromycin is recommended as an alternative treatment for travelers from areas where conventional therapies are compromised by the emergence of antimicrobial resistance (DuPont, 1995; Kuschner et al., 1995). This article focuses on in vitro and in vivo models used to study some of the pathogenic properties of C. jejuni. More specifically, we discuss the in vitro models used to examine the interactions of C. jejuni with nonprofessional and professional phagocytic cells, and the in vivo models to examine C. jejuni colonization and infection. Finally, we discuss advances in genetic manipulation techniques for C. jejuni; application of these techniques should permit a rapid increase in our understanding of the pathogenesis of this organism in the near future. INTERACTIONS OF C. JEJUNI WITH NONPROFESSIONAL PHAGOCYTIC CELLS Investigators have utilized a variety of assays to assess the pathogenic properties of C. jejuni isolates in vitro. Here we discuss the model systems used to examine C. jejuni adherence, invasion, and translocation. Adherence The ability of enteropathogenic bacteria to bind to nonprofessional phagocytic cells is considered an important virulence determinant for bacteria that colonize the intestinal tract of a host because it prevents the organism from being swept away by mechanical cleansing forces such as peristalsis and fluid flow. In some instances, binding is also a prerequisite for entry into a host cell

87 68 (Roberts, 1990; Alrutz and Isberg, 1998). Invading a host cell is advantageous to organisms that can survive intracellularly, because they are protected from the host immune responses (Isberg and Tran Van Nhieu, 1994). The ability of C. jejuni to bind to host cells is hypothesized to play an early role in the development of campylobacteriosis. This hypothesis is supported by the finding that C. jejuni isolates cultured from individuals with fever and diarrhea adhere to cultured cells at a greater efficiency than those isolated cultured from asymptomatic individuals (Fauchere et al., 1986). In vitro adherence assays have been used extensively to assess the binding potential of C. jejuni isolates (Fauchere et al., 1986; McSweegan and Walker, 1986; de Melo and Pechère, 1990; Konkel et al., 1997; Pei et al., 1998). A typical assay involves inoculating a monolayer of undifferentiated eukaryotic cells of human or nonhuman origin with a bacterial suspension containing a known number of bacterial-colony-forming units (Fig. 1). Researchers have used a variety of cell lines when performing this assay, but the INT 407 (Henle, a human intestinal epithelial cell line) and Caco-2 (a human colonic cell line) cell lines are frequently used. These eukaryotic cells are considered to be reflective of those that the organism encounters in vivo. After inoculation of the eukaryotic cells with the bacteria and subsequent centrifugation to promote contact and synchronize infection, the inoculated cells are incubated in a 37 C humidifed, CO 2 -enriched incubator to allow the bacteria to bind to the host cells. Following incubation, the monolayers are rinsed to remove the non-adherent bacteria, and epithelial cells are lysed with a detergent such as Triton-X 100 or sodium deoxycholate. The viable number of adherent bacteria is determined by plating serial dilutions of the lysates on a solid medium, incubating the cultures under the appropriate set of conditions, and then determining bacterial cellular numbers. Each isolate is usually tested in triplicate to obtain a mean and standard deviation for the number of viable bacteria bound to a cell monolayer. Advantages of this assay include its simplicity and cost (inexpensive); also, it allows for several parameters to be manipulated (for example, the temperature at which the bacteria are cultured prior to the assay, the number of bacteria that are used to inoculate a cell monolayer, the temperature at which the assay is performed, and the addition of drugs to inhibit various host cell processes).

88 69 The number of C. jejuni bound to host cells can also be determined by microscopy examination of infected cell monolayers using an immunofluorescence assay (Newell et al., 1985a). After the infected cells are incubated, the monolayers are rinsed to remove the non-adherent bacteria. The cell monolayers are then treated with a fixative, such as methanol, rather than lysing the cells with a detergent. The inoculated cell monolayers are incubated with a primary antibody that reacts with the bound bacteria, followed by a fluorescence-labeled secondary antibody. An advantage of this assay is that the binding of nonviable bacteria to host cells can be quantitatively assessed. Other methods to visualize the binding of C. jejuni to host cells include scanning and transmission electron microscopy (Newell et al., 1985a). C. jejuni synthesize a set of surface-exposed molecules that facilitate binding to host cells. The molecules that promote the binding of C. jejuni to host cell receptors are collectively referred to as adhesins. Because metabolically inactive (heat-killed or sodium azide-killed) C. jejuni bind to cultured cells at levels equivalent to metabolically active organisms (Konkel and Cieplak, 1992), the C. jejuni adhesins appear to be synthesized constitutively rather than induced upon the organism's co-cultivation with mammalian cells. It is currently unknown whether the expression of the genes encoding the adhesins is upregulated in response to changes in culture conditions. The best-characterized C. jejuni adhesins include PEB1, CadF, and JlpA. PEB1 is 28 kda protein that shares homology with membrane proteins that function in amino acid transport. The gene encoding this protein was initially identified by screening a C. jejuni genomic - λgt11 library with a hyperimmune antibody raised against the 28 kda protein (Pei and Blaser, 1993). This work was preceded by another study in which de Melo and Pechère (1990) identified four outer membrane proteins (omps), with apparent molecular masses of 28, 32, 36, and 42 kda, which demonstrated adhesive properties as judged by a ligand-binding assay (de Melo and Pechère, 1990). Evidence indicating that PEB1 is an adhesin includes the finding that a C. jejuni peb1a null mutant exhibits a reduction in the duration of mouse intestinal colonization when compared to the C. jejuni parental isolate (Pei et al., 1998). CadF (Campylobacter adhesion to Fibronectin) is a C. jejuni 37

89 70 kda omp (Konkel et al., 1997). CadF is conserved among all C. jejuni and C. coli isolates tested to date (Konkel et al., 1999a). Whether the 36 kda protein identified by de Melo and Pechère (de Melo and Pechère, 1990) and the 37 kda CadF protein represent the same protein is not known. A C. jejuni cadf mutant lacks the ability to colonize the cecum of newly hatched leghorn chickens (Ziprin et al., 1999). Jin et al. (Jin et al., 2001) identified a 42.3 kda lipoprotein termed JlpA (jejuni lipoprotein A), which mediates the binding of C. jejuni to HEp-2 cells. A mutation in the jlpa gene resulted in an 18 to 19.4 percent reduction in adherence when compared to the C. jejuni wild-type isolate, but had no effect on C. jejuni invasion. In addition, pretreatment of Hep-2 cells with recombinant JlpA reduced the binding of C. jejuni to the cells in a dose-dependent fashion. Other molecules proposed to act as adhesins include the flagellum (Pavlovskis et al., 1991; Wassenaar et al., 1991; Grant et al., 1993; Wassenaar et al., 1993), lipopolysaccharide (McSweegan and Walker, 1986), the major outer membrane protein (MOMP, also called OmpE) (Moser et al., 1997; Schröder and Moser, 1997), and a protein termed P95 (Kelle et al., 1998). In summary, evidence supports the hypothesis that adhesins play a role in C. jejuni colonization. Future studies should examine whether C. jejuni utilizes the same set of adhesins in a human host versus that of cattle, chickens, and other reservoir hosts in which C. jejuni is considered commensal flora. In addition, future studies should also focus on other factors that influence the ability of C. jejuni to colonize a host, including the organism's motility, chemotactic behavior, and surface charge. Invasion The ability of C. jejuni to enter, survive, and replicate in mammalian cells has been studied extensively using tissue culture models (Newell et al., 1985b; de Melo et al., 1989; Konkel and Joens, 1989; Konkel et al., 1990; Wassenaar et al., 1991; Everest et al., 1992; Konkel et al., 1992a; Konkel et al., 1992b; Grant et al., 1993; Oelschlaeger et al., 1993; Yao et al., 1994; Doig et al., 1996a; Pei et al., 1998; Konkel et al., 1999b). The most commonly used experimental assay for assessing

90 71 invasion involves determining the number of bacteria protected from an aminoglycosidic antibiotic, such as gentamicin, that does not penetrate eukaryotic cell membranes (Fig. 1). Everest et al. (Everest et al., 1992) found a statistically significant difference in the level of HeLa and Caco-2 cell invasion between C. jejuni isolates taken from individuals with colitis versus those isolated from individuals with noninflammatory diarrhea. In colonic biopsy specimens obtained after inoculation of primates, C. jejuni were observed to be in the process of being taken up by epithelial cells (Russell et al., 1993). C. jejuni were also observed within vacuoles and within the cytoplasm of damaged cells. The investigators concluded that early mucosal damage, occurring prior to any inflammatory response, resulted from C. jejuni invading the colonic epithelial cells. The relative ability of C. jejuni to invade cultured cells is strain-dependent (Newell et al., 1985b; Konkel and Joens, 1989; Everest et al., 1992). Newell et al. (1985) found that environmental isolates were much less invasive than clinical isolates as determined by immunofluorescence and electron microscopy examination of C. jejuni-infected HeLa cells (Newell et al., 1985b). The percent of the inoculum internalized for C. jejuni , a strain isolated from a milkborne outbreak of diarrheal illness, has been reported to range between 0.8 to 1.8 percent% (Yao et al., 1994; Doig et al., 1996b; Yao et al., 1997). Parasite-directed endocytosis is a process in which a microorganism synthesizes the proteins required to promote internalization. The internalization of C. jejuni is inhibited by chloramphenicol, a specific inhibitor of bacterial protein synthesis (Konkel and Cieplak, 1992; Oelschlaeger et al., 1993). In addition, metabolically inactive (sodium azide-killed) but intact C. jejuni are not internalized (Konkel and Cieplak, 1992). One and two-dimensional electrophoretic analyses of metabolically labeled C. jejuni cultured in the presence and absence of epithelial cells has also demonstrated that a number of proteins are synthesized, exclusively or preferentially, in the presence of epithelial cells, whereas others are selectively repressed (Konkel and Cieplak, 1992; Konkel et al., 1993). The newly synthesized proteins are distinct from those proteins induced by thermal stress (Konkel et al., 1998). Independently, Panigrahi et al. (1992) reported that C. jejuni synthesized a set

91 72 of proteins when grown in rabbit ileal loops that were not produced when the organism was cultured on standard laboratory medium (Panigrahi et al., 1992). Two of the newly synthesized proteins, with apparent molecular masses of 84 and 47 kda, were detectable using convalescent sera from C. jejuni-infected individuals. Combined, these results suggest that C. jejuni synthesize entry-promoting proteins upon exposure to host cells and that these proteins are part of a coordinated bacterial response, resulting from the regulation of expression of a defined subset of genes, to the epithelial cell microenvironment. Studies in our laboratory have demonstrated that a subset of the de novo synthesized proteins, termed Cia(s) for Campylobacter invasion antigens, are secreted upon co-cultivation of C. jejuni with intestinal cells. By differentially screening C. jejuni genomic DNA-phage expression libraries with antisera generated in rabbits against C. jejuni cultured in the presence of INT 407 cells (Cj+INT) and C. jejuni cultured in the absence of INT 407 cells (Cj-INT), plaques were identified that reacted only with the Cj+INT antiserum. This screening resulted in the identification of a gene termed ciab (Konkel et al., 1999b). In vitro assays revealed that a C. jejuni ciab null mutant bound to INT 407 cells in numbers equal to or greater than the wild-type isolate, but exhibited a significant reduction (~100-fold) in internalization. Confocal microscopy studies using an anti-ciab serum revealed an intense fluorescence signal in the cytoplasm of the C. jejuni-infected INT 407 cells but not in the cytoplasm of mock-infected INT 407 cells. Nonetheless, the specific function of CiaB is not known. Other work has revealed that Cia protein synthesis and secretion are separable, and that secretion is the rate-limiting step in these processes (Rivera-Amill et al., 2001). Cia protein synthesis is induced in response to bile salts and various eukaryotic host cell components, whereas the latter is capable of also inducing Cia protein secretion. We propose that the incubation of C. jejuni in the presence of 0.1 percent of sodium deoxycholate, which is similar to the concentration of bile salts found in the lumen of the small intestine (Pope et al., 1995), provides an environment that partially mimics that which C. jejuni would encounter in vivo. Also noteworthy is that culturing C. jejuni on

92 73 deoxycholate-supplemented medium retarded the inhibitory effect of chloramphenicol on epithelial invasion as judged by the gentamicin-protection assay (Rivera-Amill et al., 2001). These data, in combination with the observation that C. jejuni cultured with eukaryotic cells synthesize a subset of new proteins not synthesized by organisms cultured in the absence of eukaryotic cells (Konkel and Cieplak, 1992; Konkel et al., 1993), suggest that the coordinate expression of the genes encoding the Cia proteins is subject to environmental regulation. Preliminary data has been generated in our laboratory suggesting that the Cia proteins are secreted via the flagellar type III secretion apparatus (Konkel et al., unpublished observations). Mutations in the genes encoding components of the C. jejuni flagellar apparatus abolished Cia protein secretion. A caveat of this work is that cia and flagellar-structural genes may be co-regulated. The secretion of virulence proteins via the flagellar apparatus would seem to be beneficial to C. jejuni, eliminating additional energy expenditure for the synthesis of a distinct Cia secretory apparatus. There is precedent for protein secretion through the flagellar apparatus; Yersinia spp. secrete proteins termed flagellar outer proteins (Fops) from this apparatus (Macnab, 1999; Young et al., 1999). Bacon et al. (2000) reported the presence of a plasmid termed pvir in C. jejuni (Bacon et al., 2000). Nucleotide sequence of four open reading frames located on the 35 kbp pvir plasmid showed similarity with Helicobacter pylori proteins (ranging from percent similarity), and percent similarity to proteins that comprise type IV secretion systems from a diverse range of Gram-negative bacteria. Mutations in two of the four genes (comb3 and virb11) affected the virulence of C. jejuni strain as judged by in vitro binding and internalization assays and in vivo assays utilizing the ferret diarrhea model. Independent mutations in each gene resulted in a reduction of adherence (29 percent of wild type with comb3, 13 percent of wild type with virb11) with a corresponding reduction (28 percent and 9 percent, respectively) in invasion using INT 407 cells. Compared to the wild type C. jejuni strain , the virb11 mutant was significantly affected in its capacity to cause diarrhea in the ferret model at both high (9 x 1010 to 8 x

93 bacterial cells) and low (8 x 109 to 8 x 1010 bacterial cells) doses. Hybridization experiments determined that 10 percent (six of fifty-eight samples) of C. jejuni clinical isolates obtained from Thailand contained the virb11 gene. Based on this finding, the investigators concluded that distinct pathogenic mechanisms differentiate strains of C. jejuni. The relationship between the virb11 system and the Cia secretion proteins from C. jejuni is currently undefined. Translocation Investigators have utilized a cell culture system to assess the ability of pathogens to migrate, or translocate, across a cell barrier (Finlay and Falkow, 1990; Cruz et al., 1994; Kops et al., 1996; Nataro et al., 1996). Translocation is considered an important virulence attribute for certain pathogens because it permits them access to underlying tissues and may allow for their dissemination throughout a host. The in vitro assay to assess bacterial translocation involves culturing eukaryotic cells on a permeable membrane contained within a plastic insert (Fig. 2). The inserts are placed in a plastic tray (for example, a twenty-four-well tissue culture plate), thereby establishing apical and basolateral chambers. Culture media is added to each chamber to promote cell growth and differentiation. Cell differentiation results in a polarized cell monolayer with distinct apical and basolateral surfaces. Depending on the cell line used, the apical cell surfaces are characterized by well-developed microvilli and brush borders. Human adenocarcinoma cell lines HT29, Caco-2, and T84 have all been used in this assay system, because these cells will form a polarized cell monolayer when cultured under appropriate conditions. The ability of bacteria to traverse cell monolayers is assessed after the addition of the bacteria to the apical chamber of the insert, followed by plating of the basolateral chamber medium on agar plates after various periods of time. This culture system enables investigators to monitor the integrity of a cell monolayer by measuring the transepithelial electrical resistance (TER). A decrease, or loss of TER, indicates disruption of cellular tight junctions (Finlay et al., 1988; Finlay and Falkow, 1990).

94 75 Everest et al. (1992) noted that 86 percent of Campylobacter isolates from individuals with colitis were able to translocate across polarized Caco-2 cells, versus 48 percent of strains isolated from individual with non-inflammatory disease (Everest et al., 1992). Translocation of C. jejuni cells across polarized Caco-2 cell monolayers was determined using the culture system described previously. It was also noted that six C. jejuni isolates, characterized as non-invasive, as judged by the gentamicin protection assay using Caco-2 cells, were able to translocate across the polarized cell monolayers. Harvey et al. (1999) compared four C. jejuni isolates with differing abilities to invade Caco-2 cells in their ability to translocate across polarized epithelial membranes (Harvey et al., 1999). In this report, invasiveness of C. jejuni did not quantitatively correlate with the ability to translocate across tissue cell culture monolayers. Although the investigators detected fluctuations in the measurable TER with the different C. jejuni isolates over the course of the six-hour assay, cell culture monolayer integrity was maintained, and final TER values were comparable to starting baseline values. Maintenance of monolayer integrity, at least over a relatively short period of time (eight hours), has also been reported by others (Konkel et al., 1992c; Brás and Ketley, 1999). Noteworthy is that Brás et al. (1999) detected a loss in TER of Caco-2 cells inoculated with C. jejuni after twenty-four hours, indicating an eventual disruption of cellular tight junctions. These investigators proposed that the loss in monolayer integrity was the result of long-term effects of translocation and/or invasion or the accumulation of a bacterial toxin(s). The aforementioned studies argue that the genes encoding the products responsible for invasion in C. jejuni are distinct from those that confer translocation ability. The predominant route of C. jejuni translocation across polarized cell monolayers is unclear. The presence of intracellular bacteria is supporting evidence for a transcellular (through a cell) route of passage. So is the observation that C. jejuni-cellular translocation is reduced at 20ºC (Konkel et al., 1992c). Temperatures of 18-22ºC preferentially inhibit eukaryotic endocytic and phagocytic processes (Silverstein et al., 1977). In contrast, evidence exists that supports the paracellular (between cell) route of passage, including the observation that C. jejuni can be recovered from the

95 76 basolateral chamber as early as fifteen minutes post-inoculation of a polarized cell monolayer (Konkel et al., 1992c). In addition, the invasiveness of C. jejuni isolates does not quantitatively correlate with translocation efficiency (Harvey et al., 1999). Studies have demonstrated that tight junctions temporally relax to allow regulated passage of both solutes and neutrophils (Madara, 1998). Cellular tight junctions also reseal following penetration by certain pathogens (Takeuchi, 1967). Therefore it is plausible that C. jejuni utilizes a paracellular route of passage, with only a transient change in cellular integrity. Future studies will be required to clarify the major route of C. jejuni translocation through the epithelium. The in vivo relevance of C. jejuni translocation across the intestinal epithelium is not known. Whether C. jejuni reaches the lamina propria via migration across the epithelial cells lining the intestinal tract or exclusively via M cells is unclear. In the lamina propria, C. jejuni would have access to different cellular receptors and professional phagocytic cells that are likely to play a role in infection. Nevertheless, the incidence of bacteremia in individuals with C. jejuni infection is 0.4 percent (Allos and Blaser, 1995), suggesting that the host's immune system is effective in preventing the spread of the infection. Movement of C. jejuni from the apical to basolateral epithelial cell surface is clearly important for pathogenesis. However, whether translocation through epithelial cells plays a more significant role than M cell sampling has yet to be determined. INTERACTIONS OF C. JEJUNI WITH PROFESSIONAL PHAGOCYTIC CELLS Macrophages provide intestinal pathogens with an unoccupied space for acquiring nutrients and a shelter from immune surveillance. If a pathogen can survive within these cells, it is also possible for it to propagate and disseminate throughout a host. Subsequently, the pathogen may establish a carrier state in the host and serve as a source of infection for other individuals. As well as the tissue culture studies described in the preceding section, animal studies have demonstrated that C. jejuni cells invade intestinal mucosal epithelial cells and translocate to the lamina propria and deeper

96 77 submucosa of the host (Babakhani et al., 1993). Deeper tissue involvement results in a hemorrhagic necrosis in the lamina propria, formation of crypt abscesses, and infiltration of inflammatory cells (Blaser et al., 1980; Duffy et al., 1980). At this stage in a C. jejuni infection, the organism is likely engulfed by macrophages. Engulfment Field et al. (1991) demonstrated a correlation between the uptake of C. jejuni by mouse peritoneal macrophages and the organism's clearance from the bloodstream of mice (Field et al., 1991). In addition, avirulent strains of C. jejuni, defined by their inability to invade the chorioallantoic membrane of chicken embryos, were cleared in vivo and engulfed in vitro by peritoneal macrophages from Balb/c mice at significantly higher rates than virulent strains. Additional experiments revealed that these findings were independent of complement-mediated opsonization, suggesting that C. jejuni virulence was associated with resistance to phagocytosis (Field et al., 1991). Resistance to phagocytosis was also demonstrated by C. jejuni isolates exposed to guinea-pig resident peritoneal macrophages when compared to C. coli strains that were engulfed in significantly higher numbers (Banfi et al., 1986). These authors proposed that the difference in uptake of the two species of Campylobacter was caused by the presence of an antiphagocytic capsular-like material in C. jejuni that was absent in C. coli. In 1997, Wassenaar et al. reported a linear correlation between infective dose and internalized C. jejuni in monocytes from human blood donors (Wassenaar et al., 1997). This experiment supported previous evidence obtained by Myszewski and Stern (1991), who found that C. jejuni was readily internalized by chicken macrophages within a thirty-minute incubation period with increased internalization when serum or macrophages from previously colonized hosts were added to the assay (Myszewski and Stern, 1991). Recently, Day et al. (2000) demonstrated that the uptake of the clinical C. jejuni M129 isolate in mouse and porcine peritoneal macrophages, as well as a mouse macrophage cell line, occurred in the absence of serum and complement (Day et al., 2000) (Fig.

97 ). These data suggest that the internalization of C. jejuni by phagocytes is strain dependent. Interestingly, work from Guerry et al. (2000, 2002) strongly implicates sialylation of LOS as important for serum resistance (Guerry et al., 2000; Guerry et al., 2002). Guerry et al. (2002) created mutations in neuc1, a gene encoding an N-acetylglucosamine(GlcNAc)-6-phosphate 2- epimerase/glcnac-6-phosphatase, necessary for sialylation, as well as the LOS structural gene cgta (a UDP-N-acetylgalactosaminyl transferase) (Guerry et al., 2002). Loss of sialylation impaired resistance to normal human sera, but had no effect on invasion of INT 407 cells. Alteration of the LOS structure enhanced invasion. This is clearly an area that will receive more attention in the future. Survival Wassenaar et al. (1997) examined the ability of sixteen clinical and laboratory-adapted isolates of C. jejuni to resist killing by activated human peripheral monocytes (Wassenaar et al., 1997). The investigators found that the majority of C. jejuni were killed by the monocytes within twenty-four to forty-eight hours. However, the monocytes from approximately 10 percent of donorindividuals demonstrated normal uptake of C. jejuni but failed to kill the bacterium. Myszewski and Stern (1991) examined the ability of both a C. jejuni high-passage clinical isolate and a C. jejuni chicken isolate to resist killing by macrophages (Myszewski and Stern, 1991). They found that both C. jejuni isolates were killed by chicken peritoneal macrophages within a six-hour period. These observations suggest that the macrophage cell type is important for C. jejuni intracellular survival. Evidence suggesting that strain variation does have a significant role in intracellular survival has been gathered by several investigators. Joens et al. (unpublished data), in examining the survival of five environmental isolates of C. jejuni exposed to a mouse macrophage cell line (J774A.1), found that two of five isolates were inactivated at twenty-four hours. The remaining three isolates survived for seventy-two hours after engulfment. Day et al. (2000) examined the survival of a C. jejuni clinical isolate in various host macrophages (Day et al., 2000). The clinical isolate M129 was able to survive for seventy-two hours in porcine and murine peritoneal macrophages and in the J774A.1

98 79 macrophage cell line, although there was a noticeable reduction in the number of C. jejuni recovered from the three phagocytic cell types at seventy-two hours post-inoculation. Similar findings were reported by Kiehlbauch et al. (1985) when the survival of the C. jejuni 2964 clinical isolate was examined in different macrophage lineages (Kiehlbauch et al., 1985). This group was able to recover the clinical C. jejuni isolate from three phagocytic cell types over a six-day period. Together, these findings support the view that the particular C. jejuni isolate, as well as the macrophage cell type, are both important factors involved in intracellular survival. Role of Phagolysosome Processing in Macrophage Survival Although an intracellular existence provides bacteria with an unoccupied space and shelter from the host immune system, phagocytosed bacteria must be able to survive antigen processing in professional phagocytes. Reactive products such as the superoxide radical and hydrogen peroxide are produced in the phagolysosome in response to the respiratory burst and are extremely toxic to microorganisms. However, intracellular bacteria produce proteins that can effectively neutralize these toxic products (De Groote et al., 1997). To address the role of superoxide dismutase in C. jejuni intra-macrophage survival, assays were performed with a C. jejuni sodb mutant and the J774A.1 macrophage cell-line (Joens et al., unpublished data). A sodb mutant created in C. jejuni was able to survive in the J774A.1 macrophage cell-line at a level equal to that of the C. jejuni wild-type strain. In contrast, catalase production was shown to be important in C. jejuni intra-macrophage survival (Day et al., 2000). More specifically, a kata mutant of C. jejuni was found to be more susceptible to killing by J774A.1 cells than was the C. jejuni wild-type isolate. When the respiratory burst or production of nitric oxide were inhibited, the C. jejuni kata mutant and wild-type isolate were recovered in equal numbers from the J774A.1 cells following a seventy-two-hour period of incubation. Many bacteria are able to circumvent antigen processing by altering the intracellular trafficking of the phagosome. Mycobacteria spp. prevent acidification of the phagosome, thereby preventing phagosome-lysosome fusion. Salmonella enterica also interfere with antigen processing

99 80 and the endocytic pathway by preventing the maturation of the phagosome. Pitts et al. analyzed the intracellular trafficking of a C. jejuni clinical isolate in J774A.1 macrophages (Pitts et al., unpublished data). Macrophages were infected with the C. jejuni isolate at a multiplicity of infection of ten bacteria per macrophage and the phagosome examined for the presence of early and late protein markers over a seventy-two-hour time period. This experiment revealed that approximately percent of the C. jejuni internalized co-localized with early and late proteins of the endocytic pathway, and that 90 percent of the phagosomes were acidified as determined by staining with lysotracker dye. Thus, the majority of the bacteria appear to be processed through the normal endocytic pathway. It is not known whether the remaining bacteria modify the phagosome in a manner that promotes their survival. IN VIVO COLONIZATION MODELS Mice and chickens are the two animals most commonly used to study C. jejuni colonization. Oral inoculation of most inbred and outbred strains of mice with C. jejuni leads to intestinal tract colonization without significant clinical disease (Field et al., 1981; Blaser et al., 1983). However, diarrhea and clinical disease have been reported in scid mice orally infected with C. jejuni low passage clinical isolates (Hodgson et al., 1998). Systemic infection has also been reported in mice, with recovery of the organism from extraintestinal sites (Vuckovic et al., 1998). An early debate focused on whether C. jejuni adhere to the mouse intestinal mucus or directly to the intestinal epithelial cells. However, a reduction in the duration of mouse intestinal colonization was observed upon inoculation of mice with a C. jejuni cell-binding factor (Peb1A) mutant when compared to the C. jejuni wild-type isolate (Pei et al., 1998). This finding suggests that C. jejuni are capable of binding to the intestinal cells of mice. Two of the advantages of the mouse model are that the system is relatively inexpensive to maintain and operate, and various mutant lines are available for study. In addition, numerous reagents (such as monoclonal antibodies) are available to study the host response

100 81 following colonization with C. jejuni. C. jejuni cells normally colonize chicks ten to twenty-one days after hatching, following the depletion of maternal antibody. Experimentally, chicks are colonized with 1 x 1010 C. jejuni per gram of fecal content within three days of oral inoculation (Wassenaar et al., 1993). Colonization occurs throughout the bird's intestinal tract, with the highest number of cells being found in the cecum. Similar to C. jejuni-colonized mice, chickens colonized with C. jejuni are most often asymptomatic (Shanker et al., 1990). It is unclear whether C. jejuni bind to the mucus or mucusproducing epithelial cells in the chick intestinal tract. However, Ziprin et al. (1999) noted that a C. jejuni CadF mutant did not colonize the cecum of newly hatched leghorn chickens (Ziprin et al., 1999). CadF is a C. jejuni 37 kda outer membrane protein conserved among C. jejuni and C. coli isolates (Konkel et al., 1997). There are reports of extraintestinal recovery of C. jejuni in colonized chicks (Beery et al., 1988), but whether this dissemination results from the organism's translocation or direct invasion of intestinal epithelium is not known. Because poultry are naturally colonized with C. jejuni, this animal model seems ideal to study the mechanism of Campylobacter adherence. The major disadvantage of this system is the lack of reagents to study the effect of colonization within the host. IN VIVO INFECTION MODELS As previously discussed, campylobacteriosis in humans is frequently characterized by one to three days of fever followed by intense abdominal pain and watery or hemorrhagic diarrhea (Price et al., 1979; Drake et al., 1981; Blaser et al., 1983; Allos and Blaser, 1995). The severity of the disease is strain related and is usually self-limiting, with shedding of C. jejuni lasting from days to months (Blaser and Reller, 1981; Black et al., 1988; Black et al., 1992). During the acute phase of the disease, lesions can be found throughout the intestine, but are more prominent in the terminal ileum and colon. Endoscopic lesions range from hyperemia, edema to frank hemorrhage with ulceration

101 82 (Lambert et al., 1979; Colgan et al., 1980; Lambert et al., 1982; McKendrick et al., 1982; Mee et al., 1985). Histological lesions consist of acute inflammation of the mucosa with edema, inflammatory infiltrates into the lamina propria, and crypt abscesses (Van Spreeuwel et al., 1985). Although C. jejuni has been isolated from the feces of many animals, most animals do not suffer from campylobacteriosis. Discussed next are the most frequently used laboratory animal models to study the pathogenesis of Campylobacter organisms. Small-Animal Models The most frequently used laboratory animal model to study Campylobacter pathogenesis has been the young weanling ferret (Fox et al., 1987; Bell and Manning, 1991). Oral inoculation with C. jejuni results in intestinal colonization lasting from two to twelve days with the presence of a mild to moderate diarrhea. The diarrhea, sometimes mucoid with occult blood, can last for two to three days (Fox et al., 1987). Histologically, the lesions seen in ferrets infected with C. jejuni suggest a mild colitis. This model has been used to assess the pathogenicity of a variety of C. jejuni mutants including pspa, chey, virb11 and asta, as well as the C. jejuni NCTC sequence strain and the strain (Doig et al., 1996b; Yao et al., 1997; Bacon et al., 2000). A reduction in virulence (assessed by the absence of diarrhea) of C. jejuni pspa, chey, and virb11 mutants was observed when compared to the parental C. jejuni isolate. CheY is involved in C. jejuni chemotaxis (Yao et al., 1997). The function of the PspA protein is not known (Gaynor et al., 2001). The virb11 gene product has been linked to Type IV secretion, based on a low level of sequence similarity as previously discussed in this review. NCTC and the mutation in asta, however, were found to be avirulent when examined in this model (Yao et al., 1997; Bacon et al., 2000). Although the relevance of the ferret model in terms of human disease has yet to be established, it is certain that large inocula, (108 to 1011 bacterial cells) are required to induce diarrhea. In addition, control animals may excrete feces containing green mucus, which is a symptom displayed by infected animals (Bell and Manning, 1991). Nevertheless, this laboratory model appears to have potential in defining

102 83 adherence mechanisms used by the bacterium to colonize the intestinal tract of mammals. Large-Animal Models Of the large, nonprimate models used to assess the pathogenicity of Campylobacter organisms, the use of piglets versus other animals has the advantage in that these animals develop symptoms of infection and disease similar to that observed in humans infected with C. jejuni. The reason for this is two-fold: 1) the human and porcine physiological system and gastrointestinal tract are similar and, 2) Campylobacter-induced enteritis occurs naturally in weaned pigs (Taylor and Olubunmi, 1981). Taylor and Olubunmi (1981) observed inflammatory lesions and shortening of the villous epithelium in the small and large intestines of weaned piglets that resulted from a natural infection with Campylobacter fetus subspecies coli (Taylor and Olubunmi, 1981). Histopathology results from this study were similar to those lesions described in human biopsies from Campylobacter-induced enteritis (Price et al., 1979; Van Spreeuwel et al., 1985). Gnotobiotic pigs have also been used to study the pathogenesis of Campylobacter organisms (Boosinger and Powe, 1988). In a study by Boosinger and Powe (1988), C. jejuniinfected gnotobiotic pigs exhibited watery diarrhea that was yellowish in color two days postinoculation (Boosinger and Powe, 1988). The diarrhea lasted throughout the course of the twelveday experiment. Lesions were present in the cecum and colon. The lesions were characterized by the sloughed epithelial cells, distended crypts, severe edema of the submucosa, and diffuse infiltration of inflammatory cells. Although the use of gnotobiotic pigs as a model system for campylobacteriosis is artificial because of the lack of intestinal flora, the pigs develop lesions consistent with those in humans. The main disadvantage of using these animals is that they are expensive and require specialized housing. In 1993, Babakhani et al. described the use of a colostrum-deprived newborn piglet as a model to study the pathogenesis of C. jejuni (Babakhani et al., 1993). Although these piglets were free of maternal antibodies, competing intestinal microflora were present. In the study, piglets were

103 84 inoculated orally and observed for clinical signs over a nine-day period. Watery diarrhea with the presence of blood and mucus was observed one day post-infection. Lesions were detected primarily in the large intestine and consisted of a subacute, diffuse, mild to moderate, erosive colitis and typhlitis. Histologically, mucosal epithelial cells were rounded with exfoliation into the lumen, which resulted in a generalized necrosis of the lamina propria. Extensive infiltrates of inflammatory cells were found in the lamina propria and submucosa. The lumen and the crypts of pigs with severe lesions contained Campylobacter organisms. More recently, the newborn pig has been used to examine the virulence of two different isogenic mutants (Joens, unpublished data). Piglets were infected with a C. jejuni M129 kata mutant and the isogenic wild-type isolate. The kata mutant was able to colonize the inoculated pigs but failed to produce intestinal lesions. In contrast, the C. jejuni wild-type isolate produced lesions in both the small and large intestines of the piglets. The lesions were characterized by epithelial cell sloughing, necrosis, inflammatory infiltrates, and occasional hemorrhage (Fig. 21.4). Konkel et al. tested the virulence of a C. jejuni ciab mutant in the piglet model and noted fewer lesions in piglets inoculated with the mutant when compared to the piglets inoculated with the C. jejuni F38011 wildtype isolate and C. jejuni ciab isolate complemented in trans (unpublished data). The clinical signs and lesions produced by the wild-type strains in these two experiments were similar to those reported in the original study (Babakhani et al., 1993). Collectively, these studies illustrate the usefulness of the piglet model in defining the pathogenesis of campylobacteriosis and confirm its ability to mimic the disease found in humans. MOLECULAR APPROACHES USED IN CAMPYLOBACTER RESEARCH Although the vectors used to generate defined C. jejuni mutants have been available to investigators for some time (Labigne-Roussel et al., 1987; Yao et al., 1993; Richardson and Park, 1997), the tools to generate random mutations in C. jejuni have only recently become available

104 85 (Bleumink-Pluym et al., 1999; Golden et al., 2000; Colegio et al., 2001; Hendrixson et al., 2001). The development of methods to genetically manipulate the organism and the accessibility of the C. jejuni NCTC genome sequence will likely lead to significant advances in the field of Campylobacter research. The availability of the C. jejuni genome also coincides with advances in bioinformatics, protein chemistry, and microarray technologies. We refer the reader to additional papers on this topic to gain a broader understanding of the implications that sequencing the C. jejuni genome will have on future Campylobacter research (Dorrell et al., 2001; Wren et al., 2001). Methods Used to Generate Defined C. jejuni Mutants The method of allelic replacement is the most frequently used technique to generate defined C. jejuni mutants (see, for example, (Yao and Guerry, 1996; Bacon et al., 2000; Guerry et al., 2002; Pumbwe and Piddock, 2002). This method employs the use of a C. jejuni suicide vector, which by definition can replicate in a heterologous host such as E. coli but cannot replicate in Campylobacter. The procedure involves ligation of a gene into a suicide vector, and then the disruption of this target gene by the insertion of an antibiotic resistance cassette that functions in C. jejuni. Various methods, including high-voltage electroporation, conjugation, or natural transformation, can be used to introduce the suicide vector or just the Campylobacter DNA harboring the modified gene into C. jejuni (Labigne-Roussel et al., 1988; Miller et al., 1988; Wang and Taylor, 1990). If conjugation is to be used to introduce the vector from E. coli into C. jejuni, it is necessary to use a suicide vector containing additional genetic elements (Labigne-Roussel et al., 1988). Following introduction of the vector into C. jejuni, the bacteria are cultured under nonselective conditions to allow for expression of the antibiotic resistance gene. Transformants are then selected by transferring the electroporation or conjugation mixtures onto medium supplemented with the appropriate antibiotic. Mutation of the chromosomal gene is generated by homologous recombination via a double cross-over event, thereby resulting in incorporation of the allele containing the antibiotic resistance cassette into the chromosome. Disruption of the gene of interest can be confirmed by Southern hybridization

105 86 analysis using a gene-specific probe or PCR using gene-specific primers. Although allelic replacement has been used extensively to generate C. jejuni mutants, a disadvantage of the method is the use of an intermediate host such as E. coli, in which the gene of interest is cloned in its entirety. This has been demonstrated to be an obstacle in some instances because the C. jejuni gene may be toxic when expressed in E. coli. As an alternative to allelic replacement, one strategy to generate a defined mutant involves disrupting a gene by a second type of insertional inactivation (see, for example, (Konkel et al., 1997; Konkel et al., 1998; Konkel et al., 1999a). More specifically, the chromosomal gene is disrupted by homologous recombination via a single cross-over event between itself and an internal fragment of the target gene (constructed to have deletions at both the 5' and the 3' end of the coding region of the gene) on a suicide vector. In this strategy, the recombinant suicide vector is introduced into C. jejuni by electroporation or conjugation. A single cross-over event creating two defective alleles of the target gene is generated in the chromosome. Insertional mutants are initially identified by the acquisition of antibiotic resistance located on the suicide vector. The specific mutation can be confirmed by either Southern hybridization analysis using a gene-specific probe or PCR using gene-specific primers. Assessment of gene function in C. jejuni is most often performed by comparing the phenotypes of a mutant with the wild-type isolate. However, insertion mutagenesis as described previously can result in disruption of downstream genes located in the same operon (polar effect). To alleviate this concern, investigators have complemented the modified gene by introducing a functional copy of the gene into the mutated isolate in trans using a shuttle vector. The term shuttle vector is used to define plasmids that are capable of replicating in C. jejuni and an heterologous host such as E. coli. The first Campylobacter shuttle vector was generated by Labigne-Roussel et al. (Labigne-Roussel et al., 1987). Today, Campylobacter spp. shuttle vectors are available that harbor a variety of antibiotic resistance genes and other markers (Wang and Taylor, 1990; Yao et al., 1993; Yao and Guerry, 1996; Park, 1999; Miller et al., 2000). Some of these vectors also harbor the genes allowing for their introduction into Campylobacter organisms via conjugation.

106 87 An alternative technique for generating nonpolar, in-frame mutations has recently been described by Hendrixson et al. (2001) and has its basis in two previously published methods (Higuchi, 1990; Pei and Blaser, 1993; Skorupski and Taylor, 1996). The technique was demonstrated by the disruption of flia and rpon, which encode specialized sigma factors, as well as in-frame deletions in ceta (Cj1190c) and cetb (Cj1189c), which are proposed to be involved in an energy taxis response. Several separate cloning steps were required for this method. Initially, a streptomycin-resistant mutant of C. jejuni strain NCTC was selected on a medium containing a gradient of streptomycin. The rpsl gene, conferring streptomycin-resistance, was amplified from the mutant by the polymerase chain reaction and cloned into the suicide vector puc19. A C. jejuni rpslsm mutant was generated by allellic exchange (double cross-over) of this fragment, selecting for streptomycin-resistant mutants. Following the generation of the C. jejuni rpslsm, DNA fragments containing the entire coding region of flia, rpon, ceta, and cetb, as well as 500 bp of upstream and downstream flanking DNA, were PCR amplified from C. jejuni and cloned individually into a suicide plasmid (puc19 or pbr322). Target genes were disrupted by insertion of a cat-rpsl (cat, chloramphenicol-resistance) cassette into an individual target gene harbored within a suicide plasmid. In addition to the chloramphenicol-resistance marker, each of these constructs had the NCTC wild-type (streptomycin-sensitive) rpsl gene. Chloramphenicol-resistant transformants were generated in rpslsm using each recombinant plasmid. Thus, through a double cross-over event, the wild-type target gene was replaced with the mutant allele. Transformants at this stage were referred to as "intermediate strains" with respect to their streptomycin sensitivity. Intermediate level of resistance to streptomycin is the result of the recessive nature of the rpslsm mutation. The next step of the procedure involved generating additional suicide vectors harboring in-frame deletions of each of the target genes. The in-frame deletions were constructed using the SOEing reaction (splicing by overlap extension). The SOEing reaction involves amplifying 5' and 3' fragments of a gene by PCR, annealing of the two amplified

107 88 fragments, and a second round of PCR amplification to generate fusions of the upstream and downstream DNA segments of each gene (Higuchi, 1990). Products of the SOEing reactions were cloned into a suicide plasmid, and the resultant recombinant plasmids electroporated into the corresponding C. jejuni rpslsm intermediate strain. The transformants, in which allellic replacement had again occurred, were identified based on their resistance to streptomycin (loss of the NCTC rpsl gene) and sensitivity to chloramphenicol. Each mutation was confirmed by PCR using gene-specific primers. Methods Used to Generate Random C. jejuni Mutants (Transposons) None of the previous methods described allow for the random mutagenesis of the C. jejuni genome. This lack has severely hampered the genetic analysis of pathogenesis in this organism. Golden et al. (2000) was the first to demonstrate random transposon mutagenesis in C. jejuni using a mariner-based transposon (Golden et al., 2000). The construction of pothm mini-transposon mariner-based transposon vector, which is unable to replicate in C. jejuni, involved replacement of the resident chloramphenicol resistance cassette with one that would be functional in both E. coli and C. jejuni. In addition, a C. jejuni-specific promoter was cloned upstream of the Himar1 transposase gene to drive its expression in C. jejuni. Electroporation of C. jejuni with the pothm vector typically resulted in chloramphenicol-resistant colonies. Southern blot analysis of nineteen transformants concluded that random insertion within the chromosome of C. jejuni had indeed occurred. Additionally, twelve mutants were selected and the mini-transposon chromosomal junctions were sequenced to map the sites of insertion of the cat gene. Transposon insertion appeared random in the C. jejuni genome without a site preference other than the invariant TA dinucleotide directing insertion. Hendrixson et al. (2001) also generated random mutants in C. jejuni using a derivative of the mariner-based transposon (Hendrixson et al., 2001). Two mini-transposons were constructed, termed pfalcon (Kanr) and penterprise (Cmr). In contrast to Golden et al., the

108 89 transposition reactions were performed in vitro with the pfalcon and penterprise transposons and purified chromosomal DNA from C. jejuni strain , whereby the purified Himar1 MarC9 transposase enzyme was added separately to each reaction. The transposed chromosomal DNA was subsequently introduced into C. jejuni by biphasic natural transformation. Typically, 600-1,450 mutants were obtained per transposition reaction with both the pfalcon and penterprise transposons. The investigators also demonstrated the functionality of the pfalcon mutagenesis system by identifying mutants defective in motility and sequencing flanking regions within the chromosome by inverse PCR. The sequences obtained were used to search the C. jejuni NCTC genome to determine the location of transposition events. Mutations were identified in genes encoding flagellar structural components, motor proteins, chemotaxis proteins, and proteins involved in flagellar gene transcription (flia and rpon). Colegio et al. (2001) developed an in vitro transposon system for the generation of random mutants in C. jejuni utilizing a strategy based on the Staphylococcus aureus transposon Tn552 (Colegio et al., 2001). This transposon requires the coding of a single-subunit transposase and a single accessory protein for in vivo transposition. The Tn552 transposon has an advantage in that a transposition event requires only the 48-bp terminal inverted repeats. Additionally, Tn552 also displays no target site preference for insertion. A derivative of Tn552 was constructed by replacing the cat cassette with an apha-3 kanamycin resistance gene. The resultant plasmid, harboring Tn552 with the new apha-3 gene, was referred to as psb1698. Transpositions Transposition reactions were initially performed in vitro using a protocol similar to that used by Hendrixson et al. Purified chromosomal DNA from C. jejuni was mixed with agarose gel-purified Tn552 harboring the apha-3 gene from psb1698 and the purified His-tagged TnpA transposase enzyme. The reaction mixture was allowed to incubate at 37ºC for one hour, ethanol precipitated, and then introduced into C. jejuni via electroporation. Approximately one hundred kanamycin-resistant mutants were recovered from the reaction. The relatively low recovery of mutants was hypothesized to be a result

109 90 of restriction of the E. coli-grown psb1698 transposon vector. Therefore, the Tn552 transposon with apha-3 kanamycin resistance gene was cloned into a Campylobacter shuttle plasmid, resulting in plasmid psb1699. The psb1699 shuttle plasmid was introduced into C. jejuni via conjugation, and then purified. The transposition reactions were subsequently performed with C. jejuni chromosomal DNA, agarose gel-purified Tn552 harboring the apha-3 gene from psb1699, and the purified His-tagged TnpA transposase enzyme. Approximately 3,000-8,000 C. jejuni kanamycin-resistant colonies were obtained per reaction. The insertions appeared at random sites within the C. jejuni genome as judged by sequencing and searching the C. jejuni NCTC chromosome with the sequences obtained. The investigators also identified nine nonmotile mutants by screening 205 of the kanamycin-resistant colonies. Sequence data from six of the nine nonmotile mutants revealed transposon insertions in genes known or thought to be associated with motility (flad, chea, flip, fliy, rpon, and flge). The remaining three mutants had insertions in genes of unknown function. NOVEL MOLECULAR APPROACHES As stated previously, the availability of the C. jejuni genome sequence will have a significant impact on our understanding of Campylobacter biology because it provides new investigative approaches. For example, a comparison of C. jejuni strains by whole genome-microarray analysis revealed extensive genetic diversity (Dorrell et al., 2001). Based on the comparison of eleven C. jejuni strains, 21 percent of the genes in the C. jejuni NCTC 1168 sequence strain were proposed to be dispensable because they were either absent or highly divergent among the other isolates included in the study. Dorrell et al. (2001) also noted that many of the virulence genes identified to date are conserved in C. jejuni strains (Dorrell et al., 2001). Included among the strain-variable genes are those that are involved in iron acquisition, DNA restriction/modification, sialylation, flagellar biosynthesis, lipo-oligosaccharide biosynthesis, and capsular biosynthesis (Parkhill et al., 2000;

110 91 Wren et al., 2001). The investigators proposed that the variable genes may encode factors that contribute to different disease presentations and that allow the organism to establish unique ecological niches.

111 92 SUMMARY In the past few years, significant advances have been made in the tools used to generate random C. jejuni mutants. The ability to generate large numbers of random C. jejuni mutants will accelerate the rate of research progress in this area, ultimately resulting in a better understanding of the biology of Campylobacter organisms. It is also likely that our understanding of the pathogenic mechanisms utilized by C. jejuni organisms will increase given that researchers have access to the C. jejuni genome sequence and advances in bioinformatics, protein chemistry, and microarray technologies. Still needed is the development of a small-animal model that can be widely utilized by Campylobacter researchers. Given the prevalence of C. jejuni infections worldwide and the interest in understanding the biology of this organism, the next few years will certainly prove to be an exciting time in the field of Campylobacter research.

112 93 REFERENCES Adak, G.K., Cowden, J.M., Nicholas, S. and Evans, H.S The Public Health Laboratory Service national case-control study of primary indigenous sporadic cases of Campylobacter infection. Epidemiol. Infect. 115: Allos, B.M Campylobacter jejuni infections: Update on emerging issues and trends. Clin. Infect. Dis. 32: Allos, B.M. and Blaser, M.J Campylobacter jejuni and the expanding spectrum of related infections. Clin. Infect. Dis. 20: Alrutz, M.A. and Isberg, R.R Involvement of focal adhesion kinase in invasin-mediated uptake. Proc. Natl. Acad. Sci. USA 95: Altekruse, S.F., Stern, N.J., Fields, P.I. and Swerdlow, D.L Campylobacter jejuni an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Altekruse, S.F., Swerdlow, D.L. and Stern, N.J Microbial food borne pathogens: Campylobacter jejuni. Vet. Clinics N. Am.:Food Animal Pract. 14: Anonymous Surveillance data. N. Z. Public Health Rept. 8: 6. Babakhani, F.K., Bradley, G.A. and Joens, L.A Newborn piglet model for campylobacteriosis. Infect. Immun. 61: Bacon, D.J., Alm, R.A., Burr, D.H., Hu, L., Kopecko, D.J., Ewing, C.P., Trust, T.J. and Guerry, P Involvement of a plasmid in virulence of Campylobacter jejuni Infect. Immun. 68: Banfi, E., Cinco, M. and Zabucchi, G Phagocytosis of Campylobacter jejuni and C. coli by peritoneal macrophages. J. Gen. Microbiol. 132: Beery, J.T., Hugdahl, M.B. and Doyle, M.P Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Appl. Environ. Microbiol. 54: Bell, J.A. and Manning, D.D Evaluation of Campylobacter jejuni colonization of the domestic ferret intestine as a model of proliferative colitis. Am. J. Vet. Res. 52: Black, R.E., Levine, M.M., Clements, M.L., Hughes, T.P. and Blaser, M.J Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157: Black, R.E., Perlman, D., Clements, M.L., Levine, M.M. and Blaser, M.J Human volunteer studies with Campylobacter jejuni. In: Nachamkin, I., Blaser, M.J. and Tompkins, L. (eds.), Campylobacter jejuni: current status and future trends. American Society for Microbiology, Washington, D.C., pp Blaser, M.J Epidemiologic and clinical features of Campylobacter jejuni infections. J. Infect. Dis. 176(Supplement 2): S103-S105.

113 94 Blaser, M.J., Berkowitz, I.D., LaForce, F.M., Cravens, J., Reller, L.B. and Wang, W.L Campylobacter enteritis: clinical and epidemiological features. Ann. Intern. Med. 91: Blaser, M.J., Parsons, R.B. and Wang, W.L Acute colitis caused by Campylobacter fetus subsp. jejuni. Gastroenterol. 78: Blaser, M.J., Perez Perez, G., Smith, P.F., Patton, C., Tenover, F.C., Lastovica, A.J. and Wang, W.- l.l Extraintestinal Campylobacter jejuni and Campylobacter coli infections: host factors and strain characteristics. J. Infect. Dis. 153: Blaser, M.J. and Reller, L.B Campylobacter enteritis. N. Engl. J. Med. 305: Blaser, M.J., Wells, J.G., Feldman, R.A., Pollard, R.A. and Allen, J.R Campylobacter enteritis in the United States. A multicenter study. Ann. Intern. Med. 98: Bleumink-Pluym, N.M.C., Verschoor, F., Gaastra, W., van der Zeijst, B.A.M. and Fry, B.N A novel approach for the construction of a Campylobacter mutant library. Microbiol. 145: Boosinger, T.R. and Powe, T.A Campylobacter jejuni infections in gnotobiotic pigs. Am. J. Vet. Res. 49: Boucher, S.N., Slater, E.R., Chamberlain, A.H. and Adams, M.R Production and viability of coccoid forms of Campylobacter jejuni. J. Appl. Bacteriol. 77: Brás, A.M. and Ketley, J.M Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: Brieseman, M.A A further study of the epidemiology of Campylobacter jejuni infections. N. Z. Med. J. 103: Butzler, J.P. and Skirrow, M.B Campylobacter enteritis. Clin. Gastroenterol. 8: Chang, N. and Taylor, D.E Use of pulsed-field agarose gel electrophoresis to size genomes of Campylobacter species and to construct a SalI map of Campylobacter jejuni UA580. J. Bacteriol. 172: Colegio, O.R., Griffin, T.J. 4th., Grindley, N.D. and Galan, J.E In vitro transposition system for efficient generation of random mutants of Campylobacter jejuni. J Bacteriol 183: Colgan, T., Lambert, J.R., Newman, A. and Luk, S.C Campylobacter jejuni enterocolitis. A clinicopathologic study. Arch. Pathol. Lab. Med. 104: Cowden, J Campylobacter: epidemiological paradoxes. Br. Med. J. 305: Cruz, N., Lu, Q., Alvarez, X. and Deitch, E.A Bacterial translocation is bacterial species dependent: results using the human Caco-2 intestinal cell line. J. Trauma 36: Day, W.A., Sajecki, J.L., Pitts, T.M. and Joens, L.A Role of catalase in Campylobacter jejuni intracellular survival. Infect. Immun. 68:

114 De Groote, M.A., Ochsner, U.A., Shiloh, M.U., Nathan, C., McCord, J.M., Dinaure, M.C., Libby, S.J., Vasquez-Torres, A., Xu, Y. and Fang, F.C Periplasmic superoxide dismutase protects Salmonella from products of phagocyte NADPH-oxidase and nitric oxidase synthase. Proc. Natl. Acad. Sci. USA 94: de Melo, M.A., Gabbiani, G. and Pechère, J.-C Cellular events and intracellular survival of Campylobacter jejuni during infection of HEp-2 cells. Infect. Immun. 57: de Melo, M.A. and Pechère, J.-C Identification of Campylobacter jejuni surface proteins that bind to eucaryotic cells in vitro. Infect. Immun. 58: Dekeyser, P., Gossuin-Detrain, M., Butzler, J.P. and Sternon, J Acute enteritis due to related vibrio: first positive stool cultures. J. Infect. Dis. 125: Deming, M.S., Tauxe, R.V., Blake, P.A., Dixon, S.E., Fowler, B.S., Jones, T.S., Lockamy, E.A., Patton, C.M. and Sikes, R.O Campylobacter enteritis at a university: transmission from eating chicken and from cats. Am. J. Epidemiol. 126: Doig, P., Kinsella, N., Guerry, P. and Trust, T.J. 1996a. Characterization of a post-translational modification of Campylobacter flagellin: identification of a sero-specific glycosyl moiety. Mol. Microbiol. 19: Doig, P., Yao, R., Burr, D.H., Guerry, P. and Trust, T.J. 1996b. An environmentally regulated piluslike appendage involved in Campylobacter pathogenesis. Mol. Microbiol. 20: Dorrell, N., Mangan, J.A., Laing, K.G., Hinds, J., Linton, D., Al-Ghusein, H., Barrell, B.G., Parkhill, J., Stoker, N.G., Karlyshev, A.V., Butcher, P.D. and Wren, B.W Whole genome comparison of Campylobacter jejuni human isolates using a low-cost microarray reveals extensive genetic diversity. Genome Res. 11: Drake, A.A., Gilchrist, M.J., Washington, J.A.n., Huizenga, K.A. and Van Scoy, R.E Diarrhea due to Campylobacter fetus subspecies jejuni. A clinical review of 63 cases. Mayo Clin Proc 56: Duffy, M.C., Benson, J.B. and Rubin, S.J Mucosal invasion in Campylobacter enteritis. Am. J. Clin. Path. 73: DuPont, H.L Antimicrobial-resistant Campylobacter species a new threat to travelers to Thailand. Clin. Infect. Dis. 21: Eberhart-Philips, J., Walker, N., Garrett, N., Bell, D., Sinclair, D., Rainger, W. and Bates, M Campylobacteriosis in New Zealand:results of a case-control study. J. Epidemiol. Community Health 51: Engberg, J., Aaerstrup, F.M., Taylor, D.E., Gerner-Smidt, P. and Nachamkin, I Quinolone and macrolide resistance in Campylobacter jejuni and Campylobacter coli: resistance mechanisms and trends in human isolates. Emerg. Infect. Dis. 7: Everest, P.H., Goossens, H., Butzler, J.P., Lloyd, D., Knutton, S., Ketley, J.M. and Williams, P.H Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J. Med. Microbiol. 37:

115 Fauchere, J.L., Rosenau, A., Veron, M., Moyen, E.N., Richard, S. and Pfister, A Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infect. Immun. 54: Federighi, M., Tholozan, J.L., Cappelier, J.M., Tissier, J.P. and Jouve, J Evidence of noncoccoid viable but non-culturable Campylobacter jejuni cells in microcosm water by direct viable count, CTC-DAPI double staining, and scanning electron microscopy. Food Microbiol. 15: Field, L.H., Underwood, J.L., Payne, S.M. and Berry, L.J Virulence of Campylobacter jejuni for chicken embryos is associated with decreased bloodstream clearance and resistance to phagocytosis. Infect. Immun. 59: Field, L.H., Underwood, J.L., Pope, L.M. and Berry, L.J Intestinal colonization of neonatal animals by Campylobacter fetus subsp. jejuni. Infect. Immun. 33: Finlay, B.B. and Falkow, S Salmonella interactions with polarized human intestinal Caco-2 epithelial cells. J. Infect. Dis. 162: Finlay, B.B., Gumbiner, B. and Falkow, S Penetration of Salmonella through polarized Madin-Darby canine kidney epithelial cell monolayer. J. Cell. Biol. 107: Fox, J.G., Ackerman, J.I., Taylor, N., Claps, M. and Murphy, J.C Campylobacter jejuni infection in the ferret: an animal model of human campylobacteriosis. Am. J. Vet. Res. 48: Gaynor, E.C., Ghori, N. and Falkow, S Bile-induced "pili" in Campylobacter jejuni are bacteria-independent artifacts of the culture medium. Mol. Microbiol. 39: Gibreel, A., Sjögren, E., Kaijser, B., Wretlind, B. and Sköld, O Rapid emergence of high-level resistance to quinolones in Campylobacter jejuni associated with mutational changes in gyra and parc. Antimicrob. Agents Chemother. 42: Glass, R.I., Stoll, B.J., Hug, M.I., Struelens, M.J., Blaser, M.J. and Kibrya, A.K.M.G Epidemiologic and clinical features of endemic Campylobacter jejuni infections in Bangladesh. J. Infect. Dis. 148: Golden, N.J., Camilli, A. and Acheson, D.W.K Random transposon mutagenesis of Campylobacter jejuni. Infect. Immun. 68: Grant, C.C.R., Konkel, M.E., Cieplak, W.J. and Tompkins, L.S Role of flagella in adherence, internalization, and translocation of Campylobacter jejuni in nonpolarized and polarized epithelial cell cultures. Infect. Immun. 61: Guerry, P., Ewing, C.P., Hickey, T.E., Prendergast, M.M. and Moran, A.P Sialylation of lipooligosaccharide cores affects immunogenicity and serum resistance of Campylobacter jejuni. Infect. Immun. 68: Guerry, P., Szymanski, C.M., Prendergast, M.M., Hickey, T.E., Ewing, C.P., Pattarini, D.L. and P., M.A Phase variation of Campylobacter jejuni lipooligosaccharide affects ganglioside mimicry and invasiveness in vitro. Infect. Immun. 70:

116 97 Harris, N.V., Weiss, N.S. and Nolan, C.M The role of poultry and meats in the etiology of Campylobacter jejuni/coli enteritis. Am. J. Public Health 76: Harvey, P., Battle, T. and Leach, S Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Hendrixson, D.R., Akerley, B.J. and DiRita, V.J Transposon mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol. Microbiol. 40: Higuchi, R Recombinant PCR. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J. (eds.), PCR protocols:a guide to methods and applications. Academic Press, Inc., San Diego, CA, pp Hodgson, A.E., McBride, B.W., Hudson, M.J., Hall, G. and Leach, S.A Experimental Campylobacter infection and diarrhoea in immunodeficient mice. J. Med. Microbiol. 47: Hopkins, R.S., Olmsted, R. and Istre, G.R Endemic Campylobacter jejuni infection in Colorado: identified risk factors. Am. J. Public Health 74: Humphrey, T.J. and Hart, R.J Campylobacter and Salmonella contamination of unpasteurized cows' milk on sale to the public. J. Appl. Bacteriol. 65: Isberg, R.R. and Tran Van Nhieu, G Two mammalian cell internalization strategies used by pathogenic bacteria. Annu. Rev. Genet. 28: Jensen, L.B. and Aarestrup, F.M Macrolide resistance in Campylobacter coli of animal origin in Denmark. Antimicrob. Agents Chemother. 45: Jin, S., Joe, A., Lynett, J., Hani, E.K., Sherman, P. and Chan, V.L JlpA, a novel surfaceexposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Kapperud, G. and Aasen, S Descriptive epidemiology of infections due to thermotolerant Campylobacter spp. in Norway, Acta Pathol. Microbiol. Immunol. Scand. 100: Karlyshev, A.V., Henderson, J., Ketley, J.M. and Wren, B.W An improved physical and genetic map of Campylobacter jejuni NCTC (UA580). Microbiol. 144: Karmali, M.A. and Fleming, P.C Campylobacter enteritis in children. J. Pediatr. 94: Kelle, K., Pages, J.M. and Bolla, J.M A putative adhesin gene cloned from Campylobacter jejuni. Res. Microbiol. 149: Kiehlbauch, J.A., Albach, R.A., Baum, L.L. and Chang, K.-P Phagocytosis of Campylobacter jejuni and its intracellular survival in mononuclear phagocytes. Infect. Immun. 48: Konkel, M.E., Babakhani, F. and Joens, L.A Invasion-related antigens of Campylobacter jejuni. J. Infect. Dis. 162:

117 Konkel, M.E. and Cieplak, W., Jr Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M.E., Corwin, M.D., Joens, L.A. and Cieplak, W., Jr. 1992a. Factors that influence the interaction of Campylobacter jejuni with cultured mammalian cells. J. Med. Microbiol. 37: Konkel, M.E., Garvis, S.G., Tipton, S.L., Anderson, D.E., Jr. and Cieplak, W., Jr Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M.E., Gray, S.A., Kim, B.J., Garvis, S.G. and Yoon, J. 1999a. Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadf virulence gene and its product. J. Clin. Microbiol. 37: Konkel, M.E., Hayes, S.F., Joens, L.A. and Cieplak, W., Jr. 1992b. Characteristics of the internalization and intracellular survival of Campylobacter jejuni in human epithelial cell cultures. Microbial Path. 13: Konkel, M.E. and Joens, L.A Adhesion to and invasion of HEp-2 cells by Campylobacter spp. Infect. Immun. 57: Konkel, M.E., Kim, B.J., Klena, J.D., Young, C.R. and Ziprin, R Characterization of the thermal stress response of Campylobacter jejuni. Infect. Immun. 66: Konkel, M.E., Kim, B.J., Rivera-Amill, V. and Garvis, S.G. 1999b. Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32: Konkel, M.E., Mead, D.J. and Cieplak, W., Jr Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Konkel, M.E., Mead, D.J., Hayes, S.F. and Cieplak, W., Jr. 1992c. Translocation of Campylobacter jejuni across human polarized epithelial cell monolayer cultures. J. Infect. Dis. 166: Kops, S.K., Lowe, D.K., Bement, W.M. and West, A.B Migration of Salmonella typhi through intestinal epithelial monolayers: an in vitro study. Microbiol. Immunol. 40: Kuschner, R.A., Trofa, A.F., Thomas, R.J., Hoge, C.W., Pitarangsi, C., Amato, S., Olafson, R.P., Echeverria, P., Sadoff, J.C. and Taylor, D.N Use of azithromycin for the treatment of Campylobacter enteritis in travelers to Thailand, an area where ciprofloxacin resistance is prevalent. Clin. Infect. Dis. 21: Labigne-Roussel, A., Courcoux, P. and Tompkins, L Gene disruption and replacement as a feasible approach for mutagenesis of Campylobacter jejuni. J. Bacteriol. 170: Labigne-Roussel, A., Harel, J. and Tompkins, L Gene transfer from Escherichia coli to Campylobacter species: development of shuttle vectors for genetic analysis of Campylobacter jejuni. J. Bacteriol. 169:

118 99 Lambert, M., Marion, E., Coche, E. and Butzler, J.P Campylobacter enteritis and erythema nodosum. Lancet i: Lambert, M.E., Schofield, P.F., Ironside, A.G. and Mandal, B.K Campylobacter colitis. Br. Med. J. 1: Macnab, R.M The bacterial flagellum: reversible rotary propeller and type III export apparatus. J. Bacteriol. 181: Madara, J.L Regulation of the movement of solutes across tight junctions. Annu. Rev. Physiol. 60: McKendrick, M.W., Geddes, A.M. and Gearty, J Campylobacter enteritis: a study of clinical features and rectal mucosal changes. Scand. J. Infect. Dis. 14: McNicholas, A.M., Bates, M., Kiddle, E. and Wright, J Is New Zealand's recent increase in campylobacteriosis due to changes in laboratory procedures? A survey of 69 medical laboratories. N. Z. Med. J. 108: McSweegan, E. and Walker, R.I Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infect. Immun. 53: Mee, A.S., Shield, M. and Burke, M Campylobacter colitis: differentiation from acute inflammatory bowel disease. J. R. Soc. Med. 78: Miller, J.F., Dower, W.J. and Tompkins, L.S High-voltage electroporation of bacteria: genetic transformation of Campylobacter jejuni with plasmid DNA. Proc. Natl. Acad. Sci. USA 85: Miller, W.G., Bates, A.H., Horn, S.T., Brandl, M.T., Wachtel, M.R. and Mandrell, R.E Detection on surfaces and in Caco-2 cells of Campylobacter jejuni cells transformed with new gfp, yfp, and cfp marker plasmids. Appl. Environ. Microbiol. 66: Mishu, B., Ilyas, A.A., Koski, C.L., Vriesendrop, F., Cook, S.D., Mithen, F.A. and Blaser, M.A Serological evidence of previous Campylobacter jejuni infection in patients with the Guillain- Barrè syndrome. Ann. Intern. Med. 118: Moser, I., Schroeder, W. and Salnikow, J Campylobacter jejuni major outer membrane protein and a 59-kDa protein are involved in binding to fibronectin and INT 407 cell membranes. FEMS Microbiol. Lett. 157: Myszewski, M.A. and Stern, N.J Phagocytosis and intracellular killing of Campylobacter jejuni by elicited chicken peritoneal macrophages. Avian Dis. 35: Nachamkin, I., Allos, B.M. and Ho, T Campylobacter species and Guillian-Barré syndrome. Clin. Microbiol. Rev. 11: Nataro, J.P., Hicks, S., Phillips, A.D., Vial, P.A. and Sears, C.L T84 cells in culture as a model for enteroaggregative Escherichia coli pathogenesis. Infect. Immun. 64:

119 Newell, D.G., McBride, H. and Dolby, J.M. 1985a. Investigations on the role of flagella in the colonization of infant mice with Campylobacter jejuni to human epithelial cell lines. J. Hyg. 95: Newell, D.G., McBride, H., Saunders, F., Dehele, Y. and Pearson, A.D. 1985b. The virulence of clinical and environmental isolates of Campylobacter jejuni. J. Hyg. 94: Oelschlaeger, T.A., Guerry, P. and Kopecko, D.J Unusual microtubule-dependent endocytosis mechanisms triggered by Campylobacter jejuni and Citrobacter freundii. Proc. Natl. Acad. Sci. USA 90: Panigrahi, P., Losonky, G., DeTolla, L.J. and Morris, J.G Human immune response to Campylobacter jejuni proteins expressed in vitro. Infect. Immun. 60: Park, S.F The use of hipo, encoding benzoylglycine amidohydrolase (hippuricase), as a reporter of gene expression in Campylobacter coli. Lett. Appl. Microbiol. 28: Parkhill, J., Wren, B.W., Mungall, K., Ketley, J.M., Churcher, C., Basham, D., Chillingworth, T., Davies, R.M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A.V., Moule, S., Pallen, M.J., Penn, C.W., Quall, M.A., Rajandrean, M.A., Rutherford, K.M., Van Vliet, A.H.M., Whitehead, S. and Barrell, B.G The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403: Pavlovskis, O.R., Rollins, D.M., Haberberger, R.L.J., Green, A.E., Habash, L., Strocko, S. and Walker, R.I Significance of flagella in colonization resistance of rabbits immunized with Campylobacter spp. Infect. Immun. 59: Pebody, R.G., Ryan, M.J. and Wall, P.G Outbreaks of Campylobacter infection: rare events for a common pathogen. Commun. Dis. Rep. CDR Rev. 7: R33-R37. Pei, Z. and Blaser, M.J PEB1, the major cell-binding factor of Campylobacter jejuni, is a homolog of the binding component in Gram-negative nutrient transport systems. J. Biol. Chem. 268: Pei, Z., Burucoa, C., Grignon, B., Baqar, S., Huang, X.-Z., Kopecko, D.J., Bourgeois, A.L., Fauchere, J.-L. and Blaser, M.J Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Pope, L.M., Reed, K.E. and Payne, S.M Increased protein secretion and adherence to HeLa cells by Shigella spp. following growth in the presence of bile salts. Infect. Immun. 63: Potter, M.E., Blaser, M.J., Sikes, R.K., Kaufmann, A.F. and Wells, J.G Human Campylobacter infection associated with certified raw milk. Am. J. Epidemiol. 117: Price, A.B., Jewkes, J. and Sanderson, P.J Acute diarrhoea: Campylobacter colitis and the role of rectal biopsy. J. Clin. Path. 32: Pumbwe, L. and Piddock, L.J.V Identification and molecular characterisation of CmeB, a Campylobacter jejuni multidrug efflux pump. FEMS Microbiol. Lett. 206:

120 Richardson, P.T. and Park, S.F Integration of heterologous plasmid DNA into multiple sites on the genome of Campylobacter coli following natural transformation. J. Bacteriol. 179: Rivera-Amill, V., Kim, B.J., Seshu, J. and Konkel, M.E Secretion of the virulence associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J. Infect. Dis. 183: Roberts, D.D Interactions of respiratory pathogens with host cell surface and extracellular matrix components. Am. J. Respir. Cell Mol. Biol. 3: Robinson, D.A Infective dose of Campylobacter jejuni in milk. Br. Med. J. 282: Russell, R.G., O'Donnoghue, M., Blake, D.C., Jr., Zulty, J. and DeTolla, L.J Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Salloway, S., Mermel, L.A., Seamans, M., Aspinall, G.O., Shin, J.E.N., Kurjanczyk, L.A. and Penner, J.L Miller-Fisher syndrome associated with Campylobacter jejuni bearing lipopolysaccharide molecules that mimic human ganglioside GD3. Infect. Immun. 64: Savill, M.G., Hudson, J.A., Ball, A., Klena, J.D., Scholes, P., Whyte, R.J., McCormick, R.E. and Jankovic, D Enumeration of Campylobacter in New Zealand recreational and drinking waters. J. Appl. Microbiol. 91: Schröder, W. and Moser, I Primary structure analysis and adhesion studies on the major outer membrane protein of Campylobacter jejuni. FEMS Microbiol. Lett. 150: Shanker, S., Lee, A. and Sorrell, T.C Horizontal transmission of Campylobacter jejuni amongst broiler chicks: experimental studies. Epidemiol. Infect. 104: Silverstein, S.C., Steinman, R.M. and Cohn, Z.A Endocytosis. Annu. Rev. Biochem. 46: Skirrow, M.B Campylobacter enteritis: a "new" disease. Br. Med. J. 2: Skirrow, M.B A demographic survey of Campylobacter, Salmonella and Shigella infections in England. A Public Health Laboratory Service Survey. Epidemiol. Infect. 99: Skirrow, M.B Epidemiology of Campylobacter enteritis. Int. J. Food Microbiol. 12: Skorupski, K. and Taylor, R.K Positive selection vectors for allelic exchange. Gene 169: Smith, K.E., Besser, J.M., Hedberg, C.W., Leano, F.T., Bender, J.B., Wicklund, J.H., Johnson, B.P., Moore, K.A. and Osterholm, M.T Quinolone resistant Campylobacter jejuni infections in Minnesota, N. Eng. J. Med. 340: Svedhem, A. and Kaijser, B Campylobacter fetus subspecies jejuni: a common cause of diarrhea in Sweden. J. Infect. Dis. 142: Takeuchi, A Electron microscopic studies of experimental Salmonella infection. Penetration into the intestinal epithelium by Salmonella typhimurium. Am. J. Pathol. 50:

121 Tauxe, R.V Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Nachamkin, I., Blaser, M.J. and Tompkins, L. (eds.), Campylobacter jejuni:current status and future trends. American Society for Microbiology, Washington, D.C., pp Tauxe, R.V., Hargrett-Bean, N., Patton, C.M. and Wachsmuth, I.K Campylobacter isolates in the United States, MMWR CDC Surveill. Summ. 37: Taylor, D.J. and Olubunmi, P.A A re-examination of the role of Campylobacter fetus subspecies coli in enteric disease of the pig. Vet. Rec. 109: van den Bogaard, A.E. and Stobberingh, E.E Epidemiology of resistance to antibiotics: links between animals and humans. Int. J. Antimicrob. Agents 14: Van Spreeuwel, J.P., Duursma, G.C., Meijer, C.J.L.M., Bax, R., Rosekrans, P.C.M. and Lindeman, J Campylobacter colitis: histological immunohostochemical and ultrastructural findings. Gut 26: Vogt, R.L., Sours, H.E., Barrett, T., Feldman, R.A., Dickinson, R.J. and Witherell, L Campylobacter enteritis associated with contaminated water. Ann. Int. Med. 96: Vuckovic, D., Abram, M. and Doric, M Primary Campylobacter jejuni infection in different mice strains. Microb. Path. 24: Wang, Y. and Taylor, D.E Natural transformation in Campylobacter species. J. Bacteriol. 172: Wassenaar, T.M., Bleumink-Pluym, N.M. and van der Zeijst, B.A Inactivation of Campylobacter jejuni flagellin genes by homologous recombination demonstrates that flaa but not flab is required for invasion. EMBO J. 10: Wassenaar, T.M., Engelskirchen, M., Park, S. and Lastovica, A Differential uptake and killing potential of Campylobacter jejuni by human peripheral monocytes/macrophages. Med. Microbiol. Immunol. 186: Wassenaar, T.M., van der Zeijst, B.A., Ayling, R. and Newell, D.G Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol. 139: Weisblum, B Erythromycin resistance by ribosome modification. Antimicrob. Agents Chemother. 39: Wren, B.W., Linton, D., Dorrell, N. and Karlyshev, A.V Post genome analysis of Campylobacter jejuni. J. Appl. Microbiol. 90: 36S-44S. Yao, R., Alm, R.A., Trust, T.J. and Guerry, P Construction of new Campylobacter cloning vectors and a new mutational cat cassette. Gene 130: Yao, R., Burr, D.H., Doig, P., Trust, T.J., Niu, H. and Guerry, P Isolation of motile and nonmotile insertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukaryotic cells. Mol. Microbiol. 14:

122 103 Yao, R., Burr, D.H. and Guerry, P CheY-mediated modulation of Campylobacter jejuni virulence. Mol. Microbiol. 23: Yao, R. and Guerry, P Molecular cloning and site-specific mutagenesis of a gene involved in arylsulfatase production in Campylobacter jejuni. J. Bacteriol. 178: Young, G.M., Schmiel, D.H. and Miller, V.L A new pathway for the secretion of virulence factors by bacteria: the flagellar export apparatus functions as a protein-secretion system. Proc. Natl. Acad. Sci. USA 96: Ziprin, R.L., Young, C.R., Hume, M.E. and Konkel, M.E The absence of cecal colonization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectin-binding protein. Avian Dis. 43:

123 Figure 1. Diagram showing the adherence and invasion assay. 104

124 105 Figure 1. Suspension of Bacteria Cell-Monolayer 24-well plate Non-Adherent Bacteria Centrifuge Rinse 2-3 Hr Lyse Epithelial Cells Serial Dilutions Quantify Cell-Associated Bacteria Rinse Add Gentamicin Rinse 3 Hr Lyse Epithelial Cells Serial Dilutions Quantify Internalized Bacteria Internalized Bacteria Survive Treatment

125 106 Figure 2. Diagram showing cross-section of a polarized cell monolayer. The paracellular and transcellular routes of translocation are indicated (arrows). TER = Transepithelial electrical resistance.

126 107 Figure 2. TER Apical Chamber Translocation Paracellular Transcellular (intracellular) Tight Junction Brush Border Permeable Membrane Basal Chamber

127 108 Figure 3. Transmission electron micrograph of J774A.1 macrophage infected with C. jejuni M129. The sample was processed 6 hr post-infection. After sectioning, the sample was stained with aqueous uranyl acetate and lead citrate, and examined with a JEOL electron microscope (model CX2) at 80V. Bar = 5 µm.

128 Figure

129 110 Figure 4. Hematoxylin and eosin stained section of the small intestine of a C. jejuni-inoculated piglet. Arrow indicates an area of luminal exudate and hyperemia. Also note the shortening of the villous epithelium due to necrosis. Bar = 100 µm.

130 Figure

131 Chapter Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface Marshall R. Monteville and Michael E. Konkel* School of Molecular Biosciences, Washington State University, Pullman, WA Published in Infection and Immunity 2002, 70(12):

132 113 ABSTRACT Previous studies have indicated that the ability to bind to fibronectin is a key feature in successful cell invasion by Campylobacter jejuni. Given the spatial distribution of fibronectin and the architecture of the epithelium, this suggests the possibility that C. jejuni cell invasion might preferentially occur at the basolateral cell surface. To test this hypothesis, we examined the interaction of C. jejuni with T84 human colonic cells. When grown under the appropriate conditions, T84 cells form a polarized cell monolayer. C. jejuni translocation of a T84 cell monolayer appeared to occur via a paracellular (extracellular) route as opposed to a transcellular (intracellular) route based on the finding that a C. jejuni non-invasive mutant translocated as efficiently as its isogenic parent. Additional studies revealed that two distinct C. jejuni wild-type isolates could compete with one another for host cell receptors, whereas a C. jejuni fibronectin-binding deficient mutant could not compete with a wild-type isolate for host cell receptors. Further, C. jejuni adherence and internalization were significantly inhibited by anti-fibronectin antibodies, but only when cells were first treated with EGTA to expose basolateral cell surfaces. Together, these results support the theory that C. jejuni invasion occurs preferentially at the basolateral surface of eukaryotic cells.

133 114 INTRODUCTION Campylobacter jejuni is one of the leading causes of human gastrointestinal disease in the United States (1, 2). The ability of C. jejuni to cause disease is dependent upon multiple factors including motility (6, 30, 45), chemotaxis (39, 46), host cell-translocation (7, 11, 14, 24), host cell-adherence (17, 20, 33), host cell-invasion (11, 18, 23, 36), and toxin production (34, 44). Of particular significance to the present study is whether C. jejuni translocation, or migration across the intestinal epithelium, is an important virulence attribute since the pathology of Campylobacter-mediated enteritis is generally confined to the intestinal epithelium. Cell adherence by C. jejuni is multifactorial, with a number of adhesins identified. The best characterized C. jejuni adhesins to date include CadF, JlpA, and PEB1 (17, 20, 33). With one exception, the targets of these binding proteins remain unknown. The target of the CadF adhesin is fibronectin (Fn), a component of the extracellular matrix (ECM) (20). Fibronectin appears to be a common host cell target as numerous pathogens, including C. jejuni (20, 27), Staphylococcus aureus (26, 37), Streptococcus pyogenes (16, 31), Salmonella enteritidis (5), Escherichia coli (13, 43), Neisseria gonorrhoeae (42), Mycobacterium avium (38), and Treponema species (9, 10, 41), possess Fn-binding ability. To date, the in vitro studies performed to determine the role of CadF, and all other C. jejuni adhesins, have been limited to the use of non-polarized cells. Unfortunately, the architecture of cells grown on a plastic substrate differs substantially from that of cells in vivo, where Fn is localized to the basolateral cell surface. While the intestinal epithelium provides a primary defense against invading organisms, several pathogenic bacteria possess the ability to translocate an epithelial or endothelial cell barrier (12, 25). Such translocation is an important virulence attribute as it allows the invader access to underlying tissues and may permit the organism to disseminate throughout the host. The Caco-2, HT29, and

134 115 T84 human colonic cell lines posses the ability to form polarized cell monolayers when grown under appropriate conditions, thereby affording a model to assess the ability of bacteria to translocate across an intact epithelial cell barrier (8). Polarized cells are characterized by defined apical and basolateral cell surfaces separated by tight junctions, which limit the passage of solutes through the paracellular spaces (28). Transepithelial electrical resistance (TER) is frequently used as an index of tight junction permeability and monolayer integrity. Disruption of the intercellular tight junctions results in a decrease in TER. Previous work has revealed that C. jejuni can translocate a Caco-2 polarized cell monolayer without a concomitant loss in TER (11, 14, 24), indicating that C. jejuni can translocate across a cell monolayer whose integrity remains intact. A consensus is yet to be reached among investigators as to the mechanism of translocation. More specifically, whether C. jejuni translocate via a paracellular (migration from the apical to basolateral cell surface by passage between cells) or a transcellular route (migration from the apical to basolateral cell surface by host cell uptake, followed by intracellular trafficking) remains debateable. This study was initiated to further examine the binding, internalization, and translocation properties of C. jejuni using a polarized cell model system. We specifically chose T84 cells for their phenotypic similarity to colonic crypt cells (32). Histological examination of C. jejuni infected humans and animals has revealed pathology primarily in the colon (3, 6, 36). We were thus able to characterize the role of the CadF outer-membrane protein in promoting the interactions of C. jejuni with cells whose architecture models that of the organism s in vivo target.

135 116 MATERIALS AND METHODS Bacterial isolates and growth conditions. C. jejuni F38011, (TetR) (4), and the F38011 isogenic cadf mutant (KanR), ciab mutant (KanR), and ciab transformant harboring the pmek100 shuttle plasmid (TetR and KanR) were cultured on Mueller-Hinton (MH) agar plates supplemented with antibiotics (12.5 µg/ml of tetracycline, 200 µg/ml of kanamycin) under microaerophilic conditions at 37 C. The pmek100 shuttle plasmid contains a 2,248-bp fragment of DNA harboring the entire ciab gene from C. jejuni F38011 (35). A C. jejuni F38011 (Strep/NalR) isolate was also cultured on MH agar plates supplemented with 200 µg/ml of streptomycin and 50 µg/ml of nalidixic acid. All isolates were subcultured every 24 to 48 h. Escherichia coli MRF and Salmonella enterica ssp. Typhimurium SL1344 (S. typhimurium) were cultured in Luria-Bertani (LB) broth (10 g of Bacto-tryptone, 5 g of yeast extract, and 10 g of sodium chloride per liter) and on LB agar plates (LB broth with 15 g Bacto-agar per liter) in a 37 C incubator. Tissue culture. Stock cultures of T84 cells (human colonic cell line, ATCC CCL248) were grown in Eagle s Minimal Essential Media (EMEM) supplemented with 10% (v/v) fetal bovine serum (FBS; Hyclone Laboratories Inc., Logan, UT) and maintained at 37 C in a humidified, 5% CO 2 incubator. T84 non-polarized adherence and internalization assays. Adherence and internalization assays were performed using T84 cells in 24-well tissue culture trays (Costar, Cambridge, MA) as outlined elsewhere (19). To determine the number of adherent bacteria, T84 cell monolayer cultures were inoculated with 5 x 107 CFU of a bacterial suspension in pre-warmed EMEM plus 1% FBS (EMEM-1% FBS), centrifuged at 600 x g for 5 min to promote bacteria-host cell contact, and incubated for 30 min in a humidified, 5% CO 2 incubator at 37 C. Monolayers were rinsed three

136 117 times with phosphate-buffered saline (PBS), epithelial cells lysed with a solution of 0.25% (w/v) sodium deoxycholate, and serial dilutions of the lysates plated on MH/blood plates. The number of adherent bacteria was determined by counting the resulting colonies. Internalization assays were performed similarly. Infected cell monolayers were incubated for 3 h, rinsed three times with EMEM-1% FBS, and incubated for an additional 3 h with EMEM-1% FBS and a bactericidal concentration of gentamicin (Invitrogen, Carlsbad, CA). After incubation, cell monolayers were rinsed three times with PBS, T84 cells lysed with 0.25% (w/v) sodium deoxycholate, and dilutions of the lysates plated on MH/blood agar plates. The number of internalized bacteria was determined by counting the resultant colonies. Where indicated, a final concentration of 10 µm of ethylene glycol-bis(β-aminoethyl ether)-n,n,n,n -tetraacetic acid (EGTA, Sigma, St. Louis, MO) was added to the cells 15 min prior to inoculation with C. jejuni to induce cellular retraction. EGTA was removed prior to inoculation of T84 cells with C. jejuni. T84 cell viability, following EGTA treatment, was assessed by rinsing the T84 cells twice with PBS, and staining the cells for 5 min with 0.5% trypan blue. The cells were then rinsed twice with PBS, counter-stained for 5 min with 0.5 ml of 0.5% phenol red, and visualized with an inverted microscope to assess viability (15). Anti-human Fn antibodies (Telios Pharmaceuticals, Inc., San Diego, CA) were used at a 1:50 dilution and assays conducted in EMEM minus FBS. For assays in which S. typhimurium was used, the bacteria were pre-incubated anaerobically at 37 C for 3 h prior to infection. Results are presented as the mean ± standard deviation of the number of viable adherent or internalized bacteria. T84 polarized translocation, adherence and internalization assays. The BIOCOAT Intestinal Epithelium Differentiation Environment was used to prepare polarized T84 cells according to manufacturer s specifications (Becton Dickinson, Bedford, MA). Translocation experiments were performed by adding 5 x 107 CFU (multiplicity of infection (MOI) of approximately 200) in pre-warmed Entero-STIM to the apical chamber of a transwell unit and incubating in a humidified,

137 118 5% CO 2 incubator at 37 C. At time points indicated, the transwells were transferred to a 24-well plate containing fresh pre-warmed Entero-STIM medium. Bacteria which translocated across the cell monolayer into the basolateral medium were enumerated by plating 10-fold serial dilutions of the basolateral medium on MH/blood agar plates. Cell-associated bacteria were quantified by washing the apical and basolateral surfaces of a transwell membrane three times with PBS. Following the washes, the transwell membranes were excised with a sterile scalpel and placed in a solution of 0.25% (w/v) sodium deoxycholate to lyse the T84 cells. Internalized bacteria were quantified in a similar manner with modifications. Following apical and basolateral washes, Entero-STIM medium containing a bactericidal concentration of gentamicin was placed in both the apical and basolateral chambers to kill all extracellular bacteria. Following incubation, the transwell membranes were rinsed with PBS, excised, and internalized bacteria enumerated. To determine if C. jejuni preferentially remain cell-associated verses translocating into the basolateral medium, two identical sets of C. jejuni-inoculated transwell units were used. The first set was used to determine the number of cell-associated C. jejuni at the 4 h time point as outlined above. The second set of samples were identically inoculated, with the apical and basolateral surface of each cell monolayer being washed three times with Entero-STIM at the 4 h time point to remove non-adherent bacteria. The transwell units were then placed in a 24-well plate containing fresh Entero-STIM medium. The number of bacteria present in the basolateral chamber medium was determined after each additional hour of incubation, at which time the transwell units were transferred to a 24-well plate containing fresh Entero-STIM medium. The number of remaining cell-associated C. jejuni was determined at the 7 h time point. Where indicated, EGTA (10 µm) was added 15 min prior to infection to induce cellular retraction of T84 cells. EGTA was removed prior to inoculation of T84 cells with C. jejuni. Anti-human Fn antibody (Telios Pharmaceuticals, Inc., San Diego, CA) was used at a 1:50 dilution with Entero-STIM in the absence of manufacturer-provided serum supplement. Translocation kinetic studies using C. jejuni F38011 and the isogenic cadf mutant

138 119 were performed at an MOI of approximately 3, 30, and 300. E. coli MRF served as a control for integrity of the T84 polarized cell monolayers over the course of most assays. E. coli MRF, inoculated in an identical manner to C. jejuni, was not recovered from the basolateral medium. TER measurements were made using a Millicell-ERS voltohmmeter (Millipore, Bedford, MA). Competitive binding assays. Polarized cell monolayers were prepared as outlined above. C. jejuni (TetR) was added to the apical chamber of the transwell unit alone, or in combination with a 25 to 90-fold excess of C. jejuni F38011 (Strep/NalR) and the C. jejuni F38011 isogenic cadf mutant (KanR). Following a 4 h incubation, transwells were rinsed and membranes excised as indicated above. The number of each cell-associated isolate was determined by diluting lysates and plating bacteria on MH/Blood plates containing the appropriate selective antibiotic. Statistical analysis. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data. Two-tailed P values were determined for each sample, and a P value < 0.01 considered significant.

139 120 RESULTS Translocation across T84 polarized cells is independent of Cia protein secretion. While some investigators have proposed that C. jejuni translocate across polarized cells via a transcellular route, others have proposed a paracellular route of translocation. To address the route of translocation, a binding and internalization experiment was initially performed with a C. jejuni ciab mutant and T84 non-polarized cells as the phenotypic properties of the C. jejuni ciab mutant with these cells had not been assessed previously. Consistent with previous work with INT 407 cells (22), adherence assays revealed no significant differences in the numbers of the C. jejuni F38011 wild-type isolate, ciab mutant, and the ciab mutant transformed with pmek100 bound to the T84 cells (Table 1). However, a significant difference was noted in the invasiveness between the ciab mutant when compared to either the complemented ciab mutant or F38011 isolate (Table 1). Previous work has revealed that CiaB is a protein synthesized and secreted by C. jejuni upon co-cultivation with eukaryotic cells, and that a C. jejuni ciab mutant does not secrete any proteins (22, 35). The present findings suggest that one or more of the Cia proteins are required for maximal invasion of T84 non-polarized cells by C. jejuni. Assays were then performed with the C. jejuni F38011 wild-type isolate and isogenic ciab mutant to determine if C. jejuni translocate a T84 cell monolayer via a paracellular versus a transcellular route. The C. jejuni wild-type isolate and C. jejuni ciab transformant harboring pmek100 were included as controls. All C. jejuni isolates displayed similar kinetics of translocation (Table 2), suggesting a paracellular route of translocation. Given this finding, the adherence and invasive potential of each isolate was assessed. The results obtained with the T84 polarized cells mirrored those obtained with the T84 non-polarized cells with respect to the number of cell-associated and internalized bacteria. No difference was noted in the binding of the C. jejuni ciab mutant when compared to the C. jejuni F38011 wild-type isolate (Table 2, row 2). To assess cell invasion,

140 121 gentamicin was added to both the apical and the basolateral chambers and the C. jejuni-inoculated cell monolayers incubated for an addition 3 h to ensure killing of all extracellular bacteria. In contrast to adherence, differences were observed in the invasiveness of the C. jejuni isolates tested (Table 2, row 3). The invasive potential of the C. jejuni and F38011 clinical isolates was found to be the greatest, followed by the C. jejuni ciab-pmek100 transformant, and then the C. jejuni ciab mutant. These results suggest that Cia proteins are required for the maximal invasion of polarized cells, yet play no role in C. jejuni translocation. Prior to the experiment outlined above, the relationship between C. jejuni invasion and translocation of T84 polarized cells was investigated by determining how many bacteria remain associated with a polarized cell monolayer over the course of a 3 h incubation period. This assay was necessary to assess the invasive potential of C. jejuni with polarized cells as an extended incubation period with gentamicin is required to kill extracellular bacteria. Moreover, if the intracellular bacteria transcytosed the cells within a 3 h time period, an accurate assessment of the number of intracellular bacteria could not have been determined as they would have been exposed to gentamicin-containing medium. For this assay, one set of T84 polarized cells was inoculated with the C. jejuni clinical isolates F38011 and and incubated for 4 h, after which time the monolayers were rinsed and cell-association quantified (Table 3, row 1). A second set of the C. jejuni inoculated polarized cells was rinsed after the 4 h incubation period and placed in a 24-well plate containing fresh media. C. jejuni present in the basolateral media were quantified hourly for an additional 3 h period (Table 3, attributes shown in row 2). Following this incubation, cell-associated C. jejuni were determined (Table 3, row 3). The results of this assay revealed that the majority of the cell-associated C. jejuni remained cell-associated rather than translocating to the medium in the basolateral chamber (Table 3, rows 4 and 5). Indeed, 96.9% of C. jejuni F38011 and 85.2% of C. jejuni that were associated with the polarized cell monolayers after 4 h of incubation remained cell-associated after the additional 3 h incubation period.

141 122 Collectively, these data indicate that: 1) C. jejuni translocate across a cell monolayer via a paracellular route; 2) the Cia proteins play no role in C. jejuni cellular translocation; 3) the Cia proteins are required for the maximal invasion of polarized cells; and 4) the bacteria that become associated with the polarized cells remain cell-associated. Translocation results from saturation of host-cell receptors. Based on the finding that C. jejuni preferentially remain cell-associated (apically or basolaterally) rather than translocating to the medium in the basolateral chamber (Table 3), assays were conducted to compare the translocation kinetics of the C. jejuni F38011 wild-type isolate and the C. jejuni isogenic cadf mutant (20). Experiments were simultaneously performed to determine host-cell association at 4 h at different MOIs. At an MOI of 300, both C. jejuni isolates translocated across the T84 polarized cell monolayers, into the basolateral medium, with approximately the same efficiency. As the MOI was decreased from 300 to 3, a significant decrease was noted in the ability of C. jejuni F38011 to translocate into the basolateral medium compared to the cadf mutant (Fig. 1). We propose that at the lowest MOI, the C. jejuni wild-type isolate does not translocate to the basolateral medium as efficiently as the C. jejuni cadf mutant because the former binds to Fn. The difference in translocation is masked at higher MOIs because the accessible host-cell receptors to which the C. jejuni wild-type isolate binds are occupied. Also noteworthy is that at an MOI of 30, a significant difference was observed in the translocation of the C. jejuni wild-type isolate versus the cadf mutant at the 1 h interval. However, after the first hour of incubation, no significant difference was observed in the translocation kinetics of the two isolates. This finding suggests that Fn is available for C. jejuni to bind throughout the first hour of incubation, after which time all available host-receptor molecules, including Fn, are bound. Consistent with previous work using INT 407 cells (29), a significant decrease in adherence to T84 polarized cells was observed for the C. jejuni cadf mutant, when compared to the C. jejuni wild-type isolate, regardless of MOI (Table 4).

142 123 Competition assays using polarized cells were conducted to further confirm Fn serves as a host-cell receptor for C. jejuni clinical isolates F38011 and It was possible to specifically determine the number of each bacterial isolate associated with the epithelial cells based on their antibiotic resistances. C. jejuni , along with a 25 to 90-fold excess of either C. jejuni F38011 or the isogenic cadf mutant, were used to inoculate the apical surface of the T84 polarized cells. The number of each C. jejuni isolate associated with the monolayer was then determined as outlined in Materials and Methods. C. jejuni was not significantly inhibited from associating with the polarized monolayer in the presence of 50 and 90-fold excess of the cadf mutant (Table 5). In contrast, C. jejuni F38011 significantly reduced the association of C. jejuni with the T84 polarized cells in a dose-dependent manner. Noteworthy, the total number of adherent C. jejuni , at an MOI of approximately 3, was comparable to that previously observed for C. jejuni F38011 at the same MOI (Table 4). These data further support the notion that the binding of C. jejuni to Fn promotes bacterial-host cell association. Invasion of T84 cells occurs primarily via the basolateral surface. Because the two wild-type isolates competed for cell-association (adherence and internalization), additional experiments were performed to determine whether adherence and internalization could be retarded by blocking receptors with anti-fn antibodies. Initial experiments to test this hypothesis were conducted using T84 non-polarized cells. However, the adherence and invasion of T84 non-polarized cells by C. jejuni were only reduced to 81% and 76%, respectively (Table 6), in the presence of the anti-fn antibodies when compared to the control. Because Fn would be expected to predominate at cell-substrate contacts, monolayers were pre-treated with EGTA prior to infection to expose the ECM. Upon treatment of T84 cells with EGTA, C. jejuni F38011 adherence and invasion increased 144% and 288% respectively, when compared to untreated monolayers (Table 6). When anti-fn antibodies were included in the medium following EGTA treatment, the number of adherent and internalized bacteria was significantly reduced from 144% to 50% and 288% to 65%, respectively.

143 124 These data suggest that increased exposure of C. jejuni to Fn promotes maximal host-cell interaction, which in turn leads to increased invasion into non-polarized T84 cells. S. typhimurium was included as a control to ensure that the increase in C. jejuni adherence and invasion did not result directly from an increase in exposed cell surface area following treatment with EGTA. No significant difference in S. typhimurium adherence or invasion was noted following EGTA treatment of host-cells or in the presence of anti-fn antibodies. Furthermore, transient chelation of extracellular divalent cations did not appear to affect host-cell endocytic activity as judged by efficient uptake of S. typhimurium over the course of the assay or T84 cell viability as judged by staining with trypan blue. Identical experiments were also performed with T84 polarized cells to determine the affects of EGTA treatment on C. jejuni adherence and invasion. Anti-Fn antibodies added to the non-egta treated polarized cells had no effect on C. jejuni adherence and invasion, presumably due to the lack of antibody access to Fn localized to the basolateral surface. As observed with non-polarized cells, pre-treatment of the cell monolayers with EGTA resulted in a significant increase in the number of adherent and internalized bacteria to 185% and 367%, respectively (Table 7). Thus a 1.8-fold increase in binding was observed and a disproportional increase of 3.6-fold noted in internalization, suggesting that a greater proportion of the bacteria are bound to receptors involved in C. jejuni uptake. This observation was also noted when using non-polarized cells (Table 6). Inclusion of anti-fn antibodies with the infection inoculum, following EGTA treatment of T84 polarized cells, significantly decreased adherence and invasion from 185% to 35% and 367% to 67%, respectively. Noteworthy is that the addition of anti-fn antibodies following EGTA treatment of the polarized cells resulted in a proportional decrease in adherence (5.3-fold decrease) and internalization (5.5-fold decrease).

144 125 DISCUSSION The goal of this study was to further characterize the interaction of C. jejuni with cells using a model that mimics the in vivo environment that the organism encounters. Because histological examination of C. jejuni-infected humans and animals has revealed pathology in the colon (3, 6, 36), we chose to examine the interactions of C. jejuni with T84 cells. Phenotypically, T84 cells are thought to be similar to colonic crypt cells (32). The primary purpose of this study was to characterize the binding, entry, and translocation of polarized cells by C. jejuni. In agreement with Harvey et al. (14) and Everest et al. (11), but in contrast to Brás et al. (7), we found no correlation in an isolate s invasive and translocation potential. Harvey et al. (14) compared the ability of four clinical C. jejuni isolates to translocate across polarized Caco-2 cells with their ability to invade non-polarized cells. They also noted that the integrity of the cell monolayer remained over the course of the assay (14). We also noted that the final TER values were comparable to starting baseline values (not shown). Taken together, these data suggest that C. jejuni can translocate across an intact cell monolayer. Work by others has revealed that cellular tight junctions can reseal following bacterial penetration (40), thus providing a basis for C. jejuni to utilize a paracellular route of passage without the long-term disruption of the integrity of the cell monolayer. Previous studies have also demonstrated that tight junctions can temporally relax to allow regulated passage of both solutes and neutrophils (28). Investigators have utilized polarized cell culture systems to address the virulence properties of a wide variety of pathogens (12). For example, Kops et al. (25) found that while both S. typhi and S. typhimurium translocate across polarized cell monolayers, S. typhi translocates much earlier and in greater numbers than S. typhimurium. The translocation of S. typhi across the cell monolayers was accompanied by cell death and extrusion, which was reflected in rapid decrease in TER values. The

145 126 authors proposed that their findings were reflective of in vivo disease presentations, as S. typhi is commonly septic in nature and S. typhimurium is a localized superficial invader of the intestinal mucosa. Finlay and Falkow (12) also noted that the morphological effects of S. typhimurium on Caco-2 polarized cells was similar to those obtained using animal infection models. While the results presented herein indicate that C. jejuni possess the ability to translocate across a cell monolayer, they also suggest that the migration of the organism from the apical (e.g. lumen) to the basolateral (e.g. lamina propria) chamber may not, per se, be reflective of the organism s preference. More specifically, when the inoculum was removed from the apical chamber and the monolayers rinsed, the majority of C. jejuni were found to remain associated with the polarized cells rather than translocate into the basolateral chamber medium. Based on this observation, we believe that the ability of C. jejuni to translocate across a cell barrier is relevant with respect to the organism gaining access to receptor molecules located on the basolateral surface of a host cell. Our data suggest that the C. jejuni detected in the basolateral chamber represent those organisms that continue to migrate beyond the cell monolayer as the available host-cell receptor molecules to which they bind are saturated with bound bacteria. This finding is further supported by the fact that translocation of the C. jejuni F38011 and the cadf mutant differs. At a high MOI (300 bacteria per cell), Fn appears to become saturated as suggested by similar translocation kinetics for the C. jejuni F38011 parental isolate and cadf mutant. However, at the lowest MOI (3 bacteria per cell), C. jejuni F38011 translocation is retarded, which may be due to delayed saturation of available Fn binding sites. A similar phenomena may occur in vivo where early in an infection, most bacteria colonize the intestinal mucosa and begin to multiply. Thereafter, a significant number of C. jejuni may reach the lamina propria due to an increase in the bacterial load at the site of infection. While the functions of the Cia proteins are not known, the results presented herein supports a link between Cia secretion and C. jejuni-host cell invasion. The relationship between Cia protein secretion and C. jejuni invasion was evident regardless of whether the T84 host cells were cultured

146 127 using conventional methods or using conditions that allowed for the establishment of a polarized cell monolayer. The results are consistent with those previously obtained using INT 407 (22, 35) and Caco-2 (unpublished data) cells. In summary, C. jejuni appears to preferentially invade the basolateral surface of polarized cells based on competition assays whereby two C. jejuni wild-type isolates, in contrast to the C. jejuni cadf mutant, competed with one another for host-cell receptors. Pre-treatment of polarized cells with EGTA to increase exposure of cell-bound Fn resulted in a significant increase in both C. jejuni adherence and invasion. The increase in C. jejuni invasion of EGTA-treated cells was not due to an increase in cell surface area as S. typhimurium adherence and invasion frequencies were not affected when host-cells were pre-treated with EGTA. The specificity of the binding of C. jejuni to Fn was further demonstrated using antibodies that specifically reacted against Fn. Based on these data, we propose a model whereby C. jejuni translocate the intestinal epithelia, bind to Fn, and then invade the target cell. Interestingly, studies with Neisseria gonorrhoeae have suggested that the OpaA adhesin serves as a molecular-bridge between Fn and β 1 -integrins, which leads to bacterial uptake (42). Work in our laboratory has indicated that the binding of C. jejuni to Fn via CadF promotes the phosphorylation of paxillin, a focal adhesion protein (29). Based on the data presented herein, CadF also appears to play a role in promoting the binding of C. jejuni to polarized cells. Because cell-binding and the stimulation of cell-signaling events are a prerequisite for C. jejuni uptake by non-professional phagocytic cells, CadF may be an essential virulence determinant. In support of this notion, CadF has been found to be conserved among C. jejuni isolates (21). Further studies will be conducted to determine whether C. jejuni environmental isolates display the same virulence attributes as the C. jejuni clinical isolates F38011 and

147 128 ACKNOWLEDGEMENTS We thank Drs. John Klena, Anthony Garza, and Ray Larsen for reviewing this manuscript. This work was supported by a grants from the NIH (DK58911) and USDA National Research Initiative Competitive Grants Program (USDA/NRICGP, ) awarded to MEK.

148 129 REFERENCES 1. Allos, B. M Campylobacter jejuni infections: update on emerging issues and trends. Clin. Infect. Dis. 32: Altekruse, S. F., N. J. Stern, P. I. Fields, and D. L. Swerdlow Campylobacter jejuni -- an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Babakhani, F. K., G. A. Bradley, and L. A. Joens Newborn piglet model for campylobacteriosis. Infect. Immun. 61: Bacon, D. J., R. A. Alm, D. H. Burr, L. Hu, D. J. Kopecko, C. P. Ewing, T. J. Trust, and P. Guerry Involvement of a plasmid in virulence of Campylobacter jejuni Infect. Immun. 68: Baloda, S. B., A. Faris, G. Fröman, and T. Waldström Fibronectin binding to Salmonella strains. FEMS Microbiol. Lett. 28: Black, R. E., M. M. Levine, M. L. Clements, T. P. Hughes, and M. J. Blaser Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157: Brás, A. M., and J. M. Ketley Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: Cruz, N., Q. Lu, X. Alvarez, and E. A. Deitch Bacterial translocation is bacterial species dependent: results using the human Caco-2 intestinal cell line. J. Trauma. 36: Dawson, J. R., and R. P. Ellen Clustering of fibronectin adhesins toward Treponema denticola tips upon contact with immobilized fibronectin. Infect. Immun. 62: Dawson, J. R., and R. P. Ellen Tip-oriented adherence of Treponema denticola to fibronectin. Infect. Immun. 58: Everest, P. H., H. Goossens, J. P. Butzler, D. Lloyd, S. Knutton, J. M. Ketley, and P. H. Williams Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J. Med. Microbiol. 37: Finlay, B. B., and S. Falkow Salmonella interactions with polarized human intestinal Caco-2 epithelial cells. J. Infect. Dis. 162: Fröman, G., L. M. Switalski, A. Faris, T. Wadström, and M. Höök Binding of Escherichia coli to fibronectin. J. Biol. Chem. 259: Harvey, P., T. Battle, and S. Leach Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Humason, G. L Animal Tissue Techniques, vol. Fourth. W. H. Freeman and Company, San Francisco.

149 16. Jaffe, J., S. Natanson-Yaron, M. G. Caparon, and E. Hanski Protein F2, a novel fibronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol. Microbiol. 21: Jin, S., A. Joe, J. Lynett, E. K. Hani, P. Sherman, and V. L. Chan JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Konkel, M. E., and W. Cieplak, Jr Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M. E., M. D. Corwin, L. A. Joens, and W. Cieplak, Jr Factors that influence the interaction of Campylobacter jejuni with cultured mammalian cells. J. Med. Microbiol. 37: Konkel, M. E., S. G. Garvis, S. L. Tipton, D. E. Anderson, Jr., and W. Cieplak, Jr Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M. E., S. A. Gray, B. J. Kim, S. G. Garvis, and J. Yoon Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadf virulence gene and its product. J. Clin. Microbiol. 37: Konkel, M. E., B. J. Kim, V. Rivera-Amill, and S. G. Garvis Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32: Konkel, M. E., D. J. Mead, and W. Cieplak, Jr Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Konkel, M. E., D. J. Mead, S. F. Hayes, and W. Cieplak, Jr Translocation of Campylobacter jejuni across human polarized epithelial cell monolayer cultures. J. Infect. Dis. 166: Kops, S. K., D. K. Lowe, W. M. Bement, and A. B. West Migration of Salmonella typhi through intestinal epithelial monolayers: an in vitro study. Microbiol. Immunol. 40: Kuusela, P Fibronectin binds to Staphylococcus aureus. Nature (Lond). 276: Kuusela, P., A. P. Moran, T. Vartio, and T. U. Kosunen Interaction of Campylobacter jejuni with extracellular matrix components. Biochem. Biophys. Acta. 993: Madara, J. L Regulation of the movement of solutes across tight junctions. Annu. Rev. Physiol. 60:

150 29. Monteville, M. R., J. Yoon, and M. E. Konkel Maximal adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer membrane protein and microfilament reorganization. Manuscript in preparation. 30. Morooka, T., A. Umeda, and K. Amako Motility as an intestinal colonization factor for Campylobacter jejuni. J. Gen. Microbiol. 131: Myhre, E. B., and P. Kuusela Binding of human fibronectin to group A, C, and G Streptococci. Infect. Immun. 40: Nataro, J. P., S. Hicks, A. D. Phillips, P. A. Vial, and C. L. Sears T84 cells in culture as a model for enteroaggregative Escherichia coli pathogenesis. Infect. Immun. 64: Pei, Z., C. Burucoa, B. Grignon, S. Baqar, X.-Z. Huang, D. J. Kopecko, A. L. Bourgeois, J.-L. Fauchere, and M. J. Blaser Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Pickett, C. L., and C. A. Whitehouse The cytolethal distending toxin family. Trends in Microbiol. 7: Rivera-Amill, V., B. J. Kim, J. Seshu, and M. E. Konkel Secretion of the virulence associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J. Infect. Dis. 183: Russell, R. G., M. O'Donnoghue, D. C. Blake, Jr., J. Zulty, and L. J. DeTolla Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Rydén, C., K. Rubin, P. Speziale, M. Höök, M. Lindberg, and T. Wadström Fibronectin receptors from Staphylococcus aureus. J. Biol. Chem. 258: Schorey, J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown Characterization of the fibronectin-attachment protein of Mycobacterium avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Microbiol. 21: Takata, T., S. Fujimoto, and K. Amako Isolation of nonchemotactic mutants of Campylobacter jejuni and their colonization of the mouse intestinal tract. Infect. Immun. 60: Takeuchi, A Electron microscopic studies of experimental Salmonella infection. Penetration into the intestinal epithelium by Salmonella typhimurium. Am. J. Pathol. 50: Thomas, D. D., J. B. Baseman, and J. F. Alderete Fibronectin mediates Treponema pallidum cytadherence through recognition of fibronectin cell-binding domain. J. Exp. Med. 161:

151 42. van Putten, J. P., T. D. Duensing, and R. L. Cole Entry of OpaA+ gonococci into HEp-2 cells requires concerted action of glycosaminoglycans, fibronectin and integrin receptors. Mol. Microbiol. 29: Visai, L., S. Bozzini, T. E. Petersen, L. Speciale, and P. Speziale Binding sites in fibronectin for an enterotoxigenic strain of E. coli B342289c. FEBS Lett. 290: Wassenaar, T. M Toxin production by Campylobacter spp. Clin. Microbiol. Rev. 10: Wassenaar, T. M., B. A. van der Zeijst, R. Ayling, and D. G. Newell Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol. 139: Yao, R., D. H. Burr, and P. Guerry CheY-mediated modulation of Campylobacter jejuni virulence. Mol. Microbiol. 23:

152 133 Table 1. Cia proteins are required for maximal invasion of T84 non-polarized cells by C. jejuni Isolate No. of bacteria per wella Adherence Internalization C. jejuni F38011 ( ) x 106 ( ) x 104 C. jejuni ciab- ( ) x 106 ( ) x 103b C. jejuni ciab- + pmek100 ( ) x 106 ( ) x 104 C. jejuni ( ) x 106 ( ) x 104 aadherence and internalization assays were performed as outlined in Materials and Methods. Results are presented as the mean + standard deviation of triplicate determinations. bthe value was significantly different ( P value < 0.01) from that obtained using the C. jejuni F38011 wild-type isolate and C. jejuni ciab- transformant harboring pmek100.

153 134 Table 2. The ability of C. jejuni to translocate across T84 polarized cells is independent of invasive potential Attribute C. jejuni C. jejuni F38011 C. jejuni ciab- C. jejuni ciab- + pmek No. of bacteria in basolateral medium T = 0 to 1 h ( ) x 106 ( ) x 106 ( ) x 106 ( ) x 106 T = 1 to 2 h ( ) x 106 ( ) x 106 ( ) x 106 ( ) x 106 T = 2 to 3 h ( ) x 106 ( ) x 105 ( ) x 106 ( ) x 106 T = 3 to 4 h ( ) x 106 ( ) x 105 ( ) x 106 ( ) x No. of cell-associated bacteriaa ( ) x 106 ( ) x 106 ( ) x 106 ( ) x 106 T = 4 h 3. No. of internalized bacteriab ( ) x 104 ( ) x 104 ( ) x 103 c ( ) x 104 T = 7 h arepresents C. jejuni-inoculated T84 cells that were rinsed 3 times in EMEM after a 4 h incubation period. The number of cell-associated (adherent) bacteria was determined as outlined in Materials and Methods. brepresents C. jejuni-inoculated T84 cells that were rinsed 3 times in EMEM after a 4 h, followed by 3 h incubation with gentamicin to kill extracellular bacteria as outlined in Materials and Methods. cmean value is significantly different (P < 0.01) from that obtained with the C. jejuni wild-type isolate and C. jejuni ciab- transformant harboring pmek100#6.

154 135 Table 3. C. jejuni that become cell-associated remain associated with T84 polarized cells Attribute C. jejuni F38011 C. jejuni No. of cell-associated bacteria ( ) x 105 ( ) x 105 T = 4 h 2. No. of bacteria in basolateral medium T = 4 to 5 h ( ) x 104 ( ) x 104 T = 5 to 6 h ( ) x 103 ( ) x 104 T = 6 to 7 h ( ) x 103 ( ) x No. of cell-associated bacteria ( ) x 105 ( ) x 105 T = 7 h 4. Percent of cell-associated bacteria that translocated to basolateral mediuma 5. Percent of cell-associated bacteria remaining cell-associatedb a[total of translocated bacteria (sum of attributes in row 2) divided by cell-associated bacteria (row 1)] X 100. b[number of cell-associated bacteria at 7 h (row 3) divided by number of cell-associated bacteria at 4 h (row 1)] X 100.

155 136 Table 4. The CadF protein is required for maximal T84 cell-association regardless of MOI No. of cell-associated bacteria at an increasing MOIa C. jejuni F38011 ( ) x 104 ( ) x 105 ( ) x 106 C. jejuni cadf mutant ( ) x 104 b ( ) x 105 b ( ) x 106 b aassays were performed as outlined in Materials and Methods. C. jejuni cell-association was measured at 4 h post-infection using polarized T84 cells. bthe value was significantly different ( P value < 0.01) from that obtained using the C. jejuni F38011 wild-type isolate.

156 Table 5. The functional role of CadF as an adhesin is shared among the C. jejuni clinical isolates and F Competitor Number of the C. jejuni wild-type isolate bounda No competitor ( ) x % C. jejuni cadf mutant 50-fold excess ( ) x % 90-fold excess ( ) x % C. jejuni F fold excess ( ) x %b 50-fold excess ( ) x %b avalues represent the number of C. jejuni bound in the presence of the competing organism divided by the number of C. jejuni bound in the absence of the competitor, multiplied by 100. bthe value was significantly different ( P value < 0.01) from that obtained using the C. jejuni wild-type isolate in the absence of the competitor.

157 Table 6. C. jejuni association with fibronectin maximizes bacterial adherence and invasion of T84 non-polarized cells 138 Percent of adherent and internalized bacteriaa -- α-fn EGTA EGTA + α-fn C. jejuni F38011 Adherent b c Internalized b c S. typhimurium SL1344 Adherent Internalized aassays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates alone. bthe value was significantly different ( P value < 0.01) from that obtained from the inoculation of T84 cells with the C. jejuni F38011 wild-type isolate. cthe value was significantly different ( P value < 0.01) from that obtained from inoculation of the EGTA-treated T84 cells with the C. jejuni F38011 wild-type isolate.

158 Table 7. C. jejuni association with fibronectin localized to the basolateral surface of T84 polarized cells maximizes bacterial adherence and invasion 139 Percent of adherent and internalized bacteriaa -- α-fn EGTA EGTA + α-fn C. jejuni F38011 Adherent b c Internalized b c aassays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolate alone. bthe value was significantly different ( P value < 0.01) from that obtained from the inoculation of T84 cells with the C. jejuni F38011 wild-type isolate. cthe value was significantly different ( P value < 0.01) from that obtained from inoculation of the EGTA-treated T84 cells with the C. jejuni F38011 wild-type isolate.

159 140 Figure 1. Translocation kinetics of C. jejuni F38011 and the isogenic cadf mutant across polarized T84 cells at various MOIs. Each bar represents the number of C. jejuni recovered from the basolateral chamber over the course of each 1 h interval. * The value was significantly different ( P value < 0.01) from the C. jejuni F38011 wild-type isolate.

160 141 No. C. jejuni recovered in basolateral medium * C. jejuni F38011 C. jejuni cadf mutant * * * * hr 1-2 hr 2-3 hr 3-4 hr 0-1 hr 1-2 hr 2-3 hr 3-4 hr 0-1 hr 1-2 hr 2-3 hr 3-4 hr MOI = 300 MOI = 30 MOI = 3

161 142 Chapter 5 Maximal adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer membrane protein and microfilament reorganization Marshall R. Monteville, Julie E. Yoon, and Michael E. Konkel* School of Molecular Biosciences, Washington State University, Pullman, WA Published in Microbiology 2003,149:

162 143 ABSTRACT The binding of Campylobacter jejuni to fibronectin (Fn), a component of the extracellular matrix, is mediated by a 37 kda outer membrane protein termed CadF for Campylobacter adhesion to fibronectin. The specificity of C. jejuni binding to Fn, via CadF, was demonstrated using antibodies reactive against Fn and CadF. More specifically, the anti-cadf antibody reduced the binding of two C. jejuni clinical isolates to immobilized Fn by greater than 50%. Furthermore, a C. jejuni wild-type isolate, in contrast to the isogenic CadF mutant, was found to compete with another C. jejuni wild-type isolate for host cell receptors. Given the relationship between the pericellular fibronectin matrix and the cytoskeleton, the involvement of host cell cytoskeletal components in C. jejuni internalization was also examined. Cytochalasin D and mycalolide B microfilament depolymerizing agents resulted in a significant reduction in C. jejuni invasion. Studies targeting paxillin, a focal adhesion signaling molecule, identified an increased level of tyrosine phosphorylation upon C. jejuni infection of INT 407 cells. Collectively, these data suggest CadF promotes the binding of C. jejuni to fibronectin, which in turn stimulates a signal transduction pathway involving paxillin.

163 144 INTRODUCTION The ability of pathogenic bacteria to bind to host tissues is important as it represents an early event in the establishment of an in vivo niche. In some instances, such binding is also a prerequisite for host cell invasion, where the organisms are protected from the humoral and cellular immune responses (Alrutz & Isberg, 1998; Isberg & Tran Van Nhieu, 1994; Roberts, 1990). An emerging theme among pathogenic microorganisms is their ability to utilize host cell molecules during the infectious process to facilitate their binding and entry into host cells (Watarai et al., 1996). Campylobacter jejuni is a Gram-negative, microaerophilic bacterium, and is recognized as the most frequent cause of gastrointestinal disease in developed countries. The ability of C. jejuni to colonize the gastrointestinal tract of humans is proposed to be essential for disease (Allos & Blaser, 1995; Wassenaar & Blaser, 1999). Fauchère et al. (Fauchère et al., 1986) found that C. jejuni isolated from individuals with fever and diarrhea exhibited greater binding to epithelial cells than strains isolated from individuals without fever and diarrhea. Numerous studies have been done to identify and characterize C. jejuni adhesins (De Melo & Pechère, 1990; Fauchère et al., 1989; Jin et al., 2001; Kelle et al., 1998; Kervella et al., 1993; Konkel et al., 1997; McSweegan & Walker, 1986; Moser & Schröder, 1995; Moser et al., 1997; Moser et al., 1992; Pei & Blaser, 1993; Pei et al., 1998; Schröder & Moser, 1997). Based on these studies, C. jejuni appears to synthesize a number of adhesive molecules. Fibronectin (Fn) is a 220 kda glycoprotein that is present at regions of cell-to-cell contact in the gastrointestinal epithelium, thereby providing a potential binding site for pathogens (Quaroni et al., 1978). Several bacterial pathogens, including Staphylococcus aureus (Kuusela, 1978; Rydén et al., 1983), Streptococcus pyogenes (Jaffe et al., 1996; Myhre & Kuusela, 1983), Salmonella enteritidis (Baloda et al., 1985), Escherichia coli (Fröman et al., 1984; Visai et al., 1991), Neisseria

164 145 gonorrhoeae (van Putten et al., 1998), Mycobacterium avium (Schorey et al., 1996), and Treponema species (Dawson & Ellen, 1990; Dawson & Ellen, 1994; Thomas et al., 1985), bind fibronectin. We identified a 37 kda outer membrane protein in C. jejuni that mediates the organism s binding to the extracellular matrix component Fn (Konkel et al., 1997). The cadf (Campylobacter adhesion to Fn) gene has thus far been found to be conserved among C. jejuni and C. coli isolates (Konkel et al., 1999). In vivo studies have suggested that the CadF protein is required for the colonization of chickens by C. jejuni (Ziprin et al., 1999). Published work indicates a relationship between the pericellular fibronectin matrix and the cytoskeleton (Gumbiner, 1996; Miyamoto et al., 1998; van der Flier & Sonnenberg, 2001). In mammalian cells, the actin cytoskeleton is necessary for a variety of cellular processes including control of cell-to-cell and cell-to-substrate interactions. Actin nucleation occurs at membraneassociated sites called focal adhesions. Focal adhesions are the sites at which the bundles of actin filaments (stress fibers) are cross-linked with membrane-associated adhesion molecules (e.g., integrins) and extracellular molecules (Gumbiner, 1996; Sarkar, 1999). The integrin molecules bind extracellularly to matrix components and intracellularly associate with protein complexes consisting of vinculin, talin, α-actinin, paxillin, tensin, zyxin, and focal adhesion kinase (FAK) (Miyamoto et al., 1998; Tachibana et al., 1995). The integrins are transmembrane glycoprotein receptors comprised of heterodimeric αβ subunits (Danen & Yamada, 2001; Hynes, 1992; van der Flier and Sonnenberg, 2001; Vuori, 1998). The two subunits are noncovalently associated with one another (1:1) in the membrane. There are eighteen known α subunits and eight known β subunits (van der Flier and Sonnenberg, 2001). Different α and β subunit combinations dictate the specificity of cell-to-cell and cell-to-extracellular matrix (ECM) recognition. The α 5 β 1 integrin receptor specifically binds fibronectin (Hynes, 1992; Miyamoto et al., 1998). Integrin occupancy and clustering is associated with tyrosine phosphorylation of cellular cytoplasmic proteins including FAK and paxillin, and is a means of regulating host signal transduction events leading to

165 146 actin rearrangements (Miyamoto et al., 1998; Tachibana et al., 1995). As stated earlier, certain bacterial pathogens are known to utilize host cell extracellular matrix and cytoskeleton components to their benefit. For example, N. gonorrhoeae appear to utilize a fibronectin-mediated uptake pathway involving integrin receptors (van Putten et al., 1998). The binding of N. gonorrhoeae to Fn is proposed to trigger the organism s uptake via integrin receptors by stimulating the host cell signaling pathways that are responsible for cytoskeletal rearrangement. Because C. jejuni isolates possess a minimum of three different adhesive molecules including CadF, JlpA, and PEB1 (Jin et al., 2001; Konkel et al., 1997; Pei et al., 1998), we sought to determine the step of the infectious process in which the CadF outer membrane protein participates. We investigated the binding properties of two C. jejuni clinical isolates, F38011 and , to the INT 407 cell line (human embryonic intestinal cells) and the participation of the INT 407 cells in C. jejuni uptake. While the host cell cytoskeletal components involved in C. jejuni F38011 uptake have not been examined previously, C. jejuni has been reported to be internalized via a novel pathway exclusively involving microtubules (Bacon et al., 2000; Hu & Kopecko, 1999; Hu & Kopecko, 2000; Kopecko et al., 2001; Oelschlaeger et al., 1993). Based on the experiments performed herein implicating microfilaments in C. jejuni uptake, additional experiments were done to determine if the host cell signaling events known to be associated with cytoskeletal rearrangement occur during C. jejuni entry.

166 147 MATERIALS AND METHODS Bacterial isolates and growth conditions. C. jejuni F38011, 81116, and (TetR) wild-type clinical isolates were cultured on Mueller-Hinton agar plates containing 5% bovine citrated blood (MH/blood) at 37 C under microaerophilic conditions. C. jejuni (TetR) was cultured on plates supplemented with 12.5 µg/ml of tetracycline (Bacon et al., 2000). C. jejuni F38011 isogenic cadf and Cj1477c mutants were cultured on MH/blood agar plates supplemented with 200 µg/ml of kanamycin. The C. jejuni F38011 (StrepR/NalR) isolate was cultured on MH/blood agar plates supplemented with 200 µg/ml of streptomycin and 50 µg/ml of nalidixic acid. Isolates were passaged every 24 to 48 hr. Citrobacter freundii 8090 and Salmonella enterica ssp. Typhimurium SL1344 (S. typhimurium) were cultured aerobically on Luria-Bertani (LB) agar plates at 37 C. Binding of C. jejuni to immobilized ECM. Binding of C. jejuni isolates to human plasma Fn (Sigma, St. Louis, MO) was assessed as previously described (Konkel et al., 1997). Specificity of binding was determined by preincubating Fn coated coverslips with a 1:50 dilution of a rabbit anti-human Fn antibody (Telios Pharmaceuticals, Inc., San Diego, CA), preincubating C. jejuni isolates with a 1:50 dilution of a goat anti-c. jejuni 37 kda serum, or by the addition of 100 µg/ml Fn. For each coverslip, the bacteria in each of 3 randomly chosen fields were counted. Gel electrophoresis and immunoblot analysis. Bacterial whole-cell extracts (an equivalent of 0.1 OD 600 units) were solubilized in single strength electrophoresis sample buffer and incubated at 95 C for 5 min. Proteins were separated in SDS-12.5% PAGE minigels as previously described (Laemmli, 1970), and electrophoretically transferred to polyvinylidene fluoride (PVDF) membranes (Immobilon P; Millipore Corp., Bedford, MA). The membranes were washed 3 times in

167 148 phosphate-buffered saline (PBS) and incubated for 18 hr at 4 C with a 1:500 dilution of the goat anti-c. jejuni 37 kda serum in PBS ph 7.4/0.01% Tween-20 containing 20% fetal bovine serum (FBS). Bound antibodies were detected using peroxidase-conjugated rabbit anti-goat IgG (Sigma) at a 1:2,000 dilution and 4-chloro-1-napthol (Sigma) as the chromogenic substrate. Bactericidal concentration of gentamicin. Assays were performed to determine the concentration of gentamicin and length of incubation time required to kill each isolate used in this study. C. jejuni isolates, C. freundii and S. typhimurium were suspended in Minimal Essential Medium (MEM) (GIBCO Invitrogen Corporation, Carlsbad, CA) supplemented with 1% fetal bovine serum (MEM-1% FBS) or 10% FBS (MEM-10% FBS) at an approximate concentration of 2 X 106 CFU/ml. Aliquots (1 ml) of the bacterial suspensions were placed in a 24-well plate and incubated at 37 C in a humidified, 5% CO 2 incubator for 2 hr. Gentamicin (Life Technologies Gaithersburg, MD) was added to the media at a final concentration of 0, 100, 250, 500 and 750 µg/ml, and the suspensions again incubated as above. Following incubation (2 or 3 hr), the bacterial suspensions (1 ml) were removed and bacteria pelleted by centrifugation at 6,000 x g for 10 min. The supernatant fluids were discarded to remove the gentamicin and bacteria resuspended in 1 ml of PBS. Suspensions were serially diluted and the number of viable bacteria assessed. Each isolates was tested at least 3 times in triplicate. INT 407 binding and internalization assays. A stock culture of INT 407 cells (human embryonic intestine, ATCC CCL 6) was obtained from the American Type Culture Collection. This cell line was cultured in MEM-10% FBS at 37 C in a humidified, 5% CO 2 incubator. For experimental assays, each well of a 24-well tissue culture tray was seeded with 1.4 X 105 cells/well and incubated for 18 hr at 37 C in a humidified, 5% CO 2 incubator. The cells were rinsed with MEM-1% FBS and inoculated with approximately 5 X 107 CFU of a bacterial suspension. Tissue

168 149 culture trays were centrifuged at 600 x g for 5 min, and incubated at 37 C in a humidified, 5% CO 2 incubator. For binding, the infected monolayers were incubated for 2 hr, rinsed 3 times with PBS, and the epithelial cells lysed with a solution of 0.1% (vol/vol) Triton X-100 (Calbiochem, La Jollo, CA). The suspensions were serially diluted and the number of viable, adherent bacteria determined by counting the resultant colonies on MH/blood plates. To measure bacterial internalization, the infected monolayers were incubated for 2 hr, rinsed 3 times with MEM-1% FBS, and incubated for an additional 3 hr in MEM-1% FBS containing a bactericidal concentration of gentamicin. The number of internalized bacteria was determined as outlined above. Unless otherwise stated, the reported values represent the mean counts + std. deviations derived from triplicate wells. All assays in this study using cytoskeletal inhibitors were performed at a multiplicity of infection (MOI) ranging between 50 and 500 to ensure reproducibility, and repeated a minimum of 3 times. Regardless of whether a MOI of 50 to 1 or 500 to 1 was used, the effect of the cytoskeletal inhibitory drugs was the same. Competitive inhibition assays. INT 407 cells were inoculated with a suspension containing the C. jejuni (TetR) isolate and the C. jejuni cadf mutant (KanR) as well as the C. jejuni (TetR) isolate and the C. jejuni F38011 (StrepR/NalR) isolate. Binding assays were performed as mentioned above except that an MOI of approximately 10 was used for the C. jejuni (TetR) wild-type isolate. To determine the number of bacteria of a particular isolate that became bound, serial dilutions of the suspensions were plated on MH/blood agar plates supplemented with the appropriate selective antibiotic. No evidence of horizontal gene transfer was apparent as judged by lack of recovery of StrepR/NalR/TetR isolates upon mixing experiments with the C. jejuni (TetR) and C. jejuni F38011 (StrepR/NalR) isolates. Inhibitor studies. INT 407 cells were preincubated for 45 min in MEM-1% FBS with

169 150 nocodazole, cytochalasin D, or mycalolide B at the concentrations indicated in the text. When performing experiments looking at the combined effects of inhibitors, nocodazole and cytochalasin D were both added to INT 407 cells during the preincubation step. Following incubation, cells were infected with approximately 5 X 107 CFU of each isolate while maintaining indicated inhibitor concentrations. S. typhimurium was preincubated anaerobically at 37 C for 3 hr prior to infection. Binding and internalization assays were performed as outlined above. The values reported represent the percent of adherent and internalized bacteria relative to the control (untreated sample) or the mean + std. deviation of adherent and internalized bacteria. INT 407 cell viability, following inhibitor treatment, was assessed by rinsing the INT 407 cells twice with PBS, and staining the cells for 5 min with 0.5% trypan blue. The cells were then rinsed twice with PBS and counter-stained for 5 min with 0.5 ml of 0.5% phenol red (Humason, 1979). The INT 407 cells were visualized with an inverted microscope. Reversibility of cytochalasin D. INT 407 cells were pretreated with cytochalasin D and infected with bacteria as described above. To asses the reversibility of cytochalasin D, the INT 407 cell monolayers were rinsed after 2 hr with MEM-1% FBS. A set of the cytochalasin D-treated cells were then incubated for an additional 2 hr in MEM-1% FBS containing cytochalasin D or in the presence of medium alone. The number of adherent and internalized bacteria was determined as outlined above. Preparation of the polyclonal antisera. Female New Zealand White rabbits were subcutaneously injected with 100 µg of each bacterial whole cell extract in TiterMax Gold (CytRx Corporation, Norcross, GA). A booster injection of 50 µg of whole cell extract in Freund s incomplete adjuvant (Sigma) was given after 4 weeks. Blood was collected prior to first and second immunizations, and 2 weeks after the second immunization. Sera were stored at -20 C.

170 151 Confocal microscopy examination of C. jejuni infected cells. INT 407 cells (5 X 104 cells) were cultured on 13 mm circular, glass coverslips for 18 hr at 37 C in a humidified, 5% CO 2 incubator. The cells were infected by the addition of 0.5 ml of a bacterial suspension (1 X 106 CFU per well) in MEM. Mock-infected cells were used in certain instances as a negative control. Prior to infection, cell monolayers were rinsed 1 time with MEM. Following incubation (45 min), the cell monolayers were rinsed 3 times with PBS and fixed with 3.0% glutaraldehyde or 3.0% paraformaldehyde in 0.1 M phosphate buffer (ph 7.2). Cells were permeabilized with 0.1% (vol/vol) Triton-X 100. Tubulin was stained using a 1:250 dilution of FITC-conjugated monoclonal antibody against tubulin (clone DM 1A; Sigma). Actin was stained using FITC-labelled phalloidin (Sigma) at a concentration of 0.4 µg/ml. Primary antibodies directed towards bacteria were used at a 1:250 dilution followed by a secondary goat anti-rabbit IgG-rhodamine F(Ab ) 2 fragment antibody at a 1:500 dilution. Immunoprecipitation. INT 407 cells were cultured in six-well tissue culture trays as outlined above. INT 407 cell monolayers were rinsed once with MEM and inoculated with a suspension of the C. jejuni F38011 wild-type isolate and the isogenic cadf mutant (MOI = 100). Tissue culture trays were centrifuged and incubated as stated previously. At timepoints indicated, cell monolayers were rinsed 3 times with PBS and lysed by the addition of ice-cold lysis buffer (125 mm Tris ph 8.0, 137 mm NaCl, 10% glycerol, 0.5% sodium deoxycholate, 1% NP-40, 2 mm EDTA, 1 mm PMSF, 1 mm Na 3 VO 4, and 1.76 TIU aprotonin). The insoluble and soluble fractions were separated by centrifugation at 14,000 x g for 10 min at 4 C. Immunoprecipitation was performed by adding a mouse anti-paxillin antibody (Clone 349, Transduction Labs, Lexington, KY) to the soluble fraction and incubating at 4 C for 2 hr with end-over-end mixing. Following mixing, prewashed protein G agarose beads (Gibco BRL, Gaithersburg, MD) were added and samples mixed for an additional 2 hr at 4 C. The precipitate was rinsed 4 times in ice-cold lysis buffer and

171 152 once with PBS. Samples were analyzed by SDS-PAGE and immunoblot analysis as outlined above. Immunoblot detection of phosphorylated paxillin was performed using a 1:2,000 dilution of a mouse anti-phosphotyrosine antibody (PT-66, Sigma) and a 1:80,000 dilution of a peroxidase-conjugated rabbit anti-mouse IgG by enhanced chemiluminescence (Renaissance; NEN Life Science Products, Boston, MA). Detection of the total pool of paxillin was performed using a 1:10,000 dilution of a mouse anti-paxillin antibody and a 1:80,000 dilution of a peroxidase-conjugated rabbit anti-mouse IgG. Other analytical procedures. To determine the minimal inhibitory concentration (MIC) of gentamicin sulphate, C. jejuni strains , and F38011 were cultured on MHB plates for 24 h at 37 C under microaerophilic conditions. Bacteria were suspended in 3 ml of PBS (OD 540 = 0.17), and 10 µl of each cell suspension was placed onto a MHB plate containing gentamicin sulphate. Gentamicin concentrations ranged from 1 µg/ml through 512 µg/ml. Before and after spotting the suspensions on the antibiotic containing plates, 10 µl of each cell suspension was also placed on a MHB plate without antibiotic to ensure that bacterial viability was maintained. MICs were assessed by determining the lowest concentration of gentamicin in which no growth was observed following a 24 hr incubation. To ensure that the phenotype associated with the C. jejuni CadF mutant was due specifically to the mutation of the cadf gene, a knockout was generated in Cj1477c. Cj1477c lies downstream of the cadf gene in C. jejuni F38011, and encodes for a putative hydrolase (Cj1477c) (Lyngstadaas et al., 1999; Parkhill et al., 2000). Cj1477c was disrupted by homologous recombination via a single crossover event between the putative hydrolase gene on the chromosome and an internal fragment of the hydrolase gene on a suicide vector using methods described previously (Konkel et al., 1997). A 512 bp fragment, which is internal to the 648 bp coding region of Cj1477c, was amplified using the primers CJ1477cKOF (5 -GCA CTT TAA TTG ATT CTA CTG ATG-3 ) and

172 153 CJ1477cRTR(KO) (5 -GGC CAA AGG AGT GAT ATT AGC AG-3 ). The resultant fragment was ligated into the pcrii cloning vector (TA Cloning System, Invitrogen, San Diego, CA), and the ligation mixture used to transform E. coli MRF. Following plasmid purification, the cloned 512 bp fragment was then excised by restriction endonuclease digestion with EcoR I, gel-purified, and ligated into pbluescript SK+ (pbskii+) containing the apha3 gene encoding kanamycin resistance. The pbskii+ vector was digested with EcoR I and treated with calf intestinal alkaline phosphatase prior to ligation. The resultant plasmid was introduced into C. jejuni F38011 by electroporation. Potential insertional mutants were identified by the acquisition of kanamycin resistance. Disruption of the hydrolase gene in C. jejuni F38011 was confirmed by PCR using gene specific primers. Plasmid carriage in C. jejuni was confirmed by the polymerase chain reaction using the teto (forward primer, 5 -TTG ACA AAT AAA GGG T TA AGG-3 ; reverse primer, 5 -CCT TTC AAA TCT CAT TTT ATA CG-3 ) and virb11 (forward primer, 5 -GAA CAG GAA GTG GAA AAA CTA GC-3 ; reverse primer, 5 -TTC CGC ATT GGG CTA TAT G-3 ) gene specific primers, followed by sequencing of the PCR-amplified products (Bacon et al., 2000).

173 154 RESULTS CadF promotes the binding of C. jejuni to Fn. Previous work in our laboratory revealed that a 37 kda outer membrane protein, termed CadF, promoted the binding to C. jejuni F38011 to Fn (Konkel et al., 1997). To build on this initial work, experiments were performed to determine the role of CadF using the C. jejuni clinical isolates F38011 and Consistent with our previous work showing that CadF is conserved among C. jejuni isolates (Konkel et al., 1997; Konkel et al., 1999), immunoblot analysis with a goat anti-c. jejuni CadF serum revealed a 37 kda reactive band in the whole cell lysates of both C. jejuni F38011 and clinical isolates. A band corresponding to the 37 kda CadF protein was not detected in the lysate of a C. jejuni F38011 cadf mutant (Fig. 1). The C. jejuni wild-type isolate was also found to bind to Fn-coated coverslips (Table 1). The binding to Fn by both C. jejuni F38011 and was judged to be specific because it was significantly reduced by an anti-fn serum, an anti-cadf serum, and the addition of exogenous Fn (Table 1). The C. jejuni cadf mutant bound to Fn at levels less than 10% of the C. jejuni F38011 and clinical isolates. Additionally, no difference was noted in the binding of the C. jejuni cadf mutant to Fn-coated coverslips when compared to the BSA-coated coverslips. Collectively, these data suggest that CadF is the primary Fn-binding constituent in C. jejuni F38011 and CadF is required for the maximal binding and internalization of C. jejuni to INT 407 cells. To determine whether CadF plays a role in promoting bacteria-host cell interactions, binding and internalization assays were performed with INT 407 cells and the C. jejuni F38011 wild-type isolate and cadf mutant. At a MOI of 30:1, the C. jejuni cadf knockout showed a 59% reduction in adherence to INT 407 cells when compared to the C. jejuni F38011 wild-type isolate (data not shown). To determine whether the phenotype displayed by the C. jejuni cadf knockout was solely due to the absence of CadF protein, a knockout was generated in Cj1477c. Cj1477c lies

174 155 downstream of the cadf gene in C. jejuni F38011, and encodes for a putative hydrolase (Cj1477c). In contrast to the C. jejuni cadf knockout, no reduction was noted in the binding of the Cj1477c knockout to the INT 407 cells when compared to the C. jejuni F38011 wild-type isolate (data not shown). This finding indicated that the reduction in adherence noted with C. jejuni cadf knockout was due solely to the absence of the CadF protein, and was not the result of indirect effects involving the expression of adjacent genes. To determine if the C. jejuni clinical isolate utilizes CadF as an adhesin for host cells, competitive inhibition assays were performed using the C. jejuni F38011 wild-type isolate and C. jejuni F38011 cadf mutant (Table 2). A significant decrease was noted in the number of C. jejuni bound to the INT 407 cells in the presence of a 44 and 88-fold excess of the C. jejuni F38011 wild-type isolate. Moreover, increasing the number of competing organisms resulted in a dose-dependent decrease in the number of C. jejuni bound. In contrast, a statistically significant difference was not observed in the number of C. jejuni bound to the INT 407 cells in the presence of the C. jejuni cadf mutant. Based on these findings, we concluded that CadF serves as an adhesin for the C. jejuni F38011 and clinical isolates. C. jejuni internalization involves actin cytoskeletal reorganization. Previous studies with C. jejuni have suggested that this isolate is internalized via a unique pathway requiring only microtubules (Oelschlaeger et al., 1993). Because C. jejuni binds to Fn, we reexamined the role of microfilaments in C. jejuni binding and internalization in the presence of microfilament inhibitors. Cytochalasin D serves as an actin capping compound which binds to the barbed end of actin filaments, thereby shifting the polymerization- depolymerization equilibrium leading to depolymerization of microfilaments (Cooper, 1987). S. typhimurium and C. freundii were used as controls (Biswas et al., 2000; Oelschlaeger et al., 1993). Prior to performing the assays, preliminary experiments were conducted to ensure that cytochalasin D, at the concentrations

175 156 used here, had no effect on INT 407 cell viability. In addition, preliminary experiments were conducted with each bacterial isolate to ensure that a bactericidal concentration of gentamicin was used in the invasion assay. In contrast to previous reports (Bacon et al., 2000; Hu and Kopecko, 1999; Oelschlaeger et al., 1993), it was found necessary to treat C. jejuni with a concentration of 500 µg/ml of gentamicin for 3 hr to kill the organism. A 2 hr treatment of C. jejuni (range X 106 CFU) with 100 µg/ml of gentamicin resulted in the recovery of 4.8 X 104 to 2.9 X 105 CFU (n = 4 individual experiments). Bactericidal activity of gentamicin was dependent on concentration and time of exposure while independent of the percent of FBS in the media (1% or 10%). The effects of cytochalasin D on the C. jejuni F38011 and clinical isolates, S. typhimurium, and C. freundii are shown in Table 3. Cytochalasin D resulted in a dose-dependent increase in binding and a dose-dependent decrease in internalization of both C. jejuni F38011 and to INT 407 cells regardless of the MOI used in an individual experiment. Consistent with previous reports (Bacon et al., 2000; Biswas et al., 2000; Hu and Kopecko, 1999; Hu and Kopecko, 2000; Oelschlaeger et al., 1993), cytochalasin D also significantly inhibited the internalization of S. typhimurium and C. freundii by INT 407 cells. Confocal microscopy examination of the infected INT 407 cells did not reveal convincing evidence for the interaction of C. jejuni with cellular microfilaments. However, microfilament supported structures were observed in contact with C. jejuni (Fig. 2). Also noted was the expected interaction of S. typhimurium with a cellular structure supported by microfilaments. To further investigate the role of microfilaments in C. jejuni internalization, assays were performed with mycalolide B. Mycalolide B severs microfilaments (F-actin), causing them to depolymerize, and sequesters G-actin, which inhibits microfilament polymerization (Saito et al., 1994; Saito et al., 1998; Wada et al., 1998). While Mycalolide B was found to significantly inhibit the internalization

176 157 of C. jejuni and S. typhimurium (Table 4), the drug had no effect on the viability of the INT 407 cells as judged by staining with trypan blue (data not shown). These data also suggest that the uptake of C. jejuni F38011, as well as , utilize microfilaments. Our finding of the requirement for actin rearrangement in S. typhimurium internalization is consistent with that of previous reports (Finlay et al., 1991; Francis et al., 1993; Francis et al., 1992). The effect of cytochalasin D on C. jejuni internalization is reversible. Because treatment of the INT 407 cells with cytochalasin D inhibited the uptake of each organism tested (C. jejuni F38011, C. jejuni , S. typhimurium, and C. freundii), we questioned whether a small number of the INT 407 cells were being removed from the plastic substrate during the rinsing steps. If cells were detaching from the plastic substrate during either the drug-treatment step or the rinses, it could contribute to the decrease noted in internalized organisms. Thus, an assay was performed to examine the reversibility of the effect caused by cytochalasin D. The effect of cytochalasin D on C. jejuni internalization was reversible, indicating that the INT 407 cells were not detaching from the plastic substrate (Table 5). In addition, the increase in adherent bacteria caused by the drug treatment resulted in an increase in internalized organisms following the drugs removal from the assay medium, indicating that the adherent bacteria are capable of becoming internalized. C. jejuni internalization is sensitive to microtubule inhibitors. Binding and internalization assays were also performed with the C. jejuni F38011 and clinical isolates, S. typhimurium, and C. freundii in the presence of the microtubule inhibitor nocodazole (Table 6). Nocodazole binds β-tubulin, preventing tubulin polymerization, as well as enhances GTPase activity (Mejillano et al., 1996). Nocodazole had no effect on the binding of the C. jejuni isolates to the INT 407 cells, but did significantly inhibit the uptake of both isolates by the INT 407 cells in a dose-dependent fashion. While the internalization of C. freundii was also significantly inhibited by nocodazole, the drug had no effect on S. typhimurium internalization. A clear association of C. jejuni with

177 microtubules was not observed by confocal microscopy examination of the infected INT 407 cells (Fig. 3). 158 Others have suggested that C. jejuni can be internalized via two distinct uptake pathways where microtubules and microfilaments act exclusively (Hu and Kopecko, 1999; Kopecko et al., 2001; Oelschlaeger et al., 1993). Because the uptake of the C. jejuni isolates tested was found to be sensitive to both microfilament and microtubule depolymerizing agents, assays were performed to determine the combined effects of these drugs on C. jejuni internalization (Table 7). When the drugs were used in combination, the number of C. jejuni bound to the INT 407 cells was comparable to that obtained when only cytochalasin D was used. Significant differences were not observed in the number of C. jejuni internalized by the INT 407 cells in the presence of either cytochalasin D or nocodazole alone versus when the drugs were used in combination with one another. As the combination of these two drugs on C. jejuni internalization was not additive, the data suggest that microfilaments and microtubules act in concert to facilitate C. jejuni uptake (Table 7). Maximal C. jejuni entry is accompanied by the tyrosine phosphorylation of paxillin. As the data indicate that C. jejuni uptake involves cytoskeletal rearrangement, it is likely that the organism s uptake is accompanied by cytoskeletal signal transduction events. Therefore, experiments were performed to examine the levels of tyrosine phosphorylated paxillin in INT 407 cells immediately prior to and following inoculation with C. jejuni. An increased level of phosphorylated paxillin is an indicator of integrin stimulation (Watarai et al., 1996). Integrin molecules bind extracellularly to matrix components (e.g., Fn) and intracellularly associate with protein complexes that ultimately are linked to microfilaments. Here, the INT 407 cells were infected with the C. jejuni F38011 isolate at an MOI of 100 to 1 and incubated for varying lengths of time, ranging between 15 and 60 min. Following incubation, the focal adhesion-associated

178 159 protein paxillin was immunoprecipitated and subjected to SDS-PAGE coupled with immunoblot analysis using an anti-phosphotyrosine antibody. Infection of the INT 407 cells with the C. jejuni F38011 wild-type isolate led to an increase in the level of tyrosine phosphorylated paxillin at 30 and 45 min post-infection (Fig. 4). An increase in the level of phosphorylated paxillin was not observed in INT 407 cells infected with the C. jejuni cadf mutant until the MOI was increased 20-fold beyond that of the wild-type isolate, after which the pattern of phosphorylated paxillin mirrored that observed for the wild-type isolate (data not shown). Consistent with paxillin-phosphorylation, colocalization was observed between C. jejuni and host cell-tyrosine phosphorylated proteins by confocal microscopy (data not shown).

179 160 DISCUSSION Binding of C. jejuni to epithelial cells is mediated by several outer-membrane proteins including JlpA, PEB1, and CadF (Jin et al., 2001; Konkel et al., 1997; Pei et al., 1998). Although it is difficult to assess the contribution of each adhesin in the adherence of C. jejuni to host cells given variations in assay protocols, knockouts have been generated in the genes encoding the JlpA, PEB1, and CadF adhesins and the phenotypes examined. A C. jejuni jlpa knockout is reduced 19% in adherence to HEp-2 cells (Jin et al., 2001). A C. jejuni peb1 knockout is reduced 10-99%, depending on the initial inoculum, in adherence to HeLa cells (Pei et al., 1998). A C. jejuni cadf knockout is reduced 50-90%, depending on the initial inoculum, in adherence to INT 407 cells. The biological significance of the CadF adhesin has also been demonstrated in vivo. Ziprin et al. (Ziprin et al., 1999) reported that a C. jejuni F38011 cadf mutant is unable to colonize the intestinal tract of Leghorn chickens (n = 60), thus providing evidence that the 37 kda outer membrane protein plays an in vivo role in mediating the organism s binding to the intestinal epithelium. We undertook this study to more closely examine the role of CadF in C. jejuni-host cell interactions. Based on the results presented here, CadF appears to promote the binding of C. jejuni to Fn, thereby stimulating the host cell signaling events associated with bacterial uptake. Two C. jejuni clinical isolates, F38011 (Lior serotype 90) and (Lior serotype 5), were used in this study. The specificity of C. jejuni binding to Fn, via CadF, was demonstrated using antibodies reactive against Fn and CadF, as well as by the addition of exogenous Fn. The binding of C. jejuni F38011 and to Fn was inhibited by 54 and 56%, respectively, with the anti-cadf antibody. Based on these findings, the CadF protein was concluded to mediate the binding of both C. jejuni isolates to Fn. The adhesive nature of the CadF protein in C. jejuni was further revealed upon performing competitive inhibition binding assays with the C. jejuni F38011 isolate and C. jejuni F38011 cadf mutant. Here, only the C. jejuni F38011 wild-type

180 161 isolate was able to competitively inhibit the binding of C. jejuni to the INT 407 cells. Because Fn is associated with microfilaments, the role of microfilaments in C. jejuni uptake was examined. Inhibitor studies revealed that the C. jejuni F38011 and isolates require microfilament participation for efficient host cell entry. The reduction of C. jejuni uptake with cytochalasin D appeared specific as treatment of the INT 407 cells with this drug had no effect on INT 407 cell viability as judged by staining with trypan blue. Moreover, the effect of cytochalasin D, which inhibits actin polymerization and transient integrin-stimulated FAK activation (Schlaepfer et al., 1999), was reversible. Finally, treatment of the INT 407 cells with mycalolide B also inhibited C. jejuni uptake. Mycalolide B severs the microfilaments and sequesters G-actin. The sequestering of G-actin inhibits microfilament polymerization. Consistent with the notion that C. jejuni uptake requires the stimulation of host cell signaling molecules, the amount of phosphorylated paxillin significantly increased thirty minutes following C. jejuni infection. The amount of phosphorylated paxillin returned to a level equivalent to that of a noninfected control at the one hour time point. The increase in phosphorylated paxillin slightly proceeded and was concomitant with a sharp rise in the number of C. jejuni internalized, which occurs 30 to 60 minutes post-infection (Konkel et al., 1993). The inhibitory effect of microtubule-depolymerizing agents on the entry of C. jejuni strain has been noted previously (Bacon et al., 2000; Hu and Kopecko, 1999; Hu and Kopecko, 2000; Kopecko et al., 2001; Oelschlaeger et al., 1993). Based on this effect, the proposal has been put forth that C. jejuni are internalized via two pathways, one involving microtubules exclusively (considered a high efficiency uptake pathway) and the other involving microfilaments (considered a low efficiency uptake pathway) (Hu and Kopecko, 2000). While we noted a decrease in the invasiveness of C. jejuni with a microtubule inhibitor, a decrease was also noted in the number of intracellular bacteria with this organism in the presence of microfilament inhibitors. The discrepancy between our results and those reported earlier is most likely due to differences in assay

181 162 protocols. More specifically, in previous work, a 2 hr incubation with 100 µg/ml of gentamicin was used to kill the extracellular bacteria. In our hands, treatment of C. jejuni with 100 µg/ml of gentamicin for 2 hr typically resulted in the recovery of 3 to 12% of the bacteria in the original suspension (n = 4 individual experiments). Thus, even though C. jejuni is gentamicin-sensitive (e.g., MIC of 4 µg/ml), it was found necessary to increase both the time of exposure and the concentration of the antibiotic to ensure bacterial death. The protocol used by others to determine the bactericidal concentration of gentamicin was not reported (Bacon et al., 2000; Hu and Kopecko, 1999; Hu and Kopecko, 2000; Oelschlaeger et al., 1993). Also noteworthy is that an increase has been noted in the number of C. jejuni internalized in the presence of cytochalasin D (Hu and Kopecko, 1999; Hu and Kopecko, 2000; Oelschlaeger et al., 1993). One reason for the increase in C. jejuni uptake noted by others with cytochalasin D may be due to an increase in the number of organisms bound to the host cells. Increasing the number of cell-associated bacteria, while not using a bactericidal concentration of antibiotic, would mask any effect of cytochalasin D on C. jejuni uptake. While this explanation seems possible, we do not know the number of cell-associated (adherent) bacteria in earlier studies in which the effect of cytochalasin D on C. jejuni uptake was examined as it was not reported. Regardless, a sub-lethal concentration of gentamicin would not alter the net result of an assay unless there was a significant increase or decrease in the number of adherent bacteria in the test sample versus the control. To the best of our knowledge, C. jejuni is the only isolate to date that has been reported to be internalized exclusively via a microtubule-dependent pathway (Kopecko et al., 2001; Oelschlaeger et al., 1993). Using 250 µg/ml of gentamicin for 3 hr to kill the extracellular bacteria, Biswas et al. (Biswas et al., 2000) observed a reduction in entry with both microfilament and microtubule inhibitors with every C. jejuni clinical isolate tested (n = 9). Biswas et al. (Biswas et al., 2000) concluded that the most invasive isolates examined in their study utilized microfilaments.

182 163 Based on the data currently available, the most reasonable conclusion is that C. jejuni uptake involves cooperation of both microfilaments and microtubules. More specifically, uptake of C. jejuni and F38011 by INT 407 cells was reduced from 56 to 66%, respectively, in the presence of nocodazole and cytochalasin D relative to the untreated control (Table 7). These results are consistent with those reported elsewhere, even though the experimental protocols vary (Biswas et al., 2000; Oelschlaeger et al., 1993). What is unclear is why in the presence of both inhibitors a significant number of C. jejuni are still internalized. In comparison, Salmonella invasion was reduced by greater than 99% by cytochalasin D compared to untreated INT 407 cells. Comparable results for Salmonella have been reported by others (Bacon et al., 2000; Biswas et al., 2000; Hu and Kopecko, 1999; Hu and Kopecko, 2000; Oelschlaeger et al., 1993). The mechanism by which treatment of cells with microtubule inhibitors causes a reduction in C. jejuni-cell uptake is not known. However, it has been reported that microtubules regulate the turnover of focal adhesion contacts and modulate a cell s adhesive strength to the extracellular matrix (Ballestrem et al., 2000). In fact, treatment of cells with microtubule-inhibitors leads to an increase in a cell s adherence to the ECM (Ballestrem et al., 2000; Sastry & Burridge, 2000). Thus, the effect of a microtubule inhibitor on C. jejuni-cell uptake could be indirect, as the turnover of the focal adhesion sites is retarded. Alternatively, following initial microfilament-dependent uptake at the level of the plasma membrane, microtubules may be required for subsequent trafficking of the endosome to the interior of the cell as has been proposed for C. freundii (Badger et al., 1999). Regardless, there is clearly functional cooperation between host cytoskeletal elements (Ballestrem et al., 2000; Goode et al., 2000; Sastry and Burridge, 2000). Because the effects of cytochalasin D and nocodazole on C. jejuni uptake were not additive when used together, it appears most likely that the C. jejuni isolates tested here utilize microfilaments and microtubules together. C. jejuni strain has recently been reported to harbor two plasmids, one of which harbors the

183 164 tetracycline resistance gene (teto) and the other harboring a gene termed virb11 (Bacon et al., 2000). Moreover, in a recent review by Kopecko et al. (Kopecko et al., 2001), the authors suggested a possible correlation between the plasmid-borne genes and the microtubule-dependent uptake pathway. To ensure that the strain of C. jejuni used in this study harbored both plasmids, the isolate was subjected to PCR using teto and virb11 gene specific primers. The identity of the amplified products was subsequently confirmed upon sequencing of the PCRamplified products (not shown). Wooldridge et al. (Wooldridge et al., 1996) previously reported that C. jejuni uptake is reduced in the presence of protein tyrosine phosphorylation inhibitors. We chose to examine whether paxillin was phosphorylated upon infection of INT 407 cells with C. jejuni because protein tyrosine phosphorylation is one of the earliest events upon integrin stimulation (Clark & Brugge, 1995). Consistent with the idea that the binding of C. jejuni leads to integrin stimulation, an increase in phosphorylated paxillin was observed 30 and 45 min following C. jejuni F38011 infection. Noteworthy is that the increase in phosphorylated paxillin occurs just prior to and concomitant with an increase in C. jejuni internalization (Konkel et al., 1993). In contrast to the C. jejuni wild-type isolate, an increase was not observed in phosphorylated paxillin over the course of the assay with cells inoculated with the C. jejuni cadf mutant at a MOI of 100 to 1 (Fig. 4). However, upon infection of the INT 407 cells with the C. jejuni cadf mutant at an MOI of greater than 2,000 to 1, the pattern of phosphorylated paxillin in cells inoculated with the C. jejuni cadf mutant mirrored that obtained with the C. jejuni wild-type isolate. A possible explanation for this finding is that C. jejuni adherence is multifactorial, and that several adhesins simultaneously function to promote host cell binding, after which cell signaling events are stimulated. Thus, the CadF protein does not appear to be required to induce host cell signaling events, but appears to promote signaling events by facilitating the organism s binding to appropriate host cells receptors. Noteworthy is that Watarai et al. (Watarai et al., 1996) observed an increase in the tyrosine phosphorylation of FAK

184 165 and paxillin 20 and 30 min after infection of Chinese hamster ovary cells with Shigella flexneri. In summary, the data suggest that CadF promotes C. jejuni-host cell interactions. Consistent with the notion that bacterial uptake requires host cell signaling, an increase in tyrosine phosphorylated paxillin was observed upon infection of INT 407 cells with C. jejuni. Tyrosine phosphorylation of focal adhesion kinase and paxillin is a means of regulating host signal transduction events leading to actin rearrangement (Miyamoto et al., 1998; Tachibana et al., 1995). Because paxillin is an integral component of focal adhesions, as are ECM components including Fn, we speculate that CadF is responsible for promoting the initial interaction of C. jejuni with the appropriate host cell receptors involved in uptake. Future studies will be directed toward the identification and biochemical characterization of the CadF Fn-binding domain(s).

185 166 ACKNOWLEDGEMENTS We thank Dr. John Klena for performing the gentamicin-sensitivity assay, generation of the Cj1477c suicide vector, and PCR analysis to confirm plasmid carriage. We thank Joey Mickelson for assistance in generating the Cj1477c knockout. Finally, we thank Drs. John Klena (School of Molecular Biosciences, Washington State University, Pullman, Washington), Robert Heinzen (Department of Molecular Biology, University of Wyoming, Laramie, Wyoming), and Lynn Joens (Departments of Veterinary Science and Microbiology, University of Arizona, Tucson, Arizona) for reviewing this manuscript. This work was supported by a grants from the NIH (DK58911) and USDA National Research Initiative Competitive Grants Program (USDA/NRICGP, ) awarded to MEK.

186 167 REFERENCES Allos, B. M. & Blaser, M. J. (1995). Campylobacter jejuni and the expanding spectrum of related infections. Clin Infect Dis 20, Alrutz, M. A. & Isberg, R. R. (1998). Involvement of focal adhesion kinase in invasin-mediated uptake. Proc Natl Acad Sci USA 95, Bacon, D. J., Alm, R. A., Burr, D. H., Hu, L., Kopecko, D. J., Ewing, C. P., Trust, T. J. & Guerry, P. (2000). Involvement of a plasmid in virulence of Campylobacter jejuni Infect Immun 68, Badger, J. L., Stins, M. F. & Kim, K. S. (1999). Citrobacter freundii invades and replicates in human brain microvascular endothelial cells. Infect Immun 67, Ballestrem, C., Wehrle-Haller, B., Hinz, B. & Imhof, B. A. (2000). Actin-dependent lamellipodia formation and microtubule-dependent tail retraction control-directed cell migration. Mol Biol Cell 11, Baloda, S. B., Faris, A., Fröman, G. & Waldström, T. (1985). Fibronectin binding to Salmonella strains. FEMS Microbiol Lett 28, 1-5. Biswas, D., Itoh, K. & Sasakawa, C. (2000). Uptake pathways of clinical and healthy animal isolates of Campylobacter jejuni into INT-407 cells. FEMS Immunol Med Microbiol 29, Bolla, J.-M., Loret, E., Zalewski, M. & Pagès, J.-M. (1995). Conformational analysis of the Campylobacter jejuni porin. J Bacteriol 177, Clark, E. A. & Brugge, J. S. (1995). Integrins and signal transduction pathways: the road taken. Science 268, Cooper, J. A. (1987). Effects of cytochalasin and phalloidin on actin. J Cell Biol 105, Danen, E. H. & Yamada, K. M. (2001). Fibronectin, integrins, and growth control. J Cell Physiol 189, Dawson, J. R. & Ellen, R. P. (1990). Tip-oriented adherence of Treponema denticola to fibronectin. Infect Immun 58, Dawson, J. R. & Ellen, R. P. (1994). Clustering of fibronectin adhesins toward Treponema denticola tips upon contact with immobilized fibronectin. Infect Immun 62, De Melo, M. A. & Pechère, J.-C. (1990). Identification of Campylobacter jejuni surface proteins that bind to eucaryotic cells in vitro. Infect Immun 58, Fauchère, J.-L., Kervella, M., Rosenau, A., Mohanna, K. & Véron, M. (1989). Adhesion to HeLa cells of Campylobacter jejuni and C. coli outer membrane components. Res Microbiol 140,

187 Fauchère, J. L., Rosenau, A., Véron, M., Moyen, E. N., Richard, S. & Pfister, A. (1986). Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infect Immun 54, Finlay, B. B., Ruschkowski, S. & Dedhar, S. (1991). Cytoskeletal rearrangements accompanying Salmonella entry into epithelial cells. J Cell Sci 99, Francis, C. L., Ryan, T. A., Jones, B. D., Smith, S. J. & Falkow, S. (1993). Ruffles induced by Salmonella and other stimuli direct macropinocytosis of bacteria. Nature 364, Francis, C. L., Starnbach, M. N. & Falkow, S. (1992). Morphological and cytoskeletal changes in epithelial cells occur immediately upon interaction with Salmonella typhimurium grown under low-oxygen conditions. Mol Microbiol 6, Fröman, G., Switalski, L. M., Faris, A., Wadström, T. & Höök, M. (1984). Binding of Escherichia coli to fibronectin. J Biol Chem 259, Goode, B. L., Drubin, D. G. & Barnes, G. (2000). Functional cooperation between the microtubule and actin cytoskeletons. Curr Opin Cell Biol 12, Gumbiner, B. M. (1996). Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 84, Hu, L. & Kopecko, D. J. (1999). Campylobacter jejuni associates with microtubules and dynein during invasion of human intestinal cells. Infect Immun 67, Hu, L. & Kopecko, D. J. (2000). Interactions of Campylobacter with eukaryotic cells: gut luminal colonization and mucosal invasion mechanisms. In Campylobacter, 2nd Edition, pp Edited by I. Nachamkin & M. J. Blaser. Wash., D.C.: ASM. Humason, G. L. (1979). Animal Tissue Techniques. San Francisco: W. H. Freeman and Company. Hynes, R. O. (1992). Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, Isberg, R. R. & Tran Van Nhieu, G. (1994). Two mammalian cell internalization strategies used by pathogenic bacteria. Annu Rev Genet 28, Jaffe, J., Natanson-Yaron, S., Caparon, M. G. & Hanski, E. (1996). Protein F2, a novel fibronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol Microbiol 21, Jin, S., Joe, A., Lynett, J., Hani, E. K., Sherman, P. & Chan, V. L. (2001). JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol Microbiol 39, Kelle, K., Pagès, J.-M. & Bolla, J.-M. (1998). A putative adhesin gene cloned from Campylobacter jejuni. Res Microbiol 149,

188 Kervella, M., Pagès, J.-M., Pei, Z., Grollier, G., Blaser, M. J. & Fauchère, J.-L. (1993). Isolation and characterization of two Campylobacter glycine-extracted proteins that bind to HeLa cell membranes. Infect Immun 61, Konkel, M. E., Garvis, S. G., Tipton, S. L., Anderson, D. E., Jr. & Cieplak, W., Jr. (1997). Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol Microbiol 24, Konkel, M. E., Gray, S. A., Kim, B. J., Garvis, S. G. & Yoon, J. (1999). Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadf virulence gene and its product. J Clin Microbiol 37, Konkel, M. E., Mead, D. J. & Cieplak, W., Jr. (1993). Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J Infect Dis 168, Kopecko, D. J., Hu, L. & Zaal, K. J. (2001). Campylobacter jejuni - microtubule-dependent invasion. Trends Microbiol 9, Kuusela, P. (1978). Fibronectin binds to Staphylococcus aureus. Nature (Lond) 276, Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, Lyngstadaas, A., Lobner-Olesen, A., Grelland, E. & Boye, E. (1999). The gene for 2- phosphoglycolate phosphatase (gph) in Escherichia coli is located in the same operon as dam and at least five other diverse genes. Biochim Biophys Acta 1472: McSweegan, E. & Walker, R. I. (1986). Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infect Immun 53, Mejillano, M. R., Shivanna, B. D. & Himes, R. H. (1996). Studies on the nocodazole-induced GTPase activity of tubulin. Arch Biochem Biophys 336, Miyamoto, S., Katz, B.-Z., LaFrenie, R. M. & Yamada, K. M. (1998). Fibronectin and integrins in cell adhesion, signaling, and morphogenesis. Ann N.Y. Acad Sci 857, Moser, I. & Schröder, W. (1995). Binding of outer membrane preparations of Campylobacter jejuni to INT 457 cell membranes and extracellular matrix proteins. Med Microbiol Immunol 184, Moser, I., Schröder, W. & Salnikow, J. (1997). Campylobacter jejuni major outer membrane protein and a 59-kDa protein are involved in binding to fibronectin and INT 407 cell membranes. FEMS Mirobiol Lett 157, Moser, I., Schröder, W. F. K. J. & Hellmann, E. (1992). In vitro binding of Campylobacter jejuni/coli outer membrane preparations to INT 407 cell membranes. Med Microbiol Immunol 180,

189 170 Myhre, E. B. & Kuusela, P. (1983). Binding of human fibronectin to group A, C, and G Streptococci. Infect Immun 40, Oelschlaeger, T. A., Guerry, P. & Kopecko, D. J. (1993). Unusual microtubule-dependent endocytosis mechanisms triggered by Campylobacter jejuni and Citrobacter freundii. Proc Natl Acad Sci USA 90, Parkhill, J., Wren, B.W., Mungall, K., Ketley, J.M., Churcher, C., Basham, D., Chillingworth, T., Davies, R.M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A.V., Moule, S., Pallen, M.J., Penn, C.W., Quail, M.A., Rajandream, M.A., Rutherford, K.M., van Vliet, A.H., Whitehead, S. & Barrell, B.G. (2000). The genome sequence of the foodborne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, Pei, Z. & Blaser, M. J. (1993). PEB1, the major cell-binding factor of Campylobacter jejuni, is a homolog of the binding component in Gram-negative nutrient transport systems. J Biol Chem 268, Pei, Z., Burucoa, C., Grignon, B., Baqar, S., Huang, X.-Z., Kopecko, D. J., Bourgeois, A. L., Fauchere, J.-L. & Blaser, M. J. (1998). Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect Immun 66, Quaroni, A., Isselbacher, K. J. & Ruoslahti, E. (1978). Fibronectin synthesis by epithelial crypt cells of rat small intestine. Proc Natl Acad Sci USA 75, Roberts, D. D. (1990). Interactions of respiratory pathogens with host cell surface and extracellular matrix components. Am J Respir Cell Mol Biol 3, Rydén, C., Rubin, K., Speziale, P., Höök, M., Lindberg, M. & Wadström, T. (1983). Fibronectin receptors from Staphylococcus aureus. J Biol Chem 258, Saito, S., Watabe, S., Ozaki, H., Fusetani, N. & Karaki, H. (1994). Mycalolide B, a novel actin depolymerizing agent. J Biol Chem 269, Saito, S. Y., Watabe, S., Ozaki, H., Kobayashi, M., Suzuki, T., Kobayashi, H., Fusetani, N. & Karaki, H. (1998). Actin-depolymerizing effect of dimeric macrolides, bistheonellide A and swinholide A1. J Biochem (Tokyo) 123, Sarkar, S. (1999). Focal adhesions. Curr Biol 9, R428. Sastry, S. K. & Burridge, K. (2000). Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics. Exp Cell Res 261, Schlaepfer, D. D., Hauck, C. R. & Sieg, D. J. (1999). Signaling through focal adhesion kinase. Prog Biophys Mol Biol 71, Schorey, J. S., Holsti, M. A., Ratliff, T. L., Allen, P. M. & Brown, E. J. (1996). Characterization of the fibronectin-attachment protein of Mycobacterium avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol Microbiol 21,

190 Schröder, W. & Moser, I. (1997). Primary structure analysis and adhesion studies on the major outer membrane protein of Campylobacter jejuni. FEMS Microbiol Lett 150, Tachibana, K., Sato, T., D'Avirro, N. & Morimoto, C. (1995). Direct association of pp125fak with paxillin, the focal adhesion-targeting mechanism of pp125fak. J Exp Med 182, Thomas, D. D., Baseman, J. B. & Alderete, J. F. (1985). Fibronectin mediates Treponema pallidum cytadherence through recognition of fibronectin cell-binding domain. J Exp Med 161, van der Flier, A. & Sonnenberg, A. (2001). Function and interactions of integrins. Cell Tissue Res 305, van Putten, J. P., Duensing, T. D. & Cole, R. L. (1998). Entry of OpaA+ gonococci into HEp- 2 cells requires concerted action of glycosaminoglycans, fibronectin and integrin receptors. Mol Microbiol 29, Visai, L., Bozzini, S., Petersen, T. E., Speciale, L. & Speziale, P. (1991). Binding sites in fibronectin for an enterotoxigenic strain of E. coli B342289c. FEBS Lett 290, Vuori, K. (1998). Integrin signaling: tyrosine phosphorylation events in focal adhesions. J Membrane Biol 165, Wada, S., Matsunaga, S., Saito, S., Fusetani, N. & Watabe, S. (1998). Actin-binding specificity of marine macrolide toxins, mycalolide B and kabiramide D. J Biochem (Tokyo) 123, Wassenaar, T. M. & Blaser, M. J. (1999). Pathophysiology of Campylobacter jejuni infections of humans. Microbes Infect 1, Watarai, M., Funato, S. & C., S. (1996). Interaction of Ipa proteins of Shigella flexneri with α 5 β 1 integrin promotes entry of the bacteria into mammalian cells. J Exp Med 183, Wooldridge, K. G., Williams, P. H. & Ketley, J. M. (1996). Host signal transduction and endocytosis of Campylobacter jejuni. Microb Pathog 21, Ziprin, R. L., Young, C. R., Hume, M. E. & Konkel, M. E. (1999). The absence of cecal colonization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectinbinding protein. Avian Dis 43,

191 Table 1. Competitive inhibition of C. jejuni binding to Fn with antibodies reactive against either Fn or CadF and upon the addition of exogenous Fn 172 Substrate (40 µg) Inhibitor Percent of bound bacteria* C. jejuni F38011 C. jejuni Fn BSA Fn αfn Fn αcadf Fn exogenous Fn *Assays were performed as outlined in Materials and Methods. For each sample, the bacteria in each of three randomly chosen fields were counted. Results are presented as the mean + standard deviation of triplicate determinations for each sample relative to the binding of the C. jejuni wildtype isolates to Fn (C. jejuni F38011 = CFU and C. jejuni = CFU). Note: The C. jejuni cadf mutant bound to Fn and BSA at 9 + 4% and 7 + 3% compared to the C. jejuni F38011 wild-type isolate, respectively. The value was significantly different ( P value < 0.01) from that obtained for the binding of the C. jejuni F38011 and wild-type isolates to Fn in the absence of a potential inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data.

192 Table 2. The functional role of CadF as an adhesin is conserved in C. jejuni clinical isolate Competitor Number of the C. jejuni wild-type (TetR) isolate bound No competitor ( ) X % C. jejuni cadf mutant 44-fold excess ( ) X %* 94-fold excess ( ) X % C. jejuni F fold excess ( ) X % 88-fold excess ( ) X % *Values represent the number of C. jejuni bound in the presence of the competing organism divided by the number of C. jejuni bound in the absence of the competitor, multiplied by 100. The value was significantly different ( P value < 0.01) from that obtained using the C. jejuni wild-type isolate in the absence of the competitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data.

193 174 Table 3. Effect of microfilament inhibitors on C. jejuni binding and internalization* Concentration of cytochalasin D No inhibitor 0.2 µm 1 µm 2 µm C. jejuni F38011 Adherent 100 ( ) X Internalized 100 ( ) X C. jejuni Adherent 100 ( ) X Internalized 100 ( ) X S. typhimurium SL1344 Adherent 100 ( ) X Internalized 100 ( ) X C. freundii 8090 Adherent 100 ( ) X Internalized 100 ( ) X *Assays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates in the absence of the inhibitor. The value was significantly different ( P value < 0.01) from that obtained using the wild-type isolate in the absence of the inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data.

194 Table 4. Effect of mycalolide B microfilament inhibitor on C. jejuni binding and internalization* 175 Concentration of mycalolide B No inhibitor µm 0.05 µm 0.25 µm C. jejuni F38011 Adherent 100 ( ) X Internalized 100 ( ) X C. jejuni Adherent 100 ( ) X Internalized 100 ( ) X S. typhimurium SL1344 Adherent 100 ( ) X ND Internalized 100 ( ) X ND *Assays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates in the absence of the inhibitor. The value was significantly different ( P value < 0.01) from that obtained using the wild-type isolate in the absence of the inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data. ND, not done.

195 176 Table 5. Reversibility of cytochalasin D on C. jejuni binding and internalization* 1st incubation period No inhibitor Cytochalasin D (2 µm) Cytochalasin D (2 µm) 2nd incubation period No inhibitor Cytochalasin D (2 µm) No inhibitor C. jejuni F38011 Adherent ( ) X 106 ( ) X 106 ( ) X % % % Internalized ( ) X 104 ( ) X 104 ( ) X % % % C. jejuni Adherent ( ) X 105 ( ) X 105 ( ) X % % % Internalized ( ) X 104 ( ) X 104 ( ) X % % % *Assays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates in the absence of the inhibitor. The value was significantly different ( P value < 0.01) from that obtained using the wild-type isolate in the absence of the inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data.

196 177 Table 6. Effect of microtubule inhibitors on C. jejuni binding and internalization* Concentration of nocodazole No inhibitor 1 µm 10 µm 20 µm C. jejuni F38011 Adherent 100 ( ) X Internalized 100 ( ) X C. jejuni Adherent 100 ( ) X Internalized 100 ( ) X S. typhimurium SL1344 Adherent 100 ( ) X 105 ND ND Internalized 100 ( ) X 105 ND ND C. freundii 8090 Adherent 100 ( ) X Internalized 100 ( ) X *Assays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates in the absence of the inhibitor. The value was significantly different ( P value < 0.01) from that obtained using the wild-type isolate in the absence of the inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data. ND, not done.

197 178 Table 7. Combined effects of microtubule and microfilament inhibitors on C. jejuni binding and internalization* No inhibitor Nocodazole (20 µm) Cytochalasin D (2 µm) Nocodazole (20 µm) + Cytochalasin D (2 µm) C. jejuni F38011 Adherent ( ) X 106 ( ) X 106 ( ) X 107 ( ) X % % % % Internalized ( ) X 105 ( ) X 104 ( ) X 104 ( ) X % % % % C. jejuni Adherent ( ) X 106 ( ) X 106 ( ) X 106 ( ) X % % % % Internalized ( ) X 105 ( ) X 105 ( ) X 105 ( ) X % % % % *Assays were performed as outlined in Materials and Methods. Results are presented as the percent adherent and internalized bacteria relative to the values obtained for the wild-type isolates in the absence of the inhibitors. The value was significantly different ( P value < 0.01) from that obtained using the wild-type isolate in the absence of the inhibitor. Significance between samples was determined using Student s t test following logarithmic [(log x (base 10)] transformation of the data.

198 179 Figure 1. Representative gel (Panel A) and immunoblot (Panel B) showing the detection of the CadF protein in the whole-cell extracts of C. jejuni F38011, , and an F38011 cadf mutant. Bacterial whole-cell extracts (25 µg per lane) were separated in 12.5% SDS-PAGE gels, transferred to PVDF membranes, and reacted with a goat anti-37 kda serum. A 37 kda band (indicated by arrow), corresponding to the CadF protein, was detected in the whole cell extracts of both the C. jejuni F38011 and clinical isolates. The faster migrating band, with an approximate mass of 32 kda, represents the heat-modifiable form of the protein characteristic of outer membrane proteins (Bolla et al., 1995). Lanes: 1, C. jejuni (biotype I, Lior serotype 5); 2, C. jejuni F38011 (biotype I, Lior serotype 90); and 3, C. jejuni F38011 cadf mutant. The relative positions of the standards are given in kda on the left.

199 180 Figure 1. A B

200 181 Figure 2. Indirect immunofluorescence of C. jejuni, S. typhimurium, and C. freundii infected INT 407 cells. Bacteria were reacted with the appropriate polyclonal antibody and stained using a rhodamine labelled secondary antibody. Microfilaments (e.g., F-actin ) were stained using FITClabelled phalloidin. Panels: Uninfected INT 407 cells (Panel A); Cytochalasin D (2 µm) treated INT 407 cells (Panel B); C. jejuni F38011 (Panel C); C. jejuni (Panel D); S. typhimurium SL1344 (Panel E); and C. freundii 8090 (Panel F). The organism in Panel D is indicated (arrow) to ease its visualization. Panels A and B represent the same magnification, as well as panels C-F.

201 Figure

202 183 Figure 3. Indirect immunofluorescence of C. jejuni, S. typhimurium, and C. freundii infected INT 407 cells. Bacteria were reacted with the appropriate polyclonal antibody and stained using a rhodamine labelled secondary antibody. Microtubules were stained using a FITC-labelled monoclonal antibody against tubulin. Panels: Uninfected INT 407 cells (Panel A); Nocodazole (20 µm) treated INT 407 cells (Panel B); C. jejuni F38011 (Panel C); C. jejuni (Panel D); S. typhimurium SL1344 (Panel E); and C. freundii 8090 (Panel F). Panels A and B represent the same magnification, as well as panels C-F.

203 Figure

204 185 Figure 4. Binding of C. jejuni to INT 407 cells induces phosphorylation of the 68 kda focal adhesion-associated protein paxillin. The C. jejuni infected INT 407 cells were lysed and samples immunoprecipitated using an anti-paxillin antibody. The precipitated proteins were separated in 10% polyacrylamide gels and proteins transferred to PVDF membranes. Immunoblots were reacted with an anti-phosphotyrosine antibody (αpt-66) and an anti-paxillin antibody as described in Materials and Methods.

205 186 Figure 4. C. jejuni wild-type isolate α PT-66 Post-infection (min) α Paxillin Post-infection (min) phosphorylated paxillin paxillin C. jejuni cadf mutant α PT-66 Post-infection (min) α Paxillin Post-infection (min) phosphorylated paxillin paxillin

206 187 Chapter 6 Identification of a fibronectin-binding domain within the Campylobacter jejuni CadF adhesin Marshall R. Monteville, John D. Klena, and Michael E. Konkel School of Molecular Biosciences, Washington State University, Pullman, WA Manuscript in preparation

207 188 ABSTRACT The binding of Campylobacter jejuni to fibronectin (Fn), a component of the extracellular matrix, is mediated by a 37 kda outer membrane protein termed CadF for Campylobacter adhesion to fibronectin (Fn). Previous studies indicate that the maximal adherence and invasion of human intestinal epithelial cells by C. jejuni requires the organism to bind to the extracellular matrix (ECM) component Fn. To further characterize the interaction of the CadF adhesin with Fn, experiments were performed to identify the domain(s) that display Fn binding activity. Using synthesized peptides (30-mers) with overlapping regions (10-mers), maximal Fn binding activity was found to be localized to a region between amino acid residues Subsequent experiments using peptides (16-mers) with 13-mer overlaps narrowed the Fn-binding domain to amino acid residues , corresponding to residues FRSL. Peptides containing the FRSL residues competitively inhibited Fn binding to full-length recombinant CadF in a dose-dependent manner. The sequence of amino acids was also found to be essential in binding to Fn as a scrambled version of the peptide neither bound Fn or was capable of blocking Fn association with recombinant CadF. Collectively, these data suggest that the C. jejuni CadF adhesin contains a single Fn-binding domain.

208 189 INTRODUCTION Campylobacter jejuni, a Gram-negative spiral shaped organism, is recognized as the leading cause of bacterial-induced gastroenteritis in developed nations worldwide (Tauxe, 1992; Altekruse et al., 1999). Virulence attributes that contribute to the development of C. jejuni-mediated enteritis include adherence (De Melo & Pechère, 1990; Fauchère et al., 1989; Jin et al., 2001; Kelle et al., 1998; Kervella et al., 1993; Konkel et al., 1997; McSweegan & Walker, 1986; Moser & Schröder, 1995; Moser et al., 1997; Moser et al., 1992; Pei & Blaser, 1993; Pei et al., 1998; Schröder & Moser, 1997), translocation (Brás and Ketley, 1999; Everest et al., 1992; Harvey et al., 1999; Konkel et al., 1992), invasion (Everest et al., 1992; Konkel et al., 1992; Konkel et al., 1993; Russell et al., 1993), and toxin production (Pickett and Whitehouse, 1999; Wassenaar, 1997). In a broader sense, the ability of C. jejuni to colonize the intestinal tract of humans is proposed to play an early role in the development of C. jejuni-mediated enteritis. Factors that influence the ability of C. jejuni to colonize a host include the organism s motility, chemotactic behavior, and assortment of adhesins. Adhesins are surface-exposed molecules that facilitate a pathogen s attachment to host cell receptor molecules. Fauchere et al. (Fauchere et al., 1986) found that C. jejuni strains isolated from patients with fever and diarrhea adhered to cultured cells at a greater efficiency than those strains isolated from asymptomatic individuals. As such, in vitro adherence assays have been used extensively to characterize the interactions of C. jejuni with host cells and to attempt to identify the bacterial molecules that mediate host cell binding. To date, the best characterized C. jejuni adhesins are CadF, JlpA, and PEB1 (Konkel et al., 1997; Jin et al., 2001; Pei et al., 1998). The role of CadF and PEB1 as adhesins is supported by the experimental inoculation of animals with C. jejuni mutants. For example, a C. jejuni cadf mutant lacks the ability to colonize the cecum of newly hatched leghorn chickens (Ziprin et al., 1999). The C. jejuni peb1a null mutant exhibits a

209 190 reduction in the duration of mouse intestinal colonization when compared to a C. jejuni wild-type isolate (Pei et al., 1998). Recently, Jin et al. (2003) reported that JlpA binds to Hsp90α receptor on HEp-2 cells, which in turn activates signaling pathways leading to NF-kappaB and p38 MAP kinase activation. Other molecules, which are presently less well characterized than CadF, PEB1, and JlpA, that have been proposed to function as adhesins include the flagellum, lipopolysaccharide (McSweegan and Walker, 1986; Moser et al., 1992), the major outer membrane protein (MOMP, also called OmpE) (Moser et al., 1997; Schröder and Moser, 1997), and P95 (Kelle et al., 1998). The C. jejuni CadF [Campylobacter adhesion to Fibronectin (Fn)] adhesin, which is a 37 kda outer membrane protein, mediates the organism s binding to the extracellular matrix component Fn (Konkel et al., 1997). Based on in vitro assays, CadF appears to be required for the maximal adherence and invasion of INT 407 intestinal epithelial cells and T84 human colonic cells by C. jejuni (Konkel et. al., 1997; Monteville and Konkel, 2002). Current evidence indicates that CadF promotes the organism s binding to Fn, which is localized to the basolateral surface of cells, following its translocation across an intestinal cell barrier via a paracellular route (Monteville and Konkel, 2002). It also appears that the binding of C. jejuni to Fn induces host cell signalling events (e.g. phosphorylation of paxillin), which in turn leads to the rearrangement of the host cytoskeleton (Monteville et. al., 2003). Rearrangement of the cytoskeleton, involving cooperative interaction between both microtubules and microfilaments, results in the maximal uptake of C. jejuni by host cells (Monteville et al., 2003). Fibronectin (Fn) is a 220 kda glycoprotein that is present at regions of cell-to-cell contact in the gastrointestinal epithelium, thereby providing a potential binding site for pathogens (Quaroni et al., 1978). Several bacterial pathogens possess the ability to bind to Fn, including Staphylococcus aureus (Kuusela, 1978; Rydén et al., 1983), Streptococcus pyogenes (Jaffe et al., 1996; Myhre & Kuusela, 1983), Salmonella enteritidis (Baloda et al., 1985), Escherichia coli (Fröman et al., 1984;

210 191 Visai et al., 1991), Neisseria gonorrhoeae (van Putten et al., 1998), Mycobacterium avium (Schorey et al., 1996), and Treponema species (Dawson & Ellen, 1990; Dawson & Ellen, 1994; Thomas et al., 1985) indicating a common theme among these pathogens. Extending this initial work, some investigators have undertaken studies to identify Fn-binding molecules within certain pathogens as well as to determine the domain(s) within the adhesins that elicit Fn-binding activity. Schorey et. al. (Schorey et. al., 1996) identified two Fn-binding domains within the Mycobacterium avium Fn attachment protein-a (FAP-A) adhesin. These Fn-binding domains were localized to two non-contiguous 24-amino acid regions that were highly conserved within M. avium (FAP-A), Mycobacterium leprae (FAP-L), and Mycobacterium tuberculosis (FAP-TB). In a subsequent study, Zhao et. al. (Zhao et. al. 1999) found that the binding of M. avium FAP-A protein to Fn was inhibited using a single domain consisting of 12 amino acids [FAP-A-( )]. Alanine substitutions at amino acids correlated into a loss of FAP-A peptide adherence to Fn, leading to the conclusion that the 273RWFV276 amino acids play an essential role in the binding of FAP-A to Fn. This study was initiated to further characterize the C. jejuni CadF adhesin by identifying the CadF Fn-binding domain(s). Using synthesized peptides corresponding to that of the C. jejuni CadF adhesin, we were able to identify a single domain within CadF which facilitates adherence to human Fn.

211 192 MATERIALS AND METHODS Synthesis of CadF peptides. The CadF peptides were synthesized on a semimanual peptide synthesizer using standard fluorenylmethoxycarbonyl chemistry (Genemed Synthesis, Inc., San Francisco, CA). The peptides were analyzed by mass spectrometry and high performance liquid chromatography with C-18 column chromatography. Peptide-fibronectin binding assay. Greiner Bio-One (Frickenhausen, Germany) 96-well plates were coated overnight at 25 C with 1.0 µg of peptides unless otherwise indicated. Plates were then blocked with bovine serum albumin (BSA) for 1 hr at 25 C, rinsed with wash buffer (PBS, 0.1% w/v BSA, and 0.01% v/v Tween 20), followed by addition of 1.0 µg of fibronectin (Sigma, St. Louis, MO) in incubation buffer (PBS supplemented with 0.02% w/v BSA) per well. After 3 hr at 25 C, wells were rinsed with wash buffer and a 1:750 dilution of rabbit anti-fn antibody (Telios Pharmaceuticals, Inc., San Diego, CA) was added and incubated at 25 C for 1 hr. Plates were washed and bound antibody detected with a 1:1000 dilution of goat anti-rabbit horseradish peroxidase conjugate incubated at 25 C for 30 min followed by addition of TMB Substrate Solution (Endogen, Woburn, MA). Plates coated with BSA alone served as a negative control. Peptide inhibition assay. Greiner Bio-One 96-well plates were coated overnight at 25 C with 0.2 µg of recombinant CadF (rcadf) per well. Plates were then blocked with BSA as described previously. During blocking, 1.0 µg of fibronectin was added to incubation buffer in the presence of 10, 25, or 50 µg of specified peptide. Following blocking procedure, fibronectin alone (1.0 µg), or fibronectin incubated with peptides were added to wells coated with rcadf. After a 2 hr incubation at 25 C, wells were rinsed with wash buffer and the quantity of fibronectin bound to rcadf determined using primary and secondary antibodies as outlined above. Plates coated with

212 193 BSA alone served as a negative control. Expression of cadf in Escherichia coli. To express the Campylobacter jejuni F38011 cadf gene in Escherichia coli, the pbad system which uses the arabinose-inducible arac promoter, was employed. Oligonucleotide primers complimentary to C. jejuni F38011 cadf (cadf NcoI F 5 - ATA CCA TGG CAA AAA AGT TAT TAC TAT GTT TAG G; cadf Nco+1 R 5 -ATC CAT GGA TCT TAG GAT AAA TTT AGC ATC CAC T) (Invitrogen, Carlsbad, CA) were used to amplify the full-length cadf gene (967 nucleotides) without the stop codon. In addition, amplification of the C. jejuni F38011 genomic cadf gene with these primers resulted in the placement of NcoI restriction enzyme sites at each end of the PCR amplicon. For the PCR, 250 ng of C. jejuni F38011 genomic DNA was mixed with (all concentrations refer to final concentration) 2.5 mm dntps (Roche, Indianapolis, IN), 2 mm MgCl 2, 1.2 pmol/ml each primer, 1X PCR buffer (Roche), 2.5 U Taq polymerase and sterile distilled water to make a 100 ml reaction volume. Amplification was allowed to occur over 30 cycles; each cycle consisted of 1 min at 92 C, 1 min at 55 C and 1 min at 72 C, except for the final cycle in which the last step was extended for 5 min. Five percent of the PCR mixture was loaded into a 1% agarose gel, subjected to an electrical current and visualized using ethidium bromide/uv light. The PCR amplicon was passed through a Qiaquick column (Qiagen, Valencia, CA), concentrated into 30 ml of ddh 2 O and subsequently restriction digested with NcoI as per the manufacturer s instructions. NcoI-digested amplicons were gel-purified and agarose removed using a second Qiaquick column prior. The vector pbadc was digested with NcoI as per the manufacturer s instructions, precipitated with 7.5 M ammonium acetate and 100% ethanol and resuspended in sterile ddh 2 O. Calf intestinal alkaline phosphatase (Promega, Madison, WI) was added as per the manufacturer s instructions and dephosphorylated, NcoI-digested pbadc was purified using a Qiaquick spin column. Ligations,

213 194 using T4 DNA ligase (Invitrogen) were set up using the cadf amplicon and the pbadc vector at a 3:1, overnight at 14 C. Salmon sperm and ammonium acetate/100% ethanol were used to precipitate the ligation reactions, and an aliquot of the ligation mixture was used to transform E. coli INVα F cells to ampicillin-resistance. A number of transformants were screened for the presence of an NcoI insert and positive clones were subjected to nucleotide sequence analysis (Washington State Sequencing Center, Pullman, WA) to determine orientation and correctness using primers designed to the pbadc vector sequence. Purification of recombinant CadF. LMG194 pbadc::cadf was inoculated into 3 ml of Luria- Bertani (LB) containing ampicillin (50 µg/ml) with LMG194 pbadc::cadf and incubated overnight in 37 C incubator with shaking. Overnight broth was then transferred into 250 ml of LB containing ampicillin ( 50 µg/ml) and incubate at 37 C with shaking until OD 600 = 0.5. Arabinose (133 mm) was added to the broth culture and incubate as described above an additional 4 hr. Cultures were then centrifuged for 10 min at 9,000 x g. Supernatants were discarded and the pellet resuspended in 10 ml extraction/wash buffer (50 mm NaPO 4, 6 M urea, and 300 mm NaCl at ph 7.0). Suspensions were then sonicated and centrifuged for 15 min at 14,000 x g. The bicinchoninic acid assay was used to quantify protein concentration according to manufacturers specifications (Pierce, Rockford, IL). The supernatant (clarified lysate) was then placed in a 15 ml conical tube containing 3 ml talon resin/10mg protein (Clonetech, Palo Alto, CA) and incubated at 4 C overnight with rocking. Talon/resin suspension were then run through a column and washed three times with 1 ml of extraction/wash buffer. Recombinant CadF was then eluted with elution buffer (45 nm NaPO 4, 5.4 M urea, 270 mm NaCl and 150 mm imidazole at ph 6.0) and 1 ml fractions were collected to isolate purified recombinant CadF.

214 195 RESULTS The FRLS residues within CadF are essential for Fn-binding activity. To identify the domain(s) within the CadF protein that elicit Fn-binding activity, peptides (30-mers) were synthesized with overlapping regions (10-mers) corresponding to the entire CadF protein minus the cleavable N-terminal signal sequence. The amino acid residues of each of these peptides is listed in Table 1. Peptides extending from amino acids exhibited maximal Fn-binding activity as determined by enzyme-linked immunosorbant assay (ELISA) (Fig. 1). The other peptides, containing amino acid residues prior to 117 and beyond 147 exhibited less Fn binding activity. Based on these data, we concluded that the CadF protein had a single Fn-binding domain that was localized between amino acid residues To determine the residues within the CadF protein that are essential for Fn-binding activity, peptides (16-mers) with overlapping regions (4-mers) were synthesized that corresponded to amino residues As a negative control, a peptide containing the amino acids was included in the assay. A comparison of the Fn-binding activity of the 16-mers revealed that amino acids are required for Fn-binding activity (Fig. 2). The sequence 134FRLS137 was deemed specific as these four residues in scrambled orientation ( scrambled) did not display Fnbinding activity. In agreement with the data shown in Fig. 1, residues displayed only minimal levels of Fn-binding activity. The binding of rcadf to Fn is specific. To confirm the role of 134FRLS137 residues in Fnbinding activity, an ELISA assay was used whereby wells were coated with full-length rcadf, and Fn mixed with either CadF-test or control peptides to attempt to block the binding of Fn to the rcadf protein. The test peptide harbored the FRLS residues. Two peptides were included as controls. The first control peptide, which was found previously not to display any Fn-binding

215 196 activity, harbored residues (SDSLALRLETRDQINF). The second control peptide consisted of residues in scrambled orientation (LRRDLLSLSTREADFQ). The peptide that contained the FRLS residues inhibited the binding of Fn to the rcadf protein in a dosedependent manner (Fig. 3). In contrast, the peptide containing residues had no effect on the ability of Fn to bind to the rcadf protein. In addition, the scrambled peptide was unable to compete with the binding of Fn to the rcadf protein. These data suggest that the 134FRLS137 residues within the CadF protein are critical for Fn binding.

216 197 DISCUSSION A substantial amount of work indicates that the binding of Campylobacter jejuni to Fn is mediated by CadF. An anti-cadf antibody has previously been shown to reduce the binding of two C. jejuni clinical isolates to immobilized Fn by greater than 50%. Further, in vitro binding and internalization assays revealed that the binding of a C. jejuni cadf mutant to INT 407 cells is reduced by 50%, with an accompanying decrease in cell invasion, when compared to the wild-type isolate. More specifically, evidence indicates that CadF promotes the organism s binding to Fn, which is localized to the basolateral surface of cells, following its translocation across an intestinal cell barrier via a paracellular route (Monteville and Konkel, 2002). The binding of C. jejuni to Fn, via CadF, also leads to host cell invasion. More specifically, we have reported that the amount of phosphorylated paxillin increases significantly within 30 minutes post-inoculation of INT 407 cells with a wild-type isolate of C. jejuni. In contrast to this finding, an increase was not observed in the amount of phosphorylated paxillin in cells inoculated with the C. jejuni cadf mutant unless the inoculum was increased greater than 2,000 bacteria per one cell. The phosphorylation of paxillin is indicative of a host cell signalling event that can result in the rearrangement of the host cytoskeleton (Monteville et. al., 2003). Rearrangement of the cytoskeleton, involving cooperative interaction between both microtubules and microfilaments, results in the maximal uptake of C. jejuni by host cells (Monteville et al., 2003). The role of CadF as an adhesin has been further validated upon performing in vivo studies. In contrast to a C. jejuni F38011 wild-type isolate, a C. jejuni F38011 cadf mutant is unable to colonize the intestinal tract of Leghorn chickens (Ziprin et. al. 1999). In total, the data indicate that the CadF protein plays an important role in colonization of a host as well as in the development C. jejuni-mediated enteritis. We undertook this study to further define the molecular interaction of CadF binding to Fn. Based on the results presented, CadF harbors a single Fn-binding domain in which four amino

217 198 acids (134FRLS137) are crucial in facilitating adherence. Furthermore, this particular sequence of amino acids appears essential in maintaining Fn-binding activity based on competition assays. The competitive inhibition assays with the various peptides, in particular a peptide containing the 134FRLS137 domain, suggested that this domain was the sole Fn-binding domain as it was capable of completely blocking Fn association with the full-length CadF recombinant protein. If an alternative Fn-binding domain did exist within the CadF protein, then we predict that the FRLScontaining peptide would have only partially blocked the association of Fn to the CadF protein. In summary, we identified a single domain within the CadF protein that is capable of binding Fn; the FRLS sequence of amino acids in CadF appears essential for Fn-binding activity. Of interest is the fact that this specific sequence of amino acids has not been identified in other Fn-binding proteins. For example, the RWFV residues are responsible for the binding of the FAP-A protein from M. avium to Fn (Zhao et al., 1999). Additional work is currently underway in our laboratory to verify that the FRLS domain is the only Fn-binding domain within the CadF protein, as well as to determine the binding affinity of CadF to Fn.

218 199 ACKNOWLEDGEMENTS We thank Joey Mickelson for assistance in purification of recombinant CadF. This work was supported by a grants from the NIH (DK58911) and USDA National Research Initiative Competitive Grants Program (USDA/NRICGP, ) awarded to MEK.

219 200 REFERENCES Altekruse, S. F., N. J. Stern, P. I. Fields, and D. L. Swerdlow. (1999). Campylobacter jejuni - - an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Baloda, S. B., A. Faris, G. Frömam, and T. Waldström (1985). Fibronectin binding to Salmonella strains. FEMS Microbiol. Lett. 28:1-5. Brás, A. M., and J. M. Ketley Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: Dawson, J. R., and R. P. Ellen. (1990). Tip-oriented adherence of Treponema denticola to fibronectin. Infect. Immun. 58: Dawson, J. R., and R. P. Ellen (1994). Clustering of fibronectin adhesins toward Treponema denticola tips upon contact with immobilized fibronectin. Infect. Immun. 62: De Melo, M. A., and J.-C. Pechère (1990). Identification of Campylobacter jejuni surface proteins that bind to eucaryotic cells in vitro. Infect. Immun. 58: Everest, P. H., H. Goossens, J. P. Butzler, D. Lloyd, S. Knutton, J. M. Ketley, and P. H. Williams. (1992). Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J. Med. Microbiol. 37: Fauchère, J.-L., M. Kervella, A. Rosenau, K. Mohanna, and M. Véron.. (1989). Adhesion to HeLa cells of Campylobacter jejuni and C. coli outer membrane components. Res. Microbiol. 140: Fröman, G., L. M. Switalski, A. Faris, T. Wadström, and M. Höök. (1984). Binding of Escherichia coli to fibronectin. J. Biol. Chem. 259: Harvey, P., T. Battle, and S. Leach. (1999). Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Jaffe, J., S. Natanson-Yaron, M. G. Caparon, and E. Hanski. (1996). Protein F2, a novel fibronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol. Microbiol 21: Jin, S., A. Joe, J. Lynett, E. K. Hani, P. Sherman, and V. L. Chan. (2001). JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Jin S., Y. C. Song, A. Emili, P. M. Sherman, and V. L. Chan. (2003). JlpA of Campylobacter jejuni interacts with surface-exposed heat shock protein 90α and triggers signalling pathways leading to the activation of NF-kappaB and p38 MAP kinase in epithelial cells. Cell Microbiol. (3): Kelle, K., J. -M. Pagès, and J. -M. Bolla. (1998). A putative adhesin gene cloned from Campylobacter jejuni. Res. Microbiol. 149:

220 Kervella, M., J. -M. Pagès, Z. Pei, G. Grollier, M. J. Blaser, and J. -L. Fauchère. (1993). Isolation and characterization of two Campylobacter glycine-extracted proteins that bind to HeLa cell membranes. Infect. Immun. 61: Konkel, M. E., and W. Cieplak, Jr. (1992). Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M. E., S. G. Garvis, S. L. Tipton, D. E. Anderson Jr., and W. Cieplak, Jr. (1997). Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M. E., D. J. Mead, and W. Cieplak, Jr. (1993). Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Kuusela, P. (1978). Fibronectin binds to Staphylococcus aureus. Nature (Lond). 276: McSweegan, E., and R. I. Walker. (1986). Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infect. Immun. 53: Monteville, M. R., and M. E. Konkel. (2002). Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface. Infect. Immun. 70: Monteville, M. R., J. E. Yoon, and M. E. Konkel. (2003). Maximal adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer membrane protein and microfilament reorganization. Microbiol. 149: Moser, I., and W. Schröder. (1995). Binding of outer membrane preparations of Campylobacter jejuni to INT 457 cell membranes and extracellular matrix proteins. Med. Microbiol. Immunol. 184: Moser, I., W. Schröder, and J. Salnikow. (1997). Campylobacter jejuni major outer membrane protein and a 59-kDa protein are involved in binding to fibronectin and INT 407 cell membranes. FEMS Mirobiol. Lett. 157: Moser, I., W. Schröder, and E. Hellmann. (1992). In vitro binding of Campylobacter jejuni/coli outer membrane preparations to INT 407 cell membranes. Med. Microbiol. Immunol. 180: Myhre, E. B., and P. Kuusela. (1983). Binding of human fibronectin to group A, C, and G Streptococci. Infect. Immun. 40: Pei, Z., and M. J. Blaser. (1993). PEB1, the major cell-binding factor of Campylobacter jejuni, is a homolog of the binding component in Gram-negative nutrient transport systems. J. Biol. Chem. 268:

221 Pei, Z., C. Burucoa, B. Grignon, S. Baqar, X. -Z. Huang, D. J. Kopecko, A. L. Bourgeois, J. -L. Fauchere, and M. J. Blaser. (1998). Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Pickett, C. L., and C. A. Whitehouse. (1999). The cytolethal distending toxin family. Trends in Microbiol. 7: Russell, R. G., M. O'Donnoghue, D. C. Blake, Jr., J. Zulty, and L. J. DeTolla. (1993). Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Rydén, C., K. Rubin, P. Speziale, M. Höök, M. Lindberg and T. Wadström. (1983). Fibronectin receptors from Staphylococcus aureus. J. Biol. Chem. 258: Schorey J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown. (1996). Characterization of the fibronectin-attachment protein of Mycobacterium avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Microbiol. 21(2):321-9 Schröder, W., and I. Moser. (1997). Primary structure analysis and adhesion studies on the major outer membrane protein of Campylobacter jejuni. FEMS Microbiol. Lett. 150: Tauxe, R. V. (1992). Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Nachamkin, I., Blaser, M.J. and Tompkins, L. (eds.), Campylobacter jejuni:current status and future trends. American Society for Microbiology, Washington, D.C., pp Thomas, D. D., J. B. Baseman, and J. F. Alderete. (1985). Fibronectin mediates Treponema pallidum cytadherence through recognition of fibronectin cell-binding domain. J. Exp. Med. 161: van Putten, J. P., T. D. Duensing, and R. L. Cole. (1998). Entry of OpaA+ gonococci into HEp-2 cells requires concerted action of glycosaminoglycans, fibronectin and integrin receptors. Mol. Microbiol. 29: Visai, L., S. Bozzini, T. E. Petersen, L. Speciale, and P. Speziale. (1991). Binding sites in fibronectin for an enterotoxigenic strain of E. coli B342289c. FEBS Lett. 290: Wassenaar, T. M. (1997). Toxin production by Campylobacter spp. Clin. Microbiol. Rev. 10: Zhao, W., J. S. Schorey, R. Groger, P. M. Allen, E. J. Brown, and T. L. Ratliff. (1999). Characterization of the fibronectin binding motif for a unique mycobacterial fibronectin attachment protein, FAP. J. Biol. Chem. 274(8): Ziprin, R. L., C. R. Young, M. E. Hume, and M. E. Konkel. (1999). The absence of cecal colonization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectinbinding protein. Avian Dis. 43:

222 203 Table 1. Peptides (30-mers) used in this study. Amino Acids No. Residues aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa RDNNVKFEIT PTLNYNYFEG NLDMDNRYAP NLDMDNRYAP GIRLGYHFDD FWLDQLEFGL FWLDQLEFGL EHYSDVKYTN TNKTTDITRT TNKTTDITRT YLSAIKGIDV GEKFYFYGLA GEKFYFYGLA GGGYEDFSNA AYDNKSGGFG AYDNKSGGFG HYGAGVKFRL SDSLALRLET HYGAGVKFRL SDSLALRLET RDQINFNHAN RDQINFNHAN HNWVSTLGIS FGFGGKKEKA FGFGGKKEKA VEEVADTRPA PQAKCPVPSR PQAKCPVPSR EGALLDENGC EKTISLEGHF EKTISLEGHF GFDKTTINPT FQEKIKEIAK FQEKIKEIAK VLDENERYDT ILEGHTDNIG ILEGHTDNIG SRAYNQKLSE RRAKSVANEL RRAKSVANEL EKYGVEKSRI KTVGYGQDNP KTVGYGQDNP RSRNETKEGR ADNRRVDAKF GYGQDNPRSR NETKEGRADN RRVDAKFILR

223 204 Figure 1. Fn binding to CadF peptides (30-mers). CadF peptides corresponding to the amino acids indicated were coated onto ELISA plates and Fn detected as described in Materials and Methods. Each data point contains the mean + SD of a representative experiment with BSA included as a negative control. Experiments were performed a minimum of four times.

224 205 Figure A492 CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF

225 206 Figure 2. Fn binding to CadF peptides (16-mers). A. CadF peptides corresponding to the amino acids were coated onto ELISA plates and Fn detected as described in Materials and Methods. Each data point contains the mean + SD of a representative experiment with BSA included as a negative control. Experiments were performed a minimum of four times. B. Sequence of peptides (16-mers) with overlapping regions used to identify the essential amino acids required for Fn adherence.

226 207 Figure A ng Fn 10 ng Fn 5 ng Fn CadF CadF CadF CadF CadF CadF CadF CadF CadF CadF Scamble A.

227 208 Figure 2. B. SDSLALRLETRDQINF = FRLSDSLALRLETRDQ GVKFRLSDSLALRLET = = YGAGVKFRLSDSLALR = FGHYGAGVKFRLSDSL = SGGFGHYGAGVKFRLS = DNKSGGFGHYGAGVKF = AAYDNKSGGFGHYGAG = LRRDLLSLSTREADFQ = Scrambled

228 209 Figure 3. Competitive inhibition assay using peptides to block Fn binding to rcadf. Competition assays were performed as described in Materials and Methods. Fn was incubated in the presence or absence of indicated peptides for 1 hr at 25 C prior to the addition of Fn to ELISA plates coated with rcadf. Total Fn binding to rcadf or BSA served as a positive and negative control respectively.

229 210 Figure 3. 50µg CadF µg CadF non-fn Binding Peptide 10µg CadF µg CadF µg CadF Fn Binding Peptide 10µg CadF µg CadF Scrambled 25µg CadF Scrambled 10µg CadF Scrambled Fn Binding Peptide Scrambled BSA Control rcadf Alone A 492

230 211 Chapter 7 The bile salt deoxycholate alters the kinetics of Campylobacter jejuni invasion Michael E. Konkel, Debabrata Biswas, Marshall R. Monteville, and John D. Klena School of Molecular Biosciences, Washington State University, Pullman, WA Manuscript in preparation

231 212 ABSTRACT Campylobacter jejuni is a common cause of human diarrheal illness. Previous work has demonstrated that C. jejuni synthesizes proteins termed the Campylobacter invasion antigens (Cia) whose secretion is necessary for the organism s maximal invasion of INT 407 human epithelial cells. The purpose of this study was to further characterize the Cia proteins and their role in promoting bacteria-host cell interactions using the T84 human colonic cell line. One-dimensional gel electrophoresis revealed a two-fold difference in the amount of protein present in the supernatant fluids of bacteria cultured on Mueller-Hinton agar plates supplemented with deoxycholate versus a non-supplemented Mueller-Hinton agar plate. While the profile of secreted proteins did not include the flagellar FlaA and FlaB filament proteins, additional experiments revealed that a component of the flagellar export apparatus, FlhB, is required for Cia protein secretion. Culturing C. jejuni on plates supplemented with deoxycholate overcame the inhibitory effect of chloramphenicol on C. jejuni invasion of T84 non-polarized cells as judged by the gentamicin-protection assay. In addition, a significant difference was noted in the kinetics of bacterial internalization for C. jejuni that were cultured on deoxycholate-supplemented Mueller- Hinton agar plates versus those bacteria that were cultured on non-supplemented Mueller-Hinton agar plates. Collectively, the data indicate that the Cia proteins contribute to virulence of C. jejuni by maximizing the organism s invasion potential.

232 213 INTRODUCTION Campylobacter jejuni is recognized as one of the leading bacterial causes of gastrointestinal disease in humans (2, 4). Despite this fact, the mechanism(s) by which C. jejuni causes disease is ill-defined. A few of the attributes proposed to contribute to the pathogenesis of C. jejuni-mediated enteritis include motility (8, 37, 52), chemotaxis (50, 53), host cell-translocation (9, 15, 20, 28), host cell-adherence (22, 25, 43), host cell-invasion (15, 24, 27, 48), and toxin production (44, 51). C. jejuni also have the ability to modulate the expression of their surface molecules (7, 23, 38). The proteins promoting the entry of C. jejuni into eukaryotic cells are different from those that facilitate binding (24). In vitro experiments have revealed that C. jejuni synthesize unique proteins when cultured in the presence of epithelial cells, a set of which is secreted (24, 26, 27, 46). These secreted proteins, termed Campylobacter invasion antigens (Cia), are required for the maximal invasion of host epithelial cells by C. jejuni (26, 46). Additional work has revealed that Cia protein synthesis and secretion are separable, and that secretion is the rate-limiting step. Regulation of virulence gene expression is a common theme shared among pathogenic organisms, enabling them to conserve resources by expressing a given set of genes in a specific environment. Some of the environmental conditions that serve to signal the differential protein synthetic response exhibited by pathogenic bacteria include oxygen limitation, temperature, ph and osmolarity (30, 33, 35, 45). More specifically, secretion of Salmonella Sip proteins (Salmonella invasion proteins) is regulated by oxygen limitation (30) and their synthesis/secretion can be stimulated by a ph change (12), whereas virulence gene regulation in Shigella is primarily regulated by temperature (33). With respect to C. jejuni, previous work suggests that the genes encoding the Cia proteins are coordinately regulated and responsive to environmental stimuli. Alteration of temperature, ph, calcium concentration, and osmolarity does not stimulate either Cia protein synthesis or secretion.

233 214 However, the Cia proteins are synthesized and secreted when C. jejuni is cultured with epithelial cells in serum-free medium and in medium containing sera from various sources (46, 47). In addition, previous work has suggested that the synthesis of the Cia proteins is induced by including bile salts such as deoxycholate, cholate, or chenodeoxycholate, in the culture media (46). A number of enteric bacteria have evolved to use bile as a signal, thereby coordinating the expression of virulence determinants within a particular cellular location (19). This study was initiated to further assess the role of the bile salt deoxycholate in promoting Cia protein synthesis. Moreover, experiments were performed to address the question of whether culturing C. jejuni on deoxycholate-supplemented plates would alter the secretion profile and invasive phenotype of C. jejuni. The invasive phenotype of C. jejuni was examined using T84 cells, which are phenotypically similar to colonic crypt cells (39). Histological examination of C. jejuni infected humans and animals has revealed pathology primarily in the colon (5, 8, 48).

234 215 MATERIALS AND METHODS Bacterial isolates and growth conditions. C. jejuni clinical isolate (TetR) (6) was cultured on Mueller-Hinton (MH) agar plates containing 5% bovine citrated blood (MH-blood) under microaerophilic conditions at 37 C. Where appropriate, MH-blood plates were supplemented with 12.5 µg/ml of tetracycline, 200 µg/ml kanamycin, or with sodium deoxycholate (0.1% wt/vol). Isolates were subcultured every 24 to 48 h. Escherichia coli XL-1 Blue MRF (Stratagene, La Jolla, CA) and E. coli InvαF (Invitrogen, Carlsbad, CA) were cultured in Luria-Bertani (LB) broth (10 g Bacto-tryptone, 5 g yeast extract, and 5 g sodium chloride/liter) and on LB agar plates (LB broth with 15 g Bacto-agar per liter) in a 37 C incubator. LB plates were supplemented with 12.5 µg/ml tetracycline and 50 µg/ml kanamycin as appropriate. Preparation of secreted proteins. C. jejuni were harvested from either MH plates or MH-deoxycholate supplemented plates in phosphate-buffered saline (PBS), pelleted by centrifugation at 6,000 x g, washed twice in Minimal Essential Medium (MEM), and suspended in MEM to an optical density (OD 540 ) of 0.3 [approximately 5 x 108 colony forming units (CFU)]. Metabolic labeling experiments were performed in three ml of MEM lacking methionine (labeling medium; ICN Biomedicals, Inc., Aurora, OH) and [35S]-methionine (PerkinElmer Life Sciences, Inc., Boston, MA) as described elsewhere (24). Fetal bovine serum (FBS; Hyclone Laboratories Inc., Logan, UT) was used as a stimulatory signal in each of the secretion assays. Prior to each assay, the albumin was removed from the FBS using a SwellGel Blue Albumin Removal Kit (Pierce, Rockford, IL). Following a three h labeling period, bacterial cells were pelleted by centrifugation at 6,000 x g and supernatant fluids collected. One and two-dimensional protein gel electrophoresis. For one-dimensional protein gel

235 216 electrophoresis, supernatant fluids were concentrated 4-fold by the addition of 5 volumes of ice-cold 1 mm HCl-acetone. The pellets were air dried, resuspended in water, and mixed with an equal volume of double strength electrophoresis sample buffer (24). Prior to electrophoresis, the samples were heated to 95 C for 5 min and allowed to cool to ambient temperature. Proteins were resolved in sodium dodecyl sulfate-12.5% polyacrylamide gels using the discontinuous buffer system described by Laemmli (29). For two-dimensional protein gel electrophoresis, supernatant fluids were concentrated 10-fold by the addition of 5 volumes of ice-cold 1 mm HCl-acetone. The first dimension, isoelectric focusing, was performed as described elsewhere (24). The proteins were then resolved in sodium dodecyl sulfate-12.5% polyacrylamide gels. Gels were treated with Entensify (Life Sciences Products, Boston, MA) according to the supplier s instructions and dried. Labeled bacterial proteins were visualized by autoradiography using Kodak BioMax MR film at -70 C. Immunoblot analysis. Following SDS-PAGE, proteins were electrophoretically transferred to polyvinylidene fluoride membranes (Immobilon P; Millipore Corp., Bedford, MA). The membranes were washed three times in PBS and incubated for 18 h at 4 C with a 1:100 dilution of the rabbit anti-c. jejuni flagellin antibody in PBS ph 7.4/0.01% Tween-20 containing 9% non-fat dried milk. Bound antibodies were detected using peroxidase-conjugated rabbit anti-goat IgG (Sigma; St. Louis, MO) and 4-chloro-1-napthol (Sigma) as the chromogenic substrate. Isolation of a C. jejuni flhb flagellar export apparatus mutant. The flhb gene in C. jejuni was disrupted by recombination via a single crossover event between the chromosomal gene and an internal fragment of the homologous gene in a suicide vector harboring a kanamycin resistance gene (25). A 518 bp internal gene fragment of the flhb gene from C. jejuni was amplified with Taq polymerase using gene specific primers (Forward primer, 5 -AAA AAA CAG AAG AAC CCA CG; Reverse primer, 5 -GTG CAA CCA TAG TAT AAA GCT C).

236 217 The amplified product was ligated into the pcr2.1 cloning vector (TA Cloning System, Invitrogen) and transformed into E. coli XL-1 Blue MRF by electroporation with a Bio-Rad Gene Pulser with pulse controller (Bio-Rad Laboratories, Richmond, Calif.). After identifying an E. coli transformant harboring the flhb-518 bp internal gene fragment, a cleared lysate was prepared and plasmid DNA was purified on an ethidium bromide-cscl density gradient (49). The insert was excised from pcr2.1 by restriction endonuclease digestion with EcoRI, gel-purified, and ligated into pbluescript II SK + (pbii SK+) containing a Campylobacter kanamycin resistance gene (pbii SK+ -kan vector). The pbii SK+ -kan vector was digested with EcoRI and treated with calf intestinal alkaline phosphatase prior to ligation as per the manufacturer s instructions. After amplification and purification of the recombinant pbii SK+ -kan vector, it was introduced into C. jejuni by electroporation. C. jejuni flhb mutants were identified by acquisition of kanamycin resistance and flhb gene disruption confirmed by PCR. Phenotypic analysis of the C. jejuni flhb mutant. C. jejuni flhb mutants were subject to motility assays and visualized by transmission electron microscopy (TEM). Motility assays were performed using MH medium supplemented with 0.4% Select Agar (Gibco BRL, Gaithersburg, MD). A 10 µl suspension of each bacterial isolate was spotted on the surface of the medium. The plates were incubated under microaerophilic conditions for 48 h, and then scored for motility based on whether the isolate migrated from the center spot. C. jejuni isolates were also analyzed with a 1200 EX transmission electron microscope (JEOL). Briefly, bacterial suspensions were harvested from MH agar plates in PBS, and added dropwise to formvar-coated copper grids. Bacteria were stained with 1% phosphotungstic acid. Tissue culture. Stock cultures of T84 cells (human colonic cell line, ATCC CCL 248) were grown in MEM supplemented with 10% (vol/vol) FBS and maintained at 37 C in a humidified, 5%

237 218 CO 2 incubator. Binding and internalization assays. For experimental assays, each well of a 24-well tissue culture tray was seeded with 1.4 x 105 cells/well and incubated for 18 h at 37 C in a humidified, 5% CO 2 incubator. The cells were rinsed with MEM-1% FBS and inoculated with approximately 5 x 107 CFU of a bacterial suspension. Tissue culture trays were centrifuged at 600 x g for 5 min, and incubated at 37 C in a humidified, 5% CO 2 incubator. For binding, the infected monolayers were incubated for 30 min, rinsed three times with PBS, and the epithelial cells lysed with a solution of 0.1% (vol/vol) Triton X-100 (Calbiochem, La Jollo, CA). The suspensions were serially diluted and the number of viable, adherent bacteria determined by counting the resultant colonies on MH-blood plates. To measure bacterial internalization, the infected monolayers were incubated for three h, rinsed three times with MEM-1% FBS, and incubated for an additional three h in MEM-1% FBS containing a bactericidal concentration of gentamicin. The number of internalized bacteria was determined as outlined above. Unless otherwise stated, the reported values represent the mean counts + standard deviations derived from triplicate wells. All assays were performed a minimum of three times to ensure reproducibility. In certain instances, chloramphenicol (128 µg/ml) was added to inhibit bacterial protein synthesis. Other analytical procedures. Protein concentrations were determined by the bicinchoninic acid (BCA) method, with BSA as the standard, as outlined by the supplier (Pierce, Rockford, IL). Significance between samples was determined using the Student t test following logarithmic [(log x (base 10)] transformation of the data. Two-tailed P values were determined for each sample, and a P value < 0.01 considered significant.

238 219 RESULTS Deoxycholate stimulates the synthesis of the Cia proteins. Previous work in our laboratory has indicated that Cia protein synthesis and secretion are separable, and that secretion is the rate-limiting step of these processes (46). While bile salts do not stimulate Cia protein secretion, ciab transcription is induced in response to the bile salt sodium deoxycholate as judged by RT- PCR analysis (46). In an attempt to increase the amount of Cia proteins secreted, C. jejuni strain was cultured on 0.1% (wt/vol) deoxycholate-supplemented MH agar plates prior to performing the secretion assay (Fig. 1). The concentration of bile salts in the human intestinal tract has been reported to range between 0.2 to 2% (19). C. jejuni were also cultured on non-supplemented MH agar plates prior to the labeling assay as a control. As predicted, culturing C. jejuni on deoxycholate-supplemented MH agar plates prior to the labeling assay resulted in a greater amount of the Cia proteins in the supernatant fluids relative to the Cia proteins in the supernatant fluids of C. jejuni pre-cultured on non-supplemented MH agar plates (Fig. 1; compare lanes 3 and 4). While the difference in the amount of the Cia proteins in the supernatant fluids was readily apparent between samples, the secreted protein profiles of the two samples were indistinguishable. Consistent with previous work, the Cia proteins were only detected in the supernatant fluids of C. jejuni cultured in MEM containing FBS, and not in MEM alone, regardless of whether the bacteria were harvested from a deoxycholate-supplemented MH plate or a non-supplemented MH agar plate. These data support previous work, and taken together, suggest that deoxycholate induces cia gene transcription. The supernatant fluids from the samples in which C. jejuni were pre-cultured on the deoxycholatesupplemented MH agar plate and for C. jejuni that were pre-cultured on the non-supplemented MH agar plate were subject to additional analysis to quantitate the difference in proteins secreted into the medium. First, the total protein was determined in the supernatant fluids

239 220 of each sample using the BCA assay. For comparative purposes, the total protein content within the labeling medium (MEM containing 1% FBS) was also determined. The CFU present in each sample was determined to ensure that a variation in the amount of protein present in the supernatant fluids was reflective of the deoxycholate-supplement and not due to differences between samples with regards to the number of C. jejuni. Overall, a two-fold difference was observed in the amount of protein present in the supernatant fluids between the bacteria cultured on a deoxycholatesupplemented MH agar plate versus a non-supplemented MH agar plate as judged by the total protein concentration (Table 1). This finding was further supported by SDS-PAGE coupled with autoradiography, in which a two-fold difference was noted in the amount of secreted protein from bacteria cultured on a deoxycholate-supplemented MH agar plate versus a non-supplemented MH agar plate (Fig. 2). The profile of secreted proteins does not include flagellar proteins. The C. jejuni flagellar filament is comprised of two proteins, FlaA and FlaB, although it appears that FlaA is the preferred subunit (3). Both C. jejuni flagellin proteins are synthesized concomitantly, but the flaa gene is expressed at much greater levels than the flab gene (21, 41). In C. jejuni, transcription of the flaa gene is regulated by σ28 and the flab gene is regulated by σ54 (3, 21). Because culturing C. jejuni in the presence of deoxycholate is known to cause an up-regulation in flaa expression (1), experiments were performed to determine if FlaA is one component of the secreted protein profile. Simultaneously, the profile of secreted proteins was analyzed by two-dimensional electrophoresis coupled with autoradiography (Fig. 3, panels A-D). Analysis of the autoradiograph showing the secreted proteins from C. jejuni revealed a minimum of 15 proteins, ranging in size from approximately 14 to 80 kda (Fig. 3B). Nevertheless, a reactive band was not observed in the secreted proteins upon immunoblot analysis with an antiserum that contains antibodies reactive with both the FlaA and FlaB proteins (Fig. 3D). These data indicate that the FlaA and FlaB proteins, in C. jejuni , are not constituents of the Cia proteins.

240 221 Cia protein secretion is dependent upon a functional flagellar export apparatus. While the specific functions of the secreted proteins are not known, insertional mutagenesis of the gene encoding a 73 kda secreted protein (ciab/ciab) results in a reduction in the number of C. jejuni internalized when compared to a C. jejuni wild-type isolate (26). CiaB lacks the presence of an identifiable signal sequence (26). In addition, a stimulus is required to induce Cia protein secretion (47). While these characteristics are reminiscent of the type III secretion system, translation of the complete genome of the C. jejuni NCTC strain has failed to reveal proteins with homology to virulence-associated type III secretory systems. Yersinia secretes a set of proteins termed Fops, for flagellar outer proteins, via the flagellar apparatus (54). In addition, secretion of virulence-associated proteins from Bacillus thuringiensis has been found to be dependent on flha, an essential component of the flagellar export apparatus (18). Also noteworthy is that some of the proteins that comprise the classical type III secretory apparatus share amino acid sequence similarity with flagellar structural proteins (31, 34). Moreover, the flagellar protein export apparatus, which is required for secretion of flagellar rod, hook and filament proteins, is structurally and functionally analogous to that involved in type III virulence factor secretion (31). To address whether the flagellar export apparatus is required for Cia protein secretion in C. jejuni , a mutation was generated in the flhb flagellar biosynthesis gene. The flhb gene encodes a component of the flagellar export apparatus (36), and in C. jejuni, is transcribed from its own promoter (32). Matz et al. (32) have also shown that a mutation in the flhb gene in C. jejuni did not significantly affect transcription of the motb flagellar motor gene, which is located downstream and in opposite orientation of the flhb gene. We found that a C. jejuni flhb mutant lacked flagellum as judged by TEM analysis and was non-motile as judged by motility assays (not shown). Consistent with the work of Matz et al. (32), we also found that the FlaA protein was absent in a whole cell lysate of a C. jejuni flhb mutant as judged by immunoblot analysis with a rabbit anti-c. jejuni flagellin antibody (not shown). Metabolic labeling experiments also revealed that a C. jejuni flhb mutant was incapable of secreting the Cia proteins (Fig. 4). This finding indicates that a functional

241 222 FlhB product is required for Cia protein secretion, and is consistent with the notion that the Cia proteins are secreted from a flagellar type III secretory apparatus. Deoxycholate enhances C. jejuni internalization of T84 cells. Previous studies have demonstrated that chloramphenicol inhibits the uptake of C. jejuni into INT 407 cells (24, 46). To determine if deoxycholate retards the inhibitory effect of chloramphenicol on C. jejuni uptake by T84 cells, invasion assays were performed with C. jejuni that had been cultured on MH alone or MH-deoxycholate supplemented plates. Consistent with previous work (46), no significant differences were noted in the binding or internalization of C. jejuni cultured on MH agar plates when compared to C. jejuni cultured on deoxycholate-supplemented MH agar plates. However, in the presence of chloramphenicol, a statistically significant (P < 0.01) difference was noted in the number of internalized bacteria for the C. jejuni cultured on deoxycholate-supplemented MH agar plates when compared to non-supplemented MH agar plates (Table 2). These findings suggest that deoxycholate induces the synthesis of the Cia proteins, which in turn, promote C. jejuni uptake.

242 223 Pre-culturing C. jejuni with deoxycholate alters the kinetics of bacterial uptake. As deoxycholate induces cia gene expression, we hypothesized that pre-culturing C. jejuni on deoxycholate-supplemented agar plates might alter the kinetics of C. jejuni uptake by T84 cells. Thus, the invasive potential of C. jejuni was determined using bacteria that had been cultured on deoxycholate-supplemented MH agar plates as well as on non-supplemented MH agar plates. As a control, the invasive potential of the C. jejuni flhb isogenic mutant was examined after culturing the bacteria on deoxycholate-supplemented and non-supplemented MH agar plates (Fig. 5). Consistent with published work (27), a steady increase was noted in the number of C. jejuni internalized over time when the bacteria were harvested from non-supplemented MH agar plates. Over the course of the two hour assay, the number of C. jejuni internalized increased approximately one log. The kinetics of C. jejuni internalization was markedly different for those bacteria harvested from deoxycholate-supplemented MH agar plates when compared to those harvested from non-supplemented MH agar plates. Fifteen minutes post-inoculation, more than one log difference was noted in the invasiveness of the bacteria harvested from the deoxycholate-supplemented MH agar plates versus the bacteria harvested from the non-supplemented MH agar plates. In addition, only a marginal increase was noted over time in the number internalized bacteria for the bacteria that were cultured on the deoxycholate-supplemented MH agar plates. A difference was not apparent in the invasive potential of the C. jejuni flhb isogenic mutant, regardless of whether the isolate was cultured on deoxycholate-supplemented versus non-supplemented MH agar plates. Collectively, these data indicate that deoxycholate stimulates Cia protein synthesis, which in turn enable C. jejuni to immediately invade eukaryotic cells upon contact.

243 224 DISCUSSION It is well established that microorganisms that invade non-professional phagocytic cells (e.g., epithelial cells) must synthesize entry-promoting proteins (16, 17). While our understanding of the molecular mechanism(s) of epithelial cell invasion by C. jejuni is in its infancy, early work established that the internalization of C. jejuni can be significantly reduced in the presence of chloramphenicol, a specific inhibitor of bacterial protein synthesis (24). This finding, coupled with the fact that metabolically inactive (heat-killed and sodium azide-killed) C. jejuni are not internalized, suggested that C. jejuni synthesize entry-promoting proteins (24). One and two-dimensional protein electrophoretic analyses of metabolically labeled C. jejuni cultured in the presence and absence of epithelial cells revealed that a number of proteins were synthesized exclusively, or preferentially, in the presence of epithelial cells while others were selectively repressed (24, 27). Moreover, Panigrahi et al. (42) demonstrated that C. jejuni synthesized a number of proteins during growth in rabbit ileal loops that were not synthesized under standard laboratory conditions. Collectively, these findings suggested a coordinated response, whereby C. jejuni express certain genes after encountering the epithelial cell microenvironment. It was later found that a subset of the proteins synthesized by C. jejuni in the presence of cultured epithelial cells were secreted by the organism (24). More recently, the bile salt deoxycholate was found to stimulate the expression of the genes encoding the Cia secreted proteins (46). In fact, culturing C. jejuni on plates supplemented with deoxycholate prior to performing an internalization assay with INT 407 cells enables the organism to overcome the inhibitory effect of chloramphenicol on host cell-invasion (46). This study was undertaken to: 1) quantitate the amount of the Cia proteins secreted by C. jejuni in a given period of time; 2) define the relationship between deoxycholate and induction of Cia protein secretion; and 3) further clarify the relationship between deoxycholate and C. jejuni host cell invasion. We chose to examine the interactions of C. jejuni with T84 cells as these cells are reflective of those that C. jejuni encounters in vivo.

244 225 Culturing C. jejuni on deoxycholate-supplemented MH agar plates prior to a secretion assay resulted in a greater amount of the Cia proteins in the supernatant fluids relative to the Cia proteins in the supernatant fluids of C. jejuni precultured on non-supplemented MH agar plates. Indeed, a two-fold difference was noted in the amount of protein in the supernatant fluids of bacteria that were harvested from a deoxycholate-supplemented MH agar plate versus a non-supplemented MH agar plate over the course of a three hour secretion assay. The effect of deoxycholate was deemed specific with regard to inducing cia gene transcription and Cia protein synthesis as no difference was noted in the secreted protein profile of bacteria harvested from the deoxycholate-supplemented versus non-supplemented MH agar plates. In addition, the Cia proteins were only detected in the supernatant fluids when FBS was added to the medium to induce secretion. Two-dimensional gel electrophoresis revealed that the C. jejuni clinical isolate is capable of secreting a minimum of 15 proteins. Each secreted protein may be unique, or may represent a protein that has undergone either degradation or modification resulting in a shift in isoelectric point or relative mass. Differences were also noted in the concentrations of each secreted protein relative to each another. While the identities of the secreted proteins are not known with the exception of CiaB, immunoblot analysis with an anti-flagellin serum revealed that the FlaA and FlaB filament proteins are not Cia protein constituents. However, Cia protein secretion appeared dependent upon a functional flagellar export apparatus as evidenced by the inability of the C. jejuni flhb flagellar export mutant to secrete the Cia proteins. The flhb gene in C. jejuni isolates is monocistronic, so a mutation in this gene is not polar (32). One possible explanation for the lack of Cia protein secretion from the C. jejuni flhb mutant may be that the cia and flagellar-structural genes are coregulated. In this context, Matz et al. (32) reported a significant decrease in the level of flaa gene expression in a C. jejuni strain containing a defective flagellar export apparatus. Similarly, we could not detect the FlaA flagellin monomer in the whole cell lysates of the C. jejuni flhb mutant as judged by immunoblot analysis with the flagellin antiserum. This finding is

245 226 consistent with the notion that C. jejuni possess a negative regulator (e.g., FlgM) that inhibits flagellin transcription in response to a defective hook-basal body complex. In bacteria such as Salmonella, Yersinia, and Helicobacter pylori, the negative regulator FlgM inhibits flagellin transcription in response to a defective hook-basal body complex (10, 11, 13). A protein corresponding to a putative FlgM homolog has been identified in the genome of C. jejuni NCTC (Cj1464) that shares similarity with the recently identified FlgM protein from H. pylori (11). Thus, if the cia genes are transcribed from σ28 promoters, FlgM could indirectly effect cia transcription in response to a defective hook-basal body complex. Noteworthy is that in Salmonella enterica serovar Typhi, σ28 has been found to be involved in regulating gene expression of type III proteins (14). We have found that the Cia proteins are secreted in a C. jejuni flia (σ28) mutant, which eliminates the possibility that the cia genes are subject to transcriptional regulation via a mechanism involving a putative FlgM anti-sigma factor (manuscript in preparation). While the functions of the Cia proteins are not known, the results presented herein demonstrate a link between Cia secretion and C. jejuni-host cell invasion. More specifically, the relationship between Cia protein secretion and C. jejuni invasion became evident with T84 host cells when using C. jejuni that had been cultured on MH and MH-deoxycholate supplemented plates and when chloramphenicol was incorporated into the assay. Chloramphenicol is a specific inhibitor of bacterial protein synthesis, and at the concentration used in this study, immediately halts C. jejuni protein synthesis. Consistent with previous work (46), no significant differences were observed in the binding or internalization of C. jejuni cultured on MH agar plates when compared to C. jejuni cultured on deoxycholate-supplemented MH agar plates (Table 2). However, in the presence of chloramphenicol, a statistically significant (P < 0.01) difference was noted in the number of internalized bacteria for the C. jejuni cultured on deoxycholate-supplemented MH agar plates when compared to non-supplemented MH agar plates. These results suggest that deoxycholate induces

246 227 the synthesis of the Cia proteins, which in turn, promote C. jejuni uptake. Previous work with INT 407 cells has shown that the number of C. jejuni internalized from 30 minutes to three hours increases approximately two orders of magnitude, with the greatest increase in internalization occurring between 30 and 60 minutes (27). We found that by culturing the organism on a deoxycholate-supplemented agar plate, it was possible to alter the kinetics of C. jejuni uptake into T84 cells. When C. jejuni are cultured on a deoxycholate-supplemented MH agar plate, a greater proportion of the bacteria have the ability to invade an epithelial cell versus those bacteria cultured on a non-supplemented MH agar plate. A difference was not apparent in the invasive potential of the C. jejuni flhb isogenic mutant, regardless of whether the isolate was cultured deoxycholate-supplemented versus non-supplemented MH agar plates. To our knowledge, this is the first time the kinetics of C. jejuni entry has been found to be altered in response to any growth condition. These findings indicate that C. jejuni engage in a coordinated response whereby genes are expressed, a set of which is responsible for epithelial cell invasion, in response to changes in environmental conditions. In summary, two-dimensional gel electrophoresis revealed at least 15 proteins can be secreted by C. jejuni, as opposed to eight that were originally identified by one-dimensional gel electrophoresis. The secretion of the Cia proteins appears to be tightly regulated as the proteins were detected in the supernatant fluid only when the stimulus (e.g., serum) was present. Experiments are currently underway to purify the secreted proteins. In addition, we are currently testing whether C. jejuni environmental isolates, which have been reported to be less invasive than clinical isolates (40), are also capable of synthesizing and secreting the Cia proteins.

247 228 ACKNOWLEDGEMENTS We thank Gary A. Flom and Nicole Lindstrom for assistance in the preparation and examination of samples for TEM-examination. We also thank Gary A. Flom for assistance in the preparation of samples for two-dimensional gel electrophoresis. Finally, we are grateful to Brian H. Raphael for helpful discussions and for reviewing this manuscript. This work was supported by NIH grant DK58911 awarded to MEK.

248 229 REFERENCES 1. Allen, K. J., and M. W. Griffiths Effect of environmental and chemotactic stimuli on the activity of the Campylobacter jejuni flaa σ28 promoter. FEMS Microbiol. Lett. 205: Allos, B. M Campylobacter jejuni infections: Update on emerging issues and trends. Clin. Infect. Dis. 32: Alm, R. A., P. Guerry, and T. J. Trust The Campylobacter σ54 flab flagellin promoter is subject to environmental regulation. J. Bacteriol. 175: Altekruse, S. F., N. J. Stern, P. I. Fields, and D. L. Swerdlow Campylobacter jejuni -- an emerging foodborne pathogen. Emerg. Infect. Dis. 5: Babakhani, F. K., G. A. Bradley, and L. A. Joens Newborn piglet model for campylobacteriosis. Infect. Immun. 61: Bacon, D. J., R. A. Alm, D. H. Burr, L. Hu, D. J. Kopecko, C. P. Ewing, T. J. Trust, and P. Guerry Involvement of a plasmid in virulence of Campylobacter jejuni Infect. Immun. 68: Bacon, D. J., C. M. Szymanski, D. H. Burr, R. P. Silver, R. A. Alm, and P. Guerry A phase-variable capsule is involved in virulence of Campylobacter jejuni Mol. Microbiol. 40: Black, R. E., M. M. Levine, M. L. Clements, T. P. Hughes, and M. J. Blaser Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157: Brás, A. M., and J. M. Ketley Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers. FEMS Microbiol. Lett. 179: Chadsey, M. S., and K. T. Hughes A multipartite interaction between Salmonella transcription factor σ28 and its anti-sigma factor FlgM: implications for σ28 holoenzyme destabilization through stepwise binding. J. Mol. Biol. 306: Colland, F., J. C. Rain, P. Gounon, A. Labigne, P. Legrain, and H. De Reuse Identification of the Helicobacter pylori anti-σ28 factor. Mol. Microbiol. 41: Daefler, S Type III secretion by Salmonella typhimurium does not require contact with a eukaryotic host. Mol. Microbiol. 31: Daughdrill, G. W., M. S. Chadsey, J. E. Karlinsey, K. T. Hughes, and F. W. Dahlquist The C-terminal half of the anti-sigma factor, FlgM, becomes structured when bound to its target, σ28. Nat. Struct. Biol. 4:

249 14. Eichelberg, K., and J. E. Galán The flagellar sigma factor FliA (σ28) regulates the expression of Salmonella genes associated with the centisome 63 type III secretion system. Infect. Immun. 68: Everest, P. H., H. Goossens, J. P. Butzler, D. Lloyd, S. Knutton, J. M. Ketley, and P. H. Williams Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J. Med. Microbiol. 37: Falkow, S., R. R. Isberg, and D. A. Portnoy The interaction of bacteria with mammalian cells. Annu. Rev. Cell Biol. 8: Finlay, B. B., and S. Falkow Common themes in microbial pathogenecity revisited. Microbiol. Mol. Biol. Rev. 61: Ghelardi, E., F. Celandroni, S. Salvetti, D. J. Beecher, M. Gominet, D. Lereclus, A. C. L. Wong, and S. Senesi Requirement of flha for swarming differentiation, flagellin export, and secretion of virulence-associated proteins in Bacillus thuringiensis. J. Bacteriol. 184: Gunn, J. S Mechanisms of bacterial resistance and response to bile. Microb. Infect. 2: Harvey, P., T. Battle, and S. Leach Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J. Med. Microbiol. 48: Hendrixson, D. R., B. J. Akerley, and V. J. Dirita Transposon mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol. Microbiol. 40: Jin, S., A. Joe, J. Lynett, E. K. Hani, P. Sherman, and V. L. Chan JlpA, a novel surface-exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39: Karlyshev, A. V., D. Linton, N. A. Gregson, A. J. Lastovica, and B. W. Wren Genetic and biochemical evidence of a Campylobacter jejuni capsular polysaccharide that accounts for Penner serotype specificity. Mol. Microbiol. 35: Konkel, M. E., and W. Cieplak, Jr Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M. E., S. G. Garvis, S. L. Tipton, D. E. Anderson, Jr., and W. Cieplak, Jr Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M. E., B. J. Kim, V. Rivera-Amill, and S. G. Garvis Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32:

250 27. Konkel, M. E., D. J. Mead, and W. Cieplak, Jr Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J. Infect. Dis. 168: Konkel, M. E., D. J. Mead, S. F. Hayes, and W. Cieplak, Jr Translocation of Campylobacter jejuni across human polarized epithelial cell monolayer cultures. J. Infect. Dis. 166: Laemmli, U. K Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227: Lee, C. A., and S. Falkow The ability of Salmonella to enter mammalian cells is affected by bacterial growth state. Proc. Natl. Acad. Sci., USA. 87: Macnab, R. M The bacterial flagellum: reversible rotary propeller and type III export apparatus. J. Bacteriol. 181: Matz, C., A. H. M. van Vliet, J. Ketley, and C. W. Penn Mutational and transcriptional analysis of the Campylobacter jejuni flagellar biosynthesis gene flhb. Microbiol. 148: Maurelli, A. T., A. E. Hromockyj, and M. L. Bernardini Environmental regulation of Shigella virulence. Curr. Top. Microbiol. Immunol. 180: Mecsas, J. J., and E. J. Strauss Molecular mechanisms of bacterial virulence: type III secretion and pathogenicity islands. Emerg. Infect. Dis. 2: Mikulskis, A. V., I. Delor, V. H. Thi, and G. R. Cornelis Regulation of the Yersinia enterocolitica enterotoxin yst gene. Influence of growth phase, temperature, osmolarity, ph and bacterial host factors. Mol. Microbiol. 14: Minamino, T., and R. M. Macnab Components of the Salmonella flagellar export apparatus and classification of export substrates. J. Bacteriol. 181: Morooka, T., A. Umeda, and K. Amako Motility as an intestinal colonization factor for Campylobacter jejuni. J. Gen. Microbiol. 131: Nachamkin, I., B. M. Allos, and T. Ho Campylobacter species and Guillian-Barré syndrome. Clin. Microbiol. Rev. 11: Nataro, J. P., S. Hicks, A. D. Phillips, P. A. Vial, and C. L. Sears T84 cells in culture as a model for enteroaggregative Escherichia coli pathogenesis. Infect. Immun. 64: Newell, D. G., H. McBride, F. Saunders, Y. Dehele, and A. D. Pearson The virulence of clinical and environmental isolates of Campylobacter jejuni. J. Hyg. 94: Nuijten, P. J., B. A. van der Zeijst, and D. G. Newell Localization of immunogenic regions on the flagellin proteins of Campylobacter jejuni Infect. Immun. 59:

251 42. Panigrahi, P., G. Losonky, L. J. DeTolla, and J. G. Morris Human immune response to Campylobacter jejuni proteins expressed in vitro. Infect. Immun. 60: Pei, Z., C. Burucoa, B. Grignon, S. Baqar, X.-Z. Huang, D. J. Kopecko, A. L. Bourgeois, J.-L. Fauchere, and M. J. Blaser Mutation in the peb1a locus of Campylobacter jejuni reduces interactions with epithelial cells and intestinal colonization of mice. Infect. Immun. 66: Pickett, C. L., and C. A. Whitehouse The cytolethal distending toxin family. Trends Microbiol. 7: Porter, M. E., and C. J. Dorman A role for H-NS in the thermo-osmotic regulation of virulence gene expression in Shigella flexneri. J. Bacteriol. 176: Rivera-Amill, V., B. J. Kim, J. Seshu, and M. E. Konkel Secretion of the virulence associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J. Infect. Dis. 183: Rivera-Amill, V., and M. E. Konkel Secretion of Campylobacter jejuni Cia proteins is contact dependent, p In P. S. Paul and D. H. Francis (ed.), Mechanisms in the Pathogenesis of Enteric Diseases 2. Plenum Publishing Corporation, New York, NY. 48. Russell, R. G., M. O'Donnoghue, D. C. Blake, Jr., J. Zulty, and L. J. DeTolla Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J. Infect. Dis. 168: Sambrook, J., E. F. Fritsch, and T. Maniatis Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Press, Cold Spring Harbor. 50. Takata, T., S. Fujimoto, and K. Amako Isolation of nonchemotactic mutants of Campylobacter jejuni and their colonization of the mouse intestinal tract. Infect. Immun. 60: Wassenaar, T. M Toxin production by Campylobacter spp. Clin. Microbiol. Rev. 10: Wassenaar, T. M., B. A. van der Zeijst, R. Ayling, and D. G. Newell Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol. 139: Yao, R., D. H. Burr, and P. Guerry CheY-mediated modulation of Campylobacter jejuni virulence. Mol. Microbiol. 23: Young, G. M., D. H. Schmiel, and V. L. Miller A new pathway for the secretion of virulence factors by bacteria: the flagellar export apparatus functions as a protein-secretion system. Proc. Natl. Acad. Sci., USA. 96:

252 Table 1. Pre-culturing C. jejuni on deoxycholate-supplemented MH agar plates versus non-supplemented MH agar plates results in a two-fold increase in protein secretion Labelling MH MHD media + FBS + FBS + FBS No. of viable bacteria 6.5 x x 108 Protein concentration (mg/ml) Secreted protein (pg/bacterium)b ac. jejuni were subcultured overnight on non-supplemented and deoxycholate-supplemented MH agar plates at 37 C under microaerophilic conditions. Secretion assays were then performed as outlined in Materials and Methods. b [Total protein in labelling media containing bacteria and FBS minus protein in labelling media + FBS alone] divided by the number of viable bacteria included in experiment.

253 234 Table 2. Pre-culturing C. jejuni on deoxycholate-supplemented MH agar plates overcomes the inhibitory affect of chloramphenicol on T84 cell invasion. Adherence a Internalization (3 hours) -CM +CM -CM +CM MH ( ) x 106 ( ) x 106 ( ) x 105 ( ) x 104 b MH + deoxycholate ( ) x 106 ( ) x 106 ( ) x 105 ( ) x 105 ac. jejuni were subcultured overnight on non-supplemented and deoxycholate-supplemented MH agar plates at 37 C under microaerophilic conditions. Binding and internalization assays were performed as outlined in Materials and Methods. C. jejuni were incubated with (+CM) or without (-CM) chloramphenicol (128 µg/ml) at 37 C for 45 min prior to the inoculation of the T84 cell monolayers. Significant differences were not observed in the number of C. jejuni bound in the presence of CM when compared to the number of C. jejuni bound in the absence of CM. bmean value is significantly different (P < 0.01) from the mean value derived from the untreated sample as judged by the two-tailed Student s t test. The values were logarithmically transformed to achieve equality of variances.

254 235 Figure 1. Culturing C. jejuni on deoxycholate-supplemented MH agar plates increases the amount of the Cia proteins in the supernatant fluids. C. jejuni whole cell lysates (left panel) and supernatant fluids (right panel) were analyzed as outlined in the Materials and Methods. The panel on the left shows a CBB-R250 stained gel of the whole cell lysates from C. jejuni at the conclusion of the experiment. The panel on the right shows an autoradiograph of the proteins present in supernatant fluids. Lanes: 1, C. jejuni pre-cultured on MH agar plates, labeled in MEM minus FBS; 2, C. jejuni pre-cultured on deoxycholate-supplemented MH agar plates, labeled in MEM minus FBS; 3, C. jejuni pre-cultured on MH agar plates, labeled in FBS-supplemented MEM; 4, C. jejuni pre-cultured on deoxycholate-supplemented MH agar plates, labeled in FBS-supplemented MEM. Molecular mass standards, in kda, are indicated on the left.

255 236 Figure

256 237 Figure 2. Culturing C. jejuni on a deoxycholate-supplemented MH agar plate increases the amount of the Cia proteins in the supernatant fluids two-fold over that of C. jejuni cultured on a non-supplemented MH agar plate. This autoradiograph shows the proteins present in supernatant fluids. Lanes: 1, secreted protein profile of C. jejuni that was cultured on a MH agar plate prior to performing the secretion assay; 2, secreted protein profile of C. jejuni that was cultured on a deoxycholate-supplemented MH agar plate prior to performing the secretion assay; 3, a 1:2 dilution of the sample loaded in lane 2; 4, a 1:4 dilution of the sample loaded in lane 2. Molecular mass standards, in kda, are indicated on the left.

257 238 Figure

258 239 Figure 3. The profile of C. jejuni secreted proteins does not include FlaA or FlaB. The autoradiographs shown in panels A and B show the profiles of a C. jejuni whole cell lysate and secreted proteins. The immunoblots shown in panels C and D represent the corresponding gels that were transferred to nitrocellulose and then probed with an antiserum that contains antibodies reactive with both the FlaA and FlaB proteins. Panels: A, whole-cell lysate of C. jejuni ; B, secreted protein profile of C. jejuni ; C, immunoblot of a whole-cell lysate of C. jejuni reacted with the anti-flaa and anti-flab serum; and D, immunoblot of the secreted proteins of C. jejuni reacted with the anti-flaa and anti-flab serum. Molecular mass standards, in kda, are indicated on the left. IEF, isoelectric focusing.

259 240 Figure 3. secreted proteins anti-fla WCL anti-fla A SDS- PAGE IEF Acidic Basic B IEF Acidic Basic C SDS- PAGE D

260 241 Figure 4. FlhB, a component of the flagellar export apparatus, is required for Cia protein secretion. The C. jejuni wild-type isolate and isogenic flhb mutant were harvested from deoxycholate-supplemented MH agar plates and subjected to a secretion assay as outlined in the Materials and Methods. The panel on the left shows a CBB-R250 stained gel of the whole cell lysates from C. jejuni and isogenic flhb mutant. The panel on the right shows an autoradiograph of the proteins present in supernatant fluids. Lanes: 1, The C. jejuni flhb mutant was harvested from a deoxycholate-supplemented MH agar plates and subjected to a secretion assay using MEM minus FBS; 2, The C. jejuni wild-type isolate was harvested from a deoxycholate- supplemented MH agar plates and subjected to a secretion assay using MEM minus FBS; 3, The C. jejuni flhb mutant was harvested from a deoxycholate-supplemented MH agar plates and subjected to a secretion assay in MEM containing FBS; 4, The C. jejuni wild-type isolate was harvested from a deoxycholate-supplemented MH agar plates and subjected to a secretion assay in MEM containing FBS. Molecular mass standards, in kda, are indicated on the left.

261 242 Figure

262 243 Figure 5. Pre-culturing C. jejuni on deoxycholate-supplemented agar plates alters the kinetics of bacterial uptake. The number of internalized bacteria is shown for a C. jejuni wild-type isolate harvested from a deoxycholate-supplemented Mueller-Hinton agar plate ( ), a C. jejuni wild-type isolate harvested from a non-supplemented Mueller-Hinton agar plate ( ), a C. jejuni flhb mutant harvested from a deoxycholate-supplemented Mueller-Hinton agar plate ( ), and a C. jejuni flhb mutant harvested from a non-supplemented Mueller-Hinton agar plate ( ).

263 244 Figure No. of Intracellular Bacteria Time Post-Inoculation (min)

264 245 Chapter 8 Secretion of virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus Michael E. Konkel, John D. Klena, Vanessa Rivera-Amill, Marshall R. Monteville, Debabrata Biswas, and Joey Mickelson School of Molecular Biosciences, Washington State University, Pullman, WA Manuscript in preparation

265 246 ABSTRACT Campylobacter jejuni, a Gram-negative motile bacterium, secretes a set of proteins termed the- Campylobacter invasion antigens (Cia proteins). Previous work in our laboratory has demonstrated that a C. jejuni flhb flagellar protein export apparatus mutant is deficient in Cia protein secretion. The purpose of this study was to determine whether the flagellum serves as the export apparatus for the Cia proteins. As such, mutations were generated in three structural components of the flagella [i.e., flagellar basal body (flgb and flgc), hook (flge2) and filament (flaa- flab+ and flaa- flab-)] as well as in the flagellar protein export apparatus (i.e., flhb). While mutations that affected filament assembly were found to be non-motile and did not secrete Cia proteins (mot-, sec-), a filament mutant (flaa- flab+) was found to be non-motile but Cia proteinsecretion competent (mot-, sec+). Complementation of the flaa- flab- structural mutant with a shuttle plasmid harboring either the flaa or flab gene restored Cia protein secretion, suggesting that Cia export requires at least one of the two filament proteins. Infection of INT 407 human intestinal cells with the C. jejuni mutants revealed that the maximal invasion of the epithelial cells required motile bacteria that are secretion-competent. Collectively, these data suggest that the C. jejuni Cia proteins are secreted from the flagellar type III apparatus.

266 247 INTRODUCTION Campylobacter jejuni, a Gram-negative motile bacterium, is a frequent cause of human gastrointestinal infections (Tauxe, 1997). The spectrum of disease observed in C. jejuni-infected individuals ranges from asymptomatic to severe enteritis characterized by fever, severe abdominal cramping, and diarrhea with blood and mucous (Blaser et al., 1983; Allos and Blaser, 1995). Based on analogy with other more extensively characterized bacterial pathogens, the mechanism of C. jejuni-mediated enteritis is proposed to be multifactorial. Previous work has indicated that motility, as well as the presence of the flagellum, contributes to the ability of C. jejuni to colonize the intestinal tract of animals (Pavlovskis et al., 1991; Nachamkin et al., 1993; Wassenaar et al., 1993). The flagellum of C. jejuni is composed of a basal body, hook, and filament. The flagellar filament is comprised of two proteins, FlaA and FlaB, although it appears that FlaA is the preferred subunit (Alm et al., 1993). While the C. jejuni FlaA and FlaB flagellin proteins are synthesized concomitantly (Nuijten et al., 1991; Hendrixson et al., 2001), transcription of the flaa gene is expressed at much greater levels than the flab gene. In C. jejuni, the flaa gene is regulated by σ28 and the flab gene is regulated by σ54 (Alm et al., 1993; Hendrixson et al., 2001). Hendrixson et al. (2001) noted that a C. jejuni isolate deficient in σ28, which is encoded by the flia gene, is able to produce functional flagella composed exclusively of FlaB. This result suggests that the regulation of gene expression within C. jejuni differs from more intensely studied systems such as Salmonella enterica. In S. enterica, late flagellar gene expression and motility requires σ28 (Aldridge and Hughes, 2002). Previous work in our laboratory has demonstrated that C. jejuni synthesizes a set of proteins upon co-culturing with epithelial cells, some of which are secreted. The secreted proteins have been

267 248 collectively referred to as Campylobacter invasion antigens (Cia proteins) (Konkel et al., 1999). The functions of the secreted proteins are not known; however, insertional mutagenesis of the gene encoding a 73 kda secreted protein (ciab/ciab) results in a significant reduction in the number of C. jejuni internalized when compared to a C. jejuni wild-type isolate. The absence of Cia protein secretion from the C. jejuni ciab mutant was specific as the invasive phenotype of this organism was restored by complementation in trans with the ciab gene (Rivera-Amill et al., 2001). CiaB lacks the presence of an identifiable signal sequence (Konkel et al., 1999). In addition, an environmental stimulus is required to induce Cia protein secretion (Rivera-Amill et al., 2001). While these characteristics are reminiscent of virulence-associated type III secretion systems, translation of the complete genome of C. jejuni strain NCTC failed to reveal proteins with homology to a virulence-associated type III secretory system based on the functional classification of genes ( Projects/C.jejuni). Some proteins that comprise the classical type III secretory apparatus share amino acid sequence similarity with flagellar structural proteins (Mecsas and Strauss, 1996; Macnab, 1999). Moreover, evidence is beginning to accumulate that suggests components of the flagellar apparatus participate in the export of virulence determinants in several pathogens. For example, experiments by Young et al. (1999) demonstrated that Yersinia secretes flagellar outer proteins (Fops) via the flagellar apparatus. More recently, secretion of virulence-associated proteins from Bacillus thuringiensis has been found to be dependent on flha, an essential component of the flagellar export apparatus (Ghelardi et al., 2002). Consistent with the notion that components of the flagellar export apparatus can play a role in the export of virulence-associated proteins in some organisms, we noted that a C. jejuni flhb export mutant was deficient in the secretion of the Cia proteins (manuscript in preparation). The purpose of this study was to determine whether the flagellum serves as the secretory apparatus for the Cia proteins.

268 249 EXPERIMENTAL PROCEDURES Bacterial isolates and growth conditions. Human clinical C. jejuni isolates F38011 and were cultured as described previously (Rivera-Amill and Konkel, 1999). Plates were supplemented with kanamycin sulfate (200 µg/ml) as appropriate. Escherichia coli InvαF, MRF XL-1 Blue (TetR), and DH5α were cultured in Luria-Bertani broth or solid medium, supplemented with kanamycin (50 µg /ml) and tetracycline (15 µg /ml) as appropriate, in a 37 C incubator. Isolation of C. jejuni flagellar mutants. C. jejuni strain NCTC flhb, flgb, flgc, flge2, and flid gene sequences were obtained from the Sanger Center website ( Projects/C.jejuni). Two flge genes have been identified in NCTC (Cj0043 and Cj1729c), however, based on previously published work showing the sequence similarity of the flge2 product to other hook proteins (Kinsella et al., 1997; Lüneberg et al., 1998), flge2 (Cj1729c) was targeted. The flhb (Cj0335), flgb (Cj0528c), flgc (Cj0527c), flge2, and flid (Cj0548) genes in C. jejuni F38011 were disrupted by recombination via a single crossover event between the chromosomal gene and an internal fragment of the homologous gene in a suicide vector harboring a kanamycin resistance gene (Konkel et al., 1997). The C. jejuni flhb (518 bp F primer, AAA AAA CAG AAG AAC CCA CG; R primer, GTG CAA CCA TAG TAT AAA GCT C), flgb (359 bp; F primer, CAT TTA AAT CAA AAG AAC TGG; R primer; TCC ATC AAG AGC TGT TAT C), flgc (402 bp; F primer, AAG TGA TTT TGA TAT TAG TGG; R primer, GTA GCT TCA ATC AAA TCT GC), flge2 (1358 bp; F primer, AGT GGT TTA AAT ATA GGA ACT TCA AG; R primer, AAA GAA TCA TAA ATT TCA AGC C), and flid (886 bp; F primer, TGC GAA AAG AAA AGT TGT AGG; R primer, ATT TGC ATC TTC AGT AGT TCC) internal gene fragments were amplified by polymerase chain reaction (PCR) using gene specific primers. The flia (Cj0061c) gene in C. jejuni F38011 was disrupted by allelic replacement via a double crossover event between the flia

269 250 chromosomal gene and the flia gene containing an internal deletion and chloramphenicol resistance gene on a suicide vector harboring a kanamycin resistance gene (Konkel et al., 1997). The final vector was constructed using standard molecular biology techniques. Briefly, the 5 end of the flia gene with flanking DNA was amplified using the F1 and R1 primer set (F1, TTG GAT CCT TGG AAG ACA TTT TAA TAG AAG; R1, AAC CGC GGA AAG CTA GCC ACA AGC TCA TCT TGC TCT TTC). The underlined regions of the primers represent restriction sites that were used to facilitate the cloning and construction of the final vector. The flia F1 primer contained a BamHI restriction site and the flia R1 primer contained NheI and SacII restriction sites. The 3 end of the flia gene with a flanking region was amplified using a second set of primers (F2, TTG CTA GCC ACG AAG TGC TAG ATG ATC TTA AAG; R2, AAC CGC GGA TTT CTT TGA TTT CAT CTT TAT C). The flia F2 primer contained a NheI restriction site and the flia R2 primer contained a SacII restriction site. Following ligation of DNA fragments harboring the 5 and 3 regions of the flia gene, the chloramphenicol gene was ligated into the NheI restriction site. The recombinant vectors were introduced into C. jejuni F38011 by electroporation. C. jejuni F38011 mutants were identified by acquisition of kanamycin or chloramphenicol resistance/kanamycin sensitivity and specific gene disruption confirmed by PCR. Phenotypic analysis of the C. jejuni flagellar mutants. Motility assays were performed using Mueller-Hinton (MH) medium supplemented with 0.4% Select Agar (GIBCO BRL). A 10 µl suspension of each bacterial isolate was spotted on the surface of the semi-solid medium. Motility plates were incubated at 37 C under microaerophilic conditions for 48 h. C. jejuni isolates were also analyzed by transmission electron microscopy (TEM). Bacterial suspensions, prepared from MH agar plates using phosphate buffered saline, were added dropwise to formvar-coated copper grids. Bacteria were stained with 1% phosphotungstic acid. Samples were analyzed with a 1200 EX transmission electron microscope (JEOL).

270 251 Complementation analysis. A 2032-bp fragment containing the entire flaa gene and flanking DNA sequences was amplified from C. jejuni NCTC by PCR using the primers GGA TCC TAA AAC GCA TTT CAT CAC AGC (forward primer; BamHI linker is underlined in each case) and GGA TCC GAT TAA AGC AAA AAG TGT TC (reverse primer). The forward primer is 199 bp upstream of the AUG methionine initiation codon and the reverse primer extends 114 bp beyond the UAG stop codon. Following an intermediate cloning step into pcr2.1 (Invitrogen), the gel-purified insert was ligated into the BamHI site of pmek80 (Rivera-Amill et al., 2001). The resultant shuttle plasmid, designated pmek3502, was introduced into C. jejuni strain flaaflab- mutant by electroporation. Transformants were identified as described above. The C. jejuni pmek3503 shuttle vector (flab+) was generated in a similar fashion using the primers GGA TCC CAA AAT GTT TTA AGA TTA CTA CAG (forward primer) and GGA TCC TTT TTG CTT GGG TTT ATG CAC (reverse primer). The forward primer is 172 bp upstream of the AUG methionine initiation codon and reverse primer extends 161 bp beyond the UAA stop codon. The amplified PCR product containing the flab gene and flanking DNA sequences was 2052 bp. Analysis of the C. jejuni secreted proteins. C. jejuni were metabolically labeled with [35S]- methionine as described elsewhere (Konkel and Cieplak, 1992). The secreted proteins were concentrated and resolved by 12.5% SDS-PAGE using the discontinuous buffer system described by Laemmli (1970). Gels were treated with Amplify (Amersham, Life Science) according to the supplier s instructions. Autoradiography was performed with Kodak BioMax MR film at -70 C. Examination of the interactions of C. jejuni with INT 407 cells. Adherence and internalization assays were performed with INT 407 cells (human embryonic intestine, ATCC CCL 6) as previously described (Konkel et al., 1992). In a typical experiment, the number of adherent bacteria for C. jejuni F38011 was 1.0 x 106 colony forming units (CFU) and the number of

271 252 internalized bacteria was 5.0 x 104 CFU. Thus, the percentage of invading to adherent bacteria (I/A ratio) was 5.0. Significance between samples was determined using Student s t test following log 10 transformation of the data. Two-tailed P values were determined for each sample, and a P value < 0.01 was considered significant. All assays were repeated a minimum of three times. Other analytical procedures. Immunoblot analysis was performed with a 1:50 dilution of a rabbit C. jejuni anti-flagellin serum as described elsewhere (Grant et al., 1993). Bound antibodies were detected using peroxidase-conjugated goat-anti-rabbit IgG and 4-chloro-1-napthol (Sigma) as the chromogenic substrate.

272 253 RESULTS Generation of C. jejuni flagellar mutants. Experiments were performed to determine if the Cia proteins are secreted from the flagellar apparatus. Consistent with previous work in our laboratory in which a C. jejuni flhb flagellar export apparatus mutant was found to be non-motile (mot-) and Cia protein-deficient (sec-) (manuscript in preparation), the same was found true for a C. jejuni F38011 flhb mutant. The phenotype of the C. jejuni F38011 flhb mutant is summarized in Table 1A. The fact that this mutant was found to be sec- is consistent with the notion that the flagellar-export apparatus is required for Cia protein secretion. Additional experiments were undertaken to determine if the flagellum is necessary for Cia protein export. Two clinical strains of C. jejuni were used in this study. We generated mutations in the C. jejuni F38011 genes encoding the flagellar basal body (flgb and flgc) and hook (flge2). An insertion mutation was also generated in the C. jejuni F38011 flid gene encoding the putative filament cap protein. In addition, we utilized previously published isolates of C. jejuni isolates with disruptions in either flaa (GRK17) or flaa flab (GRK7) (Grant et al., 1993). For clarity, we will refer to the C. jejuni flaa mutant throughout the remainder of the text as a flaa- (flab+) mutant. After generating the mutants, we assessed whether each mutant was motile, synthesized the FlaA filament protein, and assembled a filament. Assays were initially performed to determine whether the C. jejuni mutants remained motile on semi-solid agar. Phenotypically, the C. jejuni F38011 flgb, flgc, and flge2 mutants were mot- (Table 1 and Fig. 1). Consistent with previous work, the C. jejuni strain flaa- flab- and flaa- (flab+) mutants were also mot- (Grant et al., 1993). An insertion in the flid gene of C. jejuni F38011 resulted in an organism with reduced motility when compared to the C. jejuni wild-type

273 254 isolate as judged by a reduction in the zone of bacterial migration on a motility-agar plate. To determine if flagellin is synthesized in the C. jejuni mutants, whole-cell lysates were subjected to SDS-PAGE coupled with immunoblot analysis using a flagellin polyclonal antiserum (Table 1 and Fig. 2, Panel A). In C. jejuni, the apparent mass of the flagellin monomers is 62 kda (Harris et al., 1987). As expected, an immunoreactive band was detected in the C. jejuni F38011 and C. jejuni wild-type isolates. However, an immunoreactive band of decreased intensity, likely corresponding to the FlaB protein, was detected in the C. jejuni F38011 flgb, flgc, and flge2 mutants as well as in the C. jejuni flaa- (flab+) mutant. An immunoreactive band was not detected in the C. jejuni flaa- flab- mutant. Collectively, these data indicate that the synthesis of the FlaA protein was greatly reduced in the C. jejuni F38011 flgb, flgc, and flge2 flagellar export mutants, which is consistent with the findings of Matz et al. (2002) who determined that there was a significant decrease in the level of flaa gene expression in a C. jejuni strain containing a defective flagellar export apparatus. C. jejuni organisms were negatively-stained with phosphotungstic acid and examined by TEM to assess whether the flagellin proteins were assembled into filaments. The C. jejuni wild-type F38011 and strains, when harvested from MH agar plates, produced long filaments. None of the C. jejuni F38011 flagellar structural mutants (flgb, flgc, and flge2) nor the C. jejuni flaa- flab- mutant assembled filaments as judged by TEM (not shown). A truncated filament was observed in the C. jejuni flaa- (flab+) mutant. These results are consistent with the findings of Wassenaar et al. (1991), who showed that motility correlated with the synthesis and assembly of a FlaA, but not a FlaB, filament. Wassenaar et al. (1991) also observed truncated filaments in a flaa- (flab+) mutant.

274 255 A mutation in the C. jejuni flid gene resulted in bacteria with a motility-impaired phenotype (Fig. 1). Although TEM examination revealed that the C. jejuni flid mutant displayed full-length filaments when harvested directly from motility agar plates, the same isolate displayed truncated filaments when harvested from broth cultures (not shown). In fact, the C. jejuni flid mutant grown in broth resembled that of the C. jejuni flaa- (flab+) mutant with respect to the filament structure. Yokoseki et al. (1995) found that a S. typhimurium flid mutant also formed minute swarms on motility agar plates but did not produce filaments in liquid medium. The investigators proposed that motility of the S. typhimurium flid mutant was due to flagellin monomers, which could not freely diffuse in the motility agar, being assembled into filaments (Yokoseki et al., 1995). In summary, each C. jejuni mutant exhibited a phenotype consistent with that predicted. Given this finding, metabolic labeling experiments were performed to address whether each mutant was capable of Cia protein secretion. Secretion of the Cia proteins requires the intact flagellar apparatus. Metabolic-labeling experiments were performed with labeling medium in the presence and absence of fetal bovine serum (FBS). FBS serves as an artificial signal to stimulate the synthesis and secretion of the Cia proteins (Rivera-Amill et al., 2001). The Cia proteins were readily identifiable in the supernatant fluids of C. jejuni F38011 and wild-type isolates, as well as the C. jejuni F38011 flid mutant, and the C. jejuni flaa- (flab+) mutant (Fig. 3, Panel A). No Cia protein secretion was apparent from the flgb, flgc and flge2 mutants, as well as the flaa- flab- mutant. These results suggested that an intact flagellar structure (containing the basal body, hook and at least a partial filament) is required for Cia protein secretion. In agreement with previous work indicating that a stimulus is required to induce Cia secretion, the Cia proteins were not detected in the supernatant fluids when FBS was omitted from the labeling medium (Fig. 3, Panel B).

275 256 Complementation analysis restores Cia protein secretion. To address whether the Cia proteins are secreted from a C. jejuni flaa+ flab- isolate, the C. jejuni flaa- flab- mutant was transformed with a Campylobacter shuttle plasmid harboring an intact flaa gene. As a control, the C. jejuni flaa- flab- mutant was also transformed with the same C. jejuni shuttle plasmid harboring an intact flab gene. The latter served as a control because Cia protein secretion had been observed from the C. jejuni flaa- (flab+) mutant. Immunoblot analysis revealed that one C. jejuni transformant synthesized the FlaA product while the other transformant synthesized the FlaB product (Fig. 2, Panel B). The Cia proteins were secreted from the C. jejuni flaa+ transformant, but not to the same extent as the wild-type isolate (Fig. 4). Similarly, a slightly reduced amount of the Cia proteins were secreted from the flab transformant versus the C. jejuni flaa- (flab+) mutant (Fig. 4). These observations further support the finding that an intact flagellar structure is required for C. jejuni Cia protein secretion, and indicate that the specific composition of the flagellar filament, per se, does not qualitatively effect Cia protein export. Maximal C. jejuni invasion requires Cia protein secretion and motility. In vitro assays were performed to assess the contribution of motility and Cia protein secretion in C. jejuni uptake (Table 1A). Because C. jejuni binding is a prerequisite for invasion (Konkel and Cieplak, 1992), C. jejuni-int 407 cell contact was promoted by centrifugation. Despite this effort to minimize motility-dependent effects on C. jejuni-host cell association, a reduction was noted in binding of all C. jejuni flagellar mutants (flhb, flgb, flgc, flge2, and flid) to the INT 407 cells when compared to the C. jejuni wild-type isolate. Therefore, the data are presented as the percentage of adherent bacteria that are internalized to more readily identify those organisms with deficiencies in invasive potential. With the exception of the C. jejuni flid mutant, a significant (P < 0.01) decrease was noted in internalization efficiency of the C. jejuni flhb, flgb, flgc, and flge2 flagellar mutants when compared to the C. jejuni wild-type isolate. Noteworthy is that the C. jejuni ciab mutant displayed

276 257 a mot+ sec- phenotype, but exhibited an invasion phenotype that was similar to that as the flhb, flgb, flgc, and flge2 flagellar export and structural mutants. These data suggest that Cia protein secretion, in itself, contributes more to C. jejuni host cell invasion than the mot+ phenotype alone. Binding and internalization assays were also performed with the C. jejuni wild-type isolate and isogenic flaa- flab- (mot-, sec-) and C. jejuni flaa- (flab+) (mot-, sec+) mutants (Table 1B). Significant (P < 0.01) differences were noted in the invasive potentials of C. jejuni flaa- flab- and flaa- (flab+) mutants when compared to the wild-type isolate. A significant difference was also noted in the invasiveness of the C. jejuni flaa- flab- and flaa- flab+ mutants when compared to one another, with a greater number of the flaa- (flab+) mutant organisms internalized. Wassenaar et al. (1991) also reported that a flaa- flab+ isolate was capable of invading INT 407 epithelial cells as long as cell-to-cell contact was promoted by centrifugation. Complementation of the flaa- flab- mutant with a shuttle vector, in trans, harboring either the flaa or flab gene resulted in transformants that displayed characteristics similar to that of the C. jejuni wild-type isolate or the C. jejuni flaa- (flab+) mutant, respectively. However, transformation of the C. jejuni flaa- flab- mutant with a shuttle plasmid harboring the flaa or flab gene did not fully restore Cia secretion or the percent of bacteria internalized to the expected levels. A possible explanation for the diminished secretion rates, and corresponding reduction in internalization efficiency, is that an increase in the amount of FlaA or FlaB within a cell may interfere with Cia secretion. In agreement with the results presented in Table 1A, these data suggest that the secretion of the Cia proteins from the flagellar-export apparatus significantly contributes to C. jejuni host cell invasion. The C. jejuni F38011 flid mutant displayed a similar filament structure as the C. jejuni flaa- (flab+) mutant. Given this finding, we hypothesized that the C. jejuni F38011 flid mutant grown in

277 258 liquid medium should be internalized at a level that is comparable to the C. jejuni flaa- (flab+) mutant given that both mutants displayed truncated flagellar filaments. Invasion assays were performed with the C. jejuni F38011 flid mutant grown in liquid medium using the C. jejuni F38011 flid mutant grown on solid medium as an appropriate control (Table 1C). Consistent with the results obtained in Table 1B using the C. jejuni flaa- (flab+) mutant, a two fold difference in the I/A ratio was obtained with the F38011 flid mutant cultured in broth when compared to the wild-type strain. Based on this finding, as well as that obtained with the C. jejuni fla mutants, it appears that organisms that have a filament and are mot+ are 1.5 to 2.5-fold more invasive than organisms that have a filament and are mot-. However, the C. jejuni F38011 wild-type isolate (sec+, mot+) was found to be approximately 50-fold more invasive than the C. jejuni ciab isogenic mutant (sec-, mot+), which highlights the importance of the secretion-competent phenotype. Cia protein export is independent of σ28. The C. jejuni flaa gene is transcribed from a σ28- specific promoter (Nuijten et al., 1991) and the C. jejuni flab gene is transcribed from a σ54- promoter (Hendrixson et al., 2001). A mutation was generated in the flia gene of C. jejuni F38011, which encodes σ28, to determine whether this sigma factor is responsible for cia gene transcription. As predicted, the C. jejuni flia- mutant was non-motile, most likely due to the lack of flaa transcription. Indeed, only a weakly reactive band, corresponding in mass to the FlaB protein, was detected in the mutant as judged by immunoblot analysis of whole cell lysates using the antiflagellin polyclonal serum (Fig. 2, Panel C). The C. jejuni flia mutant also resembled the C. jejuni flaa- (flab+) mutant with respect to the filament structure as judged by TEM analysis (not shown). Finally, secretion assays revealed that the C. jejuni flia- mutant was secretion competent (Fig. 5). This finding demonstrates that the cia genes are transcribed from a sigma factor other σ28.

278 259 DISCUSSION Gram-negative bacteria possess at least six different mechanisms to actively transport proteins across the bacterial membranes (reviewed in (Büttner and Bonas, 2002)). Of these six pathways, protein secretion induced upon contact of the bacteria with host cells has been referred to as the type III secretion pathway (Cornelis, 1998). Requirements of type III secretion pathways include: 1) the absence of a cleavable, hydrophobic amino-terminal signal sequence in the secreted protein; 2) the export of the protein across the bacterial inner and outer membranes without a periplasmic intermediate; and 3) a signal to induce secretion (Kubori et al., 1998). Most, but apparently not all, type III secreted proteins require chaperones (Cheng et al., 1997; Cheng and Schneewind, 2000). We have demonstrated elsewhere that C. jejuni synthesize a novel set of proteins upon co-culturing with epithelial cells, some of which are secreted (Konkel et al., 1999; Rivera-Amill et al., 2001). The secreted proteins were termed the Campylobacter invasion antigens (Cia proteins) as they were found to be required for maximal invasion of intestinal epithelial cells by C. jejuni (Konkel et al., 1999; Rivera-Amill et al., 2001; Monteville and Konkel, 2002). Because the Cia proteins are synthesized and secreted in response to an environmental stimulus and the fact that the CiaB secreted protein is not processed, the Cia proteins would appear to conform to the criteria of type III proteins. However, a BLAST search of the C. jejuni genome revealed the only apparent type III export system in C. jejuni to be the flagellar apparatus. A considerable amount of evidence exists indicating that motility, and the product of the flaa gene, is essential for the maximal colonization of animals by C. jejuni (Morooka et al., 1985; Newell, 1986; Pavlovskis et al., 1991; Nachamkin et al., 1993; Wassenaar et al., 1993). In parallel with these studies, additional work has been done to dissect the importance of motility versus the actual flagellum in the interaction of C. jejuni with cultured epithelial cells (Wassenaar et al., 1991; Grant et al., 1993; Yao et al., 1994). Investigators have targeted genes encoding various flagellar structural

279 260 components, and while discrepancies have been reported with respect to the phenotype of particular mutants (Yao et al., 1994; Golden and Acheson, 2002), there is no doubt that motility plays a role in C. jejuni pathogenesis. Moreover, motility, and the expression of the flaa gene, are clearly necessary for the maximal invasion of eukaryotic cells and for the translocation of C. jejuni across polarized cells (Wassenaar et al., 1991; Grant et al., 1993). Perhaps more relevant to this study, differences in the invasive potential of C. jejuni flaa- (flab+) and C. jejuni flaa- flab- isolates were noted in earlier studies; C. jejuni flaa- (flab+) isolates have been reported to be more invasive than a C. jejuni flaa- flab- isolate (Wassenaar et al., 1991; Grant et al., 1993). Also noteworthy is that the invasiveness of a C. jejuni flaa- (flab+) isolate is enhanced 10-fold by promoting bacteria-host cell contact via centrifugation; in contrast, the centrifugation step did not change the invasive potential of the C. jejuni wild-type isolate (Wassenaar et al., 1994). Based on the difference observed in the invasive potential of the C. jejuni flaa- (flab+) isolate versus the C. jejuni flaa- flab- isolate, Grant et al. (1993) concluded that the flagellar structure played a role in internalization that was independent of motility. Prior to this study, it was unclear how the flagellum could have any effect, other than conferring motility or acting directly as an adhesin, on C. jejuni host cell invasion. Given our previous work suggesting that the Cia proteins are secreted in a type III dependent manner and the absence of a type III secretion system dedicated to the export of virulence proteins in the C. jejuni genome, experiments were performed to determine if the flagellum serves as the Cia export apparatus. Mutations that abolished flagellin export (flhb, flgb, flgc, and flge2), protein structure (flaa, flab), and synthesis (flia) were generated. Using these mutants, we have shown that C. jejuni motility and virulence are linked in a fashion that is uncommon among other pathogens. Specifically, we demonstrate that the C. jejuni Cia proteins are secreted via the flagellar export apparatus. The secretion system utilized by C. jejuni appears unique in that the synthesis and assembly of either one of the filament proteins is required for Cia protein secretion.

280 261 To address which components of the flagellar apparatus serve as the C. jejuni Cia export apparatus, mutations were generated in the genes encoding the flagellar basal body (flgb and flgc), hook (flge2), and filament cap (flid) proteins in C. jejuni strain F Mutations that affected either the export of flagellar components (i.e., flhb) or the non-filament structural components (flgb, flgc, and flge2) resulted in a sec- phenotype. Comparable results were obtained using a second C. jejuni strain, 81116, in which the genes encoding the flagellin filament (flaa- flab-) were mutated. Although some of the mutations generated in this study had a polar effect, in each case, the downstream genes were associated with flagellar biosynthesis. Therefore we predict that the phenotype associated with polarity on the downstream genes would be similar to the targeted gene. Notably, the flhb gene in C. jejuni isolates is monocistronic, so a mutation in this gene is not polar (Matz et al., 2002). Additionally, complementation of the flagellar filament defect in with either flaa or flab restored the organism s ability to secrete the Cia proteins. Taken together, the evidence is overwhelmingly in favor of Cia protein secretion through the flagellar export system. The fact that the FlaA protein was greatly reduced in the whole cell lysates of the C. jejuni flgb, flgc, and flge2 mutants, as judged by immunoblot analysis with a flagellin antiserum, raised the possibility that C. jejuni may possess the FlgM anti-sigma factor. In bacteria such as Salmonella, Yersinia, and Helicobacter pylori, the negative regulator FlgM inhibits flagellin transcription in response to a defective hook-basal body complex (Daughdrill et al., 1997; Chadsey and Hughes, 2001; Colland et al., 2001). A protein corresponding to a putative FlgM homolog has been identified in the genome of C. jejuni NCTC (Cj1464) that shares similarity with the recently identified FlgM protein from H. pylori (Colland et al., 2001). In S. enterica serovar Typhi, σ28 is involved in regulating gene expression of type III proteins (Eichelberg and Galán, 2000). Given this fact, it is clear that the expression of virulence genes in S. enterica serovar Typhi is affected, albeit indirectly, by FlgM and the assembly of the flagellar export apparatus. Regardless, the Cia proteins are secreted in a C. jejuni flia (σ28) mutant. Therefore, the cia genes cannot be subject to

281 262 transcriptional regulation via a mechanism involving the anti-sigma factor FlgM. Our results are in agreement with that of Jagannathan et al. (2001), who observed that a C. jejuni flia mutant displayed truncated flagellin; this finding indicates that σ28 is not responsible for the transcription of the genes encoding the flagellar export apparatus in C. jejuni. It is unlikely that the Cia proteins are secreted from an export complex distinct from the flagellar apparatus given that the Cia proteins were not secreted from the C. jejuni flaa- flab- mutant; this particular mutant is one in which σ28 and σ54-dependent genes are transcribed. The sec- phenotype of the C. jejuni flaaflab- mutant was restored upon transformation of the isolate with a shuttle plasmid harboring either the flaa or flab gene, implicating the flagellar export apparatus in Cia secretion. In summary, we have shown that the C. jejuni Cia proteins are secreted via the flagellar apparatus and that synthesis of either FlaA or FlaB is sufficient for Cia secretion. Coupled with the metabolic labeling experiments in which the C. jejuni isolates were examined for protein secretion, the adherence and internalization data indicate that the difference in the invasiveness of the C. jejuni flaa- flab+ and C. jejuni flaa- flab- isolates is a result of Cia secretion. Based on the phenotype of the C. jejuni ciab mutant (mot+, sec-), it is also evident that motility, in the absence of Cia protein secretion, is not sufficient for C. jejuni invasion of epithelial cells. We believe the data presented here reveals what had formerly been unclear with respect to the Cia protein export apparatus and the relationship between C. jejuni motility and host cell-invasion.

282 263 ACKNOWLEDGEMENTS We thank Gary A. Flom for assistance in generation of the flia suicide vector, which was used to generate the C. jejuni F38011 flia mutant. We also thank Gary A. Flom (School of Molecular Biosciences, Washington State University), Nicole Lindstrom (School of Molecular Biosciences, Washington State University), and Chris Davitt (School of Biological Sciences, Washington State University, Pullman, WA) for assistance in the preparation and examination of samples for TEMexamination. We are grateful to Scott Minnich (Department of Microbiology, Molecular Biology, and Biochemistry, University of Idaho, Moscow, Idaho) for helpful discussions and for reviewing this manuscript. A portion of this work was presented at the William R. Wiley Award Research Exposition held at Washington State University in Pullman, Washington, February 21, 2001, at the 11th International Workshop on Campylobacter, Helicobacter, & Related Organisms, in Freiburg, Germany, September 1-5, 2001 and at the 82nd annual meeting of the Conference of Research Workers in Animal Diseases, St. Louis, Missouri, November 11-13, This work was supported by a grant from the National Institutes of Health (Grant DK50567) awarded to MEK.

283 264 REFERENCES Aldridge, P., and Hughes, K.T Regulation of flagellar assembly. Curr. Opin. Microbiol. 5: Allos, B.M., and Blaser, M.J Campylobacter jejuni and the expanding spectrum of related infections. Clin. Infect. Dis. 20: Alm, R.A., Guerry, P., and Trust, T.J The Campylobacter σ54 flab flagellin promoter is subject to environmental regulation. J. Bacteriol. 175: Blaser, M.J., Wells, J.G., Feldman, R.A., Pollard, R.A., and Allen, J.R Campylobacter enteritis in the United States. A multicenter study. Ann. Int. Med. 98: Büttner, D., and Bonas, U Port of entry- the type III secretion translocon. Trends Microbiol. 10: Chadsey, M.S., and Hughes, K.T A multipartite interaction between Salmonella transcription factor σ28 and its anti-sigma factor FlgM: implications for σ28 holoenzyme destabilization through stepwise binding. J. Mol. Biol. 306: Cheng, L.W., Anderson, D.M., and Schneewind, O Two independent type III secretion mechanisms for YopE in Yersinia enterocolitica. Mol. Microbiol. 24: Cheng, L.W., and Schneewind, O Type III machines of Gram-negative bacteria: delivering the goods. Trends Microbiol. 8: Colland, F., Rain, J.C., Gounon, P., Labigne, A., Legrain, P., and De Reuse, H Identification of the Helicobacter pylori anti-σ28 factor. Mol. Microbiol. 41: Cornelis, G.R The Yersinia deadly kiss. J. Bacteriol. 180: Daughdrill, G.W., Chadsey, M.S., Karlinsey, J.E., Hughes, K.T., and Dahlquist, F.W The C-terminal half of the anti-sigma factor, FlgM, becomes structured when bound to its target, σ28. Nat. Struct. Biol. 4: Eichelberg, K., and Galán, J.E The flagellar sigma factor FliA (σ28) regulates the expression of Salmonella genes associated with the centisome 63 type III secretion system. Infect. Immun. 68: Ghelardi, E., Celandroni, F., Salvetti, S., Beecher, D.J., Gominet, M., Lereclus, D., Wong, A.C.L., and Senesi, S Requirement of flha for swarming differentiation, flagellin export, and secretion of virulence-associated proteins in Bacillus thuringiensis. J. Bacteriol. 184: Golden, N.J., and Acheson, D.W.K Identification of motility and autoagglutination Campylobacter jejuni mutants by random transposon mutagenesis. Infect. Immun. 70:

284 Grant, C.C.R., Konkel, M.E., Cieplak, W. Jr., and Tompkins, L.S Role of flagella in adherence, internalization, and translocation of Campylobacter jejuni in nonpolarized and polarized epithelial cell cultures. Infect. Immun. 61: Harris, L.A., Logan, S.M., Guerry, P., and Trust, T.J Antigenic variation of Campylobacter flagella. J. Bacteriol. 169: Hendrixson, D.R., Akerley, B.J., and DiRita, V.J Transposon mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol. Microbiol. 40: Jagannathan, A., Constantinidou, C., and Penn, C.W Roles of rpon, flia and flgr in expression of flagella in Campylobacter jejuni. J. Bacteriol. 183: Kinsella, N., Guerry, P., Cooney, J., and Trust, T.J The flge gene of Campylobacter coli is under the control of the alternative sigma factor σ54. J. Bacteriol. 179: Konkel, M.E. and Cieplak, W., Jr Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect. Immun. 60: Konkel, M.E., Corwin, M.D., Joens, L.A., and Cieplak, W., Jr Factors that influence the interaction of Campylobacter jejuni with cultured mammalian cells. J. Med. Microbiol. 37: Konkel, M.E., Garvis, S.G., Tipton, S.L., Anderson, D.E., Jr., and Cieplak, W., Jr Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24: Konkel, M.E., Kim, B.J., Rivera-Amill, V., and Garvis, S.G Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32: Kubori, T., Matsushima, Y., Nakamura, D., Uralil, J., Lara-Tejero, M., Sukhan, A., Galán, J.E., and Aizawa, S.-I Supramolecular structure of the Salmonella typhimurium type III protein secretion system. Science 280: Laemmli, U.K Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: Lüneberg, E., Glenn-Calvo, E., Hartmann, M., Bär, W., and Frosch, M The central, surface-exposed region of the flagellar hook protein FlgE of Campylobacter jejuni shows hypervariability among strains. J. Bacteriol. 180: Macnab, R.M The bacterial flagellum: reversible rotary propeller and type III export apparatus. J. Bacteriol. 181: Matz, C., van Vliet, A.H.M., Ketley, J., and Penn, C.W Mutational and transcriptional analysis of the Campylobacter jejuni flagellar biosynthesis gene flhb. Microbiol. 148:

285 266 Mecsas, J., and Strauss, E.J Molecular mechanisms of bacterial virulence: type III secretion and pathogenicity islands. Emerg. Infect. Dis. 2: Monteville, M.R., and Konkel, M.E Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface. Infect. Immun. 70: Morooka, T., Umeda, A., and Amako, K Motility as an intestinal colonization factor for Campylobacter jejuni. J. Gen. Microbiol. 131: Nachamkin, I., Yang, X.H., and Stern, N.J Role of Campylobacter jejuni flagella as colonization factors for three-day-old chicks: analysis with flagellar mutants. Appl. Environ. Microbiol. 59: Newell, D.G Monoclonal antibodies directed against the flagella of Campylobacter jejuni: production, characterization and lack of effect on the colonization of infant mice. J. Hyg. 96: Nuijten, P.J.M., van der Zeijst, B.A.M., and Newell, D.G Localization of immunogenic regions on the flagellin proteins of Campylobacter jejuni Infect. Immun. 59: Pavlovskis, O.R., Rollins, D.M., Haberberger, R.L., Jr., Green, A.E., Habash, L., Strocko, S., and Walker, R.I Significance of flagella in colonization resistance of rabbits immunized with Campylobacter spp. Infect. Immun. 59: Rivera-Amill, V., Kim, B.J., Seshu, J., and Konkel, M.E Secretion of the virulence associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J. Infect. Dis. 183: Rivera-Amill, V., and Konkel, M.E.: Secretion of Campylobacter jejuni Cia proteins is contact dependent. In: Paul, P.S. and Francis, D.H. (Eds.), Mechanisms in the Pathogenesis of Enteric Diseases 2. Plenum Publishing Corporation, New York, NY., 1999, pp Tauxe, R.V Emerging foodborne diseases: an evolving public health challenge. Emerg. Infect. Dis. 3: Wassenaar, T.M., Bleumink-Pluym, N.M., Newell, D.G., Nuijten, P.J.M. and van der Zeijst, B.A.M Differential flagellin expression in a flaa flab+ mutant of Campylobacter jejuni. Infect. Immun. 62: Wassenaar, T.M., Bleumink-Pluym, N.M., and van der Zeijst, B.A.M Inactivation of Campylobacter jejuni flagellin genes by homologous recombination demonstrates that flaa but not flab is required for invasion. EMBO J. 10: Wassenaar, T.M., van der Zeijst, B.A., Ayling, R., and Newell, D.G Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol. 139:

286 Yao, R., Burr, D.H., Doig, P., Trust, T.J., Niu, H., and Guerry, P Isolation of motile and non-motile insertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukaryotic cells. Mol. Microbiol. 14: Yokoseki, T., Kutsukake, K., Ohnishi, K., and Iino, T Functional analysis of the flagellar genes in the flid operon of Salmonella typhimurium. Microbiol. 141: Young, G.M., Schmiel, D.H., and Miller, V.L A new pathway for the secretion of virulence factors by bacteria: the flagellar export apparatus functions as a proteinsecretion system. Proc. Natl. Acad. Sci. USA 96:

287 268 Table 1. Phenotypes displayed by the C. jejuni wild-type isolates and isogenic mutants A. C. jejuni flagellar protein export mutants Isolate Relevant Characteristic Flagellin synthesis Filament assembly Motility Secretion Percent adherence Percent invasion Percent adherent bacteria internalized F38011 wild-type flhb- export mutant flgb- basal body mutant flgc- basal body mutant flge2- hook mutant flid- filament cap mutant +/ ciab- Cia secretion deficient mutant

288 269 B. C. jejuni filament mutants Isolate Relevant Characteristic Flagellin synthesis Filament assembly Motility Secretion Percent adherence Percent invasion Percent adherent bacteria internalized wild-type flaa- flab- filament mutant flaa- (flab+) filament mutant + truncated flaa- flabpmek3502 (flaa+) flaa- flabpmek3503 (flab+) complemented filament mutant complemented filament mutant + truncated

289 270 C. C. jejuni filament cap mutants Isolate Relevant Characteristic Flagellin synthesis Filament assembly Motility Secretion Percent adherence Percent invasion Percent adherent bacteria internalized F38011 wild-type flid- (plate) filament cap mutant +/ flid- (broth) filament cap mutant +/- truncated Not determined Not determined flgc- basal body mutant

290 Figure 1. Assessment of C. jejuni motility on Mueller-Hinton medium supplemented with 0.4% Select Agar. 271

291 Figure C. jejuni F38011 parental isolate C. jejuni F38011 flhb - C. jejuni F38011 flgb - C. jejuni F38011 flgc- C. jejuni F38011 flge2 - C. jejuni F38011 flid- C. jejuni F38011 ciab- Export mutant Basal body Hook Cap mutants mutant mutant Cia secreted protein mutant C. jejuni parental isolate C. jejuni flaa - flab - C. jejuni flaa - (flab + ) C. jejuni flaa - flab - + pmek3502 (flaa + ) C. jejuni flaa - flab - + pmek3503 (flab + ) C. jejuni F38011 flia - Filament mutants Complemented flagellin isolates σ 28 mutant

292 273 Figure 2. Flagellin production in C. jejuni F38011 and isolates. Flagellin was detected by immunoblot analysis with a flagellin polyclonal antiserum. Panel A, Lanes: 1, C. jejuni F38011 wild-type isolate; 2, C. jejuni F38011 flgb mutant; 3, C. jejuni F38011 flgc mutant; 4, C. jejuni F38011 flge2 mutant; 5, C. jejuni F38011 flid mutant; 6, C. jejuni 81116; 7, C. jejuni flaa- (flab+) mutant; 8, C. jejuni flaa- flab- mutant. Panel B, Lanes: 1, C. jejuni wild-type isolate; 2, C. jejuni flaa- flab- mutant; 3, C. jejuni flaa- flab- mutant + pmek3502 (flaa+); and 4, C. jejuni flaa- flab- mutant + pmek3503 (flab+). Panel C, Lanes: 1, C. jejuni F38011 wild-type isolate; 2, C. jejuni F38011 flia mutant. The arrows indicated faint immunoreactive bands, corresponding in size to the predicated M r of flagellin and predicted to represent FlaB, that were detected in some lanes (Panel A, lanes 2-5 and lane 7; Panel B, lane 4; and Panel C, lane 2). Molecular mass standards, in kda, are indicated on the left.

293 Figure A B C

294 275 Figure 3. C. jejuni Cia protein secretion requires a functional flagellar apparatus. C. jejuni cells were pre-cultured on Mueller-Hinton plates (lane 1) and Mueller-Hinton plates supplemented with deoxycholate (Lanes 2-10), and labeled in Minimal Essential Medium in the presence (panel A) and absence (panel B) of FBS. Lanes: 1, C. jejuni F38011 (Mueller-Hinton plate); 2, C. jejuni F38011 (Mueller-Hinton deoxycholate plate); 3, C. jejuni F38011 flgb mutant; 4, C. jejuni F38011 flgc mutant; 5, C. jejuni F38011 flge2 mutant; 6, C. jejuni 81116; 7, C. jejuni flaa- flab- mutant; 8, C. jejuni flaa- (flab+) mutant; 9, C. jejuni F38011; and 10, C. jejuni F38011 flid mutant. Arrows indicate the readily identifiable secreted proteins. Molecular mass standards, in kda, are indicated on the left.

295 276 Figure 3. A B

296 277 Figure 4. Transformation of a C. jejuni flaa- flab- isolate with a recombinant plasmid harboring either flaa or flab restores Cia secretion. C. jejuni cells were pre-cultured on Mueller-Hinton plates supplemented with deoxycholate and labeled in Minimal Essential Medium in the presence of FBS. Lanes: 1, C. jejuni wild-type isolate; 2, C. jejuni flaa- flab- mutant harboring pmek3502 (flaa+); 3, C. jejuni flaa- flab- mutant; 4, C. jejuni flaa- flab- mutant harboring pmek3503 (flab+); and 5, C. jejuni flaa- (flab+) mutant. Molecular mass standards, in kda, are indicated on the left.

297 278 Figure