Insights Into the Virulence Determinants of the Emerging Pathogen Kingella Kingae

Insights Into the Virulence Determinants of the Emerging Pathogen Kingella kingae

Eric Allen Porsch

Department of Molecular Genetics and Microbiology Duke University

Date:______Approved:

______Joseph W. St. Geme, III, Supervisor

______Soman N. Abraham

______Margarethe J. Kuehn

______Patrick C. Seed

______Raphael H. Valdivia

Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Molecular Genetics and Microbiology in the Graduate School of Duke University

2012

ABSTRACT

Insights Into the Virulence Determinants of the Emerging Pathogen Kingella kingae

Eric Allen Porsch

Department of Molecular Genetics and Microbiology Duke University

Date:______Approved:

______Joseph W. St. Geme, III, Supervisor

______Soman N. Abraham

______Margarethe J. Kuehn

______Patrick C. Seed

______Raphael H. Valdivia

An abstract of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Molecular Genetics and Microbiology in the Graduate School of Duke University

2012

Abstract

Kingella kingae is an emerging bacterial pathogen that is being recognized increasingly as an important etiology of septic arthritis, osteomyelitis, and bacteremia, especially in young children. The pathogenesis of K. kingae disease begins with bacterial adherence to respiratory epithelium in the posterior pharynx. Previous work identified type IV pili as a critical factor for adherence to human epithelial cells. However, the finding that a significant percentage of pharyngeal isolates are non-piliated suggests that

K. kingae expresses additional surface factors that modulate interactions with host cells and likely play key roles in the pathogenesis of K. kingae disease. The purpose of this work was to increase our understanding of K. kingae virulence determinants, specifically focused on defining the surface factors and the mechanism involved in K. kingae adhesive interactions with epithelial cells. Additionally, this work aimed to further characterize components of the K. kingae type IV pilus system, namely the PilC proteins and PilA2.

We first set out to identify non-pilus factors that influence K. kingae interactions with human epithelial cells. Using targeted genetic approaches, we found that insertional inactivation of the gene encoding a predicted trimeric autotransporter protein called Knh (Kingella NhhA homolog) resulted in reduced adherence to human epithelial cells. In addition, using a variety of techniques, including morphological

analysis, cationic ferritin staining, and thin section transmission electron microscopy, we established that K. kingae elaborates a surface-associated polysaccharide capsule that requires a predicted ABC-type transporter export operon called ctrABCD for surface presentation. Furthermore, using quantitative human epithelial cell adherence assays, we discovered that the presence of surface capsule interferes with Knh-mediated adherence by non-piliated organisms and that maximal adherence in the presence of capsule requires the predicted type IV pilus retraction machinery, PilT/PilU. Based on the data presented here, we propose a novel adherence mechanism that allows K. kingae to adhere efficiently to human epithelial cells while remaining encapsulated and more resistant to immune clearance.

Having established that K. kingae produces a capsule, a large-scale polysaccharide purification technique was developed for capsule analysis of strain 269-

492. Biochemical assays determined that the purified material contained thiobarbituric and phenol-sulfuric acid reactive glycosyl residues. In collaboration with the University of Georgia Complex Carbohydrate Research Center (CCRC), mass spectrometry identified galactose, N-acetyl-galactosamine, and Kdo as the major glycosyl components of the polysaccharide preparation. NMR spectroscopy revealed that the purified material contained two distinct polysaccharides with the structures of →5)-β-Galf-(1→ and →3)-β-GalNAcp-(1→5)-β-Kdop-(2→. Further characterization of the

polysaccharides expressed by K. kingae may have implications for disease prevention strategies.

Previous work in our lab found that two PilC-like proteins called PilC1 and PilC2 influence type IV pili expression and pilus-mediated adherence. Production of either

PilC1 or PilC2 is necessary for K. kingae piliation and bacterial adherence. We set out to further investigate the role of PilC1 and PilC2 in type IV pilus-associated phenotypes.

We found that PilC1 contains a functional nine amino acid calcium-binding (Ca-binding) site with homology to the Pseudomonas aeruginosa PilY1 Ca-binding site and that PilC2 contains a functional 12 amino acid Ca-binding site with homology to the human calmodulin Ca-binding site. Using targeted mutagenesis to disrupt the Ca-binding sites, we demonstrated that the PilC1 and PilC2 Ca-binding sites are dispensable for piliation.

Interestingly, we show that the PilC1 site is necessary for twitching motility and adherence to Chang epithelial cells, while the PilC2 site has only a minor influence on twitching motility and no influence on adherence. These findings establish key differences in PilC1 and PilC2 function in K. kingae and provide insights into the biology of the PilC-like family of proteins.

Lastly, we set out to define the role of the PilA2 minor pilin in K. kingae strain

269-492. While previous studies indicated that PilA2 is not essential for pilus expression or adherence to epithelial cells, analysis of the pilin locus in a diverse set of clinical isolates revealed that the pilA2 gene sequence is highly conserved, suggesting it serves

an important function. Using targeted mutagenesis we showed that PilA2 is not essential for twitching motility and may or may not be involved in natural competence.

Western blot analysis was unable to detect PilA2 in wild type pilus preparations, indicating that it is expressed at a level beneath the assay detection limit or does not localize to the pilus. Additionally, site-directed mutagenesis was used to place pilA2 under control of the highly active pilA1 promoter and showed that PilA2 is able to be assembled into fibers that mediate intermediate adherence to epithelial cells.

Taken together, this work expands our knowledge of the K. kingae surface factor repertoire and provides insights into the roles of type IV pilus components. The mechanism of K. kingae adherence to epithelial cells is beginning to emerge. These contributions may lead to novel strategies for the prevention of invasive K. kingae disease in young children.

vii

Dedication

I dedicate this work to my wife, Kimberly, and my daughter, Carleigh, the two most important people in my life.

viii

Contents

Abstract ...... iv

List of Tables ...... xiii

List of Figures ...... xiv

Acknowledgements ...... xvi

1. Introduction ...... 1

1.1 Kingella kingae history and classification ...... 1

1.2 Microbiology ...... 2

1.3 Epidemiology ...... 3

1.3.1 Carriage and transmission ...... 3

1.3.2 Invasive disease ...... 5

1.3.3 Day care facility attendance ...... 6

1.4 Treatment ...... 7

1.5 Prevention ...... 10

1.6 Pathogenesis and immunity ...... 11

1.7 K. kingae surface factors ...... 13

1.7.1 Type IV pili ...... 13

1.7.2 Trimeric autotransporters ...... 21

1.7.3 Polysaccharide capsules ...... 26

1.8 Thesis overview ...... 31

2. Modulation of Kingella kingae adherence to human epithelial cells by type IV pili, capsule, and a novel trimeric autotransporter ...... 32

2.1 Introduction ...... 32

2.2 Methods ...... 34

2.3 Results ...... 46

2.3.1 K. kingae encodes a novel trimeric autotransporter...... 46

2.3.2 Knh is expressed in the K. kingae outer membrane and is required for full-level adherence to human epithelial cells...... 49

2.3.3 K. kingae produces a surface-associated polysaccharide capsule...... 51

2.3.4 Type IV pili, capsule, and Knh influence K. kingae adherence to human epithelial cells...... 56

2.3.5 The type IV pilus retraction ATPase system is required for full-level adherence...... 57

2.4 Discussion ...... 61

3. Kingella kingae capsular polysaccharide characterization ...... 69

3.1 Introduction ...... 69

3.2 Methods ...... 71

3.3 Results ...... 76

3.3.1 K. kingae strain 269-492 polysaccharide contains TBA and phenol-sulfuric acid reactive sugars...... 76

3.3.2 Large scale polysaccharide purification ...... 78

3.3.3 Glycosyl composition and structural analysis of strain 269-492 polysaccharide...... 80

3.3.4 Polysaccharide-protein conjugation for antiserum development ...... 82

3.4 Discussion ...... 84

4. Differential Control of Kingella kingae Type IV Pilus Phenotypes Through the Calcium- Binding Sites of PilC1 and PilC2 ...... 88

4.1 Introduction ...... 88

4.2 Methods ...... 91

4.3 Results ...... 101

4.3.1 K. kingae PilC1 and PilC2 contain a calcium-binding site...... 101

4.3.2 Influence of K. kingae PilC1 and PilC2 Ca-binding sites on type IV pilus expression...... 108

4.3.3 The PilC1 and PilC2 Ca-binding sites influence twitching motility...... 111

4.3.4 The PilC1 Ca-binding site but not the PilC2 Ca-binding site is necessary for adherence...... 113

4.4 Discussion ...... 115

5. The Role of PilA2 in Kingella kingae type IV pilus biology ...... 121

5.1 Introduction ...... 121

5.2 Methods ...... 123

5.3 Results ...... 129

5.3.1 PilA2 is not necessary for twitching motility ...... 129

5.3.2 The role of PilA2 in natural transformation is unclear...... 130

5.3.3 PilA2 can be overexpressed when under control of the PilA1 promoter ...... 132

5.3.4 PilA2 is not present or present at levels below the detection limit in purified pili...... 134

5.3.5 PilA2 can form surface fibers when overexpressed...... 135

5.3.6 PilA2 fibers can mediate significant adherence...... 136

5.4 Discussion ...... 137

6. Conclusion and future directions ...... 142

6.1 Conclusion ...... 142

6.2 K. kingae interactions with live cells ...... 143

6.3 Define the adhesive domain(s) of Knh ...... 146

6.4 Define capsule type diversity in a diverse set K. kingae clinical isolates ...... 149

6.5 Localize the PilC proteins ...... 151

6.6 Investigate the PilC/PilT relationship ...... 153

6.7 Demonstrate adhesive properties of the PilC proteins ...... 154

6.8 Evaluate and improve the K. kingae animal model ...... 156

References ...... 158

Biography ...... 184

xii

List of Tables

Table 1: Conserved components of ABC-type capsule export systems ...... 28

Table 2: Strains and plasmids used in this study ...... 35

Table 3: Primers used in this study ...... 38

Table 4: Predicted amino acid homology between K. kingae strain 296-492 and N. meningitidis MC58 Ctr proteins.* ...... 52

Table 5: Glycosyl composition of K. kingae strain 269-492 surface polysaccharide...... 81

Table 6: Strains and plasmids used in this study ...... 92

Table 7: Primers used in this study...... 95

Table 8: Strains and plasmids used in this study...... 124

xiii

List of Figures

Figure 1: Cartoon schematic of type IV pilus system components...... 18

Figure 2: Schematic representation of the K. kingae complementation construct...... 40

Figure 3: K. kingae Knh is a member of the trimeric autotransporter protein family...... 47

Figure 4: Knh is an outer membrane protein required for full-level adherence to Chang human epithelial cells...... 50

Figure 5: K. kingae has a surface capsule...... 53

Figure 6: Mutation of ctrA does not have a polar effect on bccA...... 54

Figure 7: K. kingae polysaccharide capsule can be extracted from the bacterial surface. . 55

Figure 8: Type IV pili, capsule, and Knh influence K. kingae adherence...... 57

Figure 9: PilT/PilU retraction ATPase machinery influences K. kingae piliation, twitching motility, and adherence...... 60

Figure 10: K. kingae adherence to human epithelial cells involves sequential adhesive events...... 67

Figure 11: Biochemical detection of strain 269-492 polysaccharide sugars...... 77

Figure 12: Purity of large-scale strain 269-492 polysaccharide preparation...... 79

Figure 13: Size exclusion chromatogram of strain 269-492 large-scale polysaccharide preparation on a Superose 12 column...... 80

Figure 14: Alcian blue and Coomassie stained polysaccharide-ovalbumin conjugate. ... 83

Figure 15: Comparison of K. kingae PilC1 and PilC2 with P. aeruginosa PilY1...... 103

Figure 16: K. kingae PilC2 has a separate region resembling a calcium binding domain that does not bind calcium...... 104

Figure 17: K. kingae PilC1 and PilC2 contain calcium-binding sites...... 105

xiv

Figure 18: K. kingae PilC1 mutations do not destabilize protein global structure...... 106

Figure 19: K. kingae PilC2 mutations do not destabilize protein global structure...... 107

Figure 20: Representation of the pilC1 and pilC2 loci of K. kingae strain KK03 derivatives used in this study...... 109

Figure 21: All PilC1 and PilC2 Ca-binding site mutants express surface pili...... 110

Figure 22: The PilC1 and PilC2 Ca-binding sites influence twitching motility...... 112

Figure 23: The PilC1 but not PilC2 Ca-binding site is essential for adherence...... 114

Figure 24: Twitching motility zones of KK01 and KK03 pilA2 mutants...... 130

Figure 25: Relative transformation efficiencies of KK01 and KK03 pilA2 mutants...... 131

Figure 26: Representation of the (A) wild type and (B) ΔpilA1+ pilin loci...... 132

Figure 27: Western blot analysis of KK01 and KK03 ΔpilA1+ with GP98 and GP65...... 133

Figure 28: GP65 and GP98 Western blot analysis of purfied pili...... 135

Figure 29: Summary of transmission electron microsopy of ΔpilA1+ mutants...... 136

Figure 30: Quantitative adherence of ΔpilA1+ strains to Chang epithelial cells...... 137

Figure 31: Western blot analysis of Knh expressed in E. coli outer membranes...... 147

Figure 32: Coomassie and Western blot analysis of whole cell sonicates...... 152

Acknowledgements

I first need to thank my advisor and mentor Dr. Joseph St. Geme for providing me the opportunity to complete my graduate work in his lab. He gave me the independence to pursue my own projects and develop my own research goals while providing support and encouragement when needed. His dedication to my success and always cheerful outlook provided the ideal environment to learn the essential tools of scientific inquiry.

I would like to thank past and present members of the St. Geme lab for their help and contributions through the years. Dr. Thomas Kehl-Fie introduced me to the ways of

Kingella. I never looked back. A special thanks to Sue Grass for all of her hard work keeping the lab operating smoothly and for always having an opinion. Thanks to Kim

Starr for all of her hard work exploring Kingella encapsulation. I am thankful for my benchmate Dr. Jess McCann for her scientific acumen and mentorship, not to mention the numerous hockey conversations over the years. All of St. Geme lab members made the lab an exciting, dynamic, and fun environment.

A special thanks to our UNC collaborators Dr. Michael Johnson and Dr. Matt

Redinbo. Their input and contributions to the PilC project were invaluable. I hope I have the opportunity to collaborate with them in the future.

xvi

My committee members, Dr. Pat Seed, Dr. Meta Kuehn, Dr. Raphael Valdivia, and Dr. Soman Abraham provided valuable insight and suggestions for my work. I want to thank Dr. Pat Seed specifically for the Duke Scholars in Infectious Diseases

Program. I consider my participation in the program a highlight of my graduate student tenure. In addition, it was always great to have members of the Seed lab around for thoughtful discussions and help all around.

I would like to thank my parents for their ever-present support throughout the years. They always encouraged me to pursue my interests and excel in all aspects of life.

Lastly, I want to thank my beautiful wife Kimberly and my wonderful daughter

Carleigh for their love and support. Kimberly has always been understanding of the long hours and weekend work. Their love and commitment greatly contribute to my success.

xvii

1. Introduction

1.1 Kingella kingae history and classification

K. kingae was first isolated in 1960 from blood, joint fluid, bone exudates, and throat swabs by Elizabeth O. King, a bacteriologist at the Centers for Disease Control and Prevention. King’s initial studies suggested that this organism was a novel

Moraxella species, prompting shipment of samples to Henriksen and Bovre at the

University of Oslo, Norway for their opinions (Henriksen et al., 1968). Work by these investigators confirmed King’s initial impressions and led to a publication describing the new species, which they named Moraxella kingii in honor of King (Henriksen et al.,

1968). Additional studies in the late 1960s and 1970s established significant differences between M. kingii and other members of the Moraxella genus and resulted in reclassification as a novel genus and renaming as Kingella kingae in 1976 (Bovre et al.,

1974, Henriksen et al., 1976).

Until the 1990s, K. kingae was largely ignored as a human pathogen due to its infrequent recovery in association with disease, generally in young patients with septic arthritis, osteomyelitis, or endocarditis (La Selve et al., 1986, Raymond et al., 1986,

Toshniwal et al., 1986, Woolfrey et al., 1986, Clement et al., 1988, de Groot et al., 1988,

Kennedy et al., 1988, Noftal et al., 1988, Shelton et al., 1988). As novel culture-based isolation techniques were developed in the early 1990s (Yagupsky et al., 1995b), the frequency of reported infections due to K. kingae increased. Numerous studies using

PCR-based diagnostic techniques have shown that K. kingae is present in a significant percentage of culture-negative joint fluid and bone samples from patients with suspected osteoarticular infection (Stahelin et al., 1998, Yagupsky, 1999, Moumile et al.,

2003, Chometon et al., 2007, Cherkaoui et al., 2009, Ilharreborde et al., 2009, Ceroni et al.,

2010a, Ceroni et al., 2011). Recent evidence indicates that K. kingae is the leading cause of septic arthritis and osteomyelitis in children 6-36 months of age, more common even than Staphylococcus aureus (Chometon et al., 2007).

1.2 Microbiology

K. kingae is a gram-negative bacterium that belongs to the Neisseriaceae family and is only distantly related to other members of this family. It is a facultative anaerobic, β-hemolytic, small bacillus (0.6 to 1 µm x 1 to 3 µm in size) that appears as pairs or short chains with tapered ends and often resists decolorization, sometimes resulting in misidentification as a gram-positive organism (Henriksen et al., 1968,

Yagupsky, 2004). K. kingae produces acid from glucose and maltose but not other sugars and is oxidase-positive and catalase-, urease-, and indole-negative. Growth is supported on blood and chocolate agar but not on Krigler or MacConkey agar. Optimal growth occurs at 37°C in a 5-10% CO2-enhanced atmosphere, but few strains are truly capnophilic (Yagupsky, 2004). Colonies from growth on solid agar appear as three distinct types, namely spreading/corroding, non-spreading/non-corroding, and domed

(Yagupsky, 2004, Kehl-Fie et al., 2007, Kehl-Fie et al., 2010). The spreading/corroding

colony type produces marked indentations on the agar surface, develops a peripheral fringe around the small central raised colony, and is associated with high-density piliation. The non-spreading/non-corroding colony has a smaller fringe with a large, flat, central colony and is associated with low-density piliation. The domed colony produces no fringe and is non-piliated (Kehl-Fie et al., 2010). K. kingae is most easily differentiated from K. denitrificans, K. oralis, K. potus, and Neisseria spp. by a distinct narrow zone of β-hemolysis surrounding colonies on blood agar plates.

To date, K. kingae has never been isolated from a non-human source, suggesting that the organism is human specific.

1.3 Epidemiology

1.3.1 Carriage and transmission

K. kingae is a human commensal that colonizes the posterior pharynx asymptomatically in a subset of the population and occasionally invades the bloodstream to produce invasive disease. Using a novel vancomycin-containing medium developed in the early 1990s for isolation of K. kingae from the pharynx

(Yagupsky et al., 1995b), several studies have examined carriage rates in the healthy population (Yagupsky et al., 1995a, Yagupsky et al., 2002, Yagupsky, 2004, Yagupsky et al., 2009b, Ceroni et al., 2012). Based on studies conducted primarily in Israel and

Europe, colonization with K. kingae begins around 6 months of age and increases to a rate of 9-12% in children 12-24 months of age (Yagupsky et al., 1995a, Yagupsky et al.,

2002, Ceroni et al., 2012). The carriage rate gradually decreases in older children and adults, suggesting that immune system development in healthy individuals may be responsible for upper respiratory clearance (Yagupsky et al., 1995a, Yagupsky et al.,

2002).

To achieve a better understanding of the transmission of K. kingae strains circulating in the community, pulse-field gel electrophoresis genotyping studies have been performed (Yagupsky et al., 2002). In one study, upper respiratory isolates from a cohort of healthy individuals were compared to a set of invasive isolates over a 15-year period (Yagupsky et al., 2002). Interestingly, this analysis revealed clustering of genotypically indistinguishable strains in geographical regions and households, suggesting person-to-person transmission among family members and close contacts. In addition, many strains circulating among healthy individuals matched historical strains recovered from sites of invasive disease, indicating that certain strains may persist in the population (Yagupsky et al., 2002).

In a recent study, Yagupsky and colleagues examined paired isolates from the pharynx and bloodstream in 3 children with K. kingae bacteremia and found that the pharyngeal and blood isolates were genotypically identical in each case (Yagupsky et al.,

2009a), providing evidence that colonization of the posterior pharynx is a precursor to invasive disease. Similarly, Basmaci et al. (2012) reported isolation of K. kingae from the pharynx in children with confirmed K. kingae septic arthritis.

1.3.2 Invasive disease

The most comprehensive study to date on the epidemiology of invasive disease was a multicenter study in Israel that examined a total of 321 cases of K. kingae disease

(Dubnov-Raz et al., 2010). In this report, 296 of the 312 pediatric cases involved children less than 4 years of age, representing an annual incidence of 9.4 per 100,000 children 0-4 years of age. Of note, only 4 cases involved children less than 6 months of age. Children younger than 4 years of age diagnosed with K. kingae disease were generally otherwise healthy (Dubnov-Raz et al., 2010), while older children and adults who developed K. kingae disease usually had underlying pre-disposing conditions, including malignancy, immunosuppression, or cardiac valve pathology. Of the 321 total cases, there were 140 cases of septic arthritis, 19 cases of osteomyelitis, 7 cases of osteomyelitis with septic arthritis of the adjacent joint, and 3 cases of tenosynovitis, for a total 169 skeletal infections (53% of 321). Among the remaining cases, there were 140 cases of bacteremia without a focus (44% of 321), 4 cases of bacteremia and lower respiratory infection, and 8 cases of endocarditis (Dubnov-Raz et al., 2010), Other reports have described K. kingae as rare etiology of pericarditis (Matta et al., 2007), meningitis (Van Erps et al., 1992), peritonitis (Bofinger et al., 2007), urinary tract infection (Ramana et al., 2009), and ocular infection (Carden et al., 1991).

In Israel, invasive K. kingae disease occurs throughout the year and is most common during November and December and least common between February and

April (Dubnov-Raz et al., 2008, Dubnov-Raz et al., 2010). In contrast, the occurrence of carriage is more uniform throughout the year, suggesting that the seasonal distribution of invasive K. kingae disease cannot be readily explained by the epidemiology of pharyngeal carriage. At the time of presentation with K. kingae disease, approximately

50% of children display symptoms of upper respiratory tract infection, 15-20% have stomatitis, and 15-20% have diarrhea, raising the possibility that concomitant viral infection contributes to development of invasive disease (Yagupsky et al., 2002).

1.3.3 Day care facility attendance

A longitudinal study in Israel in the mid-1990s aimed to evaluate the transmission of K. kingae strains among day care facility attendees (Slonim et al., 1998).

Fifty isolates were collected over an 11-month period by sampling two cohorts of day care facility attendees every two weeks. These isolates were compared to a collection of

60 epidemiologically diverse respiratory and invasive isolates. Based on a variety of typing techniques, children were found to routinely carry the same strain continuously or intermittently for weeks or months. Different strains were able to periodically replace the originally detected strain. Interestingly, a single strain represented 28% of the isolates and was most prevalent over the first 6 months of the study, while a different strain represented 46% of the isolates over the remaining 5 months (Slonim et al., 1998).

Epidemiologically unrelated strains showed much greater variability in relatedness than isolates from the day care facility cohorts.

To date, three reports of day care facility outbreaks of invasive K. kingae disease have been described (Kiang et al., 2005, Yagupsky et al., 2006, Sẽna et al., 2010). In

Minnesota in 2003, 3 cases of osteomyelitis/septic arthritis in children from the same 16-

24 month classroom were diagnosed (Kiang et al., 2005). Carriage analysis of other attendees of the same facility revealed an overall colonization rate of 13% (15 of 122) and a peak colonization rate of 45% (9 of 20) in the 16-24 month classroom. All three children with invasive disease had concurrent or antecedent upper respiratory illness.

In Israel in 2005, one confirmed and two suspected cases of K. kingae skeletal infections were diagnosed in children from the same day care facility (Yagupsky et al., 2006).

Evaluation of the other children at the same facility established that 4 of 11 were colonized. Pulse-field gel electrophoresis demonstrated that the pharyngeal and invasive isolates from the source day care facility were genotypically identical while pharyngeal isolates from a control facility where genotypically distinct (Yagupsky et al.,

2006). Lastly, in North Carolina in 2007, a case of confirmed K. kingae endocarditis and potential meningitis, and a case of suspected intervertebral discitis and osteomyelitis were indentified in a two-week period in children who attended the same day care facility (Sẽna et al., 2010).

1.4 Treatment

K. kingae is typically highly susceptible to penicillins and cephalosporins, although in vitro susceptibility to oxacillin is relatively reduced and β-lactamase

production has been reported in rare isolates (Jensen et al., 1994, Kugler et al., 1999,

Yagupsky et al., 2001). In addition, K. kingae is almost always susceptible to aminoglycosides, macrolides, trimethoprim/sulfamethoxazole, tetracyclines, fluoroquinolones and chloramphenicol (Jensen et al., 1994, Kugler et al., 1999, Yagupsky et al., 2001, Yagupsky, 2004). Between 40% and 100% of isolates are resistant to clindamycin and virtually all isolates are resistant to trimethoprim and glycopeptide antibiotics, including vancomycin (Yagupsky et al., 2001, Dubnov-Raz et al., 2008).

Given the lack of controlled studies, no detailed guidelines for preferred antibiotic and length of treatment course have been developed for K. kingae infections.

Typically, treatment varies based on accepted measures for other pathogens. Empiric treatment with intravenous oxacillin/nafcillin or a second-generation or third-generation cephalosporin is appropriate for suspected skeletal infections while laboratory results are pending (Lacour et al., 1991, Dagan, 1993). In areas where methicillin-resistant

Staphylococcus aureus is common, addition of vancomycin is recommended

(Saphyakhajon et al., 2008). Once K. kingae is identified and β-lactam resistance is excluded, antibiotic therapy is commonly changed to amoxicillin or cefuroxime. The switch from intravenous to oral antibiotic therapy should be evaluated based on clinical response and assessment of acute-phase reactants such as C-reactive protein (Dagan,

1993, Lundy et al., 1998, Dodman et al., 2000).

Length of antibiotic treatment for K. kingae disease has varied based on specific type of infection. Treatment has generally varied from 2 to 3 weeks for K. kingae arthritis

(Goutzmanis et al., 1991, Dodman et al., 2000), from 3 weeks to 6 months for K. kingae osteomyelitis (Goutzmanis et al., 1991), and from 3 to 12 weeks for K. kingae spondylodiscitis (Lacour et al., 1991, La Scola et al., 1998, Garron et al., 2002). Patients with septic arthritis usually respond well to antibiotic therapy and do not require repeated joint aspirations or invasive surgical procedures (Yagupsky, 2004).

Patients who present with K. kingae bacteremia typically have a relatively benign course and are usually treated initially with an intravenous β-lactam antibiotic followed by oral therapy once the clinical condition has improved (Yagupsky et al., 1994).

Antibiotic therapy is generally continued for a total of 1-2 weeks (Claesson et al., 1985,

Yagupsky et al., 1994, Moylett et al., 2000).

Children who present with K. kingae native valve endocarditis are usually treated intravenously with a β-lactam alone or in combination with an aminoglycoside

(Claesson et al., 1985, Wells et al., 2001). Typically length of treatment is 4-7 weeks

(Odum et al., 1984, Yagupsky et al., 1994, Wells et al., 2001). Due to the potential risks of life-threatening valve damage due to endocarditis, patients who do not respond to antibiotic therapy may require early surgical intervention (Wells et al., 2001, Rotstein et al., 2010, Youssef et al., 2010).

1.5 Prevention

In the 3 reported cases of day care outbreaks, oral antibiotics were administered to contacts of the infected children to prevent further spread. The antibiotic courses varied and included either rifampin 20mg/kg twice daily for 2 days (Kiang et al., 2005) or a combination of rifampin 20mg/kg twice daily for 2 days and amoxicillin 80mg/kg/day for 2 days(Sẽna et al., 2010) or 4 days (Yagupsky et al., 2006). K. kingae carriage was monitored in colonized contacts who received antibiotic prophylaxis. In the first report,

67% of colonized contacts became culture negative following prophylaxis (Yagupsky et al., 2006). In the second report, 80% of colonized individuals became culture negative after prophylaxis (Kiang et al., 2005). In the third report, carriage rates were low and prevented analysis of colonization at the end of antibiotic treatment (Sẽna et al., 2010).

Currently there are no measures available to prevent K. kingae infection in the general population. With recent evidence that K. kingae expresses a capsular polysaccharide (Bendaoud et al., 2011), it is possible that a polysaccharide-conjugate vaccine would be a viable approach to prevention of disease, similar to the situation with Haemophilus influenzae type b, Streptococcus pneumoniae, and N. meningitidis. Further research is necessary to define the heterogeneity of K. kingae polysaccharide capsule types and to determine whether selected capsule types are associated with invasive disease and thus represent attractive vaccine antigens.

1.6 Pathogenesis and immunity

The pathogenesis of disease due to K. kingae is believed to begin with colonization of the posterior pharynx. While the bacteria typically does not cause disease in the upper respiratory tract, once it breaches the respiratory epithelium it can invade the bloodstream and disseminate. The hematogeneous route of dissemination is supported by multiple reports identifying genotypically identical strains of K. kingae from respiratory swabs and blood cultures from individual with invasive disease

(Yagupsky et al., 2009a, Basmaci et al., 2012). Patients with invasive K. kingae disease frequently present with symptoms of a viral respiratory infection, evidence of herpetic gingivostomatitis, or concomitant buccal aphthous ulcers, suggesting that viral induced damage to the respiratory mucosa facilitates K. kingae invasion of the bloodstream

(Yagupsky et al., 1994, Amir et al., 1998, Yagupsky, 2004). Beyond the role of viral co- infection, K. kingae produces a potent extracellular toxin that belongs to the RTX family and is capable of lysing epithelial, synovial, and macrophage cells (Kehl-Fie et al., 2007).

This RTX toxin may facilitate disruption of the respiratory epithelium, perhaps with an enhanced effect in the setting of viral co-infection.

In patients in whom K. kingae is able to persist in the bloodstream, the organism may produce uncomplicated bacteremia or instead disseminate to joints, bones, or the endocardium. In the case of septic arthritis, K. kingae typically infects large joints such as the hip, knee, ankle, wrist, shoulder, and elbow (Yagupsky et al., 2002, Yagupsky, 2004,

Dubnov-Raz et al., 2008, Dubnov-Raz et al., 2010). In the case of osteomyelitis, the long bones, including the femur, humerus, or tibia are most often involved (Yagupsky et al.,

2002, Yagupsky, 2004, Dubnov-Raz et al., 2008, Dubnov-Raz et al., 2010). K. kingae has also been found to be the leading etiology of hematogeneous spondylodiscitis in children less than 4 years or age (Chanal et al., 1987, Garron et al., 2002, Fuursted et al.,

2008, Ceroni et al., 2010b). Many of the key questions surrounding how K. kingae exists in the bloodstream and targets the sites of invasive skeletal disease remain unanswered.

To investigate the role of the host immune system in development of K. kingae disease in healthy individuals, Slonim et al. examined serum IgA and IgG levels against outer membrane proteins in 19 children with invasive disease (Slonim et al., 2003). As expected, children convalescing from invasive disease had significant increases in IgA and IgG levels. Further analysis has demonstrated that the age incidence of disease is inversely correlated with serum IgA and IgG levels in healthy individuals. Infants under 6 months of age have undetectable serum IgA but high serum IgG levels, suggesting that protection from invasive disease in this age group is due to maternally derived IgG (Slonim et al., 2003). Children between 6 and 18 months of age have the highest age incidence of disease and the lowest serum IgG levels (Dubnov-Raz et al.,

2010). Children 24 months of age and older have a progressively decreasing incidence of disease and progressively increasing IgG and IgA levels, suggesting that carriage or exposure to K. kingae during the first two years of life may be an immunizing event.

However, K. kingae outer membrane protein epitopes are polymorphic from strain to strain, raising the possibility that the immune response is strain specific and may not prevent re-colonization by an antigenically distinct strain (Yagupsky et al., 2005).

1.7 K. kingae surface factors

To begin to understand the molecular pathogenesis of K. kingae disease, studies were initiated in our lab to define and characterize K. kingae surface factors that influence adhesive interactions with host cells.

1.7.1 Type IV pili

Type IV pili are a class of surface fibers expressed by an array of gram-negative bacteria and at least one gram-positive bacterium (Clostridium perfringens). The functions of these fibers are numerous and include adherence, natural competence, twitching motility, social gliding motility, protease secretion, and microcolony formation. The fibers are uniform in appearance as visualized using transmission electron microscopy and are typically 5-9nm in diameter and can reach several micrometers in length. The pili are composed primarily of a single protein called pilin. Pilins share common features, including a short hydrophilic leader peptide that is cleaved by a signal peptidase, a methylated N-terminal residue, and two cysteines near the C-terminus

(Strom et al., 1993a). Structural analyses have shown that pilins share a hydrophobic N- terminal alpha helix and C-terminal globular domain architecture (Parge et al., 1995,

Forest et al., 1999, Craig et al., 2004). There are two distinct classes of pilins recognized in

type IV pilus systems. The two pilin types are classified primarily on the length of the leader sequence and overall protein size. The type IVa pilins have a short leader peptide that is less than 10 amino acids and are typically 150-160 amino acids in total length

(Marrs et al., 1985, Johnson et al., 1986, Potts et al., 1988). The type IVb pilins have an extended leader sequence up to 30 amino acids and vary in overall length from ~50 amino acids up to ~200 amino acids (Shaw et al., 1990, Kachlany et al., 2001). In addition to the characteristic pilin differences, the genes encoding type IVb pili are located at a single genetic locus (Ramer et al., 2002, Kirn et al., 2003), while genes necessary for expression of type IVa pili are typically located in a number of operons or single genes scattered throughout the genome (Alm et al., 1997, Carbonnelle et al., 2006). The bundle- forming pilus of enteropathogenic E. coli (Donnenberg et al., 1992) and toxin co- regulated pilus of Vibrio cholerae (Faast et al., 1989) are examples of type IVb pilus systems, while the type IV pilus of N. meningitidis (Potts et al., 1988), P. aeruginosa

(Johnson et al., 1986), and K. kingae (Weir et al., 1992, Kehl-Fie et al., 2008) are examples of the type IVa variety.

Type IVa pili have been extensively studied in the pathogenic Neisseria spp. and

P. aeruginosa. For the purpose of clarity, the N. meningitidis nomenclature of type IV pilus components is used here, and the P. aeruginosa nomenclature is indicated in parenthesis if it differs from N. meningitidis. The main steps in pilus biogenesis are as follows: The major pilin subunit, PilE (PilA) is translated as a pre-pilin and the

hydrophilic leader peptide is cleaved by PilD, a dedicated prepilin peptidase that additionally methylates the N-terminal amino acid (typically a phenylalanine) of the mature pilin (Nunn et al., 1990, Lauer et al., 1993, Strom et al., 1993b). The mature pilin is assembled into the growing fiber by action of the cytoplasmic type IV pilus assembly

ATPase, PilF (PilB) (Turner et al., 1993, Freitag et al., 1995). The growing fiber is secreted through the outer membrane by the PilQ oligomeric secretin (Martin et al., 1993, Drake et al., 1995). The ability of the pili to retract is catalyzed by a retraction ATPase, PilT (Alm et al., 1997, Wolfgang et al., 1998a, Merz et al., 2000). Retraction of the pili is powered by

PilT-mediated disassembly of the fiber at the inner membrane back into pilin subunits

(Merz et al., 2000). Studies have estimated that disassembly occurs at a rate of 1000-1500 subunits/second (Merz et al., 2000). The force of retraction of a single pilus has been measured at ~800-150pN using molecular tweezers experiments, making it the strongest linear molecular motor described to date (Merz et al., 2000, Maier et al., 2002, Clausen et al., 2009). The action of retraction is also necessary for proper pilus function. Retraction mutants are defective in twitching motility and natural competence and show altered adherence properties (Whitchurch et al., 1991, Wolfgang et al., 1998a, Comolli et al., 1999).

However, ΔpilT mutants express fibers and are typically hyperpiliated, as the fibers are capable of being assembled but not retracted (Whitchurch et al., 1991, Pujol et al., 1999).

In addition to the major components and biogenesis steps mentioned above, type

IV pilus systems require an array of other factors for proper expression. Specifically

focusing on the pathogenic Neisseria spp., a set of 15 genes (PilE, PilD, PilF, PilM, PilN,

PilO, PilP, PilQ, PilG, PilH, PilI, PilJ, PilK, PilC1/C2, and PilW) have been identified as essential for type IV pilus biogenesis, meaning that elimination of any of these components results in ablation or a significant reduction in pilus expression and function (Carbonnelle et al., 2006). The PilM, PilN, PilO, and PilP components, along with PilQ (encoded by pilMNOPQ operon) are thought to compose a pilus assembly subcomplex, with PilM located in the cytoplasm , PilN and PilO located in the periplasm and anchored in the inner membrane, and the lipoprotein PilP also anchored in the inner membrane (Drake et al., 1997, Ayers et al., 2009). The PilG protein is a trans-inner membrane spanning protein that has homology to the type II secretion GspF secretory machinery and may serve a role in connecting the cytoplasmic biogenesis components with those in the periplasm (Tonjum et al., 1995, Collins et al., 2007). The PilH, PilI, PilJ, and PilK proteins (encoded by the pilHIKJ operon) are pilin-like proteins that appear to be incorporated into the fiber but have an unclear role in pilus biogenesis (Carbonnelle et al., 2005, Winther-Larsen et al., 2005, Carbonnelle et al., 2006, Giltner et al., 2010). The

PilCs are pilus-associated proteins that have been identified as the adhesive components of the fiber (Rudel et al., 1995b). A detailed background on the PilC-like proteins is provided in Chapter 4. Interestingly, proper localization of the PilCs in pilus fractions is dependent on PilH-PilK, suggesting these proteins may act as a complex (Winther-

Larsen et al., 2005). PilW is also a pilin-like protein that has been localized to the fibers

and is involved in pilus-dependent bacterial aggregation (Carbonnelle et al., 2005, Szeto et al., 2011). As individual mutation of most type IV pilus genes leads to a complete loss of piliation, a recent study tried to identify factors specifically necessary for pilus assembly by introducing ΔpilT mutations into a single mutant background with the understanding that the double mutant will be non-piliated if the gene of interest is involved in pilus assembly, while the double mutant will be piliated if the gene of interest acts at a biogenesis step post-assembly (Carbonnelle et al., 2006). This study found the only gene products that appear to be essential for pilus assembly are PilE,

PilD, PilF, PilM, PilN, PilO, and PilP (Carbonnelle et al., 2006). Conversely, PilC1/C2,

PilG, PilH, PilI, PilJ, PilK, and PilW appear to be involved in the process of fiber stabilization, likely through counteraction of pilus retraction by an unknown mechanism

(Wolfgang et al., 1998b, Winther-Larsen et al., 2005, Carbonnelle et al., 2006). A set of seven type IV pilus components in the N. meningitidis, including PilT, PilT2, PilU, PilV,

PilX, ComP, and PilZ are not essential for pilus biogenesis but do modulate pilus function (Carbonnelle et al., 2005, Brown et al., 2010). Like PilT, PilT2 and PilU appear to be ATPases and modulate pilus retraction but are not essential for retraction (Park et al.,

2002, Brown et al., 2010, Eriksson et al., 2012). PilV, PilX, and ComP are pilin-like proteins that can be detected in pilus preparations, suggesting they are incorporated into the fibers (Brown et al., 2010). PilV likely influences pilus retraction, as a ΔpilV mutant forms abnormal aggregates and appears to have a hypertwitching phenotype (Brown et

al., 2010). PilX has been shown to influence bacterial aggregation and natural competence (Helaine et al., 2005), with one report indicating that inter-bacterial PilX subunits interact to counteract pilus retraction (Helaine et al., 2007). ComP is essential for natural competence (Wolfgang et al., 1999, Brown et al., 2010). PilZ is located in the cytoplasm but is necessary for proper aggregation and host cell adherence via an unknown mechanism (Brown et al., 2010). A schematic of the type IV pilus components is shown in Figure 1.

Figure 1: Cartoon schematic of type IV pilus system components.

The 15 type IV pilus biosynthesis components and their genetic organization are largely conserved in type IVa pilus systems (Pelicic, 2008); however, key differences in type IV pilus biology do exist from species to species. While the pathogenic Neisseria spp. encode two PilC-like proteins at unlinked genetic loci (Jonsson et al., 1991), P. aeruginosa encodes one full-length PilC-like protein called PilY1 that is contained in an operon with the pilHIJK homologs (Alm et al., 1996). An excellent example of the species-specific roles of a type IV pilus components is PilU. A ΔpilU N. gonorrhoeae mutant is normally piliated, hyperadherent to host cells, and non-aggregative (Park et al., 2002), a ΔpilU N. meningitidis mutant has a faster rate of adherence to host cells but does form aggregates (Brown et al., 2010), a ΔpilU P. aeruginosa mutant is hyperpiliated and does not twitch (Whitchurch et al., 1994), and a ΔpilU Dichelobacter nodosus mutant is normally piliated and does not twitch (Han et al., 2008). In addition to varied functional roles, the repertoire of non-essential pilus factors can vary. For example, the PilX minor pilin that mediates aggregation in N. meningitidis is not present in the non-aggregative P. aeruginosa type IV pilus system. These examples highlight that while the core machinery necessary for the biogenesis of type IV pili is largely conserved, individual species of bacteria have modified their type IV pilus systems, likely as a consequence of their respective environmental niches and pathogenic processes.

We have shown that the fibers expressed on the surface of K. kingae are type IV pili using genetic disruptions and mass spectrometry identification of PilA1, the K.

kingae major pilin subunit (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009). Previous work in our lab also identified many of the key type IV pilus component homologs in the K. kingae strain 269-492 genome sequence, including pilF, pilD, pilMNOPQ, pilHIJK, pilTU, and pilG (Kehl-Fie et al., 2008) (Porsch and St. Geme, unpublished data). Like in other type IV pilus systems, we have shown that PilA1 and PilF are essential for pilus biogenesis (Kehl-Fie et al., 2008). In addition, K. kingae expresses two PilC-like proteins with significant primary amino acid dissimilarity referred to as PilC1 and PilC2, which appear to play complementary roles in promoting pilus biogenesis and influencing the efficiency of adherence to respiratory epithelial and synovial cells (Kehl-Fie et al., 2008).

Examination of a series of clinical isolates of K. kingae demonstrated that the majority of pharyngeal isolates are piliated and express abundant pili while strains recovered from the blood of patients with uncomplicated bacteremia are generally piliated but typically express relatively few pili (Kehl-Fie et al., 2010). In contrast, strains recovered from joint fluid, bone aspirates, or the blood of patients with endocarditis are generally non-piliated (Kehl-Fie et al., 2010). It is possible that low-density piliation facilitates a tropism for joints, bones, and the endocardium and potentiates an inflammatory response, which then selects against piliated organisms (Kehl-Fie et al.,

2010). Consistent with this possibility, pili promote efficient adherence to cultured synovial cells (Kehl-Fie et al., 2008). Similar to type IV pili expressed by other pathogens,

K. kingae pili are regulated by the σ54 transcription factor and the PilS/PilR two-

component sensor/regulator system (Kehl-Fie et al., 2009). Mutations in the PilS sensor result in reduced density of pili, similar to the relative decrease in density of pili in isolates recovered from the bloodstream compared with the oropharynx (Kehl-Fie et al.,

2009). In contrast, mutations in the PilR response regulator result in elimination of piliation, similar to the absence of pili in isolates from joints and bones (Kehl-Fie et al.,

2009). At this point, it is unclear what environmental factors influence σ54, PilS, and

PilR activity and control density of K. kingae piliation.

Many of the key questions about the roles of type IV pili and function of type IV pili components in K. kingae remain unknown. Specifically, how type IV pili influence host cell interactions in concert with other surface factors and the significance of the sequence dissimilarity of PilC1 and PilC2 are intriguing questions.

1.7.2 Trimeric autotransporters

Gram-negative bacteria have developed an array of secretion systems for protein export across the double membrane cell envelope. Substrates targeted for secretion must cross the inner membrane, the periplasmic space containing a peptidoglycan layer, and the outer membrane. To date, seven distinct secretion systems, named type I - type

VI and the chaperone-usher pathway, have been described in gram-negative bacteria.

Among the proteins that utilize the type V secretion system are integral outer membrane proteins called autotransporters, which make up the largest group of gram-negative bacterial extracellular proteins, with over 700 members (Pallen et al., 2003). The term

'autotransporter' comes from initial studies indicating that all of the information necessary for transport across the outer membrane was contained in the nascent polypeptide (Meyer et al., 1987, Pohlner et al., 1987a, Pohlner et al., 1987b). These proteins contain an N-terminal signal sequence that targets the protein to the inner membrane Sec translocator system for transit into the periplasm and is subsequently cleaved by a signal peptidase, an internal passenger domain that is exposed to the extracellular environment and has effector function, and a C-terminal β-barrel domain to anchor the protein in the outer membrane (Pohlner et al., 1987a).

The autotransporters are further broken down into two subfamilies: (1) the classical autotransporters or type Va and (2) the trimeric autotransporters or type Vc.

The classical autotransporters, including IgA1 protease from N. gonorrhoeae (Pohlner et al., 1987a), Hap from H. influenzae (St Geme et al., 1994) and pertactin from Bordetella pertussis (Leininger et al., 1991), are translated as a single polypeptide and contain a C- terminal β-barrel domain that forms a 12-stranded pore in the outer membrane. This C- terminal domain facilitates transport of the passenger domain across the outer membrane and serves as the membrane anchor. The trimeric autotransporters, including Yersinia enterocolitica YadA (Skurnik et al., 1989, Roggenkamp et al., 2003), the

E. coli immunoglobulin binding proteins (Eibs) (Sandt et al., 2000), and H. influenzae Hia and Hsf (Surana et al., 2004) are translated as a single polypeptide, like the classical autotransporters. Following translocation to the periplasm by the Sec machinery, the

short C-terminal domain containing only 4 β-sheets requires trimerization to form the complete 12-sheet pore and facilitate secretion of the anchored passenger domain to the extracellular space (Roggenkamp et al., 2003, Surana et al., 2004).

The two-partner secretion (TPS) system (type Vb) is a related type V secretion class distinct from the autotransporters. The TPS system proteins, including the high molecular weight adhesins (HMW) in H. influenzae (Barenkamp et al., 1992) and filamentous haemagglutinin (FHA) in B. pertussis (Domenighini et al., 1990), are translated as at least two separate polypeptides. The TpsA component is the extracellular effector and the TpsB component is the β-barrel pore (Grass et al., 2000,

Jacob-Dubuisson et al., 2001). Both components are translocated separately into the periplasm by the Sec machinery. The TspA subunit is thought to be secreted through the TspB pore and subsequently associates non-covalently with the pore (Grass et al.,

2000, Jacob-Dubuisson et al., 2001).

Despite the terminology, recent evidence from an array of sources suggests that autotransporters are dependent on other periplasmic and outer membrane factors for proper outer membrane localization and for surface localization of the passenger domain. Based on multiple reports, it is clear that components of the β-barrel assembly machinery (Bam) complex are involved in classical and trimeric autotransporter biogenesis (Jain et al., 2007, Sauri et al., 2009, Rossiter et al., 2011). For example, depletion of YaeT (former name of BamA) resulted in loss of secretion of the Shigella flexnari IscA

and SepA (Jain et al., 2007), the E. coli EspP (Ieva et al., 2009), and the N. meningitidis IgAP

(Voulhoux et al., 2003) autotransporters. Recently, Spahich and St. Geme (unpublished data) demonstrated that the essential Bam components BamA and BamD are required for Hia biogenesis in H. influenzae. In addition to the Bam complex, periplasmic chaperones appear to be involved in autotransporter biogenesis (Jose et al., 1996, Purdy et al., 2007, Ruiz-Perez et al., 2009, Wagner et al., 2009). One study demonstrated that the chaperones SurA and DegP directly interact with specific motifs in the E. coli EspP autotransporter (Ruiz-Perez et al., 2009). This study also showed that deletion of surA, skp, and degP impaired EspP secretion (Ruiz-Perez et al., 2009). These and other reports highlight the dependency of autotransporter biogenesis on other factors. The molecular mechanism of autotransporter secretion remains under intense investigation.

The trimeric autotransporters are classified based on their short C-terminal domain and membrane anchor-stalk-head domain structure. In addition, this autotransporter subtype is largely associated with adhesive activity to a variety of host substrates. The most extensively studied trimeric autotransporter is Y. enterocolitica

YadA. Transmission and electron microscopy and crystal structure analysis revealed a

'lollipop' architecture of the YadA passenger domain with a long coil-coil alpha helix stalk domain, a connecting neck domain, and a distal β-roll head domain (Hoiczyk et al.,

2000). While YadA is most well known as a collagen adhesin (Schulze-Koops et al.,

1992), other functions include binding to fibronectin (Tertti et al., 1992), mucus

(Paerregaard et al., 1991), and hydrophopic surfaces (Paerregaard, 1992), serum resistance (Pilz et al., 1992), autoagglutination (Kapperud et al., 1987), and invasion

(Paerregaard, 1992). Like YadA, many trimeric autotransporters have multifunctional activities and multiple binding substrates. Other trimeric autotransporter binding substrates include immunoglobulins (E. coli Eibs (Sandt et al., 2000) and Moraxella catarrhalis Hag (Pearson et al., 2002)), respiratory epithelial cells (H. influenzae Hia and

Hsf (Barenkamp et al., 1996) and N. meningitidis NadA and NhhA (Comanducci et al.,

2002, Scarselli et al., 2006)), and complement factors (M. catarrhalis UspA (Hallstrom et al., 2011)).

Given the variety of binding targets utilized by trimeric autotransporters, it is not surprising that a variety of different domains are utilized for adhesive interactions with target molecules. In the case of YadA, the collagen binding domain consists of series of left handed parallel β-rolls forming a globular structure and a connecting neck domain

(Tahir et al., 2000, Nummelin et al., 2004). The β-roles have been termed YadA-like head domains and are common across the autotransporter family. The YadA-like head domain organization has also been implicated in collagen binding in Aggregatibacter actinomycetemcomitans EmaA (Yu et al., 2008) and Bartonella henselae BadA (Kaiser et al.,

2008). H. influenzae Hia and Hsf both lack YadA-like head domains, but research has shown that the host cell binding is mediated by two (Hia) or three (Hsf) ISneck/Trp-ring domains (Meng et al., 2008). In addition to the YadA-like head, ISneck, and Trp-ring

domains, many other structurally distinct domains are conserved throughout the autotransporter family but require additional investigation to determine their function and roles in autotransporter biology.

1.7.3 Polysaccharide capsules

A variety of microorganisms, including bacteria and fungi, express a surface- associated polysaccharide called a capsule that modulates interactions with the environment. Organisms that express these structures include gram-negative and gram- positive bacteria as well as certain fungal species. Capsules are considered a 'classical' virulence factor and have been the subject of intense investigation for decades, leading to the development of effective vaccine strategies for many pathogens.

Polysaccharide capsules are highly hydrated sugar polymers, typically with an overall negative charge, covalently linked to a phospholipid associated with bacterial outer membrane in the case of gram-negative bacteria (Gotschlich et al., 1981). The structure of the polymer is highly variable due to differences not only in the glycosyl composition, but also due to the variable glycosidic linkages connecting the individual residues. The structure diversity is most strikingly highlighted in E. coli. To date, over

80 distinct K antigens (capsule types) have been described for this species alone (Orskov et al., 1977, Whitfield, 2006). Many other gram-negative (and gram-positive) bacteria express a variety of capsule types that differ in composition and linkage. In the case of

H. influenzae, six distinct capsules have been identified and designated types a, b, c, d, e

and f (Ryan et al., 2004). The type b capsule, a repeating (ribose)-(ribitol-5-phosphate) polymer, is the primary capsule type and virulence factor associated with invasive

Haemophilus disease (Moxon et al., 1981). For N. meningitidis, 13 serogroups (capsule types) have been identified, but only five serogroups are responsible for the majority of invasive meningococcal infections (Cartwright, 1995). Four of these serogroups (B, C, Y, and W-135) contain sialic acid, with serogroup B expressing α-2→8 linked polysialic acid

(Bhattacharjee et al., 1975), serogroup C expressing α-2→9 linked polysialic acid

(Bhattacharjee et al., 1975), serogroup Y expressing a repeating α-2→6 linked glucose- sialic acid disaccharide (Bhattacharjee et al., 1976), and serogroup W-135 expressing a repeating α-2→6 linked galactose-sialic acid disaccharide (Bhattacharjee et al., 1976).

The serogroup A capsule is composed of α1→6 linked N-acetylmannosamine-1- phosphate (Bundle et al., 1974). Of particular note, the H. influenzae type b and N. meningitidis serogroups A, C, Y and W-135 capsular polysaccharides, among others, are currently used as vaccine antigens in polysaccharide and polysaccharide-protein conjugate vaccines (www.cdc.gov/vaccines).

In contrast to the structurally diverse polysaccharides and the genes necessary for their synthesis, the machinery used for export of capsule in N. meningitidis, H. influenzae, and E. coli group 2 and 3 is largely conserved. While the exact genetic arrangement of capsule biogenesis components in these organisms varies, the synthesis and export components are linked genetically and are typically located adjacent to one

another (Frosch et al., 1991). The export process utilizes an ABC-type transporter system for secretion of the capsular polymer (Whitfield et al., 1999, Whitfield, 2006). Based on studies in these organisms, it is apparent that a set of six conserved components are necessary for proper secretion. The proteins and functions of the essential export components are detailed in Table 1.

Table 1: Conserved components of ABC-type capsule export systems

ABC-type capsule export system Function N. meningitidis E. coli H. influenzae CtrA KpsD BexD Outer membrane export pore CtrB KpsE BexC Inner membrane and periplasmic spanning polysaccharide co- polymerase CtrC KpsM BexB ABC transporter transmembrane domain subunit CtrD KpsT BexA ABC transporter nucleotide- binding domain subunit LipA KpsC HcsA ? LipB KpsS HcsB ?

Using the N. meningitidis nomenclature for descriptive purposes, the CtrA-like protein serves as the capsule export outer membrane pore (Frosch et al., 1992). The CtrB-like protein is inner membrane associated and is speculated to protrude through the periplasm, establishing a connection to the CtrA-like protein (Pazzani et al., 1993). The

CtrC-like and CtrD-like proteins make up the ABC-type transporter trans-inner membrane spanning and nucleotide-binding domain component, respectively (Larue et al., 2011). The functions of the LipA-like and LipB-like proteins are unclear. An early study suggested that they are involved in lipidation of the polysaccharide (Frosch et al.,

1993), but more recent reports contradict those findings (Tzeng et al., 2005, Sukupolvi-

Petty et al., 2006). Current theory speculates that these components facilitate transport across the outer membrane by an unknown mechanism (Sukupolvi-Petty et al., 2006).

The functions associated with polysaccharide capsules are numerous and diverse depending on the species under study. Many of the properties conveyed by capsule are important for survival in the host. Specifically focusing on gram-negative bacteria that invade the bloodstream during the course of infection, the double membrane envelope is intrinsically susceptible to complement-mediated lysis. Capsular polysaccharides have been demonstrated to increase serum resistance for a variety of pathogenic bacteria, including E. coli, N. meningitidis, and H. influenzae (Moxon et al., 1981, Zwahlen et al.,

1989, Mackinnon et al., 1993, Vogel et al., 1996, Kahler et al., 1998). The primary mechanism behind this phenotype is thought to be masking of alternative complement pathway activators expressed on the outer membrane of gram-negative bacteria and the resulting reduction of C3b deposition (Vogel et al., 1997). Certain capsules containing sialic acid are able to directly bind factor H, leading to formation of iC3b, a complement inhibitory complex (Meri et al., 1990, Schneider et al., 2006). In addition to serum resistance, capsule has been shown to be important for intracellular survival in E. coli, N. meningitidis, and H. influenzae (Williams et al., 1991, Kim et al., 2003, Spinosa et al., 2007).

Prevention of opsonophagocytosis has also been attributed to capsule in multiple bacterial species, likely serving as an accessibility barrier to host antibody (Domenico et

al., 1994, Roberts, 1996). While some polysaccharide capsules are immunogenic as exemplified by their use as vaccine antigens, other polymers such as the identical polysialic acid capsule of E. coli K1 and N. meningitidis serogroup B, mimic structures present on host proteins, resulting in poor capsular-specific antibody responses (Finne et al., 1983, Finne et al., 1987).

In addition to immune evasion mechanisms, capsule expression has been demonstrated to modulate adhesive interactions with host cells. In general, epithelial cells typically have a negatively charged surface and electrostatic forces dictate repulsion of a bacteria coated in a negatively charged polysaccharide capsule. In addition, capsule is believed to mask surface adhesins on the bacterial surface, thus denying access to the host cell receptor (Roberts, 1996). There are many examples of encapsulated bacteria adhering at reduced levels to epithelial cells compared to isogenic capsule mutants in N. meningitidis (Virji et al., 1992, Virji et al., 1993), E. coli (Runnels et al., 1984), and H. influenzae (St Geme et al., 1991).

Once in the bloodstream, K. kingae must be able to evade a variety of host defense mechanisms. Analysis of the genome of K. kingae strain 269-492 reveals open reading frames with homology to the N. meningitidis ctrABCD operon and lipA and lipB genes involved in capsule export. These observations suggest that K. kingae may elaborate a polysaccharide capsule that modulates interactions with the host.

1.8 Thesis overview

Chapter 2 describes how type IV pili, a novel trimeric autotransporter named

Knh, and capsule influence K. kingae interactions with human epithelial cells. Chapter 3 delves into development of a capsular polysaccharide purification technique and subsequent glycosyl composition analysis, structural analysis, and anti-capsular antibody development. Chapter 4 examines how PilC1 and PilC2 differentially influence type IV pilus phenotypes through their calcium binding sites. Chapter 5 investigates the role of the PilA2 minor pilin in K. kingae type IV pilus biology. Finally,

Chapter 6 summarizes the future directions and implications of this work.

Acknowledgments

Select portions of Chapter Sections 1.1-1.6 and Section 1.7.3 have been published or are in press and are referenced below:

Yagupsky, P., Porsch, E.A., and St. Geme III, J.W. (2011) Kingella kingae: an emerging pathogen in young children. Pediatrics 127: 557-565.

Porsch, E.A., Starr, K.F., and St. Geme III, J.W. Kingella kingae. In Cherry, Harrison, Kaplan, Steinbach, and Hotez (ed.), Feigin and Cherry's Textbook of Pediatric Infectious Diseases, 7th ed. (In Press)

2. Modulation of Kingella kingae adherence to human epithelial cells by type IV pili, capsule, and a novel trimeric autotransporter

2.1 Introduction

Kingella kingae is a gram-negative bacterium in the family Neisseriaceae and was originally considered a rare cause of human disease. However, improvements in culture-based and molecular-based diagnostics over the past two decades have led to increased recognition of this organism as an emerging pathogen (Yagupsky et al., 1992,

Yagupsky et al., 1995b, Moumile et al., 2003, Gene et al., 2004, Yagupsky, 2004, Verdier et al., 2005, Chometon et al., 2007, Lehours et al., 2011, Yagupsky et al., 2011). K. kingae is the causative agent of a number of pediatric diseases, including septic arthritis, osteomyelitis, bacteremia, and endocarditis (Yagupsky, 2004, Yagupsky et al., 2011). A recent report indicates that K. kingae is the leading cause of osteoarticular infections in children 6-36 months of age (Chometon et al., 2007).

K. kingae initiates infection by colonizing the posterior pharynx and is a common commensal organism in young children (Yagupsky et al., 1995a, Yagupsky et al., 2009a).

The organism then enters the bloodstream and disseminates to distant sites, including joints, bones, and endocardium (Yagupsky et al., 2002, Yagupsky et al., 2009b, Yagupsky et al., 2011). A key step in colonization of the respiratory tract is adherence to the respiratory epithelium. Previous studies have demonstrated that K. kingae type IV pili are essential for adherence to human epithelial cells (Kehl-Fie et al., 2008). In particular,

genetic disruption of the gene encoding the major pilin subunit called PilA1 results in a loss of piliation and complete loss of adherence (Kehl-Fie et al., 2008, Kehl-Fie et al.,

2009).

Expression of type IV pili is phase variable, allowing for evasion of the immune system. Our prototypic parent K. kingae strain 269-492 grows on solid agar as two phenotypically stable colony types, referred to as non-spreading/non-corroding colonies and spreading/corroding colonies (Froholm et al., 1972, Kehl-Fie et al., 2007). The non- spreading/non-corroding colony variant (designated KK01) expresses low levels of type

IV pili, while the spreading/corroding colony variant (designated KK03) expresses high levels of type IV pili (Kehl-Fie et al., 2007). In a collection of clinical isolates from diverse anatomic sites, we have also identified a domed colony type associated with no pili

(Kehl-Fie et al., 2010). Further analysis of this collection revealed that the majority of respiratory and non-endocarditis blood isolates were piliated, while the majority of joint fluid, bone, and endocarditis blood isolates were non-piliated, suggesting a role for pilus phase variation in vivo (Kehl-Fie et al., 2010).

While it is clear that type IV pili are important for adherence to human epithelial cells, the fact that a significant percentage of pharyngeal isolates are non-piliated suggests that K. kingae expresses additional surface factors that modulate interactions with host cells. Accordingly, to gain a more thorough understanding of the bacterial factors that potentially influence colonization of the pharynx, we set out to identify

additional determinants of adherence to human epithelial cells. This search revealed a novel trimeric autotransporter called Knh (Kingella NhhA homolog), which is necessary for full-level K. kingae adherence to human epithelial cells. Additional studies demonstrated that K. kingae elaborates a surface-associated polysaccharide capsule that interferes with Knh-mediated adherence. Analysis of a K. kingae mutant incapable of pilus retraction suggested that capsule-mediated interference is overcome by the retraction of type IV pili, presumably displacing polysaccharide and facilitating access of

Knh to the host cell surface. This adherence mechanism allows K. kingae to adhere efficiently to host cells while remaining encapsulated and resistant to host immune clearance strategies.

2.2 Methods

Bacterial strains

Strains used in this study are listed in Table 2. K. kingae strains were stored at -

80°C in brain heart infusion broth (BHI) with 30% glycerol. E. coli strains were stored at

-80°C in Luria-Bertani (LB) broth with 15% glycerol. K. kingae strains were routinely cultured at 37°C with 5% CO2 on chocolate agar supplemented with 50 µg/ml kanamycin, 1 µg/ml erythromycin, or 2 µg/ml tetracycline, as appropriate. E. coli strains were routinely cultured at 37°C in LB broth or on LB agar supplemented with 100 µg/ml ampicillin, 50 µg/ml kanamycin, 500 µg/ml erythromycin, or 12.5 µg/ml tetracycline, as appropriate.

Table 2: Strains and plasmids used in this study

Strain Description Source 269-492 K. kingae isolate from St. Louis Children’s (Kehl-Fie et Hospital al., 2007) KK03 Spreading/corroding natural variant of 269-492 (Kehl-Fie et al., 2007) KK03pilA1 Strain KK03 with an aphA3 insertion in pilA1 This work KK03knh Strain KK03 with a tetM insertion in knh This work KK03knh/(knh) Strain KK03knh with wild type knh This work chromosomally complemented KK03ctrA Strain KK03 with an ermC insertion in ctrA This work KK03ctrA/knh Strain KK03 with an ermC insertion in ctrA and a This work tetM insertion in knh KK03pilA1/knh Strain KK03 with an aphA3 insertion in pilA1 and This work a tetM insertion in knh KK03pilA1/ctrA Strain KK03 with an aphA3 insertion in pilA1 and This work an ermC insertion in ctrA KK03pilA1/ctrA/knh Strain KK03 with an aphA3 insertion in pilA1, an This work ermC insertion in ctrA, and a tetM insertion in knh KK03pilT Strain KK03 with an aphA3 insertion in pilT This work KK03pilT/(pilTU) Strain KK03pilT with wild type pilTU This work chromosomally complemented C54 Haemophilus influenzae serotype b isolate (Pichichero et al., 1982) C54 b- Spontaneous capsule deficient derivative of strain (St Geme et C54 al., 1991) Dh5α E. coli F− 80dlacZΔM15 Δ(lacZYA-argF)U169 (Sambrook et deoR recA1 endA1 hsdR17(rK− mK+) phoA supE441 al., 1989) thi-1 gyrA96 relA1 BL21(DE3) omp8 BL21(DE3) ΔlamB ompF::Tn5 ΔompA ΔompC (Prilipov et al., 1998) Plasmid Description Source pFalcon2 Source of aphA3 kanamycin resistance gene (Hendrixson et al., 2001) pHSXtetM4 Source of tetM tetracycline resistance gene (Seifert, 1997) pIDN4 Source of ermC erythromycin resistance gene (Hamilton et al., 2001) pUC19/pilA1:aphA3 pilA1 disruption construct marked with aphA3 (Kehl-Fie et al., 2008) pTrc99a/knh:tetM knh disruption construct marked with tetM This work pUC19/ctrA:ermC ctrA disruption construct marked with ermC This work pUC19/pilT:aphA3 pilT disruption construct marked with aphA3 This work 35

pCErm Chromosomal complementation construct This work marked with ermC pCE/knh pCErm containing the knh gene This work pCE/pilTU pCErm containing the pilTU operon This work pGEX6p GST fusion expression vector GE Life Sciences pGEX::Knh55-403 pGEX6p containing Knh55-403 which expresses This work GST-Knh55-403 pASK-IBA12 E. coli ompA signal sequence and strep-tag fusion IBA expression vector pASK::Knh1695-1783 pASK-IBA12 containing Knh1695-1783 which This work expresses Strep-Knh1695-1783

Genetic disruptions

Gene disruptions were introduced into K. kingae by natural transformation (Kehl-

Fie et al., 2008). Briefly, plasmid-based disruption constructs were created in E. coli, linearized, and introduced into K. kingae strain KK03 via natural transformation and plating on appropriate antibiotic-containing media. Following recovery of transformants by plating on selective media, correct localization of the gene disruptions was confirmed by Southern blotting or PCR. The pilA1 gene was disrupted as described previously (Kehl-Fie et al., 2008). All plasmids used in this study are listed in Table 2 and all primer sequences are listed in Table 3. To disrupt the knh gene, fragments corresponding to the 5’ and 3’ regions of the gene were individually amplified from K. kingae strain 269-492 using primers knh5’ F and knh5’ R and primers knh3’ F and knh3’ R, respectively. These fragments were ligated into BamHI/SalI-digested pTrc99A

(introducing an internal deletion and a ClaI site within knh), generating pTrc99A/knh::ClaI. The tetM tetracycline-resistance cassette was amplified with flanking

ClaI sites from pHSXtetM4 using primers tetM F and tetM R and was ligated into pTrc99A/knh::ClaI, generating pTrc99A/knh::tetM. To disrupt the ctrA gene, a fragment containing most of the ctrA ORF and upstream flanking sequence was amplified from K. kingae strain 269-492 with primers ctrA F and ctrA R and was ligated into

BamHI/HindIII-digested pUC19, generating pUC19/ctrA. Subsequently, the

QuikChange XL Site-Directed Mutagenesis Kit (Agilent Technologies) was used to insert a MluI site in the ctrA ORF using primers ctrA MluI F and ctrA MluI R, generating pUC19/ctrA::MluI. The erythromycin-resistance cassette ermC was amplified with flanking MluI sites from pIDN4 using primers ermC F and ermC R and was ligated into pUC19/ctrA::MluI, generating pUC19/ctrA::ermC. To disrupt the pilT gene, a fragment containing most of the pilT ORF and upstream flanking sequence was amplified from K. kingae strain 269-492 with primers pilT F and pilT R and was ligated into BamHI/HindIII digested pUC19, generating pUC19/pilT. Subsequently, site-direct mutagenesis was used to insert a MluI site in the pilT ORF using primers pilT MluI F and pilT MluI R, generating pUC19/pilT::MluI. The kanamycin-resistance cassette aphA3 was excised from pFalcon2 by digestion with MluI and was ligated into MluI digested pUC19/pilT::MluI, generating pUC19/pilT::aphA3. To rule out PilA1 amino acid sequence variation as a source of phenotypic changes, the pilA1 gene in all strains that did not contain the pilA1 disruption was sequenced and confirmed to be 100% identical to the

KK03 pilA1 sequence.

Table 3: Primers used in this study

Name Sequence knh5' F ACGTATCGATCGATTCGTTATACACCACTCTGTATGTATGATTCAT knh5' R ACGTGGATCCGGTACACATTAAAGAGACCACATTGCCCG knh3' F ACGTATCGATCCAAATTCAACTGTGAACATGGGTGGTAAC knh3' R GATCGTCGACCGAATAATGATTTGTGGCGGAGGTGC tetM F ACGTATCGATAAGATTGTAAAATAACAAATATTGGTACATGATTACAG tetM R ACGTATCGATCCACTTGTAGTTTATAATAACTATCTCCTCCTTTACAC ctrA F ACGTGGATCCCCAGCGCAACCAAAGACTTGC ctrA R ACGTAAGCTTCTCATGTGTGGTGCGTAACCACG ctrA MluI F TATGGGGTCGGGAACGCGTCAAACGGTGGCATTGC ctrA MluI R GCAATGCCACCGTTTGACGCGTTCCCGACCCCATA ermC F ACGTACGCGTGGTTACGCTTTGGGGAAATTATGAGG ermC R ACGTACGCGTGTAATCATGGTCATAGCTGTTCGATAAGC pilT F ATGCGGATCCCAACGCGGAATGCTATTATGCAGG pilT R ATGCAAGCTTATCAACGTTTGCGCGATTACGG pilT MluI F GAATTGCCCAACGTGACGCGTTTCCGTGTCA pilT MluI R TGACACGGAAACGCGTCACGTTGGGCAATTC 5'comp F AGCTCAATTGTCTAACTTAATTTGAATTTCCCTAGTTATC 5'comp R ACGTGAATTCGCGCACAGAGAAAGAATTTCAG 3'comp F ACGTAAGCTTGGTGTGCGGAATATGTAGATTG 3'comp R ACGTAAGCTTGCATGGAAAGCTGGATTCG compErm F ACGTGAATTCGGTTACGCTTTGGGGAAATTATGAGG compErm R ACGTGGTACCGTAATCATGGTCATAGCTGTTCGATAAGC compknh F ACGTGGATCCAAAGTGGCGCGATTGTATC compknh R ACGTGGTACCTGGCAAGCTGCTCTCAAAG comppilTU F ACGTGGATCCAATCGTGATTACTCAACGAATTG comppilTU R ACGTGGTACCGTCAAAAATTTCGCCGTTGG KnhAb F ATGCGAATTCGCTGTCTCCATTGGTTCAACCTCTGG KnhAb R ATGCGTCGACTTACTTACGAGCATCATCAGCACTTGC KnhBB F ACGTGAATTCGCGAACAATTTGAACGCG KnhBB R ACGTGGATCCTTACCACTGATAACCAGCACC ftsZ RT-left CCAGAGCGAACCAAAGTCTC ftsZ RT-right AAGCTATACTCGCCCTGCTG bccA RT-left GTTGGATTGGCGAAAACAGT bccA RT-right TAAAATTTTCACGCCCTTGC

Genetic complementation

To genetically complement gene disruptions, a system for chromosomal complementation was developed, as no plasmids have been identified to date for use in

K. kingae. A site within a ~1500bp region in the K. kingae strain 269-492 genome that is devoid of ORFs, tRNA genes, and other obvious genetic elements was chosen for the complementation locus. The 5' homologous targeting region was amplified from strain

269-492 with primers 5'comp F and 5'comp R, digested with MunI and EcoRI, and ligated into EcoRI-digested pUC19, generating pC5'. Digestion and sequencing were used to confirm proper orientation of the insert resulting in destruction of the correct

EcoRI site. The 3' homologous targeting region was amplified from strain 269-492 with primers 3'comp F and 3'comp R, digested with HindIII, and ligated into HindIII- digested pC5', generating pC5'3'. Sequencing was used to confirm proper orientation of the insert. The ermC erythromycin-resistance cassette was amplified from pIDN4 with primers compErm F and compErm R, digested with EcoRI and KpnI, and ligated into

EcoRI/KpnI-digested pC5'3', generating pCErm. The resulting construct retains a portion of the pUC19 multiple cloning site (MCS) containing the restriction sites KpnI,

BamHI, XbaI, SalI, PstI, and SphI for use in cloning the complementation inserts. A schematic of the complementation construct is shown in Figure 2.

Figure 2: Schematic representation of the K. kingae complementation construct.

The pUC19 vector backbone contains the 5' and 3' homologous targeting regions to direct chromosomal insertion, the ermC erythromycin resistance marker (ErmR), and a portion of the pUC19 MCS for insertion of the desired gene(s) for complementation. After the desired gene(s) is ligated into the complementation construct, the pUC19 vector backbone is linearized with a single cutter restriction enzyme and transformed into K. kingae.

To complement the knh disruption, the knh gene and ~300bps of sequence upstream of the start codon were amplified with primers compknh F and compknh R, digested with KpnI and BamHI, and ligated into KpnI/BamHI-digested pCErm, generating pCE/knh. The construct was then linearized with NdeI and transformed into

KK03knh. To complement the pilT disruption, the pilTU operon and ~300bps of sequence upstream from the start codon were amplified with primers comppilTU F and comppilTU R, digested with KpnI and BamHI, and ligated into KpnI/BamHI-digested

pCErm, generating pCE/pilTU. Due to the fact that the K. kingae pilT mutant is not competent for transformation, pCE/pilTU was linearized with NdeI and transformed into strain KK03 first, followed by targeted disruption of the WT pilT gene.

Generation of anti-Knh polyclonal antiserum

In order to generate an antiserum reactive with Knh, a 1.3Kb fragment corresponding to the 5’ coding region of knh (encoding amino acids 55-403, lacking the predicted signal sequence) was amplified with primers KnhAb F and KnhAb R and was ligated into EcoRI/SalI-digested pGEX6p (GE Healthcare Life Sciences), creating pGEX::Knh55-403, which encodes a GST fusion protein. The GST-Knh55-403 fusion protein was affinity purified on glutathione-agarose and treated with PreScission Protease (GE

Healthcare Life Sciences) to release Knh55-403 from the GST fusion. The resulting 42kDa protein was resolved on a 10% SDS-PAGE gel, stained with Coomassie blue, excised, and then injected into a guinea pig, generating antiserum GP97.

Protein analysis

A 264-bp fragment of Knh encoding amino acids 1695-1783 encompassing the predicted β-barrel membrane anchor was PCR amplified with primers KnhBB F and

KnhBB R, digested with EcoRI and BamHI, and ligated into EcoRI/BamHI digested pASK-IBA12 (IBA BioTAGnologies). The resulting construct designated pASK::Knh1695-

1783 was transformed into E. coli BL21omp8 and induced with 2 µg/ml anhydrous tetracycline for 3 hours. Outer membranes were isolated based on sarkosyl insolubility,

and an aliquot was treated with 95% formic acid overnight as previously described (St

Geme et al., 1998). Samples with or without formic acid treatment were separated on a

15% SDS-PAGE gel and transferred to nitrocellulose. Western blot analysis of recombinant Strep-tagged KnhBB was performed using the α-Strep monoclonal antibody (IBA BioTAGnologies) according to the manufacturer's recommendations.

Western blot analysis of Knh in K. kingae was performed as described previously

(Sheets et al., 2008) with Knh-specific antiserum. Briefly, outer membrane fractions were isolated based on sarkosyl insolubility and were treated overnight with 95% formic acid.

Samples were resolved on a 7.5% SDS-PAGE gel, transferred to nitrocellulose, and probed with antiserum GP97 against Knh and a secondary goat anti-guinea pig horseradish peroxidase-conjugated antibody. Western blots were developed using a chemiluminescent peroxidase substrate.

Pilus Preparations

Derivatives of K. kingae strain KK03 were incubated on chocolate agar for 17-18 hrs, and growth was suspended in 1.5mL 50mM Tris 150mM NaCl, pH 8.0 to an OD600 of

1.0, vortexed at full speed for 1min, and centrifuged at 21,000 x g for 2 min to pellet the bacteria. 1.25mL of the bacteria-free supernatant was subjected to 20% ammonium sulfate precipitation on ice for 2 hrs. Precipitated pili were collected via centrifugation at

21,000 x g for 5 min and resuspended in 1x SDS-PAGE loading buffer. Aliquots were separated on 15% SDS-PAGE gels and stained with Coomassie blue. The observed

protein band was confirmed to be the major pilin subunit, PilA1, by Western blot analysis with antiserum GP65 as described previously (Kehl-Fie et al., 2008).

Capsule extraction

Derivatives of K. kingae strain KK03 and Haemophilus influenzae type b strain C54 were grown for 17-18 hrs on chocolate agar, suspended in 1ml PBS to an OD600 of 0.8, centrifuged for 2 min at 10,000 x g, and resuspended in 0.5ml PBS. Suspensions were incubated at 55°C for 30 min to extract capsular material. Bacteria were pelleted, and supernatants were concentrated 10-fold in an Amicon Ultra 10,000 MWCO centrifuge filter. Capsule extracts were separated on a 7.5% SDS-PAGE gel and stained with the cationic dye alcian blue (Sigma) (0.125% alcian blue in 40% ethanol/5% acetic acid) for

2hrs, then destained overnight in 40% ethanol/5% acetic acid.

Cationic ferritin staining

Derivatives of K. kingae strain KK03 were incubated on chocolate agar for 17-18 hrs, and growth was suspended in 3 ml 0.1M sodium cacodylate, pH 7.0 (cacodylate buffer) to an OD600 of 0.8. Glutaraldehyde was added to a final concentration of 5%, and samples were incubated for 20 min at ambient temperature. Following centrifugation at

750 x g for 15 min, pellets were gently resuspended in 1 ml cacodylate buffer, polycationic ferritin (Sigma) was added to a final concentration of 1 mg/ml, and samples were incubated for 30 min at ambient temperature and then centrifuged at 750 x g for

15min. The pellets were washed once with 1 ml cacodylate buffer and resuspended in

cacodylate buffer with 5% glutaraldehyde and were incubated for 2 hrs at ambient temperature and then overnight at 4°C and processed for transmission electron microscopy analysis.

Transmission electron microscopy

To assess surface capsule expression, cationic ferritin stained bacteria were washed 3x with cacodylate buffer and embedded in 2% low melting point agarose followed by fixation in cacodylate buffer with 4% glutaraldehyde and 7.5% sucrose. The pellets were post-fixed in 1% Osmium tetroxide in cacodylate buffer and en-bloc stained with 0.5% uranyl acetate, dehydrated with a graded series of ethanol, and embedded into epoxy resin. Blocks were thin-section (60-90nm), placed on copper-rhodium 200 mesh grids, and examined using a Philips CM-12 electron microscope.

Twitching motility assays

Twitching motility was assessed by a modified agar plate stab assay originally developed for P. aeruginosa (Alm et al., 1995). Briefly, derivatives of K. kingae strain KK03 were incubated on chocolate agar for 17-18 hours, and growth was resuspended in 1x

PBS to an OD600 of 1.0. One µl of the bacterial suspensions was stab inoculated to the bottom of tissue culture treated 100mm plates containing chocolate agar and incubated at 37°C with 5% CO2 for 2 days. Twitching motility competent strains spread from the stab inoculation site at the plate/agar interface. The agar was carefully peeled away and

the plate was air-dried and stained with crystal violet to visualize the twitching motility zones.

Transcript levels

Quantitative real-time PCR were preformed as described previously (Kehl-Fie et al., 2009). Briefly, RNA was extracted using TRIreagent (Sigma) and the RNeasy minikit by following the lipid-rich tissue protocol (Qiagen). Residual DNA was digested with

RQ1 DNase (Fisher Scientific), which was inactivated prior to generating cDNA with random hexamers and Superscript II (Invitrogen). Primer sets used in these experiments are listed in Table S1.

Eukaryotic cell lines

Chang cells (Wong-Kilbourne derivative (D) of Chang conjunctiva, HeLa origin;

ATCC CCL-20.2) were cultivated at 37°C with 5% CO2 in media as described previously

(Kehl-fie et al., 2007).

Adherence assays

Quantitative adherence assays were performed as described previously (Kehl-Fie et al., 2008). Briefly, confluent and fixed cell monolayers in 24-well plates were inoculated with approximately 6.5x106 CFU of bacteria, and plates were centrifuged at

165 x g for 5 min and then incubated for 25 min at 37°C in 5% CO2. Monolayers were washed 4 times with PBS to remove non-adherent bacteria and were then treated with

1x trypsin-EDTA (Sigma) for 20 min at 37°C to release adherent bacteria. Appropriate

dilutions were plated on chocolate agar, and percent adherence was calculated by dividing the number of adherent CFU by the number of inoculated CFU. All experiments were performed in triplicate. Statistical analysis was performed using the unpaired t test to compare adherence levels between two strains as noted in the text.

2.3 Results

2.3.1 K. kingae encodes a novel trimeric autotransporter.

K. kingae type IV pili are essential for adherence to human epithelial cells (Kehl-

Fie et al., 2008). To identify other factors that might contribute to K. kingae adherence, we examined the draft genome sequence of strain 269-492 for putative adhesive factors by searching for homologs of genes encoding adhesins in other Neisseriaceae family members, including the Neisseria meningitidis Opa, Opc, NadA, and NhhA proteins.

This analysis revealed a 5,352 bp open reading frame (ORF) encoding a protein that has

C-terminal sequence homology to the N. meningitidis NhhA trimeric autotransporter

(Accession No. AAK09243.1) and was named Knh (Kingella NhhA homolog) (Peak et al.,

2000, Scarselli et al., 2006). Trimeric autotransporters are a family of adhesive outer membrane proteins largely classified based on amino acid sequence in the conserved C- terminal β-barrel membrane anchor domain. Sequence alignment revealed strong membrane anchor homology between Knh and NhhA as well as the Haemophilus influenzae Hia protein (Accession No. AAC43721.2) (Figure 3A). Additional analysis with SignalP 3.0 (http://www.cbs.dtu.dk/services/SignalP/) revealed a predicted N-

terminal signal peptide corresponding to amino acids 1-54 and a predicted signal peptidase cleavage site between amino acids 54 and 55, consistent with the possibility that Knh is secreted and surface localized. Domain annotation of Knh was accomplished with the Domain Annotation of Trimeric Autotransporters program

(daTAA) (http://toolkit.tuebingen.mpg.de/dataa) and demonstrated an ISneck2 domain, which has similarity with the ISneck1 domain architecture that has been associated with adhesive activity in the H. influenzae Hia and Hsf proteins (Meng et al., 2008), a series of

19 YadA-like head domains predicted to form a β-roll based on the Yersinia enterocolitica YadA head domain crystal structure (Tahir et al., 2000, Nummelin et al.,

2004), and a series of eight Trp-ring domains spaced between the YadA-like head domain region and the membrane anchor (Figure 3B).

Figure 3: K. kingae Knh is a member of the trimeric autotransporter protein family.

(A) Knh C-terminal membrane anchor amino acid sequence alignment with Neisseria meningitidis NhhA and Haemophilus influenzae Hia. (B) Annotation of Knh, highlighting the presence of numerous protein domains, including a signal peptide, a membrane anchor, YadA-like head domains, Trp ring domains, and an ISneck2 47

Figure 3 (continued) domain. (C) Strep-tag Western Blot analysis of outer membrane preparations and formic acid denatured outer membrane preparations following induction of vector control pASK-IBA12 or pASK::Knh1695-1783 which expresses the Knh β-barrel domain fused to a Strep-tag. Molecular weights (MW) are expressed in kiloDaltons.

To confirm that Knh is a trimeric autotransporter, the ability of the C-terminal β- barrel domain to trimerize in the outer membrane was examined. A 264-nucleotide fragment of knh encoding amino acids 1695-1783 corresponding to the predicted β-barrel membrane anchor was cloned in-frame into pASK-IBA12 (IBA BioTAGnology), which contains an N-terminal E. coli OmpA signal sequence fused to a Strep-tag. The resulting construct designated pASK::Knh1695-1783 was transformed into E. coli BL21 omp8 for induction. After secretion through the inner membrane and cleavage of the OmpA signal sequence, the recombinant strep-tagged Knh1695-1783 protein has a predicted monomeric molecular mass of 12kDa. As is shown in Figure 3C, induction of pASK::Knh1695-1783 resulted in outer membrane localization of a 36kDa band detected with an α-Strep monoclonal antibody. As has previously been reported with trimeric autotransporter proteins (St Geme et al., 1998), the trimer was extremely stable under standard boiling and SDS-PAGE denaturing conditions and required an additional formic acid denaturation step to dissociate into monomers (Figure 3C). These findings establish that the Knh β-barrel is sufficient for trimerization in the outer membrane and confirm that Knh is a member of the growing class of trimeric autotransporter proteins.

2.3.2 Knh is expressed in the K. kingae outer membrane and is required for full-level adherence to human epithelial cells.

To determine if Knh is expressed and present in the outer membrane in K. kingae strain KK03, sarkosyl insoluble membrane fractions were isolated and examined by

Western blotting with anti-Knh serum. Strain KK03 is a naturally occurring derivative of our prototype K. kingae strain 269-492 that expresses stable levels of type IV pili and was used as the wild type strain throughout this study. As a negative control, a Knh mutant was created by insertionally inactivating knh in strain KK03. Under standard

SDS-PAGE denaturing conditions, a high molecular mass band was evident in outer membrane preparations in strain KK03 and the non-piliated, non-adherent pilA1 mutant but not in the knh mutant (Figure 4A) While the high molecular mass of this reactive band precludes accurate size estimation, it is consistent with the trimer form of Knh, which has a predicted molecular mass of 540 kDa. Formic acid treatment of outer membranes resulted in elimination of the high molecular mass band and detection of a band corresponding to the predicted monomeric molecular mass (180 kDa) of Knh in strains KK03 and KK03pilA1 but not in strain KK03knh (Figure 4B). To complement the knh mutation, the knh ORF with the native promoter region was recombined into a separate locus in the genome in the KK03knh strain background As shown in Figs. 2A and 2B, outer membrane localization of Knh was restored in the complemented strain

KK03knh/(knh). Semi-quantitative extracellular pilus preparation analysis revealed no

difference in piliation levels between strains KK03, KK03knh, and KK03knh/(knh) using strain KK03pilA1 as a negative control (data not shown).

To determine if Knh is involved in mediating K. kingae adherence to host cells, we examined adherence by strains KK03, KK03pilA1, KK03knh, and KK03knh/(knh) in assays with Chang human epithelial cells. As shown in Figure 4C, elimination of Knh resulted in a significant reduction in the level of adherence compared to strain KK03 (P <

0.05), and complementation of the knh mutation almost completely restored adherence.

Unlike the situation when type IV pili were eliminated by disruption of the pilA1 gene encoding the major pilin subunit, the disruption of knh did not completely abrogate adherence (P < 0.05 for KK03pilA1 vs. KK03knh), suggesting that K. kingae type IV pili mediate low-level adherence and require knh for full-level adherence.

Figure 4: Knh is an outer membrane protein required for full-level adherence to Chang human epithelial cells.

(A) Western blot analysis of outer membrane preparations from K. kingae strains KK03 (WT), KK03pilA1, KK03knh, and KK03knh/(knh) under standard SDS-PAGE 50

Figure 4 (continued) reducing conditions. (B) Western blot analysis of formic acid-treated outer membrane preparations from K. kingae strains KK03 (WT), KK03pilA1, KK03knh, and KK03knh/(knh) under standard SDS-PAGE reducing conditions. Molecular weights (MW) are expressed in kiloDaltons. (C) Adherence of K. kingae stains KK03 (WT), KK03pilA1, KK03knh, and KK03knh/(knh) to Chang human epithelial cells. Adherence assays were conducted in triplicate. Error bars indicate standard error of the mean. Statistical analyses were performed using the unpaired t-test comparing adherence levels between two strains as follows: (*) P < 0.05 for KK03 vs. KK03knh; P < 0.05 for KK03pilA1 vs. KK03knh; and P < 0.05 for KK03knh vs. KK03knh/(knh).

Considered together, these data demonstrate that Knh is required for full-level adherence of K. kingae to Chang human epithelial cells and that Knh cannot mediate appreciable adherence in the absence of type IV pili.

2.3.3 K. kingae produces a surface-associated polysaccharide capsule.

The observation that type IV pili are required for Knh-mediated adherence suggested that K. kingae may express another surface factor that blocks Knh adhesive activity. Examination of K. kingae strain KK03 revealed a mucoid colony phenotype similar to that seen with other bacterial species that produce a polysaccharide capsule

(Figure 5A). To explore the possibility that K. kingae expresses a polysaccharide capsule, the draft genome sequence was searched for homologs to genes involved in capsule synthesis and export in N. meningitidis, a well-studied Neisseriaceae family member that contains the siaABCD, ctrABCD, ctrE/lipA, and ctrF/lipB capsule-associated genes. As summarized in Table 4, an intact locus with homology to the ABC-type capsule export

operon ctrABCD of N. meningitidis MC58 and unlinked genes with homology to the ctrE/lipA and ctrF/lipB genes were identified in the K. kingae genome.

Table 4: Predicted amino acid homology between K. kingae strain 296-492 and N. meningitidis MC58 Ctr proteins.*

Name %Identity / %Similarity CtrA 50 / 70 CtrB 57 / 74 CtrC 74 / 86 CtrD 81 / 92 CtrE/LipA 56 / 69 CtrF/LipB 48 / 63

*Accession numbers of N. meningitidis MC58 amino acid sequences are as follows: AAF40538.1 for CtrA, AAF40539.1 for CtrB, AAF40540.1 for CtrC, AAF40541.1 for CtrD, AAF40546.1 for CtrE/LipA, and AAF40547.1 for CtrF/LipB.

To determine if the ctrABCD operon plays a role in K. kingae encapsulation, ctrA was insertionally inactivated. As shown in Figure 5B, examination of KK03ctrA revealed non-mucoid colonies, in marked contrast to the colonies of the parent KK03 strain. To confirm that K. kingae elaborates a polysaccharide capsule, strains KK03 and KK03ctrA were stained with cationic ferritin and then examined by thin section transmission electron microscopy (TEM). Due to the highly anionic nature of most bacterial polysaccharide capsules, cationic ferritin typically binds to the capsular material, producing an electron dense rim on the bacterial surface when visualized by thin section

TEM. A thick layer of electron density was evident on the surface of WT strain KK03

(Figure 5C) and was absent in the mutant strain KK03ctrA (Figure 5D). Attempts to complement the mutation in ctrA and the likely polar effect on the ctrBCD genes were

unsuccessful, presumably reflecting the inability of E. coli to tolerate expression of the outer membrane protein encoded by ctrA in the complementation construct. However, qRT-PCR analysis revealed that transcription of the ORF downstream of the ctrABCD operon was unaffected by the ctrA insertion mutation (Figure 6), supporting the conclusion that the phenotypes observed as a result of the ctrA mutation are due to disruption of the ctrABCD operon.

Figure 5: K. kingae has a surface capsule.

(A,B) Colony phenotype of K. kingae strain KK03 (A) and strain KK03ctrA (B). (C,D) Thin section transmission electron microscopy analysis of cationic ferritin-stained K. kingae strain KK03 (C) and KK03ctrA (D). Scale bar = 500nm.

Figure 6: Mutation of ctrA does not have a polar effect on bccA.

(A) Diagram of the WT ctrABCD locus in K. kingae with the downstream gene, bccA. The ctrA gene encodes a predicted outer membrane polysaccharide export pore; ctrB encodes a predicted inner membrane and periplasmic spanning polysaccharide co- polymerase; ctrC and ctrD encode a predicted ATP-dependent (ABC-type) transport cassette. (B) qRT-PCR analysis of bccA, encoding a predicted BCCT family betaine/carnitine/choline transporter in strain KK03ctrA compared to KK03. To normalize RNA/cDNA levels, ftsZ was used as the internal control. Error bars represent standard deviation of three independent experiments.

As a complementary approach to demonstrate the presence of a capsule, surface material was released from bacterial suspensions by heat extraction at 55°C and was resolved by SDS-PAGE and stained with alcian blue (Karlyshev et al., 2001). As shown in Figure 7, high molecular mass alcian blue reactive material was present in strain KK03 but absent in strain KK03ctrA. H. influenzae strain C54, which expresses type b polysaccharide capsule, and H. influenzae strain C54b-, an isogenic mutant that is nonencapsulated, were used as controls. The alcian blue reactive material did not stain with

Coomassie blue and was unaffected by Proteinase K treatment (data not shown), confirming that the extracted material is not proteinaceous.

Together, these data indicate that K. kingae strain KK03 expresses a surface- associated polysaccharide capsule that requires the predicted ctrABCD capsule export operon for surface localization.

Figure 7: K. kingae polysaccharide capsule can be extracted from the bacterial surface.

Capsular material was extracted from strains KK03 and KK03ctrA with heat followed by separation on 7.5% SDS-PAGE gel and staining with the cationic dye, alcian blue. Encapsulated Haemophilus influenzae type b strain C54 and an isogenic nonencapsulated mutant, C54 b-, are included as controls.

2.3.4 Type IV pili, capsule, and Knh influence K. kingae adherence to human epithelial cells.

To further explore the interrelationship between type IV pili, Knh, and capsule in terms of their effect on bacterial adherence, we generated single, double, and triple mutations in pilA1, knh, and ctrA. Extracellular pilus preparations revealed no change in pilus expression levels in mutant strains with an intact pilA1 gene (data not shown). As shown in Figure 8, insertional inactivation of ctrA in strain KK03 resulted in a slight decrease in adherence that was not statistically significant. Adherence by the

KK03ctrA/knh double mutant mimicked the levels seen in the knh single mutant, indicating that encapsulation does not influence type IV pilus-mediated adherence in the absence of Knh. Interestingly, while the pilA1 mutant was non-adherent, simultaneous disruption of pilA1 and ctrA in the KK03 background restored high-level adherence (P < 0.05 for pilA1/ctrA mutant vs. pilA1 mutant), indicating that encapsulation inhibits pilus-independent Knh-mediated adherence. The pilA1/ctrA/knh triple mutant was non-adherent, providing strong evidence that Knh is an adhesin and is responsible for K. kingae pilus-independent adherence observed in the pilA1/ctrA double mutant (P < 0.05 for pilA1/ctrA/knh mutant vs. pilA1/ctrA mutant).

Figure 8: Type IV pili, capsule, and Knh influence K. kingae adherence.

Adherence to Chang human epithelial cells by K. kingae KK03 (WT), pilA1, knh, ctrA, pilA1/knh, ctrA/knh, pilA1/ctrA, and pilA1/ctrA/knh. Adherence assays were conducted in triplicate. Error bars indicate standard error of the mean. Statistical analyses were performed using the unpaired t-test comparing adherence levels between two strains as follows: (*) P < 0.05 for pilA1/ctrA vs. pilA1 and for pilA1/ctrA/knh vs. pilA1/ctrA.

2.3.5 The type IV pilus retraction ATPase system is required for full- level adherence.

The observation that capsule interferes with Knh-mediated adherence in the absence of pili but not in the presence of pili raised the possibility that pilus-mediated adherence draws the organism closer to the host cell surface, perhaps via pilus retraction. In accord with this possibility, K. kingae encodes a homolog of the PilT/PilU

retraction ATPase machinery that has been shown to be essential for pilus retraction in several well-studied type IV pilus-expressing bacteria, including N. meningitidis, N. gonorrhoeae, Pseudomonas aeruginosa, and Myxococcus xanthus (Whitchurch et al., 1991,

Wolfgang et al., 1998a, Wolfgang et al., 1998b, Pujol et al., 1999, Jakovljevic et al., 2008).

To determine if pilus retraction is necessary for full-level K. kingae adherence, pilT was insertionally inactivated in strain KK03. As the insertion in pilT likely had a polar effect on pilU, the pilTU operon with the native promoter was recombined into the chromosome to achieve complementation. Examination of KK03pilT revealed increased levels of extracellular pili as assessed by semi-quantitative pilus preparations (Figure

9A) but lacked twitching motility as assessed by a modification of the P. aeruginosa agar stab assay (Figure 9B) consistent with the hypothesis that the PilT/PilU retraction

ATPase machinery is necessary for K. kingae pilus retraction (Alm et al., 1995). As shown in Figure 9C, elimination of PilT/PilU resulted in adherence that was decreased compared to that of the parent strain KK03 (P < 0.05) but equal to that of the knh mutant.

Complementation of the pilTU operon in the pilT mutant background resulted in reduction of extracellular pilus expression to the WT levels and restoration of twitching motility and adherence (Figure 9A-C). The observation that the knh and pilT mutants have similar adherence levels suggests that both retractable and non-retractable pili mediate a basal level of adherence and that the pilT mutant is unable to utilize Knh as an

adhesin. These data support the hypothesis that retraction of type IV pili is necessary to overcome the inhibitory influence of capsule on Knh-mediated adherence.

Figure 9: PilT/PilU retraction ATPase machinery influences K. kingae piliation, twitching motility, and adherence.

K. kingae strains KK03 (WT), pilA1, knh, pilT, and pilT/(pilTU) were examined for extracellular pilus expression by quantitative pilus preparation (A), twitching motility by a modified agar stab assay (scale bar = 1cm) (B), and adherence to Chang human epithelial cells (C). Adherence assays were conducted in triplicate. Error bars indicate standard error of the mean. Statistical analyses were performed using the unpaired t-test comparing adherence levels between two strains as follows: (*) P < 0.05 for WT vs. pilT and for pilT vs. pilT/(pilTU).

2.4 Discussion

The pathogenesis of K. kingae disease is thought to begin with colonization of the upper respiratory tract. Previous work established that type IV pili promote K. kingae adherence to human epithelial cells (Kehl-Fie et al., 2008). In this study, we identified a novel K. kingae trimeric autotransporter protein called Knh that also mediates adherence to human epithelial cells. Further analysis revealed that K. kingae expresses a polysaccharide capsule that interferes with Knh adhesive activity when type IV pili are absent. We observed that type IV pili are important to overcome the polysaccharide capsule and facilitate Knh-mediated adherence in the presence of the PilT/PilU ATPase retraction machinery, presumably as a result of pilus retraction.

Trimeric autotransporters are integral outer membrane proteins and appear to be universally associated with adhesive activity (Cotter et al., 2005, Lyskowski et al., 2011).

In our study, K. kingae derivatives expressing the trimeric autotransporter, Knh, but lacking type IV pili and capsule were adherent, while those lacking Knh, type IV pili, and capsule were non-adherent, providing strong evidence that Knh is an adhesin.

Analysis of the Knh amino acid sequence revealed the presence of conserved domains possibly involved in adhesive activity. The N-terminal region of Knh is predicted to encode three clusters of β-roll degenerate repeats termed YadA-like head domains based on characterization of the Y. enterocolitica YadA trimeric autotransporter adhesin.

Studies of YadA have demonstrated that the head domain is responsible for YadA- 61

mediated adhesive activity (Tamm et al., 1993, Tahir et al., 2000), raising the possibility that this domain may be involved in Knh adhesive function as well. In addition, Knh has an ISneck2 domain, which is similar to the ISneck1 domain identified in the adhesive regions of the H. influenzae Hia and Hsf adhesins (Meng et al., 2008). It is interesting to speculate regarding the location of the adhesive domain(s) of Knh relative to the outer membrane and the presence of capsule. We have clearly demonstrated that a non-piliated K. kingae mutant that expresses Knh and capsule is non-adherent. One possible explanation for this phenotype is that the adhesive domain of Knh is buried within the capsule and is thus not accessible to the host cell receptor. The ISneck2 domain is proximal to the membrane anchor, suggesting that it is likely located close to the bacterial surface and may be masked by surface structures, including the polysaccharide capsule. The YadA-like head domains are located distally, perhaps decreasing the likelihood that they are masked by bacterial surface structures.

Encapsulation of bacterial pathogens has been shown to confer important phenotypes for survival in the host, including anti-phagocytic activity and serum resistance. In this study, we report that K. kingae expresses a surface-associated polysaccharide capsule. According to current understanding, the pathogenesis of K. kingae disease begins with colonization of the upper respiratory tract followed by hematogenous dissemination to joints, bones, or endocardium (Yagupsky et al., 2011).

This pathogenic sequence is supported by the finding that genotypically identical K.

kingae strains can be isolated from the pharynx and bloodstream of patients with systemic disease (Yagupsky et al., 2009a, Basmaci et al., 2012). We speculate that encapsulation is important for K. kingae to cause invasive disease by promoting intravascular survival, similar to observations with N. meningitidis, H. influenzae, and other encapsulated pathogens (Moxon et al., 1981, Zwahlen et al., 1989, Mackinnon et al.,

1993, Vogel et al., 1996, Kahler et al., 1998). Current studies are underway to investigate the role of the K. kingae capsule in serum resistance and anti-phagocytic activity. In addition to expressing a surface-associated polysaccharide capsule, K. kingae produces a secreted exopolysaccharide galactan that has broad anti-biofilm activity (Bendaoud et al.,

2011). At this time, the relationship between the capsular polysaccharide and the galactan is unclear.

This study established that surface expression of the K. kingae capsule requires a predicted ABC-type transporter system with homology to CtrABCD in N. meningitidis.

We also identified homologs of the N. meningitidis CtrE/LipA and CtrF/LipB proteins

(Frosch et al., 1993, Tzeng et al., 2005). Both of these proteins also have homologs in the

E. coli group 2 and group 3 capsule biosynthesis systems (KpsC and KpsS) and in encapsulated H. influenzae (HcsA and HcsB) and appear to play a role in export of capsule polymers to the bacterial surface (Bliss et al., 1996, Sukupolvi-Petty et al., 2006).

Interestingly, the genetic organization of the polysaccharide synthesis machinery in K. kingae differs from that of E. coli group 2 and group 3, H. influenzae, and N. meningitidis.

In N. meningitidis, a capsule synthesis operon (which differs in gene composition depending on the specific capsule serotype) is located immediately upstream of the ctrABCD export operon and is divergently transcribed (Frosch et al., 1991, Swartley et al.,

1996). In E. coli group 2, the synthesis operon is located immediately between the export and assembly operons, and in H. influenzae the synthesis operon is adjacent to the export operon (Kroll et al., 1989, Kroll et al., 1990, Frosch et al., 1991, Van Eldere et al., 1995,

Swartley et al., 1996, Whitfield et al., 1999, Whitfield, 2006). We have been unable to locate a putative capsule synthesis operon adjacent to ctrABCD or at any other locus in our draft K. kingae genome sequence, suggesting that K. kingae utilizes a unique genetic mechanism for synthesis of capsular polysaccharide in a system that utilizes the ABC- type transporter system for export.

We first suspected that capsule may be obstructing the ability of Knh to mediate adherence when we found that a mutant KK03 strain expressing capsule and Knh, but lacking pili, was non-adherent. In considering this possibility, elaboration of a polysaccharide capsule has been shown to interfere with adhesive interactions with the host in several other pathogens. For example, capsule interferes with adherence to human epithelial cells by H. influenzae type b (St Geme et al., 1991), with adherence to gastrointestinal epithelium by enterotoxigenic E. coli (Runnels et al., 1984), and with N. meningitidis Opa- and Opc-mediated adherence to epithelial and endothelial cells (Virji et al., 1992, Virji et al., 1993). It is interesting to compare the K. kingae adherence data

presented here with the adherence mechanism of N. meningitidis, a related encapsulated member of the Neisseriaceae family. N. meningitidis uses a two step adherence process that starts with an initial type IV pili-mediated interaction associated with microcolony formation on the host cell (Pujol et al., 1997). Type IV pili are the key surface factor for adherence of encapsulated N. meningitidis to the epithelium and endothelium (Virji et al.,

1991, Virji et al., 1992a). Following the initial adhesive interaction, type IV pili and capsule are downregulated during the transition to the second stage of host cell interaction termed intimate adherence (Pujol et al., 1997, Deghmane et al., 2000,

Deghmane et al., 2002). The shift from initial adherence to intimate adherence occurs over the course of hours, leading to host cell cytoskeletal rearrangements and cortical plaque formation under the intimately adherent bacteria (Pujol et al., 1997, Merz et al.,

1999, Deghmane et al., 2000). Interestingly, we are able to detect high-level adherence after only a 25 minute incubation of WT K. kingae with human epithelial cells, indicating that type IV pili are able to rapidly overcome the inhibitory influence of capsule on Knh- mediated adherence, apparently via PilT/PilU-mediated retraction. These results suggest that K. kingae has developed a unique mechanism to efficiently adhere to host cells without having to down regulate capsule expression and become more susceptible to immune clearance.

We propose a model for K. kingae adherence to human epithelial cells focusing on the three surface factors examined in this study: type IV pili, Knh, and capsule (Figure

10). This model involves a two-step process that begins with a low-affinity adhesive event mediated by type IV pili, as these fibers are able to extend beyond the polysaccharide capsule and engage their host cell receptor. PilT/PilU-mediated pilus retraction then brings the bacterium into close contact with the host cell, causing physical displacement of the surface polysaccharide capsule. Once capsule has been displaced, Knh is able to engage its host cell receptor, leading to the high-level adherence we observe in adherence assays. It is important to note that this model relies on strong pilus retraction forces to displace the capsule. The force generated by a single

Myxococcus xanthus type IV pilus has been measured at ~150pN, making this the strongest linear molecular motor described to date (Clausen et al., 2009). Similar measurements have been made for N. gonorrhoeae type IV pili, suggesting that the strong forces generated by these bacterial organelles may be conserved in the Neisseriaceae family (Maier et al., 2002).

Figure 10: K. kingae adherence to human epithelial cells involves sequential adhesive events.

Initial K. kingae interactions with the host respiratory epithelium are initiated by type IV pili (1). Once pili have engaged their host cell receptor, PilT-mediated pilus retraction pulls the bacterium into close contact with the host cell membrane. This action results in physical displacement of the polysaccharide capsule, exposing Knh to the host cell membrane. Unmasked Knh can then mediate high-affinity K. kingae interactions with human epithelial cells via the Knh host cell receptor without the need for genetic downregulation of capsule (2).

Our results establish a potential mechanism by which K. kingae adheres to host respiratory epithelium and promotes colonization the upper respiratory tract. However, the precise interrelationship and regulatory mechanisms of the three surface factors presented here have yet to be determined. Further investigation of the K. kingae

adherence mechanism may lead to novel strategies for the prevention of disease due to

K. kingae and to other encapsulated pathogenic bacteria.

Acknowledgements

We thank Sara Miller of the Duke Electron Microscopy Service for assistance with transmission electron microscopy and Hank Seifert for plasmids pHSXtetM4 and pIDN4.

This work has been submitted for publication as referenced below:

Porsch, E.A., Kehl-Fie, T.E., and St. Geme III, J.W. (Submitted) Modulation of Kingella kingae adherence to respiratory epithelial cells by type IV pili, capsule, and a novel trimeric autotransporter.

3. Kingella kingae capsular polysaccharide characterization

3.1 Introduction

Kingella kingae pathogenesis is believed to begin with colonization of the posterior pharynx. Recent studies suggest that approximately 10% of children less than two years of age are colonized with this organism (Yagupsky et al., 2009b, Basmaci et al.,

2012). In addition, K. kingae is a member of the HACEK (Haemophilus, Actinobacillus,

Cardiobacterium, Eikenella, Kingella) group of fastidious gram-negative bacteria responsible for 5-10% of all cases of infective endocarditis (Berbari et al., 1997). While K. kingae typically does not cause disease in the respiratory tract, current evidence suggests that the organism is able to breach the respiratory epithelium, invade the bloodstream, and disseminate to the sites of invasive disease, typically the joints, bones, or endocardium (Yagupsky, 2004, Yagupsky et al., 2011).

The hematogenous dissemination model of K. kingae invasive disease is supported by multiple lines of evidence. First, K. kingae is frequently detected in the blood of young children who present with suspected septic arthritis via diagnostic blood-culture systems (Yagupsky et al., 1992). Advancements in molecular PCR-based diagnostics using detection of the rtxA gene or 16S rRNA gene sequencing technology have led to increased detection of K. kingae, not only from the blood, but also from joint fluid samples where it has been historically difficult to detect the organism with traditional culture-based methods (Yagupsky, 1999, Chometon et al., 2007, Cherkaoui et 69

al., 2009, Ceroni et al., 2010a). Second, K. kingae strains have been identified simultaneously in the respiratory tract and the blood of patients with invasive disease

(Yagupsky et al., 2009a). Lastly, a recent report found that respiratory and invasive matched isolates from individuals with K. kingae disease possess identical rtxA gene sequences and other genotypic markers (Basmaci et al., 2012). Taken together, these studies support the model that respiratory carriage of K. kingae is a precursor of invasive disease and that the bloodstream is the route by which the organism seeds the joints, bones, and endocardium.

A necessary aspect of this model of pathogenesis is the ability of K. kingae to survive in the bloodstream. As a gram-negative bacterium, the double membrane envelope is likely susceptible to complement-mediated membrane attack complex deposition and lysis. One survival strategy employed by other gram-negative bacteria that utilize the bloodstream in various pathogenic processes, such as N. meningitidis, extra-intestinal pathogenic E. coli, and H. influenzae, is expression of a polysaccharide capsule (Cross et al., 1986, Jarvis et al., 1987, Vogel et al., 1997). Capsules are surface- associated sugar polymers that coat the surface of many microbes, including certain gram-negative bacteria, gram-positive bacteria, and fungi. In the Eubacteria kingdom, the sugar subunits and linkages that constitute the polysaccharide often vary within a given species. Capsule itself is considered a classic virulence factor contributing to multiple properties in the host, including serum-resistance, host mimicry, and immune

evasion (Roberts, 1996). However, many polysaccharide capsules are also immunogenic in the host, and antibody reactive to capsule type is the basis for strain specific serotypes

(or serogroups) in a variety of bacterial species. This property has also led to the development of polysaccharide and polysaccharide-protein conjugate vaccines. The N. meningitidis MCV4 (serogroups A, C, Y and W-135), Streptococcus pneumoniae Prevnar 13

(13 serogroups), and H. influenzae Hib (type b) are all examples of polysaccharide conjugate vaccines currently approved by the FDA and recommended by the CDC

(www.cdc.gov/vaccines).

We recently described for the first time that K. kingae strain 269-492 derivatives express a surface-associated polysaccharide capsule (See Chapter 2). In this study we expand on those findings and show that the polysaccharide capsule expressed by our prototype strain 269-492 contains a thiobarbituric acid-reactive sugar and phenol- sulfuric acid-reactive sugar. For a more thorough analysis, we developed a large-scale purification protocol that was used to determine sugar composition and structure of the polysaccharide. Lastly, the polysaccharide was conjugated to ovalbumin using CDAP chemistry for injection into a guinea pig to develop polyclonal K. kingae specific antiserum.

3.2 Methods

Bacterial strains

K. kingae strains were stored at -80°C in brain heart infusion broth (BHI) with

30% glycerol. K. kingae strains were routinely cultured at 37°C with 5% CO2 on chocolate agar supplemented with 1 µg/ml erythromycin as appropriate. Strain 269-492 is a septic arthritis clinical isolate from St. Louis Children's Hospital in St. Louis, MO, and 269-

492ctrA is a capsule-deficient derivative with a polar ermC insertion in the predicted ctrABCD capsule export operon (Chapter 2).

Surface extraction

K. kingae strain 269-492 and derivatives were grown for 17-18 hrs on chocolate agar, suspended in 1ml PBS to an OD600 of 0.8, centrifuged for 2 min at 10,000 x g, and resuspended in 0.5ml PBS. Suspensions were incubated at 55°C for 30 min to extract surface polysaccharide material. Bacteria were pelleted, and supernatants were concentrated 10-fold in an Amicon Ultra 10,000 MWCO centrifuge filter.

Biochemical analyses

A modified Warren assay was used for sialic acid detection (Warren, 1959).

Briefly, 1 µl of 25% H2S04 was added to 50 µl of extracted surface material and heated at

80°C for 30 min to hydrolyze the polysaccharide. The hydrolyzed material was then reacted with 25 µl of 25 mM sodium periodate in 0.125 N H2S04 for 30 min at 37°C.

Twenty µl of 2% Sodium Arsenite in 0.5N HCl was added and gently mixed until the yellow color dissipated, boiled in a water bath for 7.5 min, and then chilled in an ice bath for 5 min. Once cooled, 700 µl of butanol was added, the sample was vortexed for 30 sec

and centrifuged at 5000 x g for 3 min, and the upper layer was transferred to a cuvette for measurement of absorbance at OD549. Five-hundred µM sialic acid in 50 µl was used as a positive control. A 50 µl sample of 1xPBS was used as a negative control, and the

OD549 reading of this control was subtracted from the experimental readings.

Experiments were performed in triplicate.

For the phenol-sulfuric acid assays, 50 µl of extracted surface material was diluted to 150 µl in water and mixed with 150 µl of 5% phenol, and then 750 µl 96%

H2SO4 was pipetted in rapidly to ensure equal mixing of the reactants. The reactions were briefly vortexed and allowed to sit at ambient temperature for 10 min. The reactions were then transferred to a cuvette and absorbance was measured at OD490. galactose (20µg in 150µl) was used as a positive control. One-hundred and fifty µl of water was used as a negative control, and the OD490 reading of this control was subtracted from the experimental readings. Experiments were performed in triplicate.

Large-scale polysaccharide purification

After growth of strain 269-492 on 20 chocolate agar plates, a bacterial suspension was prepared in PBS pH 7.4 with 1% formaldehyde and centrifuged at 2500 x g, and the pellet was resuspended in 50mM Tris-acetate pH 5.0 and incubated for 30 minutes with shaking to release the polysaccharide from the bacterial surface. The suspension was again centrifuged to pellet the bacteria, and the supernatant was recovered and filtered through a 0.22µm pore filter. Following concentration using an Amicon 100,000

molecular weight cut-off concentrator tube, the sample was treated with DNAse and

RNAse A to digest contaminating DNA and RNA and subsequently with proteinase K to digest contaminating protein. The sample was extracted with Tris-saturated phenol and then twice with chloroform to separate the polysaccharide from the proteinase K and any remaining protein and was then dialyzed in water, concentrated, flash frozen, and lyophilized.

Size exclusion chromatography

Approximately 1mg of the purified, lyophilized polysaccharide was sent to the

University of Georgia Complex Carbohydrate Research Center (CCRC) in Athens, GA for size exclusion chromatography to determine the purity of the large-scale polysaccharide preparation. The lyophilized material was dissolved in 130 µl deionized water and filtered through a 0.22µm Ultrafree-MC HV Centrifugal Filter. A total volume of 100 µl was injected into a Agilent 1200 HPLC system with a Superose 12 column using 50mM Ammonium Acetate, pH 5.5 as the eluent at a flow rate of 0.5 min/ml. For retention time and standard curve calculations, Dextran standards (5K,

10K, 40K, 66.9K and 167K) and glucose were used.

Composition analysis

Approximately 300 µg of lyophilized capsular material was sent to the CCRC for mass spectrometry glycosyl composition analysis. Briefly, methyl glycosides were prepared from the lyophilized sample by methanolysis in 1 M HCl in methanol,

followed by re-N-acetylation with pyridine and acetic anhydride in methanol (for detection of amino sugars) and then per-O-trimethylsilylation with Tri-Sil. GC/MS analysis of the TMS methyl glycosides was performed on an Agilent 6890N GC interfaced to a 5975B MSD, using a Supelco EC-1 fused silica capillary column (30m 

0.25 mm ID)

NMR Spectroscopy

Approximately 5mg of purified polysaccharide was sent the CCRC for NMR spectroscopy analysis. Briefly, The sample was deuterium exchanged by dissolving in

D2O and lyophilization and dissolved in 0.27 mL D2O. 1-D proton and 2-D gCOSY,

TOCSY, NOESY, gHSQC, and gHMBC spectra were obtained on a Varian Inova-600

MHz spectrometer at 50 °C using standard Varian pulse sequences. Proton chemical shifts were measured relative to internal acetone (H=2.218 ppm, C=33.0 ppm).

Polysaccharide-ovalbumin CDAP conjugation and antiserum generation

In order to generate an antiserum with a high anti-capsular IgG titer, the polysaccharide was first conjugated to ovalbumin using 1-cyano-4- dimethylaminopyridinium tetrafluoroborate (CDAP) chemistry. In brief, 1mg of CDAP in 100µl of acetonitrile was added slowly to 1mg of purified polysaccharide dissolved in

100µl of H2O. After incubation for 30 sec at 25°C, 100µl of aqueous 0.2M triethylamine and then 1mg of ovalbumin dissolved in 100µl of 0.25M HEPES buffer pH 8.2 was added, and the mixture was incubated for 16 hrs at 25°C. The resulting polysaccharide-

protein conjugate was buffer exchanged with 0.9% NaCl using a 100,000 molecular weight cut-off concentration filter, facilitating removal of unconjugated ovalbumin.

Approximately 300µg of polysaccharide-ovalbumin conjugate was sent to Cocalico

Biologicals, Reamstown, PA for injection into a guinea pig.

3.3 Results

3.3.1 K. kingae strain 269-492 polysaccharide contains TBA and phenol-sulfuric acid reactive sugars.

We demonstrated previously that K. kingae strain 269-492 expresses a surface- associated polysaccharide that can be released from the bacterial surface with heat

(Chapter 2). We chose to examine the draft strain 269-492 genome annotation for clues about the sugar composition of the polysaccharide. This analysis revealed the presence of genes with homology to genes involved in sialic acid processing in other species.

These genes include neuB, which converts N-acetyl-mannosamine to sialic acid; neuA, which activates sialic acid via addition of CMP; and neuS, which polymerizes poly-sialic acid from activated sialic acid. As sialic acid is a component of some polysaccharide capsules, including E. coli K1, and N. meningitidis serogroups A, B, and C (a fellow

Neisseriaceae family member like K. kingae), we chose to investigate whether sialic acid is a component of the K. kingae capsule. Sialic acids can be detected biochemically by reaction with thiobarbituric acid (TBA) and colorimetric detection (Warren, 1959). Using free sialic acid as a control, we found that material extracted from strain 269-492 was strongly reactive with TBA, while material extracted from strain 269-492ctrA was

minimally reactive (Figure 11, panel A). However, the TBA method for sialic acid determination is also able to detect other sugars, including certain 3-deoxy sugars such as 3-deoxy-manno-octulosonic acid (Kdo) (Osborn, 1963).

Having established that the K. kingae strain 269-492 polysaccharide contains a

TBA reactive sugar, we next investigated if the polysaccharide contains simple pentose or hexose sugars. Using the phenol-sulfuric acid method for hexose and pentose detection, we showed that surface extracted material from strain 269-492 was more reactive than material extracted from strain 269-492ctrA, using galactose as a positive control (Figure 11, panel B).

Figure 11: Biochemical detection of strain 269-492 polysaccharide sugars.

Surface extracts from K. kingae strains 269-492 and 269-492ctrA were reacted with the (A) thiobarbituric acid (TBA) method for sialic acid detection and (B) phenol-sulfuric

Figure 11 (continued) acid method for pentose and hexose detection. Sialic acid and galactose were used as controls.

Taken together, these data indicate that the K. kingae strain 269-492 polysaccharide contains at least one TBA reactive sugar and at least one phenol-sulfuric acid reactive sugar. As the TBA assay does not react with simple sugars and the phenol- sulfuric acid assay does not react with sialic acids, it is likely that the assays are detecting separate sugars.

3.3.2 Large scale polysaccharide purification

In order to further characterize the strain 269-492 polysaccharide capsule, a system for large-scale purification was needed. Published protocols for capsule purification from the related Neisseriaceae family member, N. meningitidis, using polysaccharide precipitation via Cetavlon were attempted but were unsuccessful

(Pollard et al., 2001). A new protocol was developed as described in detail in the methods. Briefly, a bacterial suspension was fixed with paraformaldehyde, polysaccharide was extracted from the bacterial surface with mild acid and agitation, contaminating DNA, RNA, and protein was removed via digestion, and residual protein was eliminated by phenol-chloroform extraction. The sample was dialyzed extensively in water for 2 days and then examined for purity using SDS-PAGE and Coomassie and silver staining. As shown in Figure 12, after digestion with DNAse, RNAse, and proteinase K, the only band detectable with Coomassie was at the predicted molecular 78

weight of proteinase K (lane 1). More contaminating material was evident after silver staining. However, after extraction with phenol-chloroform, the contaminating

Coomassie and silver stained material was removed while the polysaccharide was retained.

Figure 12: Purity of large-scale strain 269-492 polysaccharide preparation.

Coomassie (left panel) and silver (right panel) stained polysaccharide preparations pre-(Lane 1) and post-(Lane 2) phenol-chloroform extraction. Lane M is the protein molecular weight marker with select sizes expressed in kDa. The prominent band at ~32kDa is likely proteinase K.

As a complementary approach to determine purity of the large-scale polysaccharide preparation, approximately 1mg of lyophilized polysaccharide was sent to the CCRC for size exclusion chromatography analysis. Using a Superose 12 column and a series of Dextran standards for size determination, the strain 269-492 polysaccharide large-scale preparation eluted in the void volume indicating that the molecular weight of the material is greater than 200 kDa (Figure 13). This high 79

molecular weight estimation is in agreement with the alcian blue and silver stained material visualized after SDS-PAGE. Only one other small peak was evident in the elution profile of the large-scale preparation and the elution time indicates that it was likely a salt (Figure 13). These results suggest that the large-scale preparation protocol purified only high molecular weight polysaccharide or polysaccharides.

Figure 13: Size exclusion chromatogram of strain 269-492 large-scale polysaccharide preparation on a Superose 12 column.

3.3.3 Glycosyl composition and structural analysis of strain 269-492 polysaccharide.

To gain a more detailed understanding of the glycosyl composition of the polysaccharide expressed on the surface of the K. kingae strain 269-492, purified polysaccharide was sent to the CCRC for analysis. Approximately 300 µg of purified polysaccharide was sent to the CCRC and an aliquot was processed for GC/MS using an

Agilent 6890N GC interfaced to a 5975B MSD as described in the methods. The results

presented in Table 5 revealed that the polysaccharide preparation contains an equimolar ratio of galactose, N-acetyl galactosamine, and Kdo. Other glycosyl residues were also identified in the analysis but were only present at mole percents of approximately 5% or less. These results indicate that K. kingae strain 269-492 expresses surface polysaccharide(s) that contain an equimolar ratio of galactose, N-acetyl galactosamine, and Kdo.

Table 5: Glycosyl composition of K. kingae strain 269-492 surface polysaccharide.

Glycosyl residue Mass (µg) Mol%1 Heptose (Hep) 0.6 5.6 Rhamnose (Rha) n.d. - Fucose (Fuc) n.d. - Xylose (Xyl) 0.3 4.6 Glucuronic acid (GlcA) n.d. - Galacturonic acid (GalA) n.d. - Mannose (Man) n.d. - Galactose (Gal) 2.4 27.1 Glucose (Glc) 0.2 1.9 N-Acetyl Galactosamine (GalNAc) 3.1 28.5 N-Acetyl Glucosamine (GlcNAc) 0.4 3.4 N-Acetyl Mannosamine (ManNAc) n.d. - Keto-deoxyoctulosonic acid (Kdo) 3.5 28.8 1Values are expressed as mole percent of total carbohydrate.

To determine the structure of the polysaccharide or polysaccharides purified from the surface of K. kingae strain 269-492, NMR spectroscopy was performed in collaboration with the CCRC as described in the Methods. This analysis revealed that the purified material consisted of two distinct polysaccharides. One-third of the sample was a galactose homopolymer with the structure of →5)-β-Galf-(1→, and two-thirds was

a repeating GalNAc-Kdo polymer with the structure of →3)-β-GalNAcp-(1→5)-β-Kdop-

(2→. These ratios are in agreement with the total glycosyl composition analysis molar ratios for the individual sugars. A structurally distinct galactose homopolymer has been described in a different K. kingae strain and has been identified as a secreted exopolysaccharide (Bendaoud et al., 2011). In conjunction with the data presented in

Chapter 2, specifically the loss of cationic ferritin staining in the ctrA mutant, the

GalNAc-Kdo polymer is likely the capsular polysaccharide.

3.3.4 Polysaccharide-protein conjugation for antiserum development

For further investigation into a variety of aspects of K. kingae encapsulation, capsular polysaccharide-specific antiserum would be a useful for many applications. As pure polysaccharide as an immunogen would drive primarily an IgM response with no memory component, we chose to use a polysaccharide-protein conjugate as the immunogen for antibody development in a guinea pig. The conjugate antigen strategy is advantageous because it leads to development of an IgG response and allows for memory generation. Both of these factors should help increase anti-capsular polysaccharide specific antibody titers following several rounds of boosting. Purified strain 269-492 polysaccharide was covalently conjugated to ovalbumin using CDAP chemistry as detailed in the Methods. To confirm conjugation, an aliquot of the conjugation reaction was separated using SDS-PAGE and stained with either alcian blue or Coomassie. As the polysaccharide-ovalbumin conjugate is higher in molecular

weight than the purified polysaccharide, we found that the material post-conjugation was shifted when stained with alcian blue and additionally stained with Coomassie

(Figure 14). These results indicate that the purified polysaccharide was successfully conjugated to ovalbumin.

Figure 14: Alcian blue and Coomassie stained polysaccharide-ovalbumin conjugate.

Purified polysaccharide (Lane 1) and polysaccharide-ovalbumin conjugate (Lane 2) were separated on a 7.5% SDS-PAGE gel and stained with either Alcian blue (Panel A) to detect polysaccharide or Coomassie (Panel B) to detect protein. Lane M is the protein marker.

The conjugate was flash frozen, lyophilized, and sent to Cocalico Biologicals,

Reamstown, PA, for injection into a guinea pig for antiserum development. After initial injection and two rounds of boosting, antiserum will be screened for reactivity against K. kingae 269-492 whole bacteria and purified polysaccharide using dot blotting with 269-

492ctrA and purified H. influenzae type b capsule as negative controls.

3.4 Discussion

While previous microbiological examination of K. kingae did not reveal the presence of a surface-associated capsule, we recently demonstrated (Chapter 2) that strain 269-492 does express surface polysaccharide that is dependent on the ctrABCD predicted capsule export operon (Chapter 2). In this study, we expand on those findings to characterize the glycosyl composition and structure of K. kingae strain 269-492 surface polysaccharide. Using biochemical sugar detection assays we show that surface extracts of K. kingae strain 269-492 does react with the TBA Warren detection method and the simple pentose and hexose phenol-sulfuric acid detection method. For a more detailed analysis, a large scale purification strategy was developed and the polysaccharide was found to contain an equimolar ratio of galactose, N-acetyl galactosamine, and Kdo using

GC/MS analysis. NMR analysis found that the material contained two separate polysaccharides and defined their structures. In addition, purified polysaccharide was conjugated to ovalbumin and antiserum generation was initiated.

During these studies, a report by Bendaoud et al. described identification of a capsular polysaccharide with structure of →3)-GlcNAc-(1→5)-Kdo-(2→ (N-acetyl- glucosamine - Kdo polymer) from K. kingae clinical isolate, PYKK181. In addition, we purified polysaccharide from an additional clinical isolate, PYKK060, and found that it contains mostly ribose and Kdo (data not shown). Based on these three analyses, it is clear that isolates of K. kingae, express different capsule types, similar to the situation

with other encapsulated bacterial species. Understanding that we have polysaccharide composition analysis of only three strains, it is still interesting that Kdo appears to be a conserved component. Kdo is most well known as the inner core sugar linking lipid A to the outer core sugars of lipopolysaccharide (LPS). However, Kdo is also a capsule polysaccharide component in a variety of gram-negative bacteria (Luderitz et al., 1984).

For example, Sinorhizobium meliloti strain Rm1021 expresses a Kdo homopolymer

(Fraysse et al., 2005), Actinobacillus pleuropneumoniae serotype 5 expresses a GlcNAc - Kdo polymer (Altman et al., 1987) (identical to that of K. kingae strain PYKK181 (Bendaoud et al., 2011)), and N. meningitidis serotype 29e expresses a GalNAc - Kdo polymer (Griffiss et al., 1983) but differs from the K. kingae strain 269-492 polymer based on the glycosidic linkages of the polymer. As noted previously (Chapter 2), the capsular synthesis genes in K. kingae have yet to be identified and the presence of Kdo in the polysaccharide may indicate an overlap between LPS and capsule synthesis.

In the same study that described the PYKK181 capsular structure, a galactose homopolymer with the structure of →3)-β-Galf-(1→6)-β-Galf-(1→ was identified and shown to have antibiofilm activity (Bendaoud et al., 2011). Interestingly, strain 269-492 also produced a galactose homopolymer but with the structure of →5)-β-Galf-(1→.

Preliminary data suggest that the genes involved the galactose homopolymer biogenesis are largely conserved in strains PYKK181 and 269-492 so their differing glycosyl linkages are intriguing (K. F. Starr and J. W. St. Geme, unpublished data).

Future studies examining the capsule composition and structure of additional strains will be necessary to determine the diversity of capsule types in K. kingae. Using a clinical isolate set that contains K. kingae isolates from both healthy carriers and patients with invasive disease, preliminary results based on alcian blue staining of surface extracts indicate that all strains are encapsulated. Of particular interest will be determining if a particular a capsule type or types is associated with specific K. kingae disease states. This information is critical for evaluation of the capsular polysaccharide as a K. kingae vaccine component.

As a reagent to further investigate K. kingae encapsulation, polysaccharide- specific antiserum generation was initiated. We learned that the purified material contained two distinct polysaccharides after antibody generation was initiated.

Therefore, the antiserum will likely have reactivity towards both polysaccharides. To generate antibody targeted only to the capsular polysaccharide, a project is underway to purify surface polysaccharide from strain 269-492 deleted for the operon responsible for the galactose homopolymer synthesis identified by previously (Bendaoud et al., 2011).

Capsule types are also frequently referred to as serotypes or serogroups, as the polysaccharide is often immunogenic in the host, and the specific antibody is useful for differentiating interspecies strain capsule diversity. Clinically, antibody reactivity to capsular polysaccharide has been used for decades to type a variety of encapsulated bacteria, including H. influenzae, N. meningitidis, and S. pneumoniae. We anticipate that

we will be able to screen our collection of clinical isolates using a dot blot to determine which strains express the same polysaccharide as strain 269-492. This reagent will also be beneficial for investigating a variety of molecular aspects of K. kingae encapsulation, including capsule-influenced interactions with host cells and differentiation of capsule synthesis and export mutants.

The results presented here are the first steps necessary to define the capsule repertoire expressed by K. kingae. Future investigation will determine the viability of capsular polysaccharide as a target for the prevention of disease due to K. kingae.

Acknowledgements

We thank Dr. Parastoo Azadi and the University of Georgia Complex

Carbohydrate Research Center for assistance with size exclusion chromatography, glycosyl composition and NMR analyses.

4. Differential Control of Kingella kingae Type IV Pilus Phenotypes Through the Calcium-Binding Sites of PilC1 and PilC2

4.1 Introduction

Kingella kingae is a gram-negative organism that belongs to the Neisseriaceae family and is being recognized increasingly as an important pathogen in young children.

Improvements in culture- and molecular-based diagnostics have identified K. kingae as a common etiology of septic arthritis, osteomyelitis, and bacteremia in the pediatric population (Yagupsky et al., 1992, Yagupsky et al., 1995b, Moumile et al., 2003, Gene et al., 2004, Yagupsky, 2004, Verdier et al., 2005, Chometon et al., 2007, Lehours et al., 2011,

Yagupsky et al., 2011). A recent study reported K. kingae as the leading cause of septic arthritis in children 6-24 months of age (Chometon et al., 2007). Based on epidemiological studies, K. kingae is believed to initiate infection by colonizing the posterior pharynx, where it typically persists for weeks without producing symptoms

(Yagupsky et al., 1995a, Yagupsky et al., 2002, Yagupsky et al., 2009a, Yagupsky et al.,

2009b). On occasion the organism will breach the epithelial barrier and enter the bloodstream and then disseminate hematogenously to the joints, bones, or endocardium

(Yagupsky, 2004, Yagupsky et al., 2011). This model is supported by reports describing genotypically identical K. kingae paired isolates from the respiratory tract and blood of patients with invasive K. kingae disease (Yagupsky et al., 2009a, Basmaci et al., 2012).

Despite increased recognition of K. kingae as an important pathogen, little is known about the molecular mechanisms used by this organism to cause disease.

Previous work established that K. kingae expresses type IV pili that are necessary for adherence to respiratory epithelial and synovial cells (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009, Kehl-Fie et al., 2010). Type IV pili are surface fibers that are widespread in gram-negative bacteria and convey many phenotypes, including adherence, twitching motility, and natural competence (Burrows, 2005, Pelicic, 2008). These fibers are composed primarily of a single protein subunit (called PilA1 in K. kingae) and require numerous other factors for proper expression. A key aspect of type IV pilus biology and the associated phenotypes is the ability of the fiber to be retracted back through the outer membrane, a process catalyzed by the retraction ATPase, PilT, and associated with twitching motility (Whitchurch et al., 1991, Anantha et al., 1998, Wolfgang et al., 1998a,

Merz et al., 2000, Chiang et al., 2008). While many of the protein factors involved in type

IV pilus biology have been well-studied, the mechanism controlling pilus extension and pilus retraction remains poorly understood. However, current data from work on the pathogenic Neisseria spp. (Neisseria meningitidis and Neisseria gonorrhoeae) and

Pseudomonas aeruginosa suggests that counteraction of retraction by PilC-like proteins is required for proper pilus expression and twitching motility (Wolfgang et al., 1998b,

Morand et al., 2004, Heiniger et al., 2010).

The PilC proteins are proposed to influence two aspects of type IV pilus biology, namely pilus biogenesis and adherence. Like N. meningitidis and N. gonorrhoeae, K. kingae encodes two PilC-like proteins called PilC1 and PilC2 (Jonsson et al., 1991, Nassif et al., 1994, Kehl-Fie et al., 2008). The PilCs of the pathogenic Neisseria spp. have been identified as the adhesive component of the fiber and are located at the tip of the fiber in

N. gonorrhoeae (Rudel et al., 1995b, Rahman et al., 1997, Scheuerpflug et al., 1999). Recent observations suggest that P. aeruginosa PilY1 (the single PilC-like protein in P. aeruginosa) is also an adhesin (Heiniger et al., 2010, Johnson et al., 2011). In earlier work, we observed a reduction in adherence to human cells by K. kingae strain 269-492 derivatives that expressed only PilC1 or PilC2, supporting a potential role for these proteins in adherence (Kehl-Fie et al., 2008). In addition, we found that elimination of both PilC1 and PilC2 in K. kingae strain 269-492 resulted in a loss of piliation, suggesting that at least one PilC protein is required for pilus expression (Kehl-Fie et al., 2008). Similar results were observed with the pathogenic Neisseria species and P. aeruginosa (Jonsson et al.,

1991, Alm et al., 1996). Recently, P. aeruginosa PilY1 was shown to regulate pilus expression and twitching motility through calcium-binding (Ca-binding) at a nine amino acid loop in the C-terminal domain (Orans et al., 2010). Using bioinformatic approaches, similar loops were identified in other PilC homologs, including N. meningitidis and N. gonorrhoeae PilC1 and PilC2 and K. kingae PilC1, but not PilC2 (Orans et al., 2010).

In contrast to the two PilC proteins in N. gonorrhoeae and N. meningitidis, K. kingae

PilC1 and PilC2 share only limited overall sequence homology (7% identity/16% similarity) (Kehl-Fie et al., 2008). In this regard, K. kingae represents a unique and intriguing system to investigate PilC-like protein biology. In this study, we examined

Ca-binding by PilC1 and PilC2 and the potential role of Ca-binding in control of type IV pilus expression, twitching motility, and adherence to Chang epithelial cells. We report that both PilC1 and PilC2 bind calcium, with PilC1 utilizing a nine amino acid PilY1-like

Ca-binding site and PilC2 utilizing a 12 amino acid human calmodulin-like Ca-binding site. While both the PilC1 and PilC2 Ca-binding sites are dispensable for pilus assembly, the PilC1 Ca-binding site is required for twitching motility and adherence.

4.2 Methods

Bacterial strains

Bacterial strains used in this study are listed in Table 6. K. kingae strains were stored at -80°C in brain heart infusion broth (BHI) with 30% glycerol. E. coli strains were stored at -80°C in Luria-Bertani (LB) broth with 15% glycerol. K. kingae strains were routinely cultured at 37°C with 5% CO2 on chocolate agar supplemented with 50 µg/mL kanamycin, 1µg/mL erythromycin, or 2 µg/mL tetracycline, as appropriate. E. coli strains were routinely cultured at 37°C in LB broth or on LB agar supplemented with 100

µg/mL ampicillin, 50 µg/mL kanamycin, 500 µg/mL erythromycin, 25 µg/mL tetracycline, 50 µg/mL streptomycin, or 50 µg/mL chloramphenicol, as appropriate.

Table 6: Strains and plasmids used in this study

Strain Description Source Dh5α E. coli F− 80dlacZΔM15 Δ(lacZYA-argF)U169 deoR (Sambrook recA1 endA1 hsdR17(rK− mK+) phoA supE441 thi-1 et al., 1989) gyrA96 relA1 BL21- E. coli B F– ompT hsdS(rB– mB–) dcm+ Tetr gal Agilent CodonPlus(DE3)-RIPL λ(DE3) endA Hte [argU proLCamr] [argU ileY leuW Strep/Specr] BL21-Gold(DE3) E. coli B F– ompT hsdS(rB– mB–) dcm+ Tetr gal Agilent λ(DE3) endA Hte KK03 Naturally occurring spreading/corroding variant (Kehl-Fie et of septic arthritis clinical isolate 269-492. al., 2007) ΔpilF KK03 with an aphA3 marked pilF deletion This work ΔpilC1 KK03 with a tetM marked pilC1 deletion This work ΔpilC2 KK03 with an aphA3 marked pilC2 deletion This work ΔpilC1/ΔpilC2 KK03 with a tetM marked pilC1 deletion and an This work aphA3 marked pilC2 deletion ΔpilC1/PilC2mark Strain KK03pilC1 with an aphA3 insertion This work immediately downstream of WT pilC2 ΔpilC2/PilC1mark Strain KK03pilC2 with an ermC insertion upstream This work of WT pilC1 ΔpilC1/PilC2D1444A Strain ΔpilC1/PilC2mark with a pilC2 mutation This work resulting in expression of PilC2D1444A ΔpilC1/PilC2D1444K Strain ΔpilC1/PilC2mark with a pilC2 mutation This work resulting in expression of PilC2D1444K ΔpilC2/PilC1D930A Strain ΔpilC2/PilC1m with a pilC1 mutation This work resulting in expression of PilC1D930A ΔpilC2/PilC1D930K Strain ΔpilC2/PilC1m with a pilC1 mutation This work resulting in expression of PilC1D930K PilC1D930A/ Strain KK03 expressing PilC1D930A and This work PilC2D1444A PilC2D1444A PilC1D930K/ Strain KK03 expressing PilC1D930K and This work PilC2D1444K PilC2D1444K Plasmid Description Source pMCSG7 Protein expression vector for 6xHis N-terminal (Stols et al., fusions and a TEV cleavage site 2002) pMCSG9 Protein expression vector for MBP 6xHis N- (Donnelly et terminal fusions and a TEV cleavage site al., 2006) pMCSG9/PilC1 For expression of MBP-PilC1739-1047 This work pMCSG9/PilC1D930A For expression of MBP-PilC1739-1047D930A This work pMCSG9/PilC1D930K For expression of MBP-PilC1739-1047D930K This work pMCSG7/PilC2 For expression of HIS-PilC2868-1502 This work pMCSG7/PilC2D1125A For expression of HIS-PilC2868-1502 D1125A This work 92

pMCSG7/PilC2D1125K For expression of HIS-PilC2868-1502 D1125K This work pMCSG7/PilC2D1444A For expression of HIS-PilC2868-1502 D1444A This work pMCSG7/PilC2D1444K For expression of HIS-PilC2868-1502 D1444K This work pFalcon2 Source of aphA3 kanamycin-resistance gene (Hendrixson et al., 2001) pHSXtetM4 Source of tetM tetracycline-resistance gene (Seifert, 1997) pIDN4 Source of ermC erythromycin-resistance gene (Hamilton et al., 2001) pUC19/ΔpilF pilF deletion construct (Kehl-Fie et al., 2008) pUC19/ΔpilC1 pilC1 deletion construct (Kehl-Fie et al., 2008) pUC19/ΔpilC2 pilC2 deletion construct This work pUC19/ErmpilC1 For introduction of ermC marked pilC1 locus This work pUC19/pilC2Kan For introduction of aphA3 marked pilC2 locus This work pUC19/PilC1D930A For introduction of PilC1D930A This work pUC19/PilC1D930K For introduction of PilC1D930K This work pUC19/PilC2D1444A For introduction of PilC2D1444A This work pUC19/PilC2D1444K For introduction of PilC2D1444K This work

Recombinant PilC constructs and protein purification

All plasmids used in this study are listed in Table 6 and all primers are listed in

Table 7. PilC1 residues 739-1047 were cloned from pUC19pilC1 with primers PilC1 738 F and PilC1 1047 R into a pMCSG9 LIC HIS MBP vector. Site-directed mutagenesis was performed to produce D930A and D930K mutations with primers PilC1D930A-

F/PilC1D930A-R and PilC1D930K-F/PilC1D930K-R, respectively. Resultant vectors were transformed into BL21 RIPL cells (Stratagene) and plated on appropriate antibiotic. A single colony was used to inoculate 50 mL of LB broth containing the previously mentioned antibiotics, and cultures were incubated overnight. Cell cultures were centrifuged at 3000 x g, and the supernatant was discarded. The resultant pellet was

used to inoculate a 1.5 L shaker flask of Terrific Broth (TB) with 50µL of antifoam

(Sigma-Aldrich) and the previously mentioned antibiotics. Cells were grown at 37°C until the density reached an OD600 of 0.6-0.8. The temperature was then reduced to

18°C, and protein expression was induced with 0.2 mM IPTG. Cells were grown overnight and harvested by centrifugation at 6000 x g at for 15 minutes at 4°C, and pellets were stored at -80°C.

Thawed PilC1 pellets were suspended in loading buffer (50 mM sodium phosphate pH 7.6, 500 mM NaCl, 25 mM imidazole) supplemented with 0.5 mM EDTA,

0.1% Triton X-100, 1 mM PMSF, one tablet of a protease inhibitor cocktail (Roche), and 1

µg/ml lysozyme. After 1 hour of gently stirring on ice, the cells were sonicated on ice for

1 min and the lysate was centrifuged at 45,000 x g for 90 min at 4°C. Using an

ÄKTAxpress™ (GE HealthCare), protein from the filtered soluble fraction was nickel purified (Elution buffer consisted of 50 mM sodium phosphate pH 7.6, 500 mM NaCl,

500 mM imidazole) and purified on an S200 gel filtration column in the final buffer 20 mM Tris-HCl pH 7.5, 250 mM NaCl, 2 mM DTT, 5% glycerol. Protein was then concentrated, flash frozen in liquid nitrogen, and stored at -80°C.

PilC2 residues 868-1502 were cloned from genomic K. kingae DNA with primers

PilC2 868 F and PilC2 end R into a pMCSG7 LIC HIS vector. Site-directed mutagenesis was performed to produce D1125A (primers PilC2D1125A-F/PilC2D1125A-R), D1125K

(primers PilC2D1125K-F/PilC2D1125K-R), D1444A (primers PilC2D1444A-

F/PilC2D1444A-R), and D1444K (primers PilC2D1444K-F/PilC2D1444K-R) mutations.

Resultant plasmids were transformed into BL21 Gold cells (Agilent), which were incubated overnight on LB plates containing ampicillin. Subsequently, a single colony was used to inoculate 100mL of LB overnight containing 50 µg/mL ampicillin, and the culture was incubated overnight. Cell cultures were treated as described above except that they were induced with 0.5 mM IPTG.

Cells pellets were thawed and resuspended in buffer consisting of 10 mM Tris pH 7.8 and 50 mM NaCl with 10 mM imidazole, DNase, and protease inhibitor tablets

(Roche). Cells were sonicated and the cell lysate was separated into soluble and insoluble fractions using high-speed centrifugation. The soluble fraction was filtered then nickel purified (with 300 mM imidazole), buffer exchanged (no imidazole), and separated using an S200 gel filtration column on an ÄKTAxpress™ (GE HealthCare). If necessary, protein and storage buffers were chelated by Chelex-100 to remove bound calcium (Bio-Rad Laboratories). Purified proteins were concentrated to ~100 µM, frozen, and stored at -80 °C.

Table 7: Primers used in this study.

Primer Sequence (5'>3') PilC1 738 F TACTTCCAATCCAATGCGTCTGCGCGTCCACATTTATTTATGATTAA TGG PilC1 1047 R TTATCCACTTCCAATGCGCTATGTTTGCTCCAACAAATTACCAGCTG TTAAAGG PilC2 868 F TACTTCCAATCCAATGCGATGTTGTTGCAATCCGTGTTCCG PilC2 end R TTATCCACTTCCAATGCGCTAGAAAATCTCGCGCCAAGATACACGT PilC1D930A-F GTGAATCGAGATGGTGTGTATGCTTTTGTCTTTGCGGGAGATTATG PilC1D930A-R ATAATCTCCCGCAAAGACAAAAGCATACACACCATCTCGATTCAC 95

PilC1D930K-F CTTGATGTGAATCGAGATGGTGTGTATAAATTTGTCTTTGCGGGA PilC1D930K-R TCCCGCAAAGACAAATTTATACACACCATCTCGATTCACATCAAG PilC2D1444A-F GTTTACCGAAGCGGCCACTTTGGTATCTGTG PilC2D1444A-R CACAGATACCAAAGTGGCCGCTTCGGTAAAC PilC2D1444K-F GGTAAGTTTACCGAAGCGAAGACTTTGGTATCTGTGGC PilC2D1444K-R GCCACAGATACCAAAGTCTTCGCTTCGGTAAACTTACC PilC2D1125A-F GACTGAAAGTGCGATGCTGTGGGAG PilC2D1125A-R CTCCCACAGCATCGCACTTTCAGTC PilC2D1125K-F CAGACTTGACTGAAAGTAAAATGCTGTGGGAGTTTAC PilC2D1125K-R GTAAACTCCCACAGCATTTTACTTTCAGTCAAGTCT pilC2Δ5'F ATGCGAATTCAGGGTTGCCACCACCGAGC pilC2Δ5'R ATGCGGATCCGGTTATCGCAAACCAAATCGTGC pilC2Δ3'F ATGCGGATCCGCACTTGTGTAAAAAGCAAGTGCCG pilC2Δ3'R ATGCAAGCTTTGTCTTCAATAGACGGACAATAGCGC aphA3FBamHI GCATGGATCCCATCTAAATCTAGGTACTAAAACAATTCATCCAG aphA3RBamHI GCATGGATCCGTTTGACAGCTTATCATCGATAAACCCAG pilC1regionF ATGCGAATTCCACGTTGGGGCAAGGCAATG pilC1regionR ATGCGTCGACATAAGCCTGATTGATTGGCTCTGCG pilC1markF CGCCGAAAACTGCTACGCGTGAAGAACCTGCAAAAGAG pilC1markR CTCTTTTGCAGGTTCTTCACGCGTAGCAGTTTTCGGCG ermC F ACGTACGCGTGGTTACGCTTTGGGGAAATTATGAGG ermC R ACGTACGCGTGTAATCATGGTCATAGCTGTTCGATAAGC pilC2markAF ATGCGAATTCCCATTGATTGCCGACAATTGGG pilC2markAR ATGCGGATCCCCGTTTGTTTTTCAGCTAATCACGA pilC2markBF ATGCGGATCCCACTTGTGTAAAAAGCAAGTGCCG pilC2markBR ATGCAAGCTTTAATCAAATCATCAACCAACACGCC

K. kingae strain construction

Gene disruptions and directed mutations were generated in K. kingae as previously described (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009). Briefly, plasmid-based disruption constructs were created in E. coli, linearized, and introduced into K. kingae strain KK03 via natural transformation and plating on appropriate antibiotic-containing media. Correct localization of disruptions was confirmed by PCR or Southern blotting and site-directed mutations were confirmed by sequencing. The ΔpilF and ΔpilC1

deletion constructs were generated as previously described (Kehl-Fie et al., 2008). As our original pilC2 disruption was an insertional transposon mutant (Kehl-Fie et al., 2008), we decided to generate a targeted ΔpilC2 deletion. PCR fragments corresponding to the 5' and 3' region of pilC2 were amplified individually from strain KK03 with primers pilC2Δ5'F/pilC2Δ5'R and pilC2Δ3'F/pilC2D3'R and were then ligated into EcoRI/HindIII- digested pUC19, generating pUC19pilC2::BamHI, which contains a large pilC2 internal deletion and a BamHI site. The kanamycin resistance cassette aphA3 was PCR amplified with primers aphA3FBamHI/aphA3RBamHI from pFalcon2 and ligated into BamHI- digested pUC19pilC2::BamHI, generating pUC19ΔpilC2.

To generate pilC alleles that encode Ca-binding site mutations, antibiotic resistance markers were inserted adjacent to the pilC gene in a pUC19 backbone, subjected to site-directed mutagenesis, sequenced, linearized, and transformed into K. kingae strain KK03. For PilC1 Ca-binding site mutations, the entire pilC1 gene and ~1000 bps of 5' and 3' flanking sequence were PCR amplified from KK03 genomic DNA with primers pilC1regionF/pilC1regionR and the ligated into EcoRI/SalI-digested pUC19, generating pUC19pilC1. The QuikChange XL-II Site-directed Mutagenesis Kit (Agilent) and primers pilC1markF/pilC1markR were used to insert an MluI site in the gene upstream of the pilC1 promoter region, a predicted ABC-type transporter, generating pUC19/MluIpilC1. The ermC erythromycin resistance cassette was amplified from pIDN4 with primers ermCF/ermCR and was ligated into MluI-digested

pUC19/MluIpilC1, generating pUC19/ErmpilC1. Restriction digestion and sequencing were used to confirm proper orientation of ermC. To confirm that the ermC marker inserted upstream of WT pilC1 did not alter the phenotypes observed in the ΔpilC2 background, a control ΔpilC2/PilC1mark (pilC2 deleted and ermC upstream of pilC1) strain was created and was confirmed to have phenotypes identical to those of the

ΔpilC2 strain. Site-directed mutagenesis with primers PilC1D930A-F/PilC1D930A-R and

PilC1D930K-F/PilC1D930K-R was used to generate the D930A and D930K mutations, respectively.

For the PilC2 Ca-binding site mutations, ~2.6 kb of the 3' region of pilC2 was amplified with primers pilC2markAF/pilC2markAR and ~1.5 kb of sequence immediately downstream of the pilC2 ORF was amplified with primers pilC2markBF/pilC2markBR individually from KK03 genomic DNA and ligated into

EcoRI/HindIII-digested pUC19, generating pUC19pilC2, which contains a BamHI site inserted 20 bp downstream of the pilC2 stop codon. The kanamycin resistance cassette aphA3 was PCR amplified with primers aphA3FBamHI/aphA3RBamHI from pFalcon2 and ligated into BamHI-digested pUC19pilC2, generating pUC19pilC2Kan. To confirm that the aphA3 marker inserted immediately downstream of WT pilC2 did not alter the phenotypes observed in the K. kingae ΔpilC1 background, a control ΔpilC1/PilC2mark

(pilC1 deleted and aphA3 downstream of pilC2) strain was created and confirmed to have phenotypes identical to those of the ΔpilC1 strain. Site-directed mutagenesis with

primers PilC2D1444A-F/PilC2D1444A-R and PilC2D1444K-F/PilC2D1444K-R was used to generate the D1444A and D1444K mutations, respectively.

Calcium binding assay

A binding curve for Oregon Green® 488 BAPTA-5N, hexapotassium salt

(Invitrogen) in 10 mM Tris pH 7.8 and 50mM NaCl was measured on a PHERAstar

(BMGLabtech) at 488 nm. With 20 µM Oregon Green and 2 µM CaCl2, purified PilC1 or

PilC2 proteins were serial diluted 1.5-3 fold from ~100 µM to ~1 nM to obtain an EC50 which was then used to calculate the respective Kd for calcium as described previously

(Orans et al., 2010).

Circular dichroism and thermal denaturation

Protein samples were exchanged into chelated 10 mM K2H4PO4 50 mM NaF pH

7.7 buffer and brought to 5 µM with or without the addition of 20 µM CaCl2. A wavelength scan from 200-260 λ was performed on a circular dichroism spectrometer 62

DS (Aviv) at 16 °C with a 10 second averaging time. Melting temperatures were measured at 214 λ from 3 °C to 95 °C at one degree increments with a 10 second averaging time.

Pilus preparations

Derivatives of K. kingae strain KK03 were incubated on chocolate agar for 17-18 hrs, and growth was suspended in 1.5 mL 50 mM Tris 150 mM NaCl, pH 8.0 to an OD600 of 1.0, vortexed at full speed for 1min, and centrifuged at 21,000 x g for 2 min to pellet

the bacteria. 1.25 ml of the bacteria-free supernatant was subjected to 20% ammonium sulfate precipitation on ice for 2 hrs. Precipitated pili were collected via centrifugation at

21,000 x g for 5 min and resuspended in 1 x SDS-PAGE loading buffer. Aliquots were separated on 15% SDS-PAGE gels and stained with Coomassie blue. The observed protein band was confirmed to be the major pilin subunit, PilA1, by Western blot analysis with antiserum GP65 as described previously (Kehl-Fie et al., 2009).

Twitching motility assays

PBS to an OD600 of 1.0. One µl of the bacterial suspensions was stab inoculated to the bottom of tissue culture-treated 100mm plates containing chocolate agar and incubated at 37 °C with 5% CO2 for 2 days. Twitching motility competent strains spread from the stab inoculation site at the plate/agar interface. The agar was carefully peeled away, and the plate was air-dried and stained with crystal violet to visualize the twitching motility zones. Three diameter measurements were taken and averaged per twitching zone. All experiments were performed in triplicate, and error bars represent standard error of the mean. Statistical analysis was performed using the unpaired t test to compare twitching motility zones between two strains as noted in the text.

Eukaryotic cell lines

100

Chang cells (Wong-Kilbourne derivative (D) of Chang conjunctiva, HeLa origin;

ATCC CCL-20.2) were cultivated at 37 °C with 5% CO2 in media as described previously

(Kehl-Fie et al., 2007).

Adherence assays

165 x g for 5 min and then incubated for 25 min at 37°C in 5% CO2. Monolayers were washed 4 times with PBS to remove non-adherent bacteria and were then treated with

1x trypsin-EDTA (Sigma) for 20 min at 37 °C to release adherent bacteria. Appropriate dilutions were plated on chocolate agar, and percent adherence was calculated by dividing the number of adherent CFU by the number of inoculated CFU. All experiments were performed in triplicate. Statistical analysis was performed using the unpaired t test to compare adherence levels between two strains as noted in the text.

4.3 Results

4.3.1 K. kingae PilC1 and PilC2 contain a calcium-binding site.

Based on sequence homology with P. aeruginosa PilY1 and N. meningitidis and N. gonorrhoeae PilC1 and PilC2, K. kingae encodes two PilC homologs called PilC1 and

PilC2. Interestingly, while PilC1 and PilC2 in the pathogenic Neisseria spp. are nearly identical, the K. kingae strain 269-492 PilC1 and PilC2 proteins share only limited

101

homology with each other and with other PilC-like protein family members (Kehl-Fie et al., 2008). Most of the similarity between members of this protein class resides in the C- terminal putative pilus-interacting half of the proteins (Orans et al., 2010). Interestingly, while K. kingae is a member of the Neisseriaceae family, the PilC1 and PilC2 C-terminal domains are more similar to P. aeruginosa PilY1 than to the Neisseria PilC1 and PilC2 proteins or even to each other (Figure 15A and C). Examination of the K. kingae PilC1 and PilC2 amino acid sequences revealed one predicted Ca-binding site in each protein

(Figure 15B and D). PilC1 contains a nine residue site (922-DVNRDGVYD-930) that is similar to the Ca-binding loop in P. aeruginosa PilY1 (Figure 15B), and PilC2 contains a

12-residue site (1433-DTNGDGKFTEAD-1444) with strong homology to the canonical

EF-hand human calmodulin Ca-binding loop (Figure 15D). An additional less conserved 12-residue calmodulin-like Ca-binding site was also identified in PilC2 (1114-

DLTNPTDLTESD-1125) (Figure 16A).

102

Figure 15: Comparison of K. kingae PilC1 and PilC2 with P. aeruginosa PilY1.

(A) Alignment of K. kingae PilC1 with P. aeruginosa PilY1 highlighting the C-terminal region with the greatest amount of homology (in black). (B) Alignment of the K. kingae PilC1 putative Ca-binding site with the calcium-chelating residues of P. aeruginosa PilY1 and the potential calcium chelating residues of PilC1 shown in bold. (C) Alignment of K. kingae PilC2 with P. aeruginosa PilY1 highlighting the C-terminal region with the greatest amount of homology (in black). (D) Alignment of the K. kingae PilC2 putative Ca-binding site with the calcium chelating residues of human calmodulin and the potential calcium-chelating residues of PilC2 shown in bold.

103

Figure 16: K. kingae PilC2 has a separate region resembling a calcium binding domain that does not bind calcium.

(A) Alignment of a K. kingae PilC2 putative Ca-binding site with the human calmodulin Ca-binding site. (B) Calcium competition binding assays using Oregon Green was performed with WT, D1125A, and D1125K recombinant PilC2868-1502.

To determine if the putative PilC1 Ca-binding site binds calcium, recombinant fragments of PilC1 were expressed in E. coli, purified, and tested for Ca-binding potential in vitro. A C-terminal PilC1 fragment containing amino acids 739-1047 was expressed as a maltose-binding protein (MBP) fusion, PilC1739-1047-MBP, and purified.

104

As shown in Fig. 2A, PilC1739-1047-MBP exhibited a binding affinity for calcium with a Kd of 530 nM, confirming that PilC1 binds calcium. Orans et al. demonstrated that mutation of the terminal bidentate aspartic acid residue in the PilY1 nine amino acid Ca-binding loop to alanine eliminated Ca-binding and created a calcium-free PilY1 state, while mutation of this residue to lysine also eliminated Ca-binding but created a potential calcium-bound mimic PilY1 state (Orans et al., 2010). As shown in Figure 17A, mutation of the terminal aspartic acid residue D930 in the K. kingae PilC1 Ca-binding site to either alanine or lysine eliminated specific Ca-binding. The mutant PilC1739-1047D930A-MBP and PilC1739-1047D930K-MBP proteins exhibited wild type circular dichroism (CD) spectra

(Figure 18), indicating that the overall structure of these fusion proteins was not altered.

Figure 17: K. kingae PilC1 and PilC2 contain calcium-binding sites.

Calcium competition binding assays using Oregon Green was performed with WT, D930A, and D930K recombinant PilC1739-1047-MBP (A); and WT, D1444A, and D1444K recombinant PilC2868-1502 (B). Binding curves were modeled to one-site competition or linear line. Error represents standard error of the mean.

105

Figure 18: K. kingae PilC1 mutations do not destabilize protein global structure.

Purified wild type PilC1739-1047-MBP and the D930A and D930K mutants were chelated or present with a fixed amount of calcium and were examined by circular dichroism. Molar ellipticity values were calculated from wavelength scans of each protein and compared.

In order to characterize PilC2, a recombinant fragment containing amino acids

868-1502 was purified. As shown in Figure 17B, purified PilC2868-1502 exhibited a binding affinity for calcium with a Kd of 5.5µM, confirming that PilC2 binds calcium. Mutation of the terminal bidentate aspartic acid residue D1444 in the PilC2 Ca-binding loop to alanine or lysine eliminated Ca-binding. The PilC2868-1502 D1444A and PilC2868-1502D1444K proteins exhibited WT circular dichroism (CD) spectra and melting temperatures (Figure

19), indicating that the overall structure of these proteins was not altered. The less conserved 12 amino acid site located at position 1114-1125 was demonstrated to have no effect on Ca-binding (Figure 16B).

106

Figure 19: K. kingae PilC2 mutations do not destabilize protein global structure.

(A) Purified wild type PilC2868-1502 and the D1444A and D1444K mutants where chelated or present with a fixed amount of calcium were examined by circular dichroism. Molar ellipticity values were calculated from wavelength scans of each protein and compared. (B,C) Wild type PilC2 and mutants where chelated or present with a fixed amount of calcium were observed at λ214 to measure the Tm of protein unfolding with (B) representing the curves and (C) representing the Tm values.

Taken together, these data indicate that PilC1 contains a functional Ca-binding site located in the region corresponding to amino acids 922-930 and PilC2 contains a functional Ca-binding site located in the region corresponding to amino acids 1433-1444.

107

4.3.2 Influence of K. kingae PilC1 and PilC2 Ca-binding sites on type IV pilus expression.

To investigate the influence and role of the PilC1 and PilC2 Ca-binding sites on type IV pilus expression and other type IV pilus phenotypes in K. kingae, we generated a series of mutants in the KK03 strain background, a naturally occurring spreading/corroding derivative of 269-492 that expresses stable levels of type IV pili.

The pilC1 and pilC2 genes are located in physically separate regions of the K. kingae genome, as highlighted in the diagrams of the mutant derivatives of KK03 in Figure 20.

To investigate the phenotypes associated with an individual pilC locus, Ca-binding site mutations were introduced into the pilC locus of interest with a nearby antibiotic resistance marker in a strain background with the other pilC gene deleted. Strains containing Ca-binding mutations in both PilC1 and PilC2 were also generated for investigation. In addition to the strains diagrammed in Figure 20, a previously described ΔpilF type IV pilus assembly ATPase mutant that is unable to assemble type

IV pili was used as a control strain (Kehl-Fie et al., 2008).

108

Figure 20: Representation of the pilC1 and pilC2 loci of K. kingae strain KK03 derivatives used in this study.

The pilC1 and pilC2 genes are present at separate loci in the genome. The ΔpilC1 and ΔpilC2 mutants are marked with a tetracycline resistance cassette (TetR) and a kanamycin resistance cassette (KanR), respectively. Both deletion mutations were combined to generated the ΔpilC1IΔpilC2 double mutant. Strains ΔpilC2/PilC1D930A and ΔpilC2/PilC1D930K, with ΔpilC2/PilC1mark (WT pilC1 with an erythromycin resistance cassette (ErmR) upstream of the promoter region) used as a control, were generated to study the role of the PilC1 Ca-binding site in the absence of pilC2. Strains ΔpilC1/PilC2D1444A and ΔpilC1/PilC2D1444K, with ΔpilC1/PilC2mark (WT pilC2 with the KanR marker immediately downstream of pilC2) used as a control, were generated to study the role of the PilC2 Ca-binding site in the absence of pilC1. Mutations were combined to generate strains PilC1D930A/PilC2D1444A and PilC1D930K/ PilC2D1444K.

109

To assess production of type IV pili, fibers were sheared from the surface of equal OD600 suspensions of K. kingae KK03 derivatives and separated using SDS-PAGE, and the major pilin subunit was stained with Coomassie blue. In earlier work we demonstrated that the major band at ~14kDa is PilA1, the major pilin subunit (Kehl-Fie et al., 2009). As shown in Figure 21, while the ΔpilF and ΔpilC1/ΔpilC2 strains were non- piliated, all strains that expressed at least one PilC protein, either WT or a Ca-binding mutant, were piliated. In some strains, there appear to be slight reductions in quantity of extracellular PilA1 relative to the parental strains, including ΔpilC2/PilC1D930A,

ΔpilC2/PilC1D930K, ΔpilC1/PilC2D1444A, and PilC1D930A/PilC2D1444A, indicating a slight defect in surface pilus expression (Figure 21). However, all strains examined did express surface pili, demonstrating that Ca-binding site mutant PilC1 and PilC2 are still able to promote piliation. These data as a whole indicate that at least one PilC protein is required to promote piliation and that the Ca-binding sites are dispensable for piliation.

Figure 21: All PilC1 and PilC2 Ca-binding site mutants express surface pili.

Pili were sheared from the surface of K. kingae strain KK03 derivatives and the major pilin subunit, PilA1, was visualized with Coomassie blue staining following SDS-

110

Figure 21 (continued)

PAGE. Strain ΔpilF and strain ΔpilC1/ΔpilC2 both express PilA1 but fail to assemble surface pili and serve as controls.

4.3.3 The PilC1 and PilC2 Ca-binding sites influence twitching motility.

We next wanted to assess the role of the PilC Ca-binding sites on twitching motility. Using a modified agar plate stab assay originally developed for use in P. aeruginosa (Alm et al., 1995), we measured the spreading zone at the plate-agar interface as a readout for twitching motility. We demonstrated the utility of this method for assessing twitching motility by showing that a strain lacking PilT, the type IV pilus retraction ATPase, still expressed surface pili but was unable to form a spreading zone

(Chapter 2). As shown in Figure 22, ΔpilC2 and the control ΔpilC2/PilC1mark strains produced statistically significantly larger twitching zones than did the parent strain (p <

0.05) while the ΔpilC1 and ΔpilC1/PilC2mark strains produced twitching zones that were slightly smaller but were not statistically significantly different compared to the parent strain. These data indicate that PilC1 and PilC2 differentially control twitching motility. Interestingly, the PilC1D930A and PilC1D930K mutations exhibited significantly reduced twitching motility compared to the parent and wild type strains (p

< 0.05), similar to the non-twitching ΔpilF and ΔpilC1/ΔpilC2 strains. In contrast, strain

ΔpilC1/PilC2D1444A demonstrated no defect in twitching motility compared to the parental strain ΔpilC1 and control ΔpilC1/PilC2mark, while strain ΔpilC1/PilC2D1444K

111

had a reduced twitching zone, although this difference was not statistically significant (p

>0.05). Reduced twitching zones were also evident in the PilC1D930A/PilC2D1444A and

PilC1D930K/PilC2D1444K double mutants compared to the parental and wild type strains. Despite the smaller twitching zones in these three mutants, the zones were significantly larger than the zones produced by the twitching-deficient controls (p <0.05), indicating reduced and not absent twitching motility. Together these data demonstrate that the PilC1 Ca-binding site is essential for twitching motility, while the PilC2 Ca- binding site has a minor influence on twitching motility.

Figure 22: The PilC1 and PilC2 Ca-binding sites influence twitching motility.

K. kingae strain KK03 and derivatives were assessed for twitching motility using a modified agar plate stab assay. Twitching zone diameters were measured in triplicate, and averages were calculated from three independent experiments. Error

112

Figure 22 (continued) bars represent standard error of the mean. Statistical analysis was performed using the unpaired t test to compare twitching motility zones between two strains as noted: *p < 0.05 for ΔpilC2 and ΔpilC2/PilC1mark vs. WT and p < 0.05 for ΔpilC2/PilC1D930A and ΔpilC2/PilC1D930K vs. ΔpilC2 and ΔpilC2/PilC1mark.

4.3.4 The PilC1 Ca-binding site but not the PilC2 Ca-binding site is necessary for adherence.

We next chose to investigate the influence of PilC1 and PilC2 Ca-binding on adherence to human epithelial cells. Using a previously established adherence assay

(Kehl-Fie et al., 2008, Kehl-Fie et al., 2009), we determined the relative adherence levels to

Chang epithelial by the series of Ca-binding site mutants. Of note, we previously observed reduced adherence by the 269-492ΔpilC1 and 269-492pilC2::aphA3 mutants compared to strain 269-492 (Kehl-Fie et al., 2008). In contrast, in this study the ΔpilC1 derivative of KK03 had only a slight defect in adherence, and the ΔpilC2 derivative of

KK03 had no defect in adherence compared to the KK03 parent (Figure 23).

Examination of the adherence to Chang cells by the PilC Ca-binding mutants revealed stark differences between PilC1 and PilC2 (Figure 23). Strains ΔpilC2/PilC1D930A and

ΔpilC2/PilC1D930K exhibited marked deficiencies in adherence compared to the ΔpilC2 and ΔpilC2/PilC1mark control strains (p <0.05) and adhered at levels similar to adherence by the non-piliated ΔpilF and ΔpilC1/ΔpilC2 controls. However, strains

ΔpilC1/PilC2D1444A and ΔpilC1/PilC2D1444K displayed normal adherence, comparable adherence by to the parental ΔpilC1 and ΔpilC1/PilC2mark control strains. The

113

PilC1D930A/PilC2D1444A and PilC1D930K/PilC2D1444K double mutants also adhered at levels similar to the ΔpilC1 strains that express either WT or mutant PilC2, indicating that PilC2 is able to promote adherence in the presence of Ca-binding mutant PilC1, regardless of PilC2 Ca-binding state. These results demonstrate that the PilC1 Ca- binding site is necessary and the PilC2 Ca-binding site is dispensable for adherence to

Chang cells.

Figure 23: The PilC1 but not PilC2 Ca-binding site is essential for adherence.

K. kingae strain KK03 and derivatives were investigated for adherence to Chang human epithelial cells using a quantitative adherence assay. Experiments were performed in triplicate, and averages were calculated from three independent experiments. Error bars represent standard error of the mean. Statistical analysis was performed using the unpaired t test to compare adherence between two strains as noted: *p < 0.05 for ΔpilC2/PilC1D930A and ΔpilC2/PilC1D930K vs. ΔpilC2 and ΔpilC2/PilC1mark. 114

4.4 Discussion

K. kingae expresses type IV pili that mediate adherence to human cells (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009). The PilC-like proteins play key roles in type IV pilus biology in K. kingae and numerous other gram-negative bacteria of medical importance

(Rudel et al., 1992, Rudel et al., 1995a, Rudel et al., 1995b, Ryll et al., 1997, Heiniger et al.,

2010, Johnson et al., 2011). In this study we found that PilC1 and PilC2 bind calcium in vitro and that significant differences in type IV pilus phenotypes exist in the PilC1 and

PilC2 Ca-binding site mutants. Both wild type and Ca-binding mutant PilC1 and PilC2 were able to promote pilus expression when expressed individually. However, Ca- binding mutant PilC1 eliminated twitching motility and adherence, while Ca-binding mutant PilC2 had little or no effect on twitching motility and no effect on adherence.

Strains that expressed both PilC1 and PilC2 Ca-binding mutants demonstrated phenotypes identical to the PilC2 single mutant. These data demonstrate important differences in PilC1 and PilC2 control of type IV pilus phenotypes in K. kingae.

In the pathogenic Neisseria spp., PilC1 and PilC2 are highly homologous to each other. N. gonorrhoeae PilC1 and PilC2 function interchangeably in promoting pilus biogenesis, natural competence, and adherence (Rudel et al., 1992, Rudel et al., 1995a). N. meningitidis PilC1 and PilC2 both promote pilus expression and natural competence but differentially promote adherence to certain cell types (Nassif et al., 1994, Rahman et al.,

1997, Ryll et al., 1997, Morand et al., 2009). Sequence analysis suggests that differences in 115

PilC1- and PilC2-mediated adherence is based in the N-terminal portion of the proteins

(Morand et al., 2001). Given the highly homologous nature of the PilC proteins in the pathogenic Neisseria species, especially in the C-terminal portion of the protein and in the nine amino acid putative Ca-binding loop, it is likely that Ca-binding is an important factor for Neisseria type IV pilus biology. In contrast, the related Neisseriaceae family member K. kingae expresses two PilCs that share only limited sequence homology and are differentially controlled by Ca-binding. We speculate that differences in host niches may be a driving factor influencing the mechanism of PilC function. K. kingae inhabits the posterior pharynx as a commensal but tends to cause disease primarily in the joints and bones, two calcium-rich sites. While normal human blood plasma has a free calcium concentration of 1.03 - 1.30 mM, calcium in human knee joint fluid has been shown to be 4 mM or higher (Maroudas, 1979). These high calcium environments may alter the Ca-dependent functions of PilC1, while leaving PilC2 unaffected and able to promote twitching motility and adherence.

The first report that calcium plays a key role in regulating type IV pilus expression and twitching motility examined PilY1, a PilC-like protein in P. aeruginosa

(Orans et al., 2010). The researchers crystallized the C-terminal domain of PilY1 and identified a nine amino acid Ca-binding site. These investigators showed that mutation of the bidentate aspartic acid residue in the Ca-binding loop to alanine (D859A) resulted in significantly reduced pilus expression and twitching motility, mimicking the effect of

116

a ΔpilY1 mutation. Mutation of the same residue to lysine resulted in increased pilus expression and reduced twitching motility, mimicking a ΔpilT pilus retraction mutation.

Interestingly, K. kingae PilC1 contains a similar nine amino acid site, and mutation of the predicted bidentate residue to alanine or lysine did eliminate twitching motility but did not have a major impact on surface pilus expression, thus contrasting with PilY1.

Hence, while control of twitching motility in PilY1 is due to the influence of the Ca- binding site on pilus expression, the K. kingae PilC1 Ca-binding site directly impacts twitching motility and adherence, not surface pilus expression. Interestingly, expression of either the alanine or lysine mutant protein in K. kingae resulted in identical phenotypes, again contrasting with PilY1. Orans et al. modeled the PilY1 D859K mutation and found that the amino group of the lysine side chain potentially mimics a calcium ion, creating a pseudo-calcium-bound PilY1 state. One potential reason for the differences observed between the K. kingae PilC1D930K mutant and the PilY1D859K mutant may be in differences in structure of the regions surrounding the Ca-binding loop in the two proteins, as the primary amino acid sequence of the loop is largely conserved (Figure 15B).

To dissect the roles of PilC1 and the Ca-binding site in K. kingae type IV pilus- mediated twitching motility and adherence, it is important to examine the relationship between the two phenotypes. We have found that disruption of the K. kingae type IV pilus retraction machinery, pilTU, resulted in increased surface piliation, significantly

117

reduced adherence, and loss of twitching motility (Chapter 2). Interestingly, adherence was reduced to approximately 50% of wild type rather than eliminated, as observed with the ΔpilC2/PilC1D930A and ΔpilC2/PilC1D930K mutants (Figure 23). Taken together, these data suggest that the adherence defect of these two mutants is not completely a result of the defect in twitching motility but is a consequence of altered

PilC1 adherence-promoting activity. In accord with these findings, Johnson et al. found that in vitro Ca-binding by P. aeruginosa PilY1 influences RGD-mediated PilY1-binding to integrins, supporting a specific role for the nine amino acid Ca-binding site in adherence in this protein class (Johnson et al., 2011).

In contrast to the situation with PilC1, the K. kingae PilC2 Ca-binding site is dispensable for twitching motility and adherence. We originally identified two potential

Ca-binding sites in PilC2 with homology to the 12-amino acid Ca-binding loop of human calmodulin and no sites with homology to the PilY1-like loop. In vitro analysis demonstrated that only one of these sites was functional. However, our data clearly show that PilC2 is able to promote pilus expression, twitching motility, and adherence when the confirmed Ca-binding site is mutated. These results suggest that K. kingae

PilC1 and PilC2 utilize different mechanisms in relation to Ca-binding to promote twitching motility and adherence.

At this time we can only speculate on the potential mechanisms by which Ca- binding by PilC1 and PilC2 influences type IV pilus phenotypes. In the case of PilC1, we

118

suggest that Ca-binding or Ca-responsiveness may be essential for proper protein structure. In this model, the Ca-binding site mutant PilC1 has an altered protein conformation that retains the ability to promote pilus assembly but is not able to promote twitching motility and adherence. In contrast, PilC2 appears to function completely independently of Ca-binding at the site identified in this report (Figure 16B).

We envision two potential scenarios to explain our findings: (1) PilC2 contains an additional cryptic Ca-binding site in the N-terminal 857 amino acids that were not included in the PilC2 fragment used for the in vitro Ca-binding studies; or (2) PilC2 utilizes a novel mechanism that promotes twitching motility and adherence regardless of functionality of the Ca-binding site. In the report by Johnson et al. that identified

RGD-mediated PilY1-binding to integrins, analysis revealed a second Ca-binding site in

PilY1 (Johnson et al., 2011). While sequence analysis of PilC2 did not reveal any additional potential Ca-binding sites, we cannot rule out this possibility.

The results presented in this report establish key differences in the role of Ca- binding in the function of the K. kingae PilC1 and PilC2 proteins. However, the precise mechanisms used by these proteins to control type IV pilus phenotypes remains unclear and is a subject of future investigation. A greater understanding of K. kingae PilC1 and

PilC2 function may provide insights into species-specific control mechanisms of type IV pilus phenotypes and suggest novel strategies for prevention or treatment of K. kingae disease.

119

Acknowledgements

This work has been submitted for publication as referenced below:

Porsch, E.A., Johnson, M.D., Broadnax, A.D., Garrett, C.K., Redinbo, M.R., and St. Geme III, J. W. Differential control of Kingella kingae type IV pilus phenotypes through the calcium-binding sites of PilC1 and PilC2. J. Bacteriol. (In Revision).

120

5. The Role of PilA2 in Kingella kingae type IV pilus biology

5.1 Introduction

Kingella kingae is a fastidious gram-negative bacterium in the Neisseriaceae family. Advances in culture-based and molecular-based diagnostics have lead to increased recognition of K. kingae as an important pathogen in the pediatric population

(Yagupsky et al., 1992, Yagupsky et al., 1995b, Moumile et al., 2003, Gene et al., 2004,

Yagupsky, 2004, Verdier et al., 2005, Chometon et al., 2007, Lehours et al., 2011, Yagupsky et al., 2011). Previous studies have shown that K. kingae expresses type IV pili, a key colonization and virulence factor in a variety of gram-negative bacteria species (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009, Kehl-Fie et al., 2010). These organelles confer multiple phenotypes important for bacterial pathogenic processes, including adherence, twitching motility and natural competence (Mattick, 2002, Burrows, 2005, Pelicic, 2008).

Type IV pilus systems are complex and involve as many as 40 gene products for proper expression and function, as exemplified in P. aeruginosa (Alm et al., 1997, Beatson et al.,

2002, Mattick, 2002). The fibers themselves are composed primarily of a single protein subunit termed pilin (encoded by pilE or pilA depending on the specific species under study), that is translated as a pre-pilin containing a signal sequence that targets the protein to PilD, an inner membrane associated pre-pilin peptidase that cleaves the signal sequence. Pilin is incorporated into fibers at the inner membrane, a process catalyzed through ATP hydrolysis by PilF, the type IV pilus assembly ATPase. The growing fibers 121

are secreted through an outer membrane pore-forming PilQ oligomer, the type IV pilus secretin. A key distinguishing feature of type IV pili is the ability of fibers to be retracted back through the outer membrane. This process is catalyzed by PilT, the type

IV pilus retraction ATPase.

While the core components necessary for type IV pilus expression are largely conserved throughout type IV pilus-expressing organisms, key differences exist in genetic arrangement, regulation, and accessory factors. Specifically, the copy number of pilin and pilin-like genes varies from species to species. For example, P. aeruginosa contains only one major pilin gene, termed pilA, that encodes the only major structural subunit of the fibers (Johnson et al., 1986). The pathogenic Neisseria species contain only one complete major pilin gene termed pilE but also contain over a dozen partial pilin gene sequences throughout their genomes (Meyer et al., 1984, Haas et al., 1986, Perry et al., 1987, Perry et al., 1988). Non-reciprocal homologous recombination events between the intact pilin locus and the partial pseudopilins serves as a source of pilin polypeptide amino acid sequence diversity in the population (Swanson et al., 1986, Seifert et al.,

1988a, Seifert et al., 1988b, Gibbs et al., 1989). In the Moraxella spp., Pseudomonas stutzeri,

Eikenella corrodens, and K. kingae, more than one intact pilin gene is present in the genome (Marrs et al., 1985, Rao et al., 1993, Graupner et al., 2001, Kehl-Fie et al., 2008).

Certain Moraxella species encode two intact pilin genes and utilize a 2.1 kb inversion event to switch between expression of the two genes (Marrs et al., 1988, Fulks et al., 1990,

122

Marrs et al., 1990). P. stutzeri, E. corrodens and K. kingae encode two pilin genes adjacent to each other, with the first gene, pilA1, encoding the major pilin (Hood et al., 1995,

Graupner et al., 2001, Kehl-Fie et al., 2008). In P. stutzeri, the downstream gene, termed pilAII, was shown to negatively influence natural genetic transformation, as a pilAII insertional mutant had an ~20-fold increase in transformation frequency compared to wild type but had wild type levels of twitching motility (Graupner et al., 2001). In contrast, disruption of the gene downstream of pilin, termed pilA2, in E. corrodens had no effect on natural competence and twitching motility (Villar et al., 2001).

The K. kingae pilin locus contains four genes, namely recJ, pilA1 (encoding pilin), pilA2, and fimB. Previous work utilizing a pilA2 mutant found no defects in type IV pilus expression and adherence to human epithelial cells (Kehl-Fie et al., 2008). In this report, we expand on the previous work and demonstrate that pilA2 is not essential for twitching motility and has an unclear impact on natural genetic transformation.

Deletion of pilA1 and the intergenic region between pilA1 and pilA2, resulting in pilA2 under control of the pilA1 promoter, demonstrated that PilA2 can form fibers that mediate intermediate adherence to Chang epithelial cells when the pilA1 promoter is highly active.

5.2 Methods

Bacterial strains and growth conditions

123

Strains used in this study are listed in Table 8. K. kingae strains were stored at

-80°C in brain heart infusion broth (BHI) with 30% glycerol. E. coli strains were stored at

-80°C in Luria-Bertani (LB) broth with 15% glycerol. K. kingae strains were routinely cultured at 37°C with 5% CO2 on chocolate agar supplemented with 50 µg/ml kanamycin as appropriate. E. coli strains were routinely cultured at 37°C in LB broth or on LB agar supplemented with 100 µg/ml ampicillin or 50 µg/ml kanamycin as appropriate.

Table 8: Strains and plasmids used in this study.

Strain or plasmid Description Source E. coli Dh5α Φ80dlacZ_M15 _lacU169 deoR Life recA endA1 Technologies K. kingae 269-492 Septic arthritis clinical isolate (prototype lab (Kehl-Fie et al., strain) 2007) 269-492pilA1 269-492 with aphA3 insertion in pilA1 (Kehl-Fie et al., 2008) KK01 Naturally occurring nonspreading (Kehl-Fie et al., /noncorroding derivative of 269-492 2007) KK01pilA1 KK01 with aphA3 insertion in pilA1 (Kehl-Fie et al., 2009) KK03 Naturally occurring spreading/corroding (Kehl-Fie et al., derivative of 269-492 2007) KK03pilA1 KK03 with aphA3 insertion in pilA1 (Kehl-Fie et al., 2009) KK01pilA2 Strain KK01 with aphA3 insertion in pilA2 This study KK03pilA2 Strain KK03 with aphA3 insertion in pilA2 This study KK01recJ Strain KK01with aphA3 insertion in recJ (Kehl-Fie et al., 2009) KK03recJ Strain KK03with aphA3 insertion in recJ (Kehl-Fie et al., 2009) KK01ΔpilA1+ Strain KK01 with aphA3 in recJ, deleted for This study

124

pilA1 and with pilA2 under control of the pilA1 promoter KK03ΔpilA1+ Strain KK03 with aphA3 in recJ, deleted for This study pilA1 and with pilA2 under control of the pilA1 promoter Plasmids pGEX6P GST-fusion expression vector GE Healthcare pGEX-PilA2 GST-PilA1 fusion expression vector This study

Strain construction

Gene disruptions were introduced into K. kingae by natural transformation.

Following recovery of transformants by plating on selective media, correct localization of the gene disruptions was confirmed by Southern blotting or PCR. All plasmids are listed in Table 8. The pilA1 and recJ genes were disrupted as described previously (Kehl-

Fie et al., 2008, Kehl-Fie et al., 2009). To generate the ΔpilA1+ mutation, plasmid pPRBK

(Kehl-Fie et al., 2009) containing the entire pilin locus was subjected to site-directed mutagenesis using primers deltapilA1 fwd and deltapilA1 rev to delete the entire pilA1

ORF beginning at the start codon and ending at the last nucleotide in the intergenic region between pilA1 and pilA2. The resulting plasmid was designated pPRBK-ΔpilA1+ and was sequenced to ensure that only the desired mutation was introduced into the pilin locus.

Type IV pilus purification

K. kingae pili were purified using a modification of the method described for the purification of Eikenella corrodens type IV pili (Hood et al., 1995). K. kingae strain 269-492

125

was grown for 18 h on 20 TSA II chocolate agar plates, and bacterial growth was scraped from the plates and suspended in 20 ml of 150 mM ethanolamine, pH 10.5 (EA buffer).

Pili were sheared from the bacterial surface by subjecting the bacterial suspension to a handheld homogenizer for 2 min. To remove bacterial cells and debris, the suspension was centrifuged for 10 min at 10,000 × g and then for 30 min at 10,000 × g. The resulting supernatant was subjected to precipitation with 10% ammonium sulfate. The precipitated pili were resuspended in EA buffer to 1/10 of the original volume and then dialyzed overnight in 2 liters of EA buffer. The suspension again was subjected to precipitation with 10% ammonium sulfate, and the purified pili were examined by SDS-

PAGE.

Generation of anti-PilA2 polyclonal antiserum

A fragment of the pilA2 gene encoding amino acids 48-133 was amplified from K. kingae strain 269-492 genomic DNA with primers pilA2Ab F and pilA2 Ab R, digested with EcoRI and BamHI, and ligated into EcoRI/BamHI-digested pGEX6P, generating pGEX-PilA248-133. The GST-PilA248-133 fusion protein was affinity purified from E. coli whole cell lysates on glutathione-agarose following induction with IPTG. After cleavage with PreScission protease (GE Healthcare) to release PilA248-133 from GST, the cleaved fragment was resolved on a 15% SDS-PAGE gel, stained with Coomassie blue, excised, and injected into a guinea pig, generating antiserum GP98 (Cocalico Biologicals).

Western Blot analysis

126

To assess PilA2 expression, K. kingae strains were grown for 17 - 18 hours on chocolate agar and resuspended to an optical density at 600 nm of 0.8 and then analyzed by Western blotting using antiserum GP98. To confirm equal loadings and uniform transfer to nitrocellulose, all membranes were stained with Ponceau S prior to Western blotting.

Transmission electron microscopy

To assess surface piliation, K. kingae strains were prepared for negative staining transmission electron microscopy as described previously (Kehl-Fie et al., 2009). Briefly, bacteria were resuspended in ammonium acetate, pH 7.4, and allowed to absorb onto

Formvar/carbon coated-200 mesh grids for 1min. The grids were subsequently stained with 1% uranyl acetate for 30 seconds, excess liquid was wicked away, and the grids were air dried and examined on a Philips CM12 Transmission Electron Microscope

(TEI) at an accelerating voltage of 80 kV.

Twitching motility assays

Twitching motility was assessed by a modified agar plate stab assay originally developed for P. aeruginosa (Alm et al., 1995). Briefly, derivatives of K. kingae strains

KK01 and KK03 were incubated on chocolate agar for 17-18 hours, and growth was resuspended in 1x PBS to an OD600 of 1.0. One µl of the bacterial suspensions was stab inoculated to the bottom of tissue culture treated 100mm plates containing chocolate agar and incubated at 37°C with 5% CO2 for 2 days. Twitching motility competent

127

strains spread from the stab inoculation site at the plate/agar interface. The agar was carefully peeled away, and the plate was air-dried and stained with crystal violet to visualize the twitching motility zones.

Transformation assays

To assess transformation efficiency of K. kingae KK01 and KK03 derivatives, the ability of the bacteria to be transformed with an antibiotic resistance cassette from exogenous DNA was determined. Approximately 6.5x106 bacteria were incubated for 30 min at ambient temperature in 100µl BHI with 1µg of genomic DNA from strain 269-

492knh::tetM, which has a tetracycline resistance cassette in the knh gene. The transformation reactions were allowed to recover by addition of 100µl 20% lysed horse blood in BHI and incubated at 37°C for 2hrs. Dilutions were plated on tetracycline- containing chocolate agar plates, and the number of tetracycline resistant CFU was divided by the total input CFU to determine transformation efficiency. Transformation efficiencies are expressed as percentage of wild type.

Eukaryotic cell lines

Chang cells (Wong-Kilbourne derivative (D) of Chang conjunctiva, HeLa origin;

ATCC CCL-20.2) were cultivated at 37°C with 5% CO2 in media as described previously

(Kehl-Fie et al., 2007).

Adherence assays

128

Quantitative adherence assays were performed as described previously (Kehl-Fie et al., 2008, Kehl-Fie et al., 2009). Briefly, confluent and fixed cell monolayers in 24-well plates were inoculated with approximately 6.5x106 CFU of bacteria, and plates were centrifuged at 165 x g for 5 min and then incubated for 25 min at 37°C in 5% CO2.

Monolayers were washed 4 times with PBS to remove non-adherent bacteria and were then treated with 1x trypsin-EDTA (Sigma) for 20 min at 37°C to release adherent bacteria. Appropriate dilutions were plated on chocolate agar, and percent adherence was calculated by dividing the number of adherent CFU by the number of inoculated

CFU. All experiments were performed in triplicate. Statistical analysis was performed using the unpaired t test to compare adherence levels between two strains as noted in the text.

5.3 Results

5.3.1 PilA2 is not necessary for twitching motility

While our lab has previously demonstrated no role of PilA2 in pilus expression and adherence, we next chose to examine the influence of PilA2 on twitching motility.

We utilized a modified agar stab assay based on methods originally developed to assay twitching motility in P. aeruginosa (Alm et al., 1995). Strains KK01 (low-piliated) and

KK03 (high-piliated) derivatives of strain 269-492 were used in these experiments to control for piliation level. Strains KK01, KK03 and two pilA2 mutants in each background KK01pilA2(a/b) and KK03pilA2(a/b) were examined with pilA1 mutants

129

used as non-twitching controls. As shown in Figure 24, both KK01pilA2 mutants spread from the stab site identically to the KK01 parental strain, while the KK01pilA1 control strain did not spread. Strain KK03pilA2 mutants likewise produced similar spreading zones compared to the KK03 parental strain (Figure 24). These data indicate that PilA2 is dispensable for twitching motility in K. kingae strains KK01 and KK03.

Figure 24: Twitching motility zones of KK01 and KK03 pilA2 mutants.

Strains KK01, KK01pilA1, KK01pilA2(a), KK01pilA2(b), KK03, KK03pilA1, KK03pilA2(a), and KK03pilA2(b) were stab inoculated in chocolate plates and the bacterial twitching zones were visualized with crystal violet staining.

5.3.2 The role of PilA2 in natural transformation is unclear.

We next investigated the role of PilA2 in natural genetic transformation. In other naturally competent type IV pilus-expressing organisms, minor pilins and other pilin- like gene products have been shown to influence the ability of the organism to be transformed with exogenous DNA. K. kingae transformation efficiency was investigated by allowing KK01 and KK03 derivatives to take up exogenous DNA containing a tetracycline resistance cassette, plating on tetracycline-containing chocolate agar, and dividing the number of tet-resistant CFU by the number of input CFU. Transformation 130

efficiencies of pilA2 mutants were then expressed as a percentage of the parent strain efficiencies. As shown in Figure 25, there was a wide degree of variation in the transformation efficiencies in the KK01pilA2(a) and KK01pilA2(b) mutants compared to the parent KK01 strain and in KK03pilA2(a) KK03pilA2(b) compared to the parent KK03.

While the transformation efficiency of strain KK01pilA2(a) was ~200% of the transformation efficiency of KK01, the transformation efficiency of strain KK01pilA2(b) was only ~20% of KK01. The KK03pilA2(a) transformation efficiency was only ~3% of

KK03 but the transformation efficiency of KK03pilA2(b) was ~47% of KK03. As expected, the control KK01pilA1 and KK03pilA1 strains were non-transformable due to the defect in type IV pilus expression. These variations make it difficult draw conclusions about the role of PilA2 in K. kingae transformation.

250%

200%

KK01 150% KK03 100%

50% Transformation efficiency Transformationefficiency ofWT % as 0% WT pilA1 pilA2(a) pilA2(b) Strain

Figure 25: Relative transformation efficiencies of KK01 and KK03 pilA2 mutants.

131

Figure 25 (continued)

Transformation efficiency of strains KK01, KK01pilA1, KK01pilA2(a), KK01pilA2(b), KK03, KK03pilA1, KK03pilA2(a), and KK03pilA2(b) determined. Transformation efficiencies are displayed as a percentage of the parental strain.

5.3.3 PilA2 can be overexpressed when under control of the PilA1 promoter

Previous work established that pilA2 is transcribed at a much lower level than is pilA1 (Kehl-Fie et al., 2008). To determine if pilA2 can substitute for pilA1, a ΔpilA1+ deletion mutation was introduced into the pilin locus of strains KK01 and KK03. The region deleted contained the entire pilA1 ORF, beginning with the start codon and ending with the last nucleotide in the intergenic region between pilA1 and pilA2. The resultant mutant pilin locus had a deletion of pilA1 and had pilA2 under transcriptional control of the pilA1 promoter (diagrammed in Figure 26).

wild type pilin locus

Figure 26: Representation of the (A) wild type and (B) ΔpilA1+ pilin loci.

132

To determine if pilA2 expression was increased in the ΔpilA1+ strains, Western blot analysis with polyclonal antiserum GP98 raised against amino acids 48-133 of PilA2 was performed. Due to sequence similarities between PilA1 and PilA2, GP98 had some cross reactivity to PilA1 in later production bleeds from the guinea pig. As shown in

Figure 27, no reactivity was evident for PilA2 in strain KK01 or KK03, indicating that

PilA2 was either not expressed or expressed at a level beneath the limit of detection for this assay. GP98 did cross react with PilA1 but only in the KK03 strain background

(Figure 27, top panel). Using the previously described anti-PilA1GP65 antiserum (Kehl-

Fie et al., 2009), PilA1 was clearly present in KK01 and KK03 but not in the ΔpilA1+ mutants, as expected (Figure 27, bottom panel). In both KK01ΔpilA1+ and KK03ΔpilA1+, a reactive band at the correct MW of PilA2 was present, indicating that PilA2 can be overexpressed when under control of the pilA1 promoter.

Figure 27: Western blot analysis of KK01 and KK03 ΔpilA1+ with GP98 and GP65.

133

Figure 27 (continued)

Whole cell sonicates of strains KK01, KK01ΔpilA1+, KK03, and KK03ΔpilA1+ were probed using anti-PilA2 antiserum GP98 (top panel) and anti-PilA1 antiserum GP65 (bottom panel) with strain KK01pilA1 used as a negative control.

5.3.4 PilA2 is not present or present at levels below the detection limit in purified pili.

While GP98 cannot detect PilA2 in the KK01 or KK03 backgrounds in whole cell sonicates, we next investigated if PilA2 is detectable in purified pili fractions. Pili were sheared from the surface and purified from strain KK03 as described previously (Kehl-

Fie et al., 2009). To remove cross-reactive PilA1 antibodies in GP98, the antisera was absorbed with an acetone powder of strain KK03pilA2. As shown in Figure 28, the pili preparation (pili prep) was reactive with GP65 (top panel), as shown previously (Kehl-

Fie et al., 2009) but no PilA2 was detected with GP98 (bottom panel). These data indicate that PilA2 is either not present in K. kingae strain 269-492 pili or is present at a low abundance that is not detectable with GP98.

134

Figure 28: GP65 and GP98 Western blot analysis of purfied pili.

Western blot analysis of strain KK03 purified pili preparation with anti-PilA1 GP65 (top panel) and anti-PilA2 GP98 (bottom panel) using strain KK03ΔpilA1+ as a positive control for PilA2.

5.3.5 PilA2 can form surface fibers when overexpressed.

Given that PilA2 can be overexpressed when under control of the pilA1 promoter, we next examined the ability of the ΔpilA1+ mutants to form surface fibers.

Using negative staining transmission electron microscopy, 20 KK01ΔpilA1+ and 20

KK03ΔpilA1+ bacteria were examined for the presence of surface fibers. These results are summarized in Figure 29 and show that only one of the 20 KK01ΔpilA1+ bacteria examined had any surface fibers. The surface fibers on this organism were abnormally short and bunched in morphology. In contrast, 10 of 20 KK03ΔpilA1+ bacteria had surface fibers. Some fibers with wild type morphology were identified, but most of the fibers were intermediate in length and bunched. In addition, the number of surface

135

fibers was drastically reduced compared to KK03. Pili from a representative

KK03ΔpilA1+ organism are shown in Figure 29. Taken together, these data indicate that

PilA2 can form surface fibers, although largely different in morphology compared to wild type fibers, when pilA2 is expressed from the highly active pilA1 promoter in the

KK03 strain background.

Figure 29: Summary of transmission electron microsopy of ΔpilA1+ mutants.

Twenty KK01ΔpilA1+ and KK03ΔpilA1+ bacteria were examined by negative staining transmission electron microscopy for the presence and morphological examination of surface fibers. Arrows indicate surface fibers. Scale bar = 100nm.

5.3.6 PilA2 fibers can mediate significant adherence.

Given the fact that the ΔpilA1+ mutation in the KK03 background results in expression of atypical surface fibers, we next examined the ability of these mutants to adhere to Chang epithelial cells. Strains KK03 and KK03pilA1 were used as positive and negative controls, respectively. The quantitative adherence results are presented in

Figure 30. Strain KK01ΔpilA1+ adhered at a background level comparable to the non- piliated negative control strain KK03pilA1. However, strain KK03ΔpilA1+ adhered to 136

Chang cells significantly greater than the negative control, although reduced approximately 2/3rds compared to KK03 (60% vs. 20% ). These data demonstrate that the fibers present on the surface of strain KK03ΔpilA1+ are able to mediate appreciable adherence to Chang epithelial cells.

80 269KK03-492

70 pilA1KK03 - pilA1 60 KK01/ KK01ΔΔpilA1+pilA1+ 50 KK03/KK03Δ ΔpilA1+pilA1+ 40 30 20

Adherence (% (% inoculum) Adherence 10 0 1

Figure 30: Quantitative adherence of ΔpilA1+ strains to Chang epithelial cells.

Adherence to Chang epithelial cells by strains KK03, KK03pilA1, KK01ΔpilA1+, KK03ΔpilA1+. Experiments were performed in triplicate, and error bars represent standard error of the mean.

5.4 Discussion

Type IV pili are surface fibers expressed by a wide range of gram-negative bacteria, including many species of clinical importance. Previous studies have established that type IV pili are necessary for multiple phenotypes important for pathogenic processes, including adherence to host cells, natural competence and twitching motility. Many of the key components of type IV pilus biology have been

137

elucidated (see Chapter 1). However, the roles of other minor factors have not been thoroughly investigated. In this report, we explore the role and function of a second pilin gene, pilA2, in the pediatric pathogen K. kingae. We report that PilA2 is not involved in twitching motility, and may or may not be involved in natural genetic transformation. We show that PilA2 is not expressed at a level detectable with a polyclonal antisera in whole cell lysates or purified pilus preparations. In addition, we demonstrate that PilA2 can form pilus-like fibers when expressed from the highly active pilA1 promoter and that the fibers mediate reduced but significant adherence to Chang epithelial cells.

While the genetic arrangement of the region encoding the major pilin subunit

(pilin locus) varies across type IV pilus-expressing bacterial species, many organisms, including K. kingae, contain two pilin genes located adjacent to one another. We previously showed that PilA1 is the major pilin subunit in K. kingae and that PilA2 does not play a role in adherence to human epithelial cells, as a pilA2 mutant adheres at wild type levels (Kehl-Fie et al., 2008). In a study focused on type IV pilus phenotypes in a set of K. kingae clinical isolates, we sequenced the pilA1 and pilA2 genes in a selection of respiratory, blood, and joint isolates (Kehl-Fie et al., 2010). Interestingly, while substantial sequence diversity was identified in the variable region of pilA1 in the clinical isolates, pilA2 was highly conserved across the entire ORF. Considering that

PilA1, the major pilin subunit, is a secreted factor, it is likely that this protein is subjected

138

to select pressures in the host and changes in PilA1 amino acid sequence may play an important role in modulating interactions of K. kingae with host cells. However, the fact that the predicted PilA2 primary amino acid sequence is largely conserved across a diverse set of clinical isolates suggests that it may not be under selective pressure to the extent that PilA1 is. It is also possible that PilA2 sequence conservation is necessary for an in vivo K. kingae pathogenic process that we have yet to appreciate in our in vitro studies.

In other type IV pilus-expressing bacteria, pilin-like genes have been shown to influence multiple phenotypes. In P. aeruginosa the FimU, PilU, PilW, PilX, and PilE minor pilins were shown to be involved in initiation of fiber biogenesis and were demonstrated to be incorporated throughout the pilus fiber (Giltner et al., 2010). The

PilX minor pilin of N. meningitidis (a different gene, not a homolog of PilX of P. aeruginosa) is not essential for piliation but serves as an aggregative fiber component necessary for bacterial adherence to host cells (Helaine et al., 2005). The ComP minor pilin-like gene product in N. gonorrhoeae is necessary for sequence-specific, type IV pilus- dependent DNA uptake (Wolfgang et al., 1999). In contrast, the P. stutzeri PilAII minor pilin negatively regulates natural competence (Graupner et al., 2001). These examples highlight the variety of roles minor pilins play in type IV pilus biology. Interestingly, we have not been able to identify a role for PilA2 in K. kingae type IV pilus phenotypes.

While we observed a reduction in transformation efficiency in some of the pilA2

139

mutants, we also observed an increase in one of the mutants, suggesting that there may be unknown phase-variable events in the pilA2 mutants that influence natural transformation efficiency. Future experiments to refine the transformation assays will be necessary to further elucidate if PilA2 is involved in this process. Of key importance will be separating the two key events involved in natural competence, (1) DNA uptake and (2) recombination. If PilA2 is involved in natural competence, it likely influences this phenotype at the stage of DNA uptake, as has been reported in other type IV pilus- expressing organisms (Biswas et al., 1977, Aas et al., 2002, Luke et al., 2004).

Of particular intrigue in these studies is the finding that PilA2 can form pilus-like structures when PilA2 is over expressed and pilA1 is deleted. While it was extremely rare to see fibers on the surface of the KK01ΔpilA1+ mutant, we found that approximately 50% of the KK03ΔpilA1+ bacteria observed did have pilus-like surface structures. We attribute this difference to the fact that the pilA1 promoter is driving more pilA2 transcription in the KK03 background based on previous transcript level studies of pilA1 in KK01 vs. KK03 (Kehl-Fie et al., 2009). Given that KK03 produces greater than 10 pili per bacteria as assessed by TEM (Kehl-Fie et al., 2007, Kehl-Fie et al.,

2008, Kehl-Fie et al., 2009), the level of piliation in the KK03ΔpilA1+ mutant is drastically reduced compared to WT. This is likely due to structural differences in PilA2 that do not allow efficient incorporation of the PilA2 subunit through the post-translational processing and pilus assembly machinery. Interestingly, the KK03ΔpilA1+ mutant is

140

capable of appreciable adherence, indicating the pilus-like fibers have some functionality. However, preliminary data indicate that this mutant is non-transformable and does not twitch (data not shown).

The results presented here indicate that PilA2 does not play an essential role in type IV pilus phenotypes in vitro. New techniques to assay contributions of K. kingae genes in an in vivo animal model may be necessary to determine if PilA2 is contributing a key role in pathogenic processes and may identify novel strategies for the prevention of K. kingae disease.

141

6. Conclusion and future directions

6.1 Conclusion

The data presented in Chapters 2 - 5 begin to establish the key surface factor repertoire and the mechanism utilized by K. kingae to adhere efficiently to human epithelial cells. We envision an encapsulated bacterium expressing Knh and type IV pili first makes contact with an epithelial cell via its surface pili. Expression of PilC1, PilC2, or both is critical for the function of the surface fibers. If this organism is expressing only PilC1, then calcium responsive PilC1 allows the pili to mediate not only an adhesive interaction with the host cell, but also controls the ability of the pili to retract.

A PilC2-expressing organism is able to mediate these properties utilizing a Ca-binding site-dispensable mechanism. PilC function also enables PilT-mediated twitching motility to bring the bacterium into close proximity to the host cell plasma membrane.

We speculate that the retractile force of retraction displaces the polysaccharide capsule and exposes Knh to its host cell receptor leading to efficient adherence. One alternative hypothesis is that the capsule is instead shed from the bacterial surface during the adherence process thus exposing Knh and enabling adhesive function. The future directions and experimental approaches outlined below will further define the adhesive mechanism of K. kingae and begin to provide insights into the in vivo contributions of pili, Knh, and capsule to K. kingae pathogenesis.

142

6.2 K. kingae interactions with live cells

Early studies aimed at investigating K. kingae interactions with host cells revealed rapid and pronounced cytotoxicity of epithelial, synovial, and macrophage-like cell monolayers when incubated with the bacteria (Kehl-Fie et al., 2007). Using a mariner transposon mutagenesis screen for non-cytotoxic mutants, a locus encoding an RTX family toxin was identified (Kehl-Fie et al., 2007). As complete destruction of the monolayer occurred within minutes, studies of K. kingae interactions with live cells were not feasible. To nullify the action of the RTX toxin on host cells, a glutaraldehyde cell fixation procedure was developed to study interactions of K. kingae with host cells. This approach led to many discoveries, including identification of type IV pili, a novel trimeric autotransporter, and the influence of capsule on host cell adherence. While this approach is useful for determining quantitative strain-to-strain adherence level differences, many other informative bacteria-cell interaction studies are not possible.

A strategy is currently underway to establish a live-cell adherence assay for K. kingae. Preliminary evidence suggests that deletion of the rtx locus reduces but does not completely eliminate cytotoxicity (data not shown). The Δrtx mutant is able to remain on cell monolayers for approximately 2 hours but then induces cell rounding and release. Given that many other organisms including N. meningitidis and H. influenzae are routinely cultured with cell monolayers for 8 hours or more (Virji et al., 1991, Virji et al.,

1992, Virji et al., 1993, St Geme et al., 1994), it is likely that K. kingae expresses another

143

toxin. Current work is underway to identify a potential second cytotoxin. However, the

Δrtx mutant will be useful for initiating live-cell adherence studies.

Of particular interest will be evaluating how type IV pili, capsule, and Knh contribute to K. kingae live cell adherence. By introducing the mutations described in

Chapter 2 into the Δrtx background, we will be able to dissect how each of these surface factors contributes to live cell adherence. It will be interesting to learn if differences exist in adherence levels between live vs. fixed cell interactions, particularly in the case of the pilT pilus retraction mutant. In N. meningitidis, the hyperpiliated pilT mutant adheres at higher levels to fixed cells than to live cells but fails to transition from immediate to intimate adherence (Pujol et al., 1999). In K. kingae the pilT mutant adheres less than the

WT strain to fixed cells, possibly because retraction is needed for capsule displacement and Knh-mediated adherence (Chapter 2). The relevance of this adherence model to K. kingae pathogenesis will be strengthened if the results are replicated on live cells. Due to possible differences in live vs. fixed cell adherence, we may be able to identify new adhesive factors that can only mediate adherence to live cells by screening a mariner transposon library made in the Δrtx background.

The live-cell studies will also enable us to explore new avenues of K. kingae-host interactions. Thin-section TEM analysis of K. kingae infected monolayers will allow us to visualize direct interactions of the bacteria with the host cell plasma membrane. Of particular interest is to determine how the known surface factors involved in adherence

144

(pili, capsule, Knh) affect the morphology of interactions. For example, wild type N. meningitidis adherence to host cells triggers pedestal formation at the bacterial/host cell membrane interface (Pujol et al., 1999). However, a pilT mutant is incapable of triggering pedestal formation (Pujol et al., 1999). Using this approach we should also be able to appreciate how capsule influences interactions in the context of pilus retraction. Based on our model, we would predict that tight interactions would be evident in the WT strain but would be absent in the pilT strain because capsule would inhibit Knh- mediated adherence in the absence of retraction.

One key aspect of K. kingae pathogenesis that remains unknown is how the organism breaches the respiratory epithelium to enter the bloodstream. One hypothesis is that the RTX toxin causes disruption of the epithelial barrier, enabling access of bacteria to the bloodstream. However, no information exists concerning expression of the RTX toxin in vivo. Another plausible route that K. kingae could take is a transcellular or paracellular route through the epithelial layer. With a live-cell interaction system, for the first time we will be able to determine if K. kingae is capable of invading host cells.

Previous studies established that K. kingae is susceptible to gentamicin, making standard invasion assays feasible (Yagupsky et al., 2001). In addition, using a polarized monolayer cultured in transwells, we should be able to appreciated if K. kingae can traverse a cell layer and using microscopy should be able to determine if the organism utilizes a transcellular or paracellular route.

145

Lastly, we will be able to initiate experiments aimed at investigating the host response to K. kingae. Initial studies should focus on select proinflammatory mediator transcript level (qRT-PCR) and secretion (ELISA) assays, such as TNF-α, IL-6, and IL-8 following K. kingae infection of epithelial cells over various times points. Infection of epithelial cells with K. kingae mutants lacking type IV pili, capsule, and/or Knh will establish how these surface factors influence the host response. In addition, contributions of type IV pilus components, such as PilT, the PilCs, or minor pilins may lead to recognition of novel contributions of these factors to the K. kingae pathogenic process. More detailed host cell gene expression (microarray) and secretome (mass spectrometry) profiles will generate a broader view of the epithelial cell response to K. kingae.

6.3 Define the adhesive domain(s) of Knh

In the work detailed in Chapter 2 we demonstrated that insertional inactivation of a gene encoding a trimeric autotransporter called Knh results in a reduction of adherence to Chang epithelial cells. Furthermore, while a pilA1/ctrA mutant lacking pili and capsule adhered at WT levels, a triple pilA1/ctrA/knh mutant was non-adherent.

These data are consistent with Knh having adhesive activity. However, no evidence of direct Knh-host cell interaction has been discovered. Preliminary studies to express Knh in non-adherent E. coli were not successful. While under standard SDS-PAGE denaturing conditions, full-length Knh is detectable only as a multimer using Western

146

blotting (Figure 4). When the Knh membrane anchor is expressed in E. coli, it is detectable as a trimer (Figure 3); however, Knh is detectable primarily as a monomer when expressed as the full-length protein in E. coli strain DH5α or E. coli strain BL21

(Figure 31).

Figure 31: Western blot analysis of Knh expressed in E. coli outer membranes.

Western blot analysis with anti-Knh GP97 of outer membrane preparations from strains DH5α/pBAD18 (lane 1), DH5α/pBAD-Knh (lane 2), and BL21/pBAD-Knh (lane 3). Molecular weights (MW) are expressed in kDa.

These data indicate that full-length Knh is not as stable in the E. coli outer membrane as in the K. kingae outer membrane. When tested for adherence, Knh expressed in Dh5α or

BL21 did not confer any adhesive activity (data not shown). Future studies should focus on two experimental directions to confirm that Knh can confer adherence to a non- adherent host. First, fragments of the Knh passenger domain fused to its membrane anchor, or a membrane anchor from a trimeric autotransporter known to function in E. coli, can be tested for adhesive activity. Second, full-length Knh should be tested for

147

ability to confer adherence in a non-adherent host other than E. coli, such as H. influenzae strain Rd.

As a complementary approach, fragments of Knh can be expressed as GST or

MBP fusion proteins and directly tested for adhesive activity. Using immunofluorescence microscopy, Radin et al. demonstrated direct interaction of the two binding domains of the H. influenzae Hsf trimeric autotransporter to epithelial cells

(Radin et al., 2009). This approach has the advantage that fragments can be selected for adhesive activity testing based on predicted structural domains using the Domain

Annotation of Trimeric Autotransporters program (daTAA, http://toolkit.tuebingen.mpg.de/dataa). For example, the N-terminal 42 kDa of Knh contains a series of YadA-like head domains that have been implicated in adherence in multiple trimeric autotransporters (Tahir et al., 2000, Nummelin et al., 2004). In addition,

Knh contains an iSneck2 domain, which is similar in predicted structure to the Hsf and

Hia binding domains (Meng et al., 2008). Recombinant Knh fragments can also be used to pretreat cell monolayers prior to addition of bacteria in adherence-blocking experiments.

The earlier finding that adherence to Chang epithelial cells is Knh-dependent in the absence of capsule and type IV pili provides us with an additional avenue to determine the adhesive domains. By generating a mutagenesis construct with an antibiotic marker upstream of knh, we can introduce selected domain deletions and test

148

for loss of adherence. This method will also be useful for fine-mapping key residues necessary for adherence. Identification of the adhesive domain(s) of Knh may provide insights into the host cell receptor.

6.4 Define capsule type diversity in a diverse set K. kingae clinical isolates

The discovery that K. kingae elaborates a polysaccharide capsule opens a potential avenue for prevention of K. kingae disease. Polysaccharide and polysaccharide- conjugate vaccines have been in use for decades for prevention of disease from a variety of bacterial pathogens. As detailed in Chapter 3, currently we have evidence of at least three different capsule types in K. kingae. To further evaluate to the K. kingae polysaccharide as a potential vaccine component, we first need to determine; (1) the diversity of capsule types expressed by K. kingae; (2) if a particular type or types is associated with invasive disease; and (3) if those capsule types are immunogenic.

We currently possess a collection of ~60 clinical isolates recovered from the respiratory tract, blood, and joint fluid. A previous study examined type IV pilus phenotypes in this collection (Kehl-Fie et al., 2010). In collaboration with the CCRC, we will perform glycosyl composition and linkage analysis on purified polysaccharide.

Once this information is in hand, we can examine the strains by capsule type and site of isolation to determine if there is a correlation. If our number of strains does not provide enough statistical power to establish significance, we will acquire more isolates from our collaborator, Pablo Yagupsky, who possesses a larger isolate set. 149

With information regarding capsule type diversity in hand, it will be necessary to determine if the capsules are immunogenic in the human host. As mentioned previously, some polysaccharides are immunogenic and have proven effective vaccine components. However, other capsule types, most notably the polysialic acid capsules of

E. coli K1 and N. meningitidis serotype B, are largely not immunogenic in humans due to the fact that they are identical in structure to the polysialic acid linked to the neural cell adhesion molecule (N-CAM) glycoprotein (Finne et al., 1983, Finne et al., 1987). One study has examined immunogenicity of K. kingae by probing outer membrane fractions with convalescent patient serum and normal human serum using Western blot analysis

(Yagupsky et al., 2005). However, this study only examined protein reactivity, as capsule was not identified by this point in time. Our plan is to acquire matched acute and convalescent serum from patients recovered from infection with a known capsule type. To examine antibody capsule reactivity and specificity, purified matched polysaccharide (capsular material from the infecting strain) and non-matched polysaccharide (capsular material from a different capsule type strain) will be probed using dot blot and modified Western blot. More detailed immunogold TEM approaches will also be employed.

If these studies are successful in identifying the major capsule types in K. kingae and these polysaccharide capsules shown to be immunogenic, future experiments will

150

aim to evaluate polysaccharide as vaccine component. These studies will require the development of an animal model of K. kingae disease (See Chapter 6.7).

6.5 Localize the PilC proteins

The PilC-like proteins remain one of the most intriguing type IV pilus components. As detailed in Chapter 4, expression of at least one PilC is necessary for K. kingae pilus biogenesis indicating a role of this protein class in pilus biogenesis. The PilC proteins also appear to directly mediate the adhesive phenotype of the fibers (Rudel et al., 1995b, Scheuerpflug et al., 1999, Kirchner et al., 2005a, Kirchner et al., 2005b, Heiniger et al., 2010). The calcium-dependent differences in the K. kingae PilCs coupled with the highly divergent primary amino sequence suggest different mechanisms of PilC1 and

PilC2 function. To begin to dissect PilC1 and PilC2 differences in K. kingae, generation of antisera against the two proteins was initiated as tools for study. The fragments used for the in vitro Ca-binding studies in Chapter 4 were sent to Cocalico Biologicals for antiserum development in a guinea pig for PilC1 and a rabbit for PilC2. Using SDS-

PAGE and Coomassie staining of whole cell sonicates, it became apparent that a band at the predicted molecular weight of PilC2 (165kDa) was not present in the ΔpilC2 mutant

(Figure 32, left panel). This band was confirmed to be PilC2 using the rabbit anti-PilC2 serum in Western blot analysis (Figure 32, right panel). In contrast, no difference was observed in the banding pattern of the ΔpilC1 mutant in the molecular weight range of

PilC1 (145kDa). In addition, we could not detect any specific reactivity of the guinea pig

151

anti-PilC1 serum (data not shown). These data suggest that PilC1 is expressed at a lower level compared to PilC2 and that PilC1 antibody development was not successful.

Figure 32: Coomassie and Western blot analysis of whole cell sonicates.

Whole cell sonicates from strains 269-492 (lane 1), ΔpilC1 (lane 2), ΔpilC2 (lane 3), and ΔpilC1/ΔpilC2 (lane 4) were examined by Coomassie staining (left panel) and Western blot (right panel) with anti-PilC2 antiserum.

Future experiments are needed to determine if PilC1 and PilC2 are co-expressed in the WT background and to determine the subcellular localization of the proteins.

Given that one intact pilC gene is necessary for piliation and that a ΔpilC1 or ΔpilC2 single mutant remains piliated, it likely that PilC1 and PilC2 are coexpressed. To confirm this conclusion, PilC1 antiserum will be essential. Scheuerpflug et al. previously described a method for purification of the Neisseria PilC proteins (Scheuerpflug et al.,

1999), and recently a collaborator has developed a new method for Neisseria PilC purification (Johnson and Redinbo, personal communication). If these methods can be optimized for expression of the K. kingae PilCs, then new antiserum development with 152

purified PilC1 can be initiated. Once specific antiserum is in hand, subcellular fractionation experiments coupled with Western blot analysis will help determine where the PilCs are located. We expect PilC1 and PilC2 to be located in pilus and outer membrane fractions, as has been shown for other organisms (Rudel et al., 1995b, Rahman et al., 1997, Heiniger et al., 2010). Ideally, immunogold TEM experiments will also pinpoint the extracellular localization of the PilCs. One particular area of interest is to determine if both PilCs are expressed on a single organism or if in a population of organisms there are subpopulations that express only PilC1 or PilC2. Co-staining with the guinea pig PilC1-specific and rabbit PilC2-specific antisera coupled with different colloidal gold size-labeled secondary antibodies should allow for this distinction to be appreciated.

6.6 Investigate the PilC/PilT relationship

In the well-studied Neisseria and P. aeruginosa type IV pilus systems, expression of at least on PilC-like protein is necessary for piliation. A ΔpilC1/ΔpilC2 K. kingae double mutant is also non-piliated, suggesting a conserved role of the PilC-like proteins in pilus expression (Kehl-Fie et al., 2008). In the Neisseria spp. and P. aeruginosa, disruption of pilT in a ΔpilC1/C2 (or ΔpilY1) background restores surface fiber expression but renders the pili non-functional in terms of adherence and twitching motility (Heiniger et al., 2010) (Wolfgang et al., 1998b). Before we can investigate the mechanism behind PilC-dependent pilus biogenesis, we first need to determine if

153

piliation is restored in the ΔpilC1/ΔpilC2 mutant after elimination of PilT expression. At this time it is not feasible to introduce a ΔpilT mutation into the ΔpilC1/ΔpilC2 background because this strain is non-piliated and thus non-transformable via the natural competence mechanism. To overcome this hurdle, current work is focused on generating an IPTG-inducible pilT strain so that ΔpilC1 and ΔpilC2 mutations can be introduced successively by inducing pilT. The lacI gene and lac promoter element were cloned immediately upstream of the predicted pilT RBS to disrupt the native promoter and place pilT under control of the lac system. This technique has been used successfully to induce GFP in K. kingae (Starr and St. Geme, unpublished data). Once the

ΔpilC1/ΔpilC2/ind-pilT strain is generated, growth on chocolate agar without IPTG should turn off pilT and create a mock ΔpilC1/ΔpilC2/ΔpilT triple mutant phenotype. A similar approach was used to study the relationship of PilT and the PilCs in N. gonorrhoeae (Wolfgang et al., 1998b). We expect the ΔpilC1/ΔpilC2/ind-pilT strain grown in the absence of IPTG to be piliated but non-adherent due to the lack of PilCs.

6.7 Demonstrate adhesive properties of the PilC proteins

Current evidence suggests that the K. kingae PilC proteins are the adhesive components of the type IV pilus fiber but this conclusion has yet to be directly demonstrated. Ideally, we would demonstrate direct interaction of purified PilC protein

(the same full-length protein mentioned in Chapter 6.4) with an epithelial cell monolayer. Several studies have shown that purified Neisseria PilCs can directly

154

interact with select cell monolayers (Scheuerpflug et al., 1999, Kirchner et al., 2005a,

Kirchner et al., 2005b). Using co-staining techniques with differentially labeled secondary antibodies, we may even be able to appreciate differences in cellular binding location. This information could be used to evaluate if PilC1 and PilC2 bind the same or different host targets. Purified PilCs will also be useful in pretreatment adherence blocking experiments, as detailed for purified Knh-fragments in Chapter 6.2. Of particular interest will be determining if purified PilC1 can block adherence by both the

ΔpilC2 and ΔpilC1 strains or if it can only block adherence by the ΔpilC2 strain and vice versa. This data will provide compelling evidence supporting either similar or different

PilC1 and PilC2 cell receptors.

As a complementary approach, we can utilize the WT, ΔpilC1, ΔpilC2, and

ΔpilC1/ΔpilC2/ΔpilT strains to evaluate the adhesive activity of sheared fibers. Instead of directly testing for adherence of the purified PilC proteins, fibers (presumably containing/associated with the PilCs) will be sheared from the bacterial surface, added to cell monolayers, and detected with anti-PilA1 GP65 antiserum. This method will also be necessary if the PilCs are not adherent when not associated with the fibers. We would expect fibers sheared from WT, ΔpilC1, and ΔpilC2 strain to be adherent, and fibers from the ΔpilC1/ΔpilC2/ind-pilT (not induced) strain to be nonadherent. Like the purified

PilCs, sheared pili can also be used in adherence blocking experiments.

155

Both of the techniques mentioned here will also allow us to evaluate the necessity of the PilC1 Ca-binding site on adherence, as detailed in Chapter 4. Based on those studies, purified PilC1D930A or PilC1D930K or sheared fibers from strains

ΔpilC2/PilC1D930A or ΔpilC2/PilC1D930K would not bind host cells and would not block adherence. Likewise, these experimental approaches will be essential for future studies aimed at localizing the PilC adhesive domains.

6.8 Evaluate and improve the K. kingae animal model

Our work has demonstrated the in vitro contributions of type IV pili, pili components, capsule, and Knh to epithelial cell adherence. However to begin to understand how these and other K. kingae factors contribute to in vivo virulence, an animal model is necessary. Establishment of animal models for fastidious gram- negative organisms has proven difficult for the research community. Pilot experiments to establish a mouse model of K. kingae disease in our laboratory utilized a variety of mouse strains, including C3/HEN, B6, Swiss, and BalbC (T. E. Kehl-Fie and J. W. St.

Geme, unpublished data). The initial studies started with 2x107 bacteria inoculated via the tail vein. The mice were subsequently evaluated for signs of illness. Several mice strains showed signs of illness and developed gait issues, including dragging of the hind legs. While these symptoms potentially suggest hip involvement, no bacteria were recovered from the animals and histological analysis revealed no joint or bone pathology. Recently, Basmaci et al. reported a juvenile rat model to evaluate clinical K.

156

kingae isolate virulence (Basmaci et al., 2012). These investigators inoculated 107 organisms IP into 5-day-old Spragues-Dawley albino rats and evaluated the animals for

5 days. They found strain differences in survival and development of ischemic-necrosis lesions (Basmaci et al., 2012). Also, bacteria were recovered from excised lesions.

The Basmaci et al. study provides a valuable step forward for the study of K. kingae pathogenesis and potential virulence factors. We will first need to attempt to replicate the results by Basmaci et al. to confirm utility of the method. Once replicated, we will initiate studies using strains mutated and complemented for potential virulence factors, including pilA1, pilT, and knh. While the IP infection route does not replicate a natural course of infection and the disease state in the juvenile rat does not replicate human disease, the model does enable us to begin to address in vivo importance of select factors. The model also falls short in that it will not allow evaluation of vaccine candidates. Other animal models, including different mouse strains, adult rats, and rabbits should be evaluated, as an intact, developed immune system will be necessary to evaluate vaccination strategies.

157

References

Aas, F. E., M. Wolfgang, S. Frye, S. Dunham, C. Lovold & M. Koomey, (2002) Competence for natural transformation in Neisseria gonorrhoeae: components of DNA binding and uptake linked to type IV pilus expression. Mol Microbiol 46: 749-760.

Alm, R. A. & J. S. Mattick, (1995) Identification of a gene, pilV, required for type 4 fimbrial biogenesis in Pseudomonas aeruginosa, whose product possesses a pre-pilin-like leader sequence. Mol Microbiol 16: 485-496.

Alm, R. A., J. P. Hallinan, A. A. Watson & J. S. Mattick, (1996) Fimbrial biogenesis genes of Pseudomonas aeruginosa: pilW and pilX increase the similarity of type 4 fimbriae to the GSP protein-secretion systems and pilY1 encodes a gonococcal PilC homologue. Mol Microbiol 22: 161-173.

Alm, R. A. & J. S. Mattick, (1997) Genes involved in the biogenesis and function of type-4 fimbriae in Pseudomonas aeruginosa. Gene 192: 89-98.

Altman, E., J. R. Brisson & M. B. Perry, (1987) Structure of the capsular polysaccharide of Haemophilus pleuropneumoniae serotype 5. European journal of biochemistry / FEBS 170: 185- 192.

Amir, J. & P. Yagupsky, (1998) Invasive Kingella kingae infection associated with stomatitis in children. Pediatr Infect Dis J 17: 757-758.

Anantha, R. P., K. D. Stone & M. S. Donnenberg, (1998) Role of BfpF, a member of the PilT family of putative nucleotide-binding proteins, in type IV pilus biogenesis and in interactions between enteropathogenic Escherichia coli and host cells. Infect Immun 66: 122-131.

Ayers, M., L. M. Sampaleanu, S. Tammam, J. Koo, H. Harvey, P. L. Howell & L. L. Burrows, (2009) PilM/N/O/P proteins form an inner membrane complex that affects the stability of the Pseudomonas aeruginosa type IV pilus secretin. J Mol Biol 394: 128-142.

Barenkamp, S. J. & E. Leininger, (1992) Cloning, expression, and DNA sequence analysis of genes encoding nontypeable Haemophilus influenzae high-molecular-weight surface- exposed proteins related to filamentous hemagglutinin of Bordetella pertussis. Infect Immun 60: 1302-1313.

Barenkamp, S. J. & J. W. St Geme, 3rd, (1996) Identification of a second family of high- molecular-weight adhesion proteins expressed by non-typable Haemophilus influenzae. Mol Microbiol 19: 1215-1223. 158

Basmaci, R., B. Ilharreborde, P. Bidet, C. Doit, M. Lorrot, K. Mazda, E. Bingen & S. Bonacorsi, (2012) Isolation of Kingella kingae in the oropharynx during K. kingae arthritis in children. Clin Microbiol Infect 18: E134-136.

Beatson, S. A., C. B. Whitchurch, J. L. Sargent, R. C. Levesque & J. S. Mattick, (2002) Differential regulation of twitching motility and elastase production by Vfr in Pseudomonas aeruginosa. J Bacteriol 184: 3605-3613.

Bendaoud, M., E. Vinogradov, N. V. Balashova, D. E. Kadouri, S. C. Kachlany & J. B. Kaplan, (2011) Broad-spectrum biofilm inhibition by Kingella kingae exopolysaccharide. J Bacteriol 193: 3879-3886.

Berbari, E. F., F. R. Cockerill, 3rd & J. M. Steckelberg, (1997) Infective endocarditis due to unusual or fastidious microorganisms. Mayo Clinic proceedings. Mayo Clinic 72: 532-542.

Bhattacharjee, A. K., H. J. Jennings, C. P. Kenny, A. Martin & I. C. Smith, (1975) Structural determination of the sialic acid polysaccharide antigens of Neisseria meningitidis serogroups B and C with carbon 13 nuclear magnetic resonance. The Journal of biological chemistry 250: 1926-1932.

Bhattacharjee, A. K., H. J. Jennings, C. P. Kenny, A. Martin & I. C. Smith, (1976) Structural determination of the polysaccharide antigens of Neisseria meningitidis serogroups Y, W-135, and BO1. Canadian journal of biochemistry 54: 1-8.

Biswas, G. D., T. Sox, E. Blackman & P. F. Sparling, (1977) Factors affecting genetic transformation of Neisseria gonorrhoeae. J Bacteriol 129: 983-992.

Bliss, J. M. & R. P. Silver, (1996) Coating the surface: a model for expression of capsular polysialic acid in Escherichia coli K1. Mol Microbiol 21: 221-231.

Bofinger, J. J., T. Fekete & R. Samuel, (2007) Bacterial peritonitis caused by Kingella kingae. J Clin Microbiol 45: 3118-3120.

Bovre, K., S. D. Henriksen & V. Jonsson, (1974) Correction of Specific Epithet kingii in Combinations Moraxella kingii Henriksen and Bovre 1968 and Pseudomonas kingii Jonsson 1970 to Kingae. Int J Syst Bacteriol 24: 307-307.

Brown, D. R., S. Helaine, E. Carbonnelle & V. Pelicic, (2010) Systematic functional analysis reveals that a set of seven genes is involved in fine-tuning of the multiple functions mediated by type IV pili in Neisseria meningitidis. Infect Immun 78: 3053-3063.

159

Bundle, D. R., I. C. Smith & H. J. Jennings, (1974) Determination of the structure and conformation of bacterial polysaccharides by carbon 13 nuclear magnetic resonance. Studies on the group-specific antigens of Neisseria meningitidis serogroups A and X. The Journal of biological chemistry 249: 2275-2281.

Burrows, L. L., (2005) Weapons of mass retraction. Mol Microbiol 57: 878-888.

Carbonnelle, E., S. Helaine, L. Prouvensier, X. Nassif & V. Pelicic, (2005) Type IV pilus biogenesis in Neisseria meningitidis: PilW is involved in a step occurring after pilus assembly, essential for fibre stability and function. Mol Microbiol 55: 54-64.

Carbonnelle, E., S. Helaine, X. Nassif & V. Pelicic, (2006) A systematic genetic analysis in Neisseria meningitidis defines the Pil proteins required for assembly, functionality, stabilization and export of type IV pili. Mol Microbiol 61: 1510-1522.

Carden, S. M., D. J. Colville, G. Gonis & G. L. Gilbert, (1991) Kingella kingae endophthalmitis in an infant. Australian and New Zealand journal of ophthalmology 19: 217- 220.

Cartwright, K., (1995) Meningococcal disease, p. viii, 316 p., 314 p. of plates. J. Wiley & Sons, Chichester ; New York.

Ceroni, D., A. Cherkaoui, S. Ferey, A. Kaelin & J. Schrenzel, (2010a) Kingella kingae osteoarticular infections in young children: clinical features and contribution of a new specific real-time PCR assay to the diagnosis. Journal of pediatric orthopedics 30: 301-304.

Ceroni, D., A. Cherkaoui, A. Kaelin & J. Schrenzel, (2010b) Kingella kingae spondylodiscitis in young children: toward a new approach for bacteriological investigations? A preliminary report. Journal of children's orthopaedics 4: 173-175.

Ceroni, D., A. Cherkaoui, C. Combescure, P. Francois, A. Kaelin & J. Schrenzel, (2011) Differentiating osteoarticular infections caused by Kingella kingae from those due to typical pathogens in young children. Pediatr Infect Dis J 30: 906-909.

Ceroni, D., V. Dubois-Ferriere, R. Anderson, C. Combescure, L. Lamah, A. Cherkaoui & J. Schrenzel, (2012) Small Risk of Osteoarticular Infections in Children with Asymptomatic Oropharyngeal Carriage of Kingella kingae. Pediatr Infect Dis J.

Chanal, C., F. Tiget, P. Chapuis, D. Campagne, M. Jan & J. Sirot, (1987) Spondylitis and osteomyelitis caused by Kingella kingae in children. J Clin Microbiol 25: 2407-2409.

160

Cherkaoui, A., D. Ceroni, S. Emonet, Y. Lefevre & J. Schrenzel, (2009) Molecular diagnosis of Kingella kingae osteoarticular infections by specific real-time PCR assay. Journal of medical microbiology 58: 65-68.

Chiang, P., L. M. Sampaleanu, M. Ayers, M. Pahuta, P. L. Howell & L. L. Burrows, (2008) Functional role of conserved residues in the characteristic secretion NTPase motifs of the Pseudomonas aeruginosa type IV pilus motor proteins PilB, PilT and PilU. Microbiology 154: 114-126.

Chometon, S., Y. Benito, M. Chaker, S. Boisset, C. Ploton, J. Berard, F. Vandenesch & A. M. Freydiere, (2007) Specific real-time polymerase chain reaction places Kingella kingae as the most common cause of osteoarticular infections in young children. Pediatr Infect Dis J 26: 377-381.

Claesson, B., E. Falsen & B. Kjellman, (1985) Kingella kingae infections: a review and a presentation of data from 10 Swedish cases. Scandinavian journal of infectious diseases 17: 233-243.

Clausen, M., V. Jakovljevic, L. Sogaard-Andersen & B. Maier, (2009) High-force generation is a conserved property of type IV pilus systems. J Bacteriol 191: 4633-4638.

Clement, J. L., J. Berard, J. P. Cahuzac & J. Gaubert, (1988) Kingella kingae osteoarthritis and osteomyelitis in children. Journal of pediatric orthopedics 8: 59-61.

Collins, R. F., M. Saleem & J. P. Derrick, (2007) Purification and three-dimensional electron microscopy structure of the Neisseria meningitidis type IV pilus biogenesis protein PilG. J Bacteriol 189: 6389-6396.

Comanducci, M., S. Bambini, B. Brunelli, J. Adu-Bobie, B. Arico, B. Capecchi, M. M. Giuliani, V. Masignani, L. Santini, S. Savino, D. M. Granoff, D. A. Caugant, M. Pizza, R. Rappuoli & M. Mora, (2002) NadA, a novel vaccine candidate of Neisseria meningitidis. The Journal of experimental medicine 195: 1445-1454.

Comolli, J. C., A. R. Hauser, L. Waite, C. B. Whitchurch, J. S. Mattick & J. N. Engel, (1999) Pseudomonas aeruginosa gene products PilT and PilU are required for cytotoxicity in vitro and virulence in a mouse model of acute pneumonia. Infect Immun 67: 3625-3630.

Cotter, S. E., N. K. Surana & J. W. St Geme, 3rd, (2005) Trimeric autotransporters: a distinct subfamily of autotransporter proteins. Trends Microbiol 13: 199-205.

Craig, L., M. E. Pique & J. A. Tainer, (2004) Type IV pilus structure and bacterial pathogenicity. Nature reviews. Microbiology 2: 363-378.

161

Cross, A. S., K. S. Kim, D. C. Wright, J. C. Sadoff & P. Gemski, (1986) Role of lipopolysaccharide and capsule in the serum resistance of bacteremic strains of Escherichia coli. J Infect Dis 154: 497-503.

Dagan, R., (1993) Management of acute hematogenous osteomyelitis and septic arthritis in the pediatric patient. Pediatr Infect Dis J 12: 88-92. de Groot, R., D. Glover, C. Clausen, A. L. Smith & C. B. Wilson, (1988) Bone and joint infections caused by Kingella kingae: six cases and review of the literature. Reviews of infectious diseases 10: 998-1004.

Deghmane, A. E., S. Petit, A. Topilko, Y. Pereira, D. Giorgini, M. Larribe & M. K. Taha, (2000) Intimate adhesion of Neisseria meningitidis to human epithelial cells is under the control of the crgA gene, a novel LysR-type transcriptional regulator. EMBO J 19: 1068- 1078.

Deghmane, A. E., D. Giorgini, M. Larribe, J. M. Alonso & M. K. Taha, (2002) Down- regulation of pili and capsule of Neisseria meningitidis upon contact with epithelial cells is mediated by CrgA regulatory protein. Mol Microbiol 43: 1555-1564.

Dodman, T., J. Robson & D. Pincus, (2000) Kingella kingae infections in children. Journal of paediatrics and child health 36: 87-90.

Domenico, P., R. J. Salo, A. S. Cross & B. A. Cunha, (1994) Polysaccharide capsule- mediated resistance to opsonophagocytosis in Klebsiella pneumoniae. Infect Immun 62: 4495-4499.

Domenighini, M., D. Relman, C. Capiau, S. Falkow, A. Prugnola, V. Scarlato & R. Rappuoli, (1990) Genetic characterization of Bordetella pertussis filamentous haemagglutinin: a protein processed from an unusually large precursor. Mol Microbiol 4: 787-800.

Donnelly, M. I., M. Zhou, C. S. Millard, S. Clancy, L. Stols, W. H. Eschenfeldt, F. R. Collart & A. Joachimiak, (2006) An expression vector tailored for large-scale, high- throughput purification of recombinant proteins. Protein expression and purification 47: 446-454.

Donnenberg, M. S., J. A. Giron, J. P. Nataro & J. B. Kaper, (1992) A plasmid-encoded type IV fimbrial gene of enteropathogenic Escherichia coli associated with localized adherence. Mol Microbiol 6: 3427-3437.

162

Drake, S. L. & M. Koomey, (1995) The product of the pilQ gene is essential for the biogenesis of type IV pili in Neisseria gonorrhoeae. Mol Microbiol 18: 975-986.

Drake, S. L., S. A. Sandstedt & M. Koomey, (1997) PilP, a pilus biogenesis lipoprotein in Neisseria gonorrhoeae, affects expression of PilQ as a high-molecular-mass multimer. Mol Microbiol 23: 657-668.

Dubnov-Raz, G., O. Scheuerman, G. Chodick, Y. Finkelstein, Z. Samra & B. Z. Garty, (2008) Invasive Kingella kingae infections in children: clinical and laboratory characteristics. Pediatrics 122: 1305-1309.

Dubnov-Raz, G., M. Ephros, B. Z. Garty, Y. Schlesinger, A. Maayan-Metzger, J. Hasson, I. Kassis, O. Schwartz-Harari & P. Yagupsky, (2010) Invasive pediatric Kingella kingae Infections: a nationwide collaborative study. Pediatr Infect Dis J 29: 639-643.

Eriksson, J., O. S. Eriksson & A. B. Jonsson, (2012) Loss of Meningococcal PilU Delays Microcolony Formation and Attenuates Virulence in vivo. Infect Immun 80: 2538-2547.

Faast, R., M. A. Ogierman, U. H. Stroeher & P. A. Manning, (1989) Nucleotide sequence of the structural gene, tcpA, for a major pilin subunit of Vibrio cholerae. Gene 85: 227-231.

Finne, J., M. Leinonen & P. H. Makela, (1983) Antigenic similarities between brain components and bacteria causing meningitis. Implications for vaccine development and pathogenesis. Lancet 2: 355-357.

Finne, J., D. Bitter-Suermann, C. Goridis & U. Finne, (1987) An IgG monoclonal antibody to group B meningococci cross-reacts with developmentally regulated polysialic acid units of glycoproteins in neural and extraneural tissues. J Immunol 138: 4402-4407.

Forest, K. T., S. A. Dunham, M. Koomey & J. A. Tainer, (1999) Crystallographic structure reveals phosphorylated pilin from Neisseria: phosphoserine sites modify type IV pilus surface chemistry and fibre morphology. Mol Microbiol 31: 743-752.

Fraysse, N., B. Lindner, Z. Kaczynski, L. Sharypova, O. Holst, K. Niehaus & V. Poinsot, (2005) Sinorhizobium meliloti strain 1021 produces a low-molecular-mass capsular polysaccharide that is a homopolymer of 3-deoxy-D-manno-oct-2-ulosonic acid harboring a phospholipid anchor. Glycobiology 15: 101-108.

Freitag, N. E., H. S. Seifert & M. Koomey, (1995) Characterization of the pilF-pilD pilus- assembly locus of Neisseria gonorrhoeae. Mol Microbiol 16: 575-586.

163

Froholm, L. O. & K. Bovre, (1972) Fimbriation associated with the spreading-corroding colony type in Moraxella kingii. Acta pathologica et microbiologica Scandinavica. Section B: Microbiology and immunology 80: 641-648.

Frosch, M., U. Edwards, K. Bousset, B. Krausse & C. Weisgerber, (1991) Evidence for a common molecular origin of the capsule gene loci in gram-negative bacteria expressing group II capsular polysaccharides. Mol Microbiol 5: 1251-1263.

Frosch, M., D. Muller, K. Bousset & A. Muller, (1992) Conserved outer membrane protein of Neisseria meningitidis involved in capsule expression. Infect Immun 60: 798-803.

Frosch, M. & A. Muller, (1993) Phospholipid substitution of capsular polysaccharides and mechanisms of capsule formation in Neisseria meningitidis. Mol Microbiol 8: 483-493.

Fulks, K. A., C. F. Marrs, S. P. Stevens & M. R. Green, (1990) Sequence analysis of the inversion region containing the pilin genes of Moraxella bovis. J Bacteriol 172: 310-316.

Fuursted, K., M. Arpi, B. E. Lindblad & L. N. Pedersen, (2008) Broad-range PCR as a supplement to culture for detection of bacterial pathogens in patients with a clinically diagnosed spinal infection. Scandinavian journal of infectious diseases 40: 772-777.

Garron, E., E. Viehweger, F. Launay, J. M. Guillaume, J. L. Jouve & G. Bollini, (2002) Nontuberculous spondylodiscitis in children. Journal of pediatric orthopedics 22: 321-328.

Gene, A., J. J. Garcia-Garcia, P. Sala, M. Sierra & R. Huguet, (2004) Enhanced culture detection of Kingella kingae, a pathogen of increasing clinical importance in pediatrics. Pediatr Infect Dis J 23: 886-888.

Gibbs, C. P., B. Y. Reimann, E. Schultz, A. Kaufmann, R. Haas & T. F. Meyer, (1989) Reassortment of pilin genes in Neisseria gonorrhoeae occurs by two distinct mechanisms. Nature 338: 651-652.

Giltner, C. L., M. Habash & L. L. Burrows, (2010) Pseudomonas aeruginosa minor pilins are incorporated into type IV pili. J Mol Biol 398: 444-461.

Gotschlich, E. C., B. A. Fraser, O. Nishimura, J. B. Robbins & T. Y. Liu, (1981) Lipid on capsular polysaccharides of gram-negative bacteria. The Journal of biological chemistry 256: 8915-8921.

Goutzmanis, J. J., G. Gonis & G. L. Gilbert, (1991) Kingella kingae infection in children: ten cases and a review of the literature. Pediatr Infect Dis J 10: 677-683.

164

Grass, S. & J. W. St Geme, 3rd, (2000) Maturation and secretion of the non-typable Haemophilus influenzae HMW1 adhesin: roles of the N-terminal and C-terminal domains. Mol Microbiol 36: 55-67.

Graupner, S. & W. Wackernagel, (2001) Pseudomonas stutzeri has two closely related pilA genes (Type IV pilus structural protein) with opposite influences on natural genetic transformation. J Bacteriol 183: 2359-2366.

Griffiss, J. M. & B. L. Brandt, (1983) Immunological relationship between the capsular polysaccharides of Neisseria meningitidis serogroups Z and 29E. J Gen Microbiol 129: 447- 452.

Haas, R. & T. F. Meyer, (1986) The repertoire of silent pilus genes in Neisseria gonorrhoeae: evidence for gene conversion. Cell 44: 107-115.

Hallstrom, T., T. Nordstrom, T. T. Tan, T. Manolov, J. D. Lambris, D. E. Isenman, P. F. Zipfel, A. M. Blom & K. Riesbeck, (2011) Immune evasion of Moraxella catarrhalis involves ubiquitous surface protein A-dependent C3d binding. J Immunol 186: 3120-3129.

Hamilton, H. L., K. J. Schwartz & J. P. Dillard, (2001) Insertion-duplication mutagenesis of Neisseria: use in characterization of DNA transfer genes in the gonococcal genetic island. J Bacteriol 183: 4718-4726.

Han, X., R. M. Kennan, J. K. Davies, L. A. Reddacliff, O. P. Dhungyel, R. J. Whittington, L. Turnbull, C. B. Whitchurch & J. I. Rood, (2008) Twitching motility is essential for virulence in Dichelobacter nodosus. J Bacteriol 190: 3323-3335.

Heiniger, R. W., H. C. Winther-Larsen, R. J. Pickles, M. Koomey & M. C. Wolfgang, (2010) Infection of human mucosal tissue by Pseudomonas aeruginosa requires sequential and mutually dependent virulence factors and a novel pilus-associated adhesin. CellularMicrobiology 12: 1158-1173.

Helaine, S., E. Carbonnelle, L. Prouvensier, J. L. Beretti, X. Nassif & V. Pelicic, (2005) PilX, a pilus-associated protein essential for bacterial aggregation, is a key to pilus- facilitated attachment of Neisseria meningitidis to human cells. Mol Microbiol 55: 65-77.

Helaine, S., D. H. Dyer, X. Nassif, V. Pelicic & K. T. Forest, (2007) 3D structure/function analysis of PilX reveals how minor pilins can modulate the virulence properties of type IV pili. Proc Natl Acad Sci U S A 104: 15888-15893.

165

Hendrixson, D. R., B. J. Akerley & V. J. DiRita, (2001) Transposon mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol Microbiol 40: 214-224.

Henriksen, S. D. & K. Bovre, (1968) Moraxella kingii sp.nov., a haemolytic, saccharolytic species of the genus Moraxella. J Gen Microbiol 51: 377-385.

Henriksen, S. D. & K. Bovre, (1976) Transfer of Moraxella-kingae Kingella-Henriksen and Bovre to Genus Gen-Nov in Family Neisseriaceae. Int J Syst Bacteriol 26: 447-450.

Hoiczyk, E., A. Roggenkamp, M. Reichenbecher, A. Lupas & J. Heesemann, (2000) Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J 19: 5989-5999.

Hood, B. L. & R. Hirschberg, (1995) Purification and characterization of Eikenella corrodens type IV pilin. Infect Immun 63: 3693-3696.

Ieva, R. & H. D. Bernstein, (2009) Interaction of an autotransporter passenger domain with BamA during its translocation across the bacterial outer membrane. Proc Natl Acad Sci U S A 106: 19120-19125.

Ilharreborde, B., P. Bidet, M. Lorrot, J. Even, P. Mariani-Kurkdjian, S. Liguori, C. Vitoux, Y. Lefevre, C. Doit, F. Fitoussi, G. Pennecot, E. Bingen, K. Mazda & S. Bonacorsi, (2009) New real-time PCR-based method for Kingella kingae DNA detection: application to samples collected from 89 children with acute arthritis. J Clin Microbiol 47: 1837-1841.

Jacob-Dubuisson, F., C. Locht & R. Antoine, (2001) Two-partner secretion in Gram- negative bacteria: a thrifty, specific pathway for large virulence proteins. Mol Microbiol 40: 306-313.

Jain, S. & M. B. Goldberg, (2007) Requirement for YaeT in the outer membrane assembly of autotransporter proteins. J Bacteriol 189: 5393-5398.

Jakovljevic, V., S. Leonardy, M. Hoppert & L. Sogaard-Andersen, (2008) PilB and PilT are ATPases acting antagonistically in type IV pilus function in Myxococcus xanthus. J Bacteriol 190: 2411-2421.

Jarvis, G. A. & N. A. Vedros, (1987) Sialic acid of group B Neisseria meningitidis regulates alternative complement pathway activation. Infect Immun 55: 174-180.

166

Jensen, K. T., H. Schonheyder & V. F. Thomsen, (1994) In-vitro activity of beta-lactam and other antimicrobial agents against Kingella kingae. The Journal of antimicrobial chemotherapy 33: 635-640.

Johnson, K., M. L. Parker & S. Lory, (1986) Nucleotide sequence and transcriptional initiation site of two Pseudomonas aeruginosa pilin genes. The Journal of biological chemistry 261: 15703-15708.

Johnson, M. D., C. K. Garrett, J. E. Bond, K. A. Coggan, M. C. Wolfgang & M. R. Redinbo, (2011) Pseudomonas aeruginosa PilY1 binds integrin in an RGD- and calcium- dependent manner. PloS one 6: e29629.

Jonsson, A. B., G. Nyberg & S. Normark, (1991) Phase variation of gonococcal pili by frameshift mutation in pilC, a novel gene for pilus assembly. EMBO J 10: 477-488.

Jose, J., J. Kramer, T. Klauser, J. Pohlner & T. F. Meyer, (1996) Absence of periplasmic DsbA oxidoreductase facilitates export of cysteine-containing passenger proteins to the Escherichia coli cell surface via the Iga beta autotransporter pathway. Gene 178: 107-110.

Kachlany, S. C., P. J. Planet, R. Desalle, D. H. Fine, D. H. Figurski & J. B. Kaplan, (2001) flp-1, the first representative of a new pilin gene subfamily, is required for non-specific adherence of Actinobacillus actinomycetemcomitans. Mol Microbiol 40: 542-554.

Kahler, C. M., L. E. Martin, G. C. Shih, M. M. Rahman, R. W. Carlson & D. S. Stephens, (1998) The (alpha2-->8)-linked polysialic acid capsule and lipooligosaccharide structure both contribute to the ability of serogroup B Neisseria meningitidis to resist the bactericidal activity of normal human serum. Infect Immun 66: 5939-5947.

Kaiser, P. O., T. Riess, C. L. Wagner, D. Linke, A. N. Lupas, H. Schwarz, G. Raddatz, A. Schafer & V. A. Kempf, (2008) The head of Bartonella adhesin A is crucial for host cell interaction of Bartonella henselae. Cellular microbiology 10: 2223-2234.

Kapperud, G., E. Namork, M. Skurnik & T. Nesbakken, (1987) Plasmid-mediated surface fibrillae of Yersinia pseudotuberculosis and Yersinia enterocolitica: relationship to the outer membrane protein YOP1 and possible importance for pathogenesis. Infect Immun 55: 2247-2254.

Karlyshev, A. V., M. V. McCrossan & B. W. Wren, (2001) Demonstration of polysaccharide capsule in Campylobacter jejuni using electron microscopy. Infect Immun 69: 5921-5924.

167

Kehl-Fie, T. E. & J. W. St Geme, 3rd, (2007) Identification and characterization of an RTX toxin in the emerging pathogen Kingella kingae. J Bacteriol 189: 430-436.

Kehl-Fie, T. E., S. E. Miller & J. W. St Geme, 3rd, (2008) Kingella kingae expresses type IV pili that mediate adherence to respiratory epithelial and synovial cells. J Bacteriol 190: 7157-7163.

Kehl-Fie, T. E., E. A. Porsch, S. E. Miller & J. W. St Geme, 3rd, (2009) Expression of Kingella kingae type IV pili is regulated by sigma54, PilS, and PilR. J Bacteriol 191: 4976- 4986.

Kehl-Fie, T. E., E. A. Porsch, P. Yagupsky, E. A. Grass, C. Obert, D. K. Benjamin, Jr. & J. W. St Geme, 3rd, (2010) Examination of type IV pilus expression and pilus-associated phenotypes in Kingella kingae clinical isolates. Infect Immun 78: 1692-1699.

Kennedy, C. A. & H. Rosen, (1988) Kingella kingae bacteremia and adult epiglottitis in a granulocytopenic host. The American journal of medicine 85: 701-702.

Kiang, K. M., F. Ogunmodede, B. A. Juni, D. J. Boxrud, A. Glennen, J. M. Bartkus, E. A. Cebelinski, K. Harriman, S. Koop, R. Faville, R. Danila & R. Lynfield, (2005) Outbreak of osteomyelitis/septic arthritis caused by Kingella kingae among child care center attendees. Pediatrics 116: e206-213.

Kim, K. J., S. J. Elliott, F. Di Cello, M. F. Stins & K. S. Kim, (2003) The K1 capsule modulates trafficking of E. coli-containing vacuoles and enhances intracellular bacterial survival in human brain microvascular endothelial cells. Cellular microbiology 5: 245-252.

Kirchner, M., D. Heuer & T. F. Meyer, (2005a) CD46-independent binding of neisserial type IV pili and the major pilus adhesin, PilC, to human epithelial cells. Infect Immun 73: 3072-3082.

Kirchner, M. & T. F. Meyer, (2005b) The PilC adhesin of the Neisseria type IV pilus- binding specificities and new insights into the nature of the host cell receptor. Mol Microbiol 56: 945-957.

Kirn, T. J., N. Bose & R. K. Taylor, (2003) Secretion of a soluble colonization factor by the TCP type 4 pilus biogenesis pathway in Vibrio cholerae. Mol Microbiol 49: 81-92.

Kroll, J. S., S. Zamze, B. Loynds & E. R. Moxon, (1989) Common organization of chromosomal loci for production of different capsular polysaccharides in Haemophilus influenzae. J Bacteriol 171: 3343-3347.

168

Kroll, J. S., B. Loynds, L. N. Brophy & E. R. Moxon, (1990) The bex locus in encapsulated Haemophilus influenzae: a chromosomal region involved in capsule polysaccharide export. Mol Microbiol 4: 1853-1862.

Kugler, K. C., D. J. Biedenbach & R. N. Jones, (1999) Determination of the antimicrobial activity of 29 clinically important compounds tested against fastidious HACEK group organisms. Diagnostic microbiology and infectious disease 34: 73-76.

La Scola, B., I. Iorgulescu & G. Bollini, (1998) Five cases of Kingella kingae skeletal infection in a French hospital. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 17: 512-515.

La Selve, H., J. Berard, G. Barbe, N. Cochat & P. Cochat, (1986) Osteoarthritis and osteomyelitis due to Kingella kingae in children. Apropos of 5 cases and review of the literature. Pediatrie 41: 297-304.

Lacour, M., M. Duarte, A. Beutler, R. Auckenthaler & S. Suter, (1991) Osteoarticular infections due to Kingella kingae in children. European journal of pediatrics 150: 612-618.

Larue, K., R. C. Ford, L. M. Willis & C. Whitfield, (2011) Functional and structural characterization of polysaccharide co-polymerase proteins required for polymer export in ATP-binding cassette transporter-dependent capsule biosynthesis pathways. The Journal of biological chemistry 286: 16658-16668.

Lauer, P., N. H. Albertson & M. Koomey, (1993) Conservation of genes encoding components of a type IV pilus assembly/two-step protein export pathway in Neisseria gonorrhoeae. Mol Microbiol 8: 357-368.

Lehours, P., A. M. Freydiere, O. Richer, C. Burucoa, S. Boisset, P. Lanotte, M. F. Prere, A. Ferroni, C. Lafuente, F. Vandenesch, F. Megraud & A. Menard, (2011) The rtxA toxin gene of Kingella kingae: a pertinent target for molecular diagnosis of osteoarticular infections. J Clin Microbiol 49: 1245-1250.

Leininger, E., M. Roberts, J. G. Kenimer, I. G. Charles, N. Fairweather, P. Novotny & M. J. Brennan, (1991) Pertactin, an Arg-Gly-Asp-containing Bordetella pertussis surface protein that promotes adherence of mammalian cells. Proc Natl Acad Sci U S A 88: 345- 349.

Luderitz, O., K. Tanamoto, C. Galanos, G. R. McKenzie, H. Brade, U. Zahringer, E. T. Rietschel, S. Kusumoto & T. Shiba, (1984) Lipopolysaccharides: structural principles and biologic activities. Reviews of infectious diseases 6: 428-431.

169

Luke, N. R., A. J. Howlett, J. Shao & A. A. Campagnari, (2004) Expression of type IV pili by Moraxella catarrhalis is essential for natural competence and is affected by iron limitation. Infect Immun 72: 6262-6270.

Lundy, D. W. & D. K. Kehl, (1998) Increasing prevalence of Kingella kingae in osteoarticular infections in young children. Journal of pediatric orthopedics 18: 262-267.

Lyskowski, A., J. C. Leo & A. Goldman, (2011) Structure and biology of trimeric autotransporter adhesins. Adv Exp Med Biol 715: 143-158.

Mackinnon, F. G., R. Borrow, A. R. Gorringe, A. J. Fox, D. M. Jones & A. Robinson, (1993) Demonstration of lipooligosaccharide immunotype and capsule as virulence factors for Neisseria meningitidis using an infant mouse intranasal infection model. Microb Pathog 15: 359-366.

Maier, B., L. Potter, M. So, C. D. Long, H. S. Seifert & M. P. Sheetz, (2002) Single pilus motor forces exceed 100 pN. Proc Natl Acad Sci U S A 99: 16012-16017.

Maroudas, A., (1979) Physiochemical properties of articular cartilage. In: Adult Articular Cartilage. M. A. R. Freeman (ed). Kent, UK: Pitman Medical, pp. 215-290.

Marrs, C. F., G. Schoolnik, J. M. Koomey, J. Hardy, J. Rothbard & S. Falkow, (1985) Cloning and sequencing of a Moraxella bovis pilin gene. J Bacteriol 163: 132-139.

Marrs, C. F., W. W. Ruehl, G. K. Schoolnik & S. Falkow, (1988) Pilin-gene phase variation of Moraxella bovis is caused by an inversion of the pilin genes. J Bacteriol 170: 3032-3039.

Marrs, C. F., F. W. Rozsa, M. Hackel, S. P. Stevens & A. C. Glasgow, (1990) Identification, cloning, and sequencing of piv, a new gene involved in inverting the pilin genes of Moraxella lacunata. J Bacteriol 172: 4370-4377.

Martin, P. R., M. Hobbs, P. D. Free, Y. Jeske & J. S. Mattick, (1993) Characterization of pilQ, a new gene required for the biogenesis of type 4 fimbriae in Pseudomonas aeruginosa. Mol Microbiol 9: 857-868.

Matta, M., D. Wermert, I. Podglajen, O. Sanchez, A. Buu-Hoi, L. Gutmann, G. Meyer & J. L. Mainardi, (2007) Molecular diagnosis of Kingella kingae pericarditis by amplification and sequencing of the 16S rRNA gene. J Clin Microbiol 45: 3133-3134.

Mattick, J. S., (2002) Type IV pili and twitching motility. Annual review of microbiology 56: 289-314.

170

Meng, G., J. W. St Geme, 3rd & G. Waksman, (2008) Repetitive architecture of the Haemophilus influenzae Hia trimeric autotransporter. J Mol Biol 384: 824-836.

Meri, S. & M. K. Pangburn, (1990) Discrimination between activators and nonactivators of the alternative pathway of complement: regulation via a sialic acid/polyanion binding site on factor H. Proc Natl Acad Sci U S A 87: 3982-3986.

Merz, A. J., C. A. Enns & M. So, (1999) Type IV pili of pathogenic Neisseriae elicit cortical plaque formation in epithelial cells. Mol Microbiol 32: 1316-1332.

Merz, A. J., M. So & M. P. Sheetz, (2000) Pilus retraction powers bacterial twitching motility. Nature 407: 98-102.

Meyer, T. F., E. Billyard, R. Haas, S. Storzbach & M. So, (1984) Pilus genes of Neisseria gonorrheae: chromosomal organization and DNA sequence. Proc Natl Acad Sci U S A 81: 6110-6114.

Meyer, T. F., R. Halter & J. Pohlner, (1987) Mechanism of extracellular secretion of an IgA protease by gram-negative host cells. Adv Exp Med Biol 216B: 1271-1281.

Morand, P. C., P. Tattevin, E. Eugene, J. L. Beretti & X. Nassif, (2001) The adhesive property of the type IV pilus-associated component PilC1 of pathogenic Neisseria is supported by the conformational structure of the N-terminal part of the molecule. Mol Microbiol 40: 846-856.

Morand, P. C., E. Bille, S. Morelle, E. Eugene, J. L. Beretti, M. Wolfgang, T. F. Meyer, M. Koomey & X. Nassif, (2004) Type IV pilus retraction in pathogenic Neisseria is regulated by the PilC proteins. EMBO J 23: 2009-2017.

Morand, P. C., M. Drab, K. Rajalingam, X. Nassif & T. F. Meyer, (2009) Neisseria meningitidis differentially controls host cell motility through PilC1 and PilC2 components of type IV Pili. PloS one 4: e6834.

Moumile, K., J. Merckx, C. Glorion, P. Berche & A. Ferroni, (2003) Osteoarticular infections caused by Kingella kingae in children: contribution of polymerase chain reaction to the microbiologic diagnosis. Pediatr Infect Dis J 22: 837-839.

Moxon, E. R. & K. A. Vaughn, (1981) The type b capsular polysaccharide as a virulence determinant of Haemophilus influenzae: studies using clinical isolates and laboratory transformants. J Infect Dis 143: 517-524.

171

Moylett, E. H., S. N. Rossmann, H. R. Epps & G. J. Demmler, (2000) Importance of Kingella kingae as a pediatric pathogen in the United States. Pediatr Infect Dis J 19: 263- 265.

Nassif, X., J. L. Beretti, J. Lowy, P. Stenberg, P. O'Gaora, J. Pfeifer, S. Normark & M. So, (1994) Roles of pilin and PilC in adhesion of Neisseria meningitidis to human epithelial and endothelial cells. Proc Natl Acad Sci U S A 91: 3769-3773.

Noftal, F., A. Mersal, Y. Yaschuk, J. Wedge & W. Albritton, (1988) Osteomyelitis due to Kingella kingae infection. Canadian journal of surgery. Journal canadien de chirurgie 31: 21-22.

Nummelin, H., M. C. Merckel, J. C. Leo, H. Lankinen, M. Skurnik & A. Goldman, (2004) The Yersinia adhesin YadA collagen-binding domain structure is a novel left-handed parallel beta-roll. EMBO J 23: 701-711.

Nunn, D., S. Bergman & S. Lory, (1990) Products of three accessory genes, pilB, pilC, and pilD, are required for biogenesis of Pseudomonas aeruginosa pili. J Bacteriol 172: 2911-2919.

Odum, L., K. T. Jensen & T. D. Slotsbjerg, (1984) Endocarditis due to Kingella kingae. European journal of clinical microbiology 3: 263-266.

Orans, J., M. D. Johnson, K. A. Coggan, J. R. Sperlazza, R. W. Heiniger, M. C. Wolfgang & M. R. Redinbo, (2010) Crystal structure analysis reveals Pseudomonas PilY1 as an essential calcium-dependent regulator of bacterial surface motility. Proc Natl Acad Sci U S A 107: 1065-1070.

Orskov, I., F. Orskov, B. Jann & K. Jann, (1977) Serology, chemistry, and genetics of O and K antigens of Escherichia coli. Bacteriological reviews 41: 667-710.

Osborn, M. J., (1963) Studies on the Gram-Negative Cell Wall. I. Evidence for the Role of 2-Keto- 3-Deoxyoctonate in the Lipopolysaccharide of Salmonella typhimurium. Proc Natl Acad Sci U S A 50: 499-506.

Paerregaard, A., F. Espersen, O. M. Jensen & M. Skurnik, (1991) Interactions between Yersinia enterocolitica and rabbit ileal mucus: growth, adhesion, penetration, and subsequent changes in surface hydrophobicity and ability to adhere to ileal brush border membrane vesicles. Infect Immun 59: 253-260.

Paerregaard, A., (1992) Interactions between Yersinia enterocolitica and the host with special reference to virulence plasmid encoded adhesion and humoral immunity. Danish medical bulletin 39: 155-172.

172

Pallen, M. J., R. R. Chaudhuri & I. R. Henderson, (2003) Genomic analysis of secretion systems. Current opinion in microbiology 6: 519-527.

Parge, H. E., K. T. Forest, M. J. Hickey, D. A. Christensen, E. D. Getzoff & J. A. Tainer, (1995) Structure of the fibre-forming protein pilin at 2.6 A resolution. Nature 378: 32-38.

Park, H. S., M. Wolfgang & M. Koomey, (2002) Modification of type IV pilus-associated epithelial cell adherence and multicellular behavior by the PilU protein of Neisseria gonorrhoeae. Infect Immun 70: 3891-3903.

Pazzani, C., C. Rosenow, G. J. Boulnois, D. Bronner, K. Jann & I. S. Roberts, (1993) Molecular analysis of region 1 of the Escherichia coli K5 antigen gene cluster: a region encoding proteins involved in cell surface expression of capsular polysaccharide. J Bacteriol 175: 5978-5983.

Peak, I. R., Y. Srikhanta, M. Dieckelmann, E. R. Moxon & M. P. Jennings, (2000) Identification and characterisation of a novel conserved outer membrane protein from Neisseria meningitidis. FEMS Immunol Med Microbiol 28: 329-334.

Pearson, M. M., E. R. Lafontaine, N. J. Wagner, J. W. St Geme, 3rd & E. J. Hansen, (2002) A hag mutant of Moraxella catarrhalis strain O35E is deficient in hemagglutination, autoagglutination, and immunoglobulin D-binding activities. Infect Immun 70: 4523-4533.

Pelicic, V., (2008) Type IV pili: e pluribus unum? Mol Microbiol 68: 827-837.

Perry, A. C., C. A. Hart, I. J. Nicolson, J. E. Heckels & J. R. Saunders, (1987) Inter-strain homology of pilin gene sequences in Neisseria meningitidis isolates that express markedly different antigenic pilus types. J Gen Microbiol 133: 1409-1418.

Perry, A. C., I. J. Nicolson & J. R. Saunders, (1988) Neisseria meningitidis C114 contains silent, truncated pilin genes that are homologous to Neisseria gonorrhoeae pil sequences. J Bacteriol 170: 1691-1697.

Pichichero, M. E., M. Loeb, Anderson & D. H. Smith, (1982) Do pili play a role in pathogenicity of Haemophilus influenzae type B? Lancet 2: 960-962.

Pilz, D., T. Vocke, J. Heesemann & V. Brade, (1992) Mechanism of YadA-mediated serum resistance of Yersinia enterocolitica serotype O3. Infect Immun 60: 189-195.

Pohlner, J., R. Halter, K. Beyreuther & T. F. Meyer, (1987a) Gene structure and extracellular secretion of Neisseria gonorrhoeae IgA protease. Nature 325: 458-462.

173

Pohlner, J., R. Halter & T. F. Meyer, (1987b) Neisseria gonorrhoeae IgA protease. Secretion and implications for pathogenesis. Antonie van Leeuwenhoek 53: 479-484.

Pollard, A. J. & M. C. J. Maiden, (2001) Meningococcal vaccines methods and protocols. In: Methods in molecular medicine 66. Totowa, N.J.: Humana Press,, pp. 1 online resource (xvii, 421 p.).

Potts, W. J. & J. R. Saunders, (1988) Nucleotide sequence of the structural gene for class I pilin from Neisseria meningitidis: homologies with the pilE locus of Neisseria gonorrhoeae. Mol Microbiol 2: 647-653.

Prilipov, A., P. S. Phale, P. Van Gelder, J. P. Rosenbusch & R. Koebnik, (1998) Coupling site-directed mutagenesis with high-level expression: large scale production of mutant porins from E. coli. FEMS Microbiol Lett 163: 65-72.

Pujol, C., E. Eugene, L. de Saint Martin & X. Nassif, (1997) Interaction of Neisseria meningitidis with a polarized monolayer of epithelial cells. Infect Immun 65: 4836-4842.

Pujol, C., E. Eugene, M. Marceau & X. Nassif, (1999) The meningococcal PilT protein is required for induction of intimate attachment to epithelial cells following pilus- mediated adhesion. Proc Natl Acad Sci U S A 96: 4017-4022.

Purdy, G. E., C. R. Fisher & S. M. Payne, (2007) IcsA surface presentation in Shigella flexneri requires the periplasmic chaperones DegP, Skp, and SurA. J Bacteriol 189: 5566- 5573.

Radin, J. N., S. A. Grass, G. Meng, S. E. Cotter, G. Waksman & J. W. St Geme, 3rd, (2009) Structural basis for the differential binding affinities of the HsfBD1 and HsfBD2 domains in the Haemophilus influenzae Hsf adhesin. J Bacteriol 191: 5068-5075.

Rahman, M., H. Kallstrom, S. Normark & A. B. Jonsson, (1997) PilC of pathogenic Neisseria is associated with the bacterial cell surface. Mol Microbiol 25: 11-25.

Ramana, K. & S. Mohanty, (2009) An adult case of urinary tract infection with Kingella kingae: a case report. Journal of medical case reports 3: 7236.

Ramer, S. W., G. K. Schoolnik, C. Y. Wu, J. Hwang, S. A. Schmidt & D. Bieber, (2002) The type IV pilus assembly complex: biogenic interactions among the bundle-forming pilus proteins of enteropathogenic Escherichia coli. J Bacteriol 184: 3457-3465.

Rao, V. K. & A. Progulske-Fox, (1993) Cloning and sequencing of two type 4 (N- methylphenylalanine) pilin genes from Eikenella corrodens. J Gen Microbiol 139: 651-660.

174

Raymond, J., M. Bergeret, F. Bargy & G. Missenard, (1986) Isolation of two strains of Kingella kingae associated with septic arthritis. J Clin Microbiol 24: 1100-1101.

Roberts, I. S., (1996) The biochemistry and genetics of capsular polysaccharide production in bacteria. Annual review of microbiology 50: 285-315.

Roggenkamp, A., N. Ackermann, C. A. Jacobi, K. Truelzsch, H. Hoffmann & J. Heesemann, (2003) Molecular analysis of transport and oligomerization of the Yersinia enterocolitica adhesin YadA. J Bacteriol 185: 3735-3744.

Rossiter, A. E., D. L. Leyton, K. Tveen-Jensen, D. F. Browning, Y. Sevastsyanovich, T. J. Knowles, K. B. Nichols, A. F. Cunningham, M. Overduin, M. A. Schembri & I. R. Henderson, (2011) The essential beta-barrel assembly machinery complex components BamD and BamA are required for autotransporter biogenesis. J Bacteriol 193: 4250-4253.

Rotstein, A., I. E. Konstantinov & D. J. Penny, (2010) Kingella-infective endocarditis resulting in a perforated aortic root abscess and fistulous connection between the sinus of Valsalva and the left atrium in a child. Cardiology in the young 20: 332-333.

Rudel, T., J. P. van Putten, C. P. Gibbs, R. Haas & T. F. Meyer, (1992) Interaction of two variable proteins (PilE and PilC) required for pilus-mediated adherence of Neisseria gonorrhoeae to human epithelial cells. Mol Microbiol 6: 3439-3450.

Rudel, T., D. Facius, R. Barten, I. Scheuerpflug, E. Nonnenmacher & T. F. Meyer, (1995a) Role of pili and the phase-variable PilC protein in natural competence for transformation of Neisseria gonorrhoeae. Proc Natl Acad Sci U S A 92: 7986-7990.

Rudel, T., I. Scheurerpflug & T. F. Meyer, (1995b) Neisseria PilC protein identified as type-4 pilus tip-located adhesin. Nature 373: 357-359.

Ruiz-Perez, F., I. R. Henderson, D. L. Leyton, A. E. Rossiter, Y. Zhang & J. P. Nataro, (2009) Roles of periplasmic chaperone proteins in the biogenesis of serine protease autotransporters of Enterobacteriaceae. J Bacteriol 191: 6571-6583.

Runnels, P. L. & H. W. Moon, (1984) Capsule reduces adherence of enterotoxigenic Escherichia coli to isolated intestinal epithelial cells of pigs. Infect Immun 45: 737-740.

Ryan, K. J., C. G. Ray & J. C. Sherris, (2004) Sherris medical microbiology : an introduction to infectious diseases, p. xiii, 979 p. McGraw-Hill, Medical Pub. Division, New York.

Ryll, R. R., T. Rudel, I. Scheuerpflug, R. Barten & T. F. Meyer, (1997) PilC of Neisseria meningitidis is involved in class II pilus formation and restores pilus assembly, natural

175

transformation competence and adherence to epithelial cells in PilC-deficient gonococci. Mol Microbiol 23: 879-892.

Sambrook, J., E. F. Fritsch & T. Maniatis, (1989) Molecular cloning : a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

Sandt, C. H. & C. W. Hill, (2000) Four different genes responsible for nonimmune immunoglobulin-binding activities within a single strain of Escherichia coli. Infect Immun 68: 2205-2214.

Saphyakhajon, P., A. Y. Joshi, W. C. Huskins, N. K. Henry & T. G. Boyce, (2008) Empiric antibiotic therapy for acute osteoarticular infections with suspected methicillin-resistant Staphylococcus aureus or Kingella. Pediatr Infect Dis J 27: 765-767.

Sauri, A., Z. Soprova, D. Wickstrom, J. W. de Gier, R. C. Van der Schors, A. B. Smit, W. S. Jong & J. Luirink, (2009) The Bam (Omp85) complex is involved in secretion of the autotransporter haemoglobin protease. Microbiology 155: 3982-3991.

Scarselli, M., D. Serruto, P. Montanari, B. Capecchi, J. Adu-Bobie, D. Veggi, R. Rappuoli, M. Pizza & B. Arico, (2006) Neisseria meningitidis NhhA is a multifunctional trimeric autotransporter adhesin. Mol Microbiol 61: 631-644.

Scheuerpflug, I., T. Rudel, R. Ryll, J. Pandit & T. F. Meyer, (1999) Roles of PilC and PilE proteins in pilus-mediated adherence of Neisseria gonorrhoeae and Neisseria meningitidis to human erythrocytes and endothelial and epithelial cells. Infect Immun 67: 834-843.

Schneider, M. C., R. M. Exley, H. Chan, I. Feavers, Y. H. Kang, R. B. Sim & C. M. Tang, (2006) Functional significance of factor H binding to Neisseria meningitidis. J Immunol 176: 7566-7575.

Schulze-Koops, H., H. Burkhardt, J. Heesemann, K. von der Mark & F. Emmrich, (1992) Plasmid-encoded outer membrane protein YadA mediates specific binding of enteropathogenic yersiniae to various types of collagen. Infect Immun 60: 2153-2159.

Seifert, H. S., R. Ajioka & M. So, (1988a) Alternative model for Neisseria gonorrhoeae pilin variation. Vaccine 6: 107-109.

Seifert, H. S. & M. So, (1988b) Genetic mechanisms of bacterial antigenic variation. Microbiological reviews 52: 327-336.

Seifert, H. S., (1997) Insertionally inactivated and inducible recA alleles for use in Neisseria. Gene 188: 215-220.

176

Sẽna, A. C., P. Seed, B. Nicholson, M. Joyce & C. K. Cunningham, (2010) Kingella kingae endocarditis and a cluster investigation among daycare attendees. Pediatr Infect Dis J 29: 86-88.

Shaw, C. E. & R. K. Taylor, (1990) Vibrio cholerae O395 tcpA pilin gene sequence and comparison of predicted protein structural features to those of type 4 pilins. Infect Immun 58: 3042-3049.

Sheets, A. J., S. A. Grass, S. E. Miller & J. W. St Geme, 3rd, (2008) Identification of a novel trimeric autotransporter adhesin in the cryptic genospecies of Haemophilus. J Bacteriol 190: 4313-4320.

Shelton, M. M., M. P. Nachtigal, D. A. Yngve, W. A. Herndon & H. D. Riley, Jr., (1988) Kingella kingae osteomyelitis: report of two cases involving the epiphysis. Pediatr Infect Dis J 7: 421-424.

Skurnik, M. & H. Wolf-Watz, (1989) Analysis of the yopA gene encoding the Yop1 virulence determinants of Yersinia spp. Mol Microbiol 3: 517-529.

Slonim, A., E. S. Walker, E. Mishori, N. Porat, R. Dagan & P. Yagupsky, (1998) Person-to- person transmission of Kingella kingae among day care center attendees. J Infect Dis 178: 1843-1846.

Slonim, A., M. Steiner & P. Yagupsky, (2003) Immune response to invasive Kingella kingae infections, age-related incidence of disease, and levels of antibody to outer- membrane proteins. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 37: 521-527.

Spinosa, M. R., C. Progida, A. Tala, L. Cogli, P. Alifano & C. Bucci, (2007) The Neisseria meningitidis capsule is important for intracellular survival in human cells. Infect Immun 75: 3594-3603.

St Geme, J. W., 3rd & S. Falkow, (1991) Loss of capsule expression by Haemophilus influenzae type b results in enhanced adherence to and invasion of human cells. Infect Immun 59: 1325-1333.

St Geme, J. W., 3rd, M. L. de la Morena & S. Falkow, (1994) A Haemophilus influenzae IgA protease-like protein promotes intimate interaction with human epithelial cells. Mol Microbiol 14: 217-233.

St Geme, J. W., 3rd, V. V. Kumar, D. Cutter & S. J. Barenkamp, (1998) Prevalence and distribution of the hmw and hia genes and the HMW and Hia adhesins among

177

genetically diverse strains of nontypeable Haemophilus influenzae. Infect Immun 66: 364- 368.

Stahelin, J., D. Goldenberger, H. E. Gnehm & M. Altwegg, (1998) Polymerase chain reaction diagnosis of Kingella kingae arthritis in a young child. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 27: 1328-1329.

Stols, L., M. Gu, L. Dieckman, R. Raffen, F. R. Collart & M. I. Donnelly, (2002) A new vector for high-throughput, ligation-independent cloning encoding a tobacco etch virus protease cleavage site. Protein expression and purification 25: 8-15.

Strom, M. S. & S. Lory, (1993a) Structure-function and biogenesis of the type IV pili. Annual review of microbiology 47: 565-596.

Strom, M. S., D. N. Nunn & S. Lory, (1993b) A single bifunctional enzyme, PilD, catalyzes cleavage and N-methylation of proteins belonging to the type IV pilin family. Proc Natl Acad Sci U S A 90: 2404-2408.

Sukupolvi-Petty, S., S. Grass & J. W. St Geme, 3rd, (2006) The Haemophilus influenzae Type b hcsA and hcsB gene products facilitate transport of capsular polysaccharide across the outer membrane and are essential for virulence. J Bacteriol 188: 3870-3877.

Surana, N. K., D. Cutter, S. J. Barenkamp & J. W. St Geme, 3rd, (2004) The Haemophilus influenzae Hia autotransporter contains an unusually short trimeric translocator domain. The Journal of biological chemistry 279: 14679-14685.

Swanson, J., S. Bergstrom, K. Robbins, O. Barrera, D. Corwin & J. M. Koomey, (1986) Gene conversion involving the pilin structural gene correlates with pilus+ in equilibrium with pilus- changes in Neisseria gonorrhoeae. Cell 47: 267-276.

Swartley, J. S., J. H. Ahn, L. J. Liu, C. M. Kahler & D. S. Stephens, (1996) Expression of sialic acid and polysialic acid in serogroup B Neisseria meningitidis: divergent transcription of biosynthesis and transport operons through a common promoter region. J Bacteriol 178: 4052-4059.

Szeto, T. H., A. Dessen & V. Pelicic, (2011) Structure/function analysis of Neisseria meningitidis PilW, a conserved protein that plays multiple roles in type IV pilus biology. Infect Immun 79: 3028-3035.

Tahir, Y. E., P. Kuusela & M. Skurnik, (2000) Functional mapping of the Yersinia enterocolitica adhesin YadA. Identification Of eight NSVAIG - S motifs in the amino- terminal half of the protein involved in collagen binding. Mol Microbiol 37: 192-206.

178

Tamm, A., A. M. Tarkkanen, T. K. Korhonen, P. Kuusela, P. Toivanen & M. Skurnik, (1993) Hydrophobic domains affect the collagen-binding specificity and surface polymerization as well as the virulence potential of the YadA protein of Yersinia enterocolitica. Mol Microbiol 10: 995-1011.

Tertti, R., M. Skurnik, T. Vartio & P. Kuusela, (1992) Adhesion protein YadA of Yersinia species mediates binding of bacteria to fibronectin. Infect Immun 60: 3021-3024.

Tonjum, T., N. E. Freitag, E. Namork & M. Koomey, (1995) Identification and characterization of pilG, a highly conserved pilus-assembly gene in pathogenic Neisseria. Mol Microbiol 16: 451-464.

Toshniwal, R., T. C. Draghi, F. E. Kocka & C. A. Kallick, (1986) Manifestations of Kingella kingae infections in adults: resemblance to neisserial infections. Diagnostic microbiology and infectious disease 5: 81-85.

Turner, L. R., J. C. Lara, D. N. Nunn & S. Lory, (1993) Mutations in the consensus ATP- binding sites of XcpR and PilB eliminate extracellular protein secretion and pilus biogenesis in Pseudomonas aeruginosa. J Bacteriol 175: 4962-4969.

Tzeng, Y. L., A. K. Datta, C. A. Strole, M. A. Lobritz, R. W. Carlson & D. S. Stephens, (2005) Translocation and surface expression of lipidated serogroup B capsular Polysaccharide in Neisseria meningitidis. Infect Immun 73: 1491-1505.

Van Eldere, J., L. Brophy, B. Loynds, P. Celis, I. Hancock, S. Carman, J. S. Kroll & E. R. Moxon, (1995) Region II of the Haemophilus influenzae type b capsulation locus is involved in serotype-specific polysaccharide synthesis. Mol Microbiol 15: 107-118.

Van Erps, J., E. Schmedding, A. Naessens & B. Keymeulen, (1992) Kingella kingae, a rare cause of bacterial meningitis. Clinical neurology and neurosurgery 94: 173-175.

Verdier, I., A. Gayet-Ageron, C. Ploton, P. Taylor, Y. Benito, A. M. Freydiere, F. Chotel, J. Berard, P. Vanhems & F. Vandenesch, (2005) Contribution of a broad range polymerase chain reaction to the diagnosis of osteoarticular infections caused by Kingella kingae: description of twenty-four recent pediatric diagnoses. Pediatr Infect Dis J 24: 692-696.

Villar, M. T., R. L. Hirschberg & M. R. Schaefer, (2001) Role of the Eikenella corrodens pilA locus in pilus function and phase variation. J Bacteriol 183: 55-62.

Virji, M., H. Kayhty, D. J. Ferguson, C. Alexandrescu, J. E. Heckels & E. R. Moxon, (1991) The role of pili in the interactions of pathogenic Neisseria with cultured human endothelial cells. Mol Microbiol 5: 1831-1841.

179

Virji, M., K. Makepeace, D. J. Ferguson, M. Achtman, J. Sarkari & E. R. Moxon, (1992) Expression of the Opc protein correlates with invasion of epithelial and endothelial cells by Neisseria meningitidis. Mol Microbiol 6: 2785-2795.

Virji, M., K. Makepeace, D. J. Ferguson, M. Achtman & E. R. Moxon, (1993) Meningococcal Opa and Opc proteins: their role in colonization and invasion of human epithelial and endothelial cells. Mol Microbiol 10: 499-510.

Vogel, U., S. Hammerschmidt & M. Frosch, (1996) Sialic acids of both the capsule and the sialylated lipooligosaccharide of Neisseria meningitidis serogroup B are prerequisites for virulence of meningococci in the infant rat. Med Microbiol Immunol 185: 81-87.

Vogel, U., A. Weinberger, R. Frank, A. Muller, J. Kohl, J. P. Atkinson & M. Frosch, (1997) Complement factor C3 deposition and serum resistance in isogenic capsule and lipooligosaccharide sialic acid mutants of serogroup B Neisseria meningitidis. Infect Immun 65: 4022-4029.

Voulhoux, R., M. P. Bos, J. Geurtsen, M. Mols & J. Tommassen, (2003) Role of a highly conserved bacterial protein in outer membrane protein assembly. Science 299: 262-265.

Wagner, J. K., J. E. Heindl, A. N. Gray, S. Jain & M. B. Goldberg, (2009) Contribution of the periplasmic chaperone Skp to efficient presentation of the autotransporter IcsA on the surface of Shigella flexneri. J Bacteriol 191: 815-821.

Warren, L., (1959) The thiobarbituric acid assay of sialic acids. The Journal of biological chemistry 234: 1971-1975.

Weir, S. & C. F. Marrs, (1992) Identification of type 4 pili in Kingella denitrificans. Infect Immun 60: 3437-3441.

Wells, L., N. Rutter & F. Donald, (2001) Kingella kingae endocarditis in a sixteen-month- old-child. Pediatr Infect Dis J 20: 454-455.

Whitchurch, C. B., M. Hobbs, S. P. Livingston, V. Krishnapillai & J. S. Mattick, (1991) Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialised protein export system widespread in eubacteria. Gene 101: 33-44.

Whitchurch, C. B. & J. S. Mattick, (1994) Characterization of a gene, pilU, required for twitching motility but not phage sensitivity in Pseudomonas aeruginosa. Mol Microbiol 13: 1079-1091.

180

Whitfield, C. & I. S. Roberts, (1999) Structure, assembly and regulation of expression of capsules in Escherichia coli. Mol Microbiol 31: 1307-1319.

Whitfield, C., (2006) Biosynthesis and assembly of capsular polysaccharides in Escherichia coli. Annual review of biochemistry 75: 39-68.

Williams, A. E., D. J. Maskell & E. R. Moxon, (1991) Relationship between intracellular survival in macrophages and virulence of Haemophilus influenzae type b. J Infect Dis 163: 1366-1369.

Winther-Larsen, H. C., M. Wolfgang, S. Dunham, J. P. van Putten, D. Dorward, C. Lovold, F. E. Aas & M. Koomey, (2005) A conserved set of pilin-like molecules controls type IV pilus dynamics and organelle-associated functions in Neisseria gonorrhoeae. Mol Microbiol 56: 903-917.

Wolfgang, M., P. Lauer, H. S. Park, L. Brossay, J. Hebert & M. Koomey, (1998a) PilT mutations lead to simultaneous defects in competence for natural transformation and twitching motility in piliated Neisseria gonorrhoeae. Mol Microbiol 29: 321-330.

Wolfgang, M., H. S. Park, S. F. Hayes, J. P. van Putten & M. Koomey, (1998b) Suppression of an absolute defect in type IV pilus biogenesis by loss-of-function mutations in pilT, a twitching motility gene in Neisseria gonorrhoeae. Proc Natl Acad Sci U S A 95: 14973-14978.

Wolfgang, M., J. P. van Putten, S. F. Hayes & M. Koomey, (1999) The comP locus of Neisseria gonorrhoeae encodes a type IV prepilin that is dispensable for pilus biogenesis but essential for natural transformation. Mol Microbiol 31: 1345-1357.

Woolfrey, B. F., R. T. Lally & R. J. Faville, (1986) Intervertebral diskitis caused by Kingella kingae. American journal of clinical pathology 85: 745-749.

Yagupsky, P., R. Dagan, C. W. Howard, M. Einhorn, I. Kassis & A. Simu, (1992) High prevalence of Kingella kingae in joint fluid from children with septic arthritis revealed by the BACTEC blood culture system. J Clin Microbiol 30: 1278-1281.

Yagupsky, P. & R. Dagan, (1994) Kingella kingae bacteremia in children. Pediatr Infect Dis J 13: 1148-1149.

Yagupsky, P., R. Dagan, F. Prajgrod & M. Merires, (1995a) Respiratory carriage of Kingella kingae among healthy children. Pediatr Infect Dis J 14: 673-678.

181

Yagupsky, P., M. Merires, J. Bahar & R. Dagan, (1995b) Evaluation of novel vancomycin- containing medium for primary isolation of Kingella kingae from upper respiratory tract specimens. J Clin Microbiol 33: 1426-1427.

Yagupsky, P., (1999) Diagnosis of Kingella kingae arthritis by polymerase chain reaction analysis. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 29: 704-705.

Yagupsky, P., O. Katz & N. Peled, (2001) Antibiotic susceptibility of Kingella kingae isolates from respiratory carriers and patients with invasive infections. The Journal of antimicrobial chemotherapy 47: 191-193.

Yagupsky, P., N. Peled & O. Katz, (2002) Epidemiological features of invasive Kingella kingae infections and respiratory carriage of the organism. J Clin Microbiol 40: 4180-4184.

Yagupsky, P., (2004) Kingella kingae: from medical rarity to an emerging paediatric pathogen. Lancet Infect Dis 4: 358-367.

Yagupsky, P. & A. Slonim, (2005) Characterization and immunogenicity of Kingella kingae outer-membrane proteins. FEMS Immunol Med Microbiol 43: 45-50.

Yagupsky, P., Y. Erlich, S. Ariela, R. Trefler & N. Porat, (2006) Outbreak of Kingella kingae skeletal system infections in children in daycare. Pediatr Infect Dis J 25: 526-532.

Yagupsky, P., N. Porat & E. Pinco, (2009a) Pharyngeal colonization by Kingella kingae in children with invasive disease. Pediatr Infect Dis J 28: 155-157.

Yagupsky, P., I. Weiss-Salz, R. Fluss, L. Freedman, N. Peled, R. Trefler, N. Porat & R. Dagan, (2009b) Dissemination of Kingella kingae in the community and long-term persistence of invasive clones. Pediatr Infect Dis J 28: 707-710.

Yagupsky, P., E. Porsch & J. W. St Geme, 3rd, (2011) Kingella kingae: an emerging pathogen in young children. Pediatrics 127: 557-565.

Youssef, D., R. Henaine & S. Di Filippo, (2010) Subtle bacterial endocarditis due to Kingella kingae in an infant: a case report. Cardiology in the young 20: 448-450.

Yu, C., T. Ruiz, C. Lenox & K. P. Mintz, (2008) Functional mapping of an oligomeric autotransporter adhesin of Aggregatibacter actinomycetemcomitans. J Bacteriol 190: 3098- 3109.

182

Zwahlen, A., J. S. Kroll, L. G. Rubin & E. R. Moxon, (1989) The molecular basis of pathogenicity in Haemophilus influenzae: comparative virulence of genetically-related capsular transformants and correlation with changes at the capsulation locus cap. Microb Pathog 7: 225-235.

183

Biography

Eric Allen Porsch was born April 24, 1984 in Harrisburg, PA. He attended

Juniata College in Huntingdon, PA and graduated cum laude with a B.S. in Molecular

Biology in 2005. In the summer of 2004 he was introduced to microbiology research by

Dr. Michael D. Boyle at Juniata College and spent several years studying a variety of aspects of Group A Streptococcus pathogenesis. In 2007, he entered the Graduate

Program in the Department of Molecular Genetics and Microbiology (MGM) at Duke

University and joined the laboratory of Joseph St. Geme, III to study virulence determinants of the pediatric pathogen Kingella kingae. He won first place in Graduate

Student Presentations at the 2010 MGM Department Retreat and presented an abstract at the 48th Annual Meeting of the Infectious Diseases Society of America in 2010 in

Vancouver, BC, Canada. In addition, he participated in the 2010-2011 Duke Scholars of

Infectious Diseases Program. In his free time he enjoys outdoor activities and spending time with his wife, Kimberly and daughter, Carleigh.

Peer-reviewed publications and works-in-progress:

Porsch, E. A., Starr, K.F., and St. Geme III, J.W.. Kingella kingae. In Cherry, Harrison, Kaplan, Steinbach, and Hotez (ed.), Feigin and Cherry's Textbook of Pediatric Infectious Diseases, 7th ed. (In Press).

184

Porsch, E.A., Johnson, M.D., Broadnax, A.D., Garret, C.K., Redinbo, M.R., and St. Geme III, J. W. Differential control of Kingella kingae type IV pilus phenotypes through the calcium-binding sites of PilC1 and PilC2. J. Bacteriol. (In Revision).

Yagupsky, P., Porsch, E.A., and St. Geme III, J.W. (2011) Kingella kingae: an Emerging Pathogen in Preschool Children. Pediatrics 127(3):557-565.

Porsch, E.A., Shertz, C.A., and Boyle, M.D. (2010) A Novel Sample Preparation for Mass Spectral Analysis of Complex Biological Samples. Curr. Proteomics 7(2):90-101.

Kehl-Fie, T.E., Yagupsky, P., Porsch, E.A., Grass, E.A., Benjamin, D., and St. Geme III, J.W. (2010) Examination of type IV pilus expression and pilus-associated phenotypes in Kingella kingae clinical isolates. Infect. Immun. 78(4):1692-1699.

Kehl-Fie, T.E., Porsch, E.A., Miller, S.E., and St. Geme III, J.W. (2009) Expression of Kingella kingae type IV pili is regulated by σ54, PilS, and PilR. J. Bacteriol. 191(15):4976- 4986.

Hess, J.L., Porsch, E.A., Shertz, C.A., and Boyle, M.D. (2007) Immunoglobulin cleavage by the streptococcal cysteine protease IdeS can be detected using Protein G capture and mass spectrometry. J. Microbiol. Meth. 70(2):284-291.

185