The Pennsylvania State University. The Graduate School. College of Agricultural Science UNDERSTANDING HOW VACCINATION AND PARTICULAR VIRULENCE

Transcription

1 The Pennsylvania State University The Graduate School College of Agricultural Science UNDERSTANDING HOW VACCINATION AND PARTICULAR VIRULENCE FACTORS CONTRIBUTE TO BORDETELLA TRANSMISSION A Dissertation in Immunology and Infectious Disease by William E. Smallridge 2014 William E. Smallridge Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy May 2014

2 ii The dissertation of William E. Smallridge was reviewed and approved* by the following: Eric Harvill Professor of Microbiology and Infectious Disease Dissertation advisor Chair of Committee Margherita Cantorna Professor of Molecular Immunology Co-chair of the Graduate Program Mary Kennett Department Head and Professor, Veterinary and Biomedical Sciences David Hughes Assistant Professor of Entomology and Biology *Signatures are on file in the Graduate School

3 iii Abstract Epidemiologists focus on preventing and controlling the spread of disease within populations; however the majority of current experimental research only explores how virulence factors interact with host immunity within or clinical therapeutics targeted to individual hosts. The defining characteristic of any infectious disease is its ability to transmit; investigation into how host and bacterial factors contribute to facilitate transmission is necessary and will contribute to novel strategies that reduce infectious disease burden. This work utilizes a recently developed model system to measure the transmission of Bordetella bronchiseptica from infected mice to naïve susceptible mice. In particular, this dissertation investigates how vaccine-induced immunity and bacterial factors affect transmission of the Bordetella. First, we examined the role of an evolutionarily-conserved virulence factor utilized by many bacterial pathogens, the Type III Secretion System (T3SS). Investigation of a mutant lacking an active T3SS failed to transmit to naïve susceptible mice. Following examination of multiple parameters of transmission, initial colonization was identified to be deficient. In the course of natural colonization, a host s first line of defense against a pathogen is its microbiota. Our analysis revealed that during infection, the T3SS is required for B. bronchiseptica to compete with microbiota in vivo and that displacement of the host microbiota by wild-type B. bronchiseptica occurs during transmission in our model system. Furthermore, prior depletion of microbiota in susceptible mouse by enrofloxacin restored the mutant s ability to transmit. Another important virulence factor utilized by the Bordetella is filamentous hemagglutinin (FHA). A B. bronchiseptica mutant lacking fhab was also unable to transmit in this system. Mice inoculated with the mutant failed to shed any bacteria during the time-course. As an

4 iv important factor in the transmission process, failure to shed dramatically decreases the chances to pass disease to new hosts. FHA is capable of modulating the immune system, and in its absence, B. bronchiseptica failed to recruit neutrophils to the site of infection, a key component previously linked to the shedding of B. bronchiseptica. Finally, we have examined how current vaccine strategies can affect transmission rates, identifying discrepancies between whole-cell and acellular vaccines ability to control transmission. Examination of mice previously vaccinated revealed that only whole-cell vaccination was able to reduce shedding from infected individuals. In particular, components of the immune system that are recruited to the site of infection play an important role in determining the amount of shedding. Increased recruitment of B cells and decreased recruitment of neutrophils to the nasal cavity of vaccinated mice correlated to the amount of shedding observed from individual mice. Combined, the data here suggest roles for both the T3SS and FHA in the transmission of Bordetella, and that current vaccines, while effective at limiting severe disease, fail to affect transmission of the Bordetella. These data are of particular interest, as the causative agent of whooping cough in humans, Bordetella pertussis, has seen resurgence in recent times. The documented rise in cases has raised particular questions concerning current vaccine strategies. Finally, this thesis will discuss the value of this model system in identifying novel mechanisms involved in transmission, as well as therapeutics to control the spread of disease.

5 v Table of Contents List of Figures... vii List of Abbreviations... viii Acknowledgements... ix Chapter 1: Introduction.... Infectious+Disease... 2 Transmission+of+Infectious+Disease... 4 Infection+of+and+Immunity+to+the+Bordetella... 6 Bordetella+Prevalence+and+Vaccine+Strategies... 6 Bordetella*virulence+strategies... 8 Host+Immunity+to+the*Bordetella...11 Bordetella+Transmission+models...12 Preface...14 References...17 Chapter 2: The Bordetella bronchiseptica type three secretion system affects transmission via interactions with host microbiota.... Abstract...27 Introduction...28 Materials+and+Methods...30 Results...32 Discussion...37 References...41 Chapter 3: Filamentous hemagglutinin stimulates the recruitment of neutrophils that is required for shedding of Bordetella bronchiseptica Materials+and+Methods...48 Results...50 Discussion...53 References...57 Chapter 4: Different effects of whole-cell and acellular vaccines on Bordetella transmission.... Abstract...61 Introduction...62 Materials+and+Methods...65 Results...68 Discussion...75 References...80 Chapter 5: Summary and Significance.... Synopsis...84 Summary+and+Implications...84 Bordetella* *B.*bronchiseptica*

6 vi Conclusions...89 References...91 Appendix A: Analysis of transmission within the Classical Bordetella Bvg+regulation+and+bacterial+factors...96 Genetic+diversity+of+B.*bronchiseptica+contribute+to+various+transmission+phenotypes...99 B.*pertussis+and+B.*parapertussis+shedding References Appendix B: Analysis of B. bronchiseptica strain Abstract Introduction Materials+and+Methods Results Discussion References Appendix C:B. bronchiseptica specific antibody titer in the nasal cavity of whole-cell or sham-vaccinated mice during B. bronchiseptica infection Appendix D: Anti-B. bronchiseptica antibodies in serum Appendix E: Effect of vaccination on shedding by TLR4 deficient mice Appendix F: Cytokines contribute to vaccine-induced control of shedding

7 vii List of Figures Figure 2-1 The T3SS is required for efficient transmission of B. bronchiseptica...34 Figure 2-2 T3SS is not required to enable shedding, growth, or persistence in mice...35 Figure 2-3 T3SS is required for competition between microbiota in the nasal cavity of mice...36 Figure 2-4 Depleting susceptible mice with antibiotics facilitates transmission of RB50 bscn..37 Figure 3-1 Filamentous hemagglutinin is required for transmission of B. bronchiseptica...52 Figure 3-2 Filamentous hemagglutinin is required for shedding, but not initial colonization..53 Figure 3-3 Recruitment of neutrophils and dendritic cells to the nasal cavity are altered during RB50 fhab infection...54 Figure 4-1 Vaccination affects B. bronchiseptica shedding, but not colonization...70 Figure 4-2 Vaccination alters the recruitment of B cells and neutrophils to the nasal cavity during B. bronchiseptica infection...71 Figure 4-3 Recruitment of B cells and neutrophils correlate with shedding output during B. bronchiseptica infection...71 Figure 4-4 Adaptive immune components limit shedding of B. bronchiseptica during infection.73 Figure 4-5 Acellular vaccination is ineffective at inhibiting shedding during B. bronchiseptica infection...74 Figure 4-6 Whole-cell vaccination, but not acellular vaccination, is effective at controlling transmission to susceptible individuals...75 Figure 5-1 The outcome of transmission, growth, and persistence in the C3/HeJ mouse model for several B. bronchiseptica mutants...87 Figure A.1 Modulation of the Bvg system in Bordetella is important for efficient shedding and transmission...98 Figure A.2 Shedding of RB50 dnt and RB50 prn Figure A.3 Shedding of 4 genetically distinct B. bronchiseptica isolates Figure A.4 Shedding of B. pertussis and B. parapertussis from C3/HeJ mice Figure B.1 B. bronchiseptica strain 980 form a more robust biofilm than strain Figure B.2 Potential secretion a biofilm-promoting molecule by B. bronchiseptica strain Figure B.3 In vivo and in vitro analysis of B. bronchiseptica strain Figure C.1 B. bronchiseptica specific antibody titer in the nasal cavity of whole-cell or shamvaccinated mice during B. bronchiseptica infection Figure D.1 Anti-B. bronchiseptica antibodies in serum Figure E.1 Effect of vaccination on shedding by TLR4 deficient mice Figure F.1 Cytokines contribute to vaccine-induced control of shedding...123

8 viii List of Abbreviations ACT: Adenylate Cyclase Toxin APCs: professional antigen presenting cells BG: Bordet Gengou Bvg: Bordetella virulence genes camp: cyclic adenosine monophosphate CFU: Colony Forming Unit CO 2 : Carbon Dioxide CyaA: adenylate cyclase DCs: dendritic cells ELISA: Enzyme Linked Immunosorbent Assay FBS: fetal bovine serum FHA: filamentous hemagglutinin FIM: fimbriae g: gravity IACUC: Institutional Animal Care and Use Committee ID50: mean infectious dose IFN: interferon Ig: Immunoglobulin IL: Interleukin LPS: Lipopolysaccharide Log: logarithm MOI: Multiplicity of Infection OD: Optical Density PAMPs: Pathogen-associated Molecular Patterns PCR: Polymerase Chain Reaction PBS: Phosphate Buffered Saline PRN: Pertactin PT: Pertussis Toxin SS: Stainer-Scholte TLR: Toll-like Receptor TNF: Tumor necrosis factor T3SS: Type III Secretion System T6SS: type VI secretion system Th: T helper Treg: T regulatory cells

9 ix Acknowledgements Chapter 4 of this dissertation, Different effects of whole-cell and acellular vaccination on transmission of the Bordetella was published in 2014 in the Journal of Infectious Disease. All permissions have been obtained regarding the reproduction of the text and figures of this manuscript within this dissertation. I would like to extend my gratitude to the USDA-AFRI fellowship in Microbial Functional Genomics, whose fellowship funded me during my dissertation research. I would like to thank the following people for their support throughout my graduate career: in particular, my adviser Dr. Eric Harvill for his guidance and patience during both the highs and lows of graduate school. Past members: Dr. Sara Hester, Dr. Laura Weyrich, Dr. Alexia Karanikas, Dr. Olivier Rolin, and Dr. Jihye Park, whose experience and advice have made me a better scientist. Current members: Laura Goodfield, Sarah Muse, Liron Bendor, and Dr. Yury Ivanov for their scientific discussions and critical analysis of my thesis work. I am particular indebted to Dr. Olivier Rolin for taking me under his wing and contributing important ideas and technical support throughout much of this thesis. I would also like to thank my committee members, Dr. Mary Kennett, Dr. Margherita Cantorna, Dr. David Hughes, staff at the Huck Core facilities, and the Veterinary and Biomedical Science department, whose support and resources were always helpful. Most of all, I would like to thank my family and friends for their love and support. A special thanks to my wife, Chelsea, for her ability to motivate me when I could not see the end and for her love and support, without which, I would have given up long ago.

10 1 Chapter 1 : Introduction

11 2 Infectious Disease Infectious diseases are still a major contributor to morbidity and mortality worldwide despite today s medical advances, and are characterized by their unique ability to transmit between hosts. Approximately 15 million deaths globally are attributed to infectious disease each year, with 4.3 million caused by respiratory infections [1]. The respiratory mucosa offers a unique environmental challenge for infectious diseases caused by bacteria in particular, and will be the focus of this dissertation. The respiratory tract is divided into two major components: the upper respiratory tract and the lower respiratory tract. Each of these components have drastically different environments, varying by temperature, oxygen and carbon dioxide levels, nutrient availability, as well as other factors [2, 3]. In order to adapt to these variations, bacterial pathogens must be able to sense the particular environment in which they find themselves. Therefore, respiratory pathogens, such as Pseudomonas, Bacillus, Francisella, and Bordetella, have evolved numerous complex sensing mechanisms that promote the regulation of specific subsets of genes to successfully colonize new hosts [4-6]. Furthermore, for successful colonization of respiratory pathogens to occur, they must overcome innate physical barriers, such as beating cilia and a thick mucus layer [7, 8]. Pathogens must also survive the secretion of a wide range of antimicrobial peptides, as well as interactions with resident microbiota [9, 10]. Lastly, respiratory pathogens must combat the host immune system. Two types of immune responses exist: the innate response and the adaptive response [11]. Innate immunity comprises host receptors that recognize highly conserved molecular patterns of pathogens. An example is the Toll-like receptor 4 (TLR4) that recognizes lipopolysaccharides (LPS) of Gram-negative bacteria [12]. TLR4 and another pathogen-associated molecular patterns

12 3 (PAMPs) stimulate the production of cytokines that guide the immune response [13]. Resident macrophages and dendritic cells (DCs) at the site of infection can phagocytize and degrade the invading bacteria [14]. These phagocytic cells can also release particular cytokines of their own that recruit other immune factors, such as neutrophils, helping to clear infection [15]. Activated DCs then migrate to the local lymph nodes where they stimulate the adaptive immune system by priming antigen specific T cells and supporting the expansion of antigen specific B cells [16, 17]. Once primed, the adaptive immune system can enhance the clearance of the infecting pathogen and respond quickly to any secondary encounter. Not surprisingly, bacterial pathogens have evolved strategies to successfully colonize new hosts. Staphylococcus aureus is able to avoid the effects of antimicrobial peptides by incorporating D-alanine or D-lysine into teichoic acid and phosphatidylgcerol respectively, while Salmonella enterica serovar Typhimurium incorporates aminoarabinose and additional fatty acids into its lipid A [18]. To overcome the host microbiota, pathogenic bacteria utilize bacteriocins or complex secretion systems like the Type 6 Secretion System (T6SS) to compete [19, 20]. Others utilize unique nutrient sources or simulate a host immune response that it is resistant to in order to out-compete the host microbiota [21, 22]. Finally, pathogenic bacteria utilize numerous virulence systems to evade the host immune system. Effectors secreted by the Type 3 Secretion System (T3SS) can affect the host immune response. Effectors from the T3SS of Yersinia pestis can directly inhibit phagocytosis [23]. Other pathogenic bacteria, such as Listeria monocytogenes, have evolved systems to break out of the phagosome and then replicate within the host cell, avoiding the immune system [24]. Pathogens also have developed ways to interfere with the generation of adaptive immunity. Salmonella typhimurium SifA inhibits expression of major histocompatibility class II (MHC-II) on antigen presenting cells (APCs),

13 4 which is necessary for the activation of the adaptive immune response [25]. In addition, pathogens can skew the immune response by manipulating the secretion of cytokines during the course of infection. The bacterial pathogen Brucella abortus is able to persist in its host by inducing the production of IL-10, an anti-inflammatory cytokine [26]. Not long ago, infectious disease was one of the defining challenges of human existence. Diphtheria, tuberculosis, influenza and others plagued early citizens of the United States [1]. One of the first breakthroughs in controlling infectious diseases was the implementation of basic sanitation and hygiene practices [27]. The discovery of antibiotics in the early 19 th century revolutionized the treatment of infectious disease caused by bacterial pathogens, saving millions of lives since their implementation [28]. Today, one of the most powerful tools in the fight against infectious diseases is vaccination [1]. Vaccination allows individuals to generate adaptive immunity to a pathogen even before the individual comes into contact with the virulent form of the disease, often through exposure to inactive forms or purified components of the pathogen. Even so, infectious diseases are continually evolving and adapting to new ecologic niches. Therefore, we will always confront new or reemerging infectious diseases. To build on the success of the past century, we must continue to use broad approaches that attack infectious disease on many different fronts, including constant surveillance, public health efforts, and implementation of new disease control therapies, discovered through multidisciplinary research. Transmission of Infectious Disease Transmission is the defining characteristic of all infectious diseases and is dependent on many factors. These factors determine whether individuals that come in contact with an infected host will ultimately acquire the infection. Host genetics can influence susceptibility to particular diseases. For example, individuals who carry polymorphisms in TLR4 are more susceptible to

14 5 bacterial infections [29]. Likewise, behavioral factors such as diet and social interactions are also intrinsically linked with the spread of disease [30-32]. Ultimately, disease transmission is dependent on host infectiousness (the ability of a host to spread disease) and is directly linked to the intensity and duration of shedding from a particular host [31]. Shedding of pathogens can depend heavily on immune-pathology [33]. During respiratory infections, the immune response can act like a double-edged sword by potentially attenuating shedding or resulting in pathology that causes the host to cough and sneeze. Immune manipulation by a pathogen may be a means to enhance its transmission, not just its growth and persistence within the host. For example, shedding of Salmonella enterica serovar Typhimurium is dependent on the immune state of the host. Lapses in the control of inflammatory neutrophil infiltration can result in a super-shedder phenotype [33]. Developing therapeutics that target factors that directly influence the shedding of particular pathogens will greatly reduce transmission of particular diseases. Vaccination has been intensely important in combating the spread of disease. Often called the father of immunology, Edward Jenner s pioneering work on smallpox vaccination has changed how infectious diseases are prevented [34]. Vaccination cannot only prevent disease within an individual, but can also control the spread of disease. Vaccination can create herd immunity which reduces infection in unimmunized individuals, as a result of immunizing a significant proportion of the population, by breaking the chain of transmission [35]. However, despite the use of vaccines and other anti-microbials, infectious disease still represents a significant cause of mortality worldwide [36]. The success of smallpox vaccination has been hard to replicate in other infectious diseases. Diseases, such as whooping cough, measles, Haemophilus influenzae, tuberculosis, and tetanus continue to circulate in vaccinated populations

15 6 [37]. Development of new models of disease transmission will help identify the intricacies of the spread of these diseases from host to host, suggesting novel prevention strategies that will lower overall disease burden. Infection of and Immunity to the Bordetella Bordetella Prevalence and Vaccine Strategies There are nine species in the Bordetella genus: Bordetella bronchiseptica, B. pertussis, B. parapertussis, B. holmesii, B. hinzii, B. ansorpii, B. avium, B. petrii, and B. trematum [38]. B. bronchiseptica, B. pertussis and B. parapertussis comprise the classical Bordetella, causing respiratory infections in a wide range of mammalian hosts and are the most commonly studied [39]. B. pertussis and B. parapertussis are thought to have independently evolved from a B. bronchiseptica-like progenitor [38]. B. bronchiseptica mainly causes snuffles in rabbits, atrophic rhinitis in pigs, and kennel cough in dogs; however, B. bronchiseptica can also cause disease in humans [38, 40]. As mainly an animal pathogen, B. bronchiseptica results in considerable economic loss to the agricultural industry, particularly within the swine industry [41, 42]. B. bronchiseptica s highly-related counterparts, B. pertussis and B. parapertussis, cause whooping cough in humans. Whooping cough is a severe respiratory disease which affects approximately 50 million individuals each year, resulting in an estimated 300,000 deaths worldwide [43]. Most of the fatalities occurred in children prior to the development of a whole-cell vaccine in the 1940 s [44]. The use of whole-cell vaccines lowered the rate of B. pertussis infection from 800 cases per 100,000 persons to 0.5 cases per 100,000 persons in the United States by 1976 [44]. In the early 1990s, the inactivated whole-cell vaccine was replaced with an acellular vaccine following reactogenicity concerns [45]. However, despite the initial success of vaccination, the incidence

16 7 rate of whooping cough has steadily climbed since the 1980s, with significant increases occurring in the last ten years [46]. The World Health Organization (WHO) has estimated a near 100% incidence of infection by age 15 [43]. Classified as a reemerging disease, research has been focused on why and how to best to stem the observed increase in cases. Many hypotheses have been drawn as to why increases in whooping cough cases are occurring, such as the evolution of new virulence strategies to circumvent immunity [47]. Recently, B. pertussis isolates, not expressing pertactin, have been observed, raising concerns of vaccine-driven evolution within the Bordetella [48]. Others have attributed the increases to short-lived vaccine protection, potentially leading to transmission of asymptomatic or atypical whooping cough infections [49, 50]. Interestingly, the reemergence of whooping cough has coincided with the development and increased use of acellular vaccines, leading to discussions concerning acellular vaccine efficiency [51]. Acellular vaccines contain some variation of the B. pertussis antigens pertactin, fimbriae type 2 and 3, pertussis toxin and filamentous hemagglutinin, while whole-cell vaccines were prepared directly from heat-inactivated cultures of B. pertussis. Acellular vaccines also uniquely contain aluminum hydroxide as an adjuvant, creating a largely Th2 and Th17 response [52, 53]. During natural infection, however, protective immunity is largely Th1 mediated. Similarly, whole-cell vaccines generate a Th1 mediated response [52, 53]. Th1 cells are able to generate an effective IFN-y response that has previously been shown to be important during B. pertussis infection [54]. During head-to-head trials comparing acellular and whole-cell vaccines, both showed efficacy at preventing severe whooping cough illness. These trials, however, did find that whole-cell had a significant increase in protection against B. pertussis carriage, suggesting that current acellular vaccines fail to elicit effective herd immunity [55]. The

17 8 discrepancy between these two cellular host responses may contribute to the sub-optimal generation of herd immunity induced by current acellular vaccines, leading to increased transmission and incidence. Bordetella virulence strategies The classical Bordetella use common strategies for virulence and their regulation. The majority of the classical Bordetella virulence factors are regulated by the BvgAS system, a two component phosphorelay system, which is composed of BvgS, a sensor kinase protein, and a DNA-binding protein BvgA, the response regulator [56]. The BvgAS regulatory system is sensitive to temperature and compounds, such as MgSO 4 and nicotinic acid [57, 58]. Regulation by the BvgAS system results in phases where particular genes are up-regulated or downregulated. During the Bvg phase, genes for virulence are decreased, while those encoding for motility and nutrient acquisition are increased. This phase is suited for life outside the host, although no known environmental reservoirs exist for the classical Bordetella. The Bvg + phase up-regulates genes, such as toxins and adherence factors, that are required during infection [59, 60]. The classical Bordetella also exhibit an intermediate phase, termed Bvg i. During the Bvg i phase, both adherence factors and motility genes are expressed, while toxins and motility genes are not [61]. Little is known about why the Bvg i phase exists; however, it is speculated to have some role in transmission, as both these subset of genes would be important for initial colonization of new hosts [62]. Overall, the BvgAS system is important for the success of the classical Bordetella, as the recognition of environmental cues allows for the classical Bordetella to efficiently colonize and grow within the highly variable respiratory tract. The classical Bordetella share a number of important adherence factors under the regulation of the BvgAS system. Filamentous hemagglutinin (FHA) plays an important role in

18 9 enabling adherence, by binding, to complement receptor type 3 of immune cells [63] and very late antigen (VLA-5) on epithelial cells through an Arg-Gly-Asp (RGD) tripeptide motif [64]. Secretion of FHA can also contribute to immune modulation. FHA can increase IL-10, leading to decreases in IL-12 production that is necessary to stimulate the differentiation of naïve T cells into Th1 cells [65, 66]. FHA can also lead to the stimulation of proinflammatory and apoptotic responses from human monocyte-like cells and bronchial epithelial cells [67]. Pertactin (PRN) similarly contains an RGD domain, as well as several proline-rich regions and leucine-rich repeats that aid in adherence of Bordetella to host tissues [68]. Finally, the fimbriae (FIM) are important adherence factors in all Gram-negative bacteria [38]. The presence of FIM is known to modulate the immune system by inducing a Th2-mediated response [38]. The secreted toxins, adenylate cyclase toxin (ACT), dermonecrotic toxin (DNT), and pertussis toxin (PT) are all encoded in the classical Bordetella genomes and are utilized during host pathogenesis [39, 69]. ACT inhibits functions of macrophages by converting ATP to camp, preventing chemotaxis, production of reactive oxygen species, and phagocytosis [70-72]. In dendritic cells, ACT inhibits the expression of the co-stimulatory molecule CD40 interfering with their ability to stimulate T cells [73]. ACT is also essential for infection, since mutants lacking the toxin fail to combat the effects of innate immunity and are cleared from individual mice rapidly [74]. The function of DNT within the classical Bordetella is less understood. B. pertussis mutants lacking DNT show no decrease in virulence during infection [75]. However, during B. bronchiseptica infection in swine, DNT contributes to turbinate atrophy, suggesting a role in immuno-pathology of the upper respiratory tract [76]. Finally, PT is only expressed in B. pertussis, despite the ptx locus being conserved in all three classical Bordetella species [77]. PT

19 10 catalyzes the ADP-ribosylation of G proteins, interfering with cell signaling, and also is the main cause of leukocytosis seen in whooping cough patients [78, 79]. Finally, the classical Bordetella express one or both of the T3SS and the T6SS, which have been studied extensively in the B. bronchispetica mouse model [39]. Both are complex needle-like structures that enable the injection of effector proteins directly into host cells. The B. bronchispetica T3SS is responsible for the dampening of the pro-inflammatory IFN-γ Th1 response, allowing for persistence within the host, and contributing to the apoptosis of epithelial cells, macrophages and dendritic cells in vitro [80, 81]. The T6SS is a newly characterized bacterial system and has recently been shown to be present in both B. bronchispetica and B. parapertussis [39]. Most of the work on the T6SS system has been based on its role in bacterial competition [19, 82]. Its role in host-pathogen interactions is less well known; however, recent studies have shed light on its role in Bordetella pathogenesis. During B. bronchispetica infection, the T6SS is required for the induction of IL-1β, IL-6, IL-17, and IL-10 in vitro, and therefore likely contributing to increases in pathology and persistence within the respiratory tract of mice [83]. The regulation of these virulence factors is dependent on the environmental clues that the classical Bordetella encounter. Utilizing the above factors, the classical Bordetella are able to adhere and colonize hosts. Much is still unknown about how these factors contribute to the overall success of the Bordetella; in particular how these factors contribute to the successful transmission of the Bordetella. Factors contributing to adherence likely are important during initial colonization of new host. Likewise, virulence factors that stimulate the host immune response can lead to symptoms that allows shedding to occur. The use of multi-disciplinary

20 11 approaches will further our knowledge of how the Bordetella are so successful, leading to the development of improved prevention strategies and disease treatments. Host Immunity to the Bordetella Animal model systems have also been used to dissect components of the immune system that are required for the control and clearance of the classical Bordetella. Many innate immune functions are important for the survival of a host infected with Bordetella. Anatomical barriers, the complement system, and inflammation are all important components of the innate immune system. One particular anatomical barrier the Bordetella must overcome is the regulated expression of mucin within the respiratory tract. Mucin production can entrap bacteria, which are then expelled by the mucociliary escalator. Both B. bronchiseptica and B. pertussis can be encased in mucin; however, this barrier is dampened by the tracheal colonization toxin of B. pertussis which paralyzes ciliated cells [84-86]. The complement system helps with the clearance of pathogens by opsonization, attracting phagocytic cells, and directly causing cell lysis. B. pertussis and B. parapertussis are able to reduce killing via the classical complement pathway through the expression of brka and O antigen respectively; however, the classical pathway still plays an important role in controlling infection of the Bordetella [87-90]. Lastly, inflammation is caused by the influx of immune cells, and secretion of cytokines produced by cells at the site of infection in response to the invading bacteria being recognized by PAMPs. PAMPs play an important role during the innate response, directing the recruitment of immune components to the site of infection. TLR4 recognizes LPS of Gram-negative bacteria and is a particular important PAMP-receptor during Bordetella infection. Mice lacking TLR4 quickly succumb to B. bronchiseptica infection, while those infected with B. pertussis show increased bacterial load, cellular infiltrate, and pathology within the respiratory tract [91, 92].

21 12 Stimulation of TLR4, as well as other PAMPs, leads to the production of cytokines. Cytokines, such as TNF-α, are required during early B. pertussis infection to limit leukocyte infiltrate caused by PT [93]. Likewise, IL-1 signaling can be important in fighting microbial infection and is required during B. pertussis infection [94]. These and other cytokines can recruit innate immune cells to the site of infection. Of the innate immune cells, neutrophils and macrophages are particularly important during Bordetella infections. Depletion of either of these subsets enhanced infection by B. pertussis [95, 96], and depletion of neutrophils during B. bronchiseptica infection led to lethal infection[93]. Lastly, DC s help bridge the innate and adaptive immune systems. Both T cell and B cells of the adaptive immune system are important to the ultimate outcome of Bordetella infection. Stimulation of DC s by ACT and FHA help produce Th17 and Treg subsets, important during infection [97, 98]. Infection by the Bordetella also induces the differentiation of Th1 and Th17 cells [97, 99]. Th1 cells generated in response to infection produce IFN-γ that is required for immunity to all the classical Bordetella [ ]. B cells contribute to the control of the classical Bordetella by producing antigen specific antibodies that contribute to respiratory tract clearance [103]. B. bronchiseptica is particularly sensitive to antibody clearance, resulting from the effects of complement and Fcγ receptors [104, 105]. B. pertussis and B. parapertussis have each evolved ways to combat rapid clearance by antibodies; however, vaccine-induced antibodies are important in protection, as Ig-defective mice vaccinated with whole-cell or acellular vaccine failed to clear B. pertussis [79, 103, 106]. Bordetella Transmission models The Bordetella are highly successful at circulating within populations and, therefore, transmission models of the Bordetella are required to understand the underlying mechanisms that

22 13 contribute to Bordetella transmission [43, 107]. Of particular interest is the transmission of the human-restricted pathogen B. pertussis. Due to its lack of no known animal or environmental reservoir, B. pertussis likely requires continuous transmission from infected to naïve individuals [108]. While the mouse model has yielded much information, it is limited in the degree of sensitivity to accurately reflect natural human infection. B. pertussis and B. parapertussis are human-adapted pathogens requiring large inocula to reliably colonize mice. It has been suggested that B. pertussis s inability to colonize the mouse model effectively is due to an inability to compete with the mouse microbiota [109]. Due to the inconsistencies involved during mouse infection, a baboon model of B. pertussis infection has recently been described which mimics human infection, producing the classical cough for which the disease is named [110]. The baboon model system experimental showed that B. pertussis transmission occurs through aerosolized droplets [108]. B. bronchispetica transmission, on the other hand, adversely affects the agricultural industry, particular swine. Therefore, swine have recently been used to understand virulence determinates that effect transmission in a natural host [111]. These powerful model systems are ideal for the study of transmission of B. pertussis and B. bronchiseptica in natural hosts, however lack the powerful immunological tools of the mouse model. Current studies to date have used high-dose infection models to elucidate interactions between infecting pathogens and host immunity. However the high-dose system is limited in its ability to reproduce how natural infections begin. A low-dose model system allows for the examination of factors that are required during natural infection and transmission processes. B. bronchiseptica naturally infects mice. As few as 5 colony forming units are needed to establish infection. Due to B. bronchiseptica s high relatedness to the other classical Bordetella in both

23 14 genomic content and virulence strategies, a low-dose model system can infer information about how the classical Bordetella circulate. Recently, experiments in the low-dose mouse model determined that TLR4 is required to limit transmission between wild-type mice [112]. In particular, TLR4 was important in both infectiousness of the individual (shedding) and host susceptibility (colonization of susceptible mice). Interestingly, neutrophil recruitment directly correlated with shedding output. The lowdose model system also allows for the determination of factors that are required for the natural infection and transmission processes. For example, the addition of glucosamine to B. bronchiseptica lipid A was found to have no defect in the high-dose model system. However, the addition of glucosamine to B. bronchiseptica lipid A was shown to be important for the transmission success of B. bronchiseptica by protecting it from antimicrobial peptides during the initial stages of colonization [113]. This experimental model can be used to identify both bacterial and immunological mechanisms involved in the transmission of the Bordetella. Discoveries within this system will complement and guide the baboon and swine models of transmission. Preface This dissertation will focus on the interactions between host and pathogen in the context of transmission, utilizing the mouse experimental transmission system, which has many tools and is easy to manipulate. Bacterial factors were assayed for their ability to contribute to particular facets of the transmission process, such as the initial colonization of naïve hosts, the growth and persistence within the host, and shedding to the external environment. Finally, vaccines effects on factors contributing to transmission were also determined and dissected.

24 15 The second chapter will examine how the B. bronchiseptica T3SS contributes to the transmission process. In the absence of a functioning T3SS (RB50 bscn), infected mice fail to transmit to naïve susceptible cage mates. No defects in shedding were observed from mice, suggesting an important role during invasion of new hosts. During experimental inoculation, as few as 5 CFU in 5 µl of RB50 bscn were able to infect individual mice; however, observations by Weyrich et al. showed an important role for the T3SS system in the displacement of host microbiota during infection by B. bronchiseptica. Depletion of nasal cavity microbiota in susceptible mice by antibiotic treatment enabled RB50 bscn to transmit from the infected mice. The data found in this chapter reveal that to establish colonization in a new host, B. bronchiseptica must be able to compete with the host microbiota, as they are a major contributor of colonization resistance. The following chapter will examine how FHA contributes to B. bronchiseptica transmission. In the absence of FHA B. bronchiseptica fails to transmit from one host to the next. A defect in shedding was observed from mice inoculated with RB50 fhab, and likely the reason for failed transmission. Further analysis revealed that in the absence of FHA, changes in the immune response in the nasal cavity resulted. Recruitment of neutrophils and dendritic cells was decreased in mice that were inoculated with RB50 fhab. Neutrophil presence in the nasal cavity is important for the shedding process of Bordetella. The results here describe the importance FHA in manipulating the immune response toward one that is conducive to the transmission process of the Bordetella. In the next chapter we describe the effects of vaccination on Bordetella transmission. In response to heat-killed whole-cell vaccination, shedding of B. bronchiseptica from the external nares was reduced, and was not a result of reduced bacterial burden in the nasal cavity of mice.

25 16 To investigate the reduction in shedding, cell recruitment to the site of infection was examined. Neutrophil numbers were reduced, while B cell numbers were increased at the site of infection. Correlations of these cell subsets with shedding output were observed, revealing their importance during the host-pathogen dynamics of transmission. Further analysis revealed the importance of both antibodies and CD4 T cells in controlling shedding. Finally, a current acellular vaccine was used in this model system and was found to be unable to reduce shedding from mice. During transmission experiments, only whole-cell vaccination resulted in protection for susceptible individuals (herd immunity). These results show potential discrepancies between whole-cell and acellular vaccination that could contribute to the increased incidence of B. pertussis. Finally, the last chapter will summarize the findings and define the significance of this work in the context of Bordetella transmission. Overall, these initial findings provide a base that will contribute to expanding fields that study Bordetella transmission, as well as provide a system to analyze novel therapeutics that prevent transmission.

26 17 References 1. Fauci AS, Morens DM. The perpetual challenge of infectious diseases. The New England journal of medicine 2012; 366: Hatta M, Hatta Y, Kim JH, et al. Growth of H5N1 influenza A viruses in the upper respiratory tracts of mice. PLoS pathogens 2007; 3: Powell FL, Geiser J, Gratz RK, Scheid P. Airflow in the avian respiratory tract: variations of O2 and CO2 concentrations in the bronchi of the duck. Respiration physiology 1981; 44: Groisman EA, Mouslim C. Sensing by bacterial regulatory systems in host and non-host environments. Nature reviews Microbiology 2006; 4: Miller JF, Mekalanos JJ, Falkow S. Coordinate regulation and sensory transduction in the control of bacterial virulence. Science 1989; 243: WHO The world health report changing history (2004).. Available at: Accessed Jan Schwab JA, Zenkel M. Filtration of particulates in the human nose. The Laryngoscope 1998; 108: Alonso JM. [Immunity and pathophysiology of respiratory tract infections]. Medecine et maladies infectieuses 2008; 38: Bevins CL. Antimicrobial peptides as agents of mucosal immunity. Ciba Foundation symposium 1994; 186:250-60; discussion Buffie CG, Pamer EG. Microbiota-mediated colonization resistance against intestinal pathogens. Nature reviews Immunology 2013; 13: Hornef MW, Wick MJ, Rhen M, Normark S. Bacterial strategies for overcoming host innate and adaptive immune responses. Nature immunology 2002; 3: Akira S, Takeda K. Toll-like receptor signalling. Nature reviews Immunology 2004; 4: Tang D, Kang R, Coyne CB, Zeh HJ, Lotze MT. PAMPs and DAMPs: signal 0s that spur autophagy and immunity. Immunological reviews 2012; 249: Medzhitov R. Recognition of microorganisms and activation of the immune response. Nature 2007; 449: Leavy O. Neutrophil recruitment on the big screen. Nature reviews Immunology 2010; 10:807.

27 Wykes M, MacPherson G. Dendritic cell-b-cell interaction: dendritic cells provide B cells with CD40-independent proliferation signals and CD40-dependent survival signals. Immunology 2000; 100: Banchereau J, Briere F, Caux C, et al. Immunobiology of dendritic cells. Annual review of immunology 2000; 18: Peschel A. How do bacteria resist human antimicrobial peptides? Trends in microbiology 2002; 10: Hood RD, Singh P, Hsu F, et al. A type VI secretion system of Pseudomonas aeruginosa targets a toxin to bacteria. Cell host & microbe 2010; 7: Cotter PD, Hill C, Ross RP. Bacteriocins: developing innate immunity for food. Nature reviews Microbiology 2005; 3: Winter SE, Thiennimitr P, Winter MG, et al. Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 2010; 467: Brown SP, Fredrik Inglis, R. and Taddei, F.. SYNTHESIS: Evolutionary ecology of microbial wars: within-host competition and (incidental) virulence. Evolutionary Applications 2009; 2: Mota LJ, Cornelis GR. The bacterial injection kit: type III secretion systems. Annals of medicine 2005; 37: Rabinovich L, Sigal N, Borovok I, Nir-Paz R, Herskovits AA. Prophage excision activates Listeria competence genes that promote phagosomal escape and virulence. Cell 2012; 150: Mitchell EK, Mastroeni P, Kelly AP, Trowsdale J. Inhibition of cell surface MHC class II expression by Salmonella. European journal of immunology 2004; 34: Xavier MN, Winter MG, Spees AM, et al. CD4+ T cell-derived IL-10 promotes Brucella abortus persistence via modulation of macrophage function. PLoS pathogens 2013; 9:e Armstrong GL, Conn LA, Pinner RW. Trends in infectious disease mortality in the United States during the 20th century. JAMA : the journal of the American Medical Association 1999; 281: Feldman HA. The beginning of antimicrobial therapy: introduction of the sulfonamides and penicillins. The Journal of infectious diseases 1972; 125:Suppl: Morre SA, Murillo LS, Spaargaren J, Fennema HS, Pena AS. Role of the toll-like receptor 4 Asp299Gly polymorphism in susceptibility to Candida albicans infection. The Journal of infectious diseases 2002; 186:1377-9; author reply 9.

28 Funk S, Salathe M, Jansen VA. Modelling the influence of human behaviour on the spread of infectious diseases: a review. Journal of the Royal Society, Interface / the Royal Society 2010; 7: Woolhouse ME, Dye C, Etard JF, et al. Heterogeneities in the transmission of infectious agents: implications for the design of control programs. Proceedings of the National Academy of Sciences of the United States of America 1997; 94: Beck MA, Levander OA. Host nutritional status and its effect on a viral pathogen. The Journal of infectious diseases 2000; 182 Suppl 1:S Gopinath S, Hotson A, Johns J, Nolan G, Monack D. The systemic immune state of supershedder mice is characterized by a unique neutrophil-dependent blunting of TH1 responses. PLoS pathogens 2013; 9:e Riedel S. Edward Jenner and the history of smallpox and vaccination. Proceedings 2005; 18: John TJ, Samuel R. Herd immunity and herd effect: new insights and definitions. European journal of epidemiology 2000; 16: Organization WH. Annes Table 2: Deaths by cause, sex and mortality stratum in WHO regions, estimates for Available at: Accessed 2/18/ Logan Brenzel LJW, Julia FOx-Rushby, Mark Miller, and Neal A. Halsey. Chapter 20: Vaccine-Preventable Disease. In: Dean T. Jamison JGB, Anthony R. Measham, George Alleyne, Mariam Claeson, David B. Evans, Prabhat Jha, Anne Mills, Philip Musgrove, ed. Disease Control Priorities in Developing Countries. World Bank; Oxford University Press USA, Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Park J, Zhang Y, Buboltz AM, et al. Comparative genomics of the classical Bordetella subspecies: the evolution and exchange of virulence-associated diversity amongst closely related pathogens. BMC genomics 2012; 13: Woolfrey BF, Moody JA. Human infections associated with Bordetella bronchiseptica. Clinical microbiology reviews 1991; 4: Nicholson TL, Brockmeier SL, Loving CL, et al. Phenotypic modulation of the virulent Bvg phase is not required for pathogenesis and transmission of Bordetella bronchiseptica in swine. Infect Immun 2012; 80: Zhao Z, Wang C, Xue Y, et al. The occurrence of Bordetella bronchiseptica in pigs with clinical respiratory disease. Veterinary journal 2011; 188:

29 Crowcroft NS, Stein C, Duclos P, Birmingham M. How best to estimate the global burden of pertussis? Lancet Infect Dis 2003; 3: Cherry JD. The epidemiology of pertussis and pertussis immunization in the United Kingdom and the United States: a comparative study. Current problems in pediatrics 1984; 14: Pichichero ME, Green JL, Francis AB, Marsocci SM, Lynd AM, Litteer T. Comparison of a three-component acellular pertussis vaccine with whole cell pertussis vaccine in two-month-old children. The Pediatric infectious disease journal 1994; 13: Reporting C-PSa. April 18, Mooi FR. Bordetella pertussis and vaccination: the persistence of a genetically monomorphic pathogen. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases 2010; 10: Queenan AM, Cassiday PK, Evangelista A. Pertactin-negative variants of Bordetella pertussis in the United States. The New England journal of medicine 2013; 368: Wendelboe AM, Van Rie A, Salmaso S, Englund JA. Duration of immunity against pertussis after natural infection or vaccination. The Pediatric infectious disease journal 2005; 24:S Long SS, Welkon CJ, Clark JL. Widespread silent transmission of pertussis in families: antibody correlates of infection and symptomatology. The Journal of infectious diseases 1990; 161: Higgs R, Higgins SC, Ross PJ, Mills KH. Immunity to the respiratory pathogen Bordetella pertussis. Mucosal immunology 2012; 5: Ryan M, Murphy G, Ryan E, et al. Distinct T-cell subtypes induced with whole cell and acellular pertussis vaccines in children. Immunology 1998; 93: Ross PJ, Sutton CE, Higgins S, et al. Relative contribution of Th1 and Th17 cells in adaptive immunity to Bordetella pertussis: towards the rational design of an improved acellular pertussis vaccine. PLoS pathogens 2013; 9:e Barbic J, Leef MF, Burns DL, Shahin RD. Role of gamma interferon in natural clearance of Bordetella pertussis infection. Infect Immun 1997; 65: Simondon F, Preziosi MP, Yam A, et al. A randomized double-blind trial comparing a twocomponent acellular to a whole-cell pertussis vaccine in Senegal. Vaccine 1997; 15: Cotter PA, Jones AM. Phosphorelay control of virulence gene expression in Bordetella. Trends in microbiology 2003; 11: Melton AR, Weiss AA. Characterization of environmental regulators of Bordetella pertussis. Infect Immun 1993; 61:

30 Lacey BW. Antigenic modulation of Bordetella pertussis. The Journal of hygiene 1960; 58: Akerley BJ, Monack DM, Falkow S, Miller JF. The bvgas locus negatively controls motility and synthesis of flagella in Bordetella bronchiseptica. Journal of bacteriology 1992; 174: Merkel TJ, Barros C, Stibitz S. Characterization of the bvgr locus of Bordetella pertussis. Journal of bacteriology 1998; 180: Vergara-Irigaray N, Chavarri-Martinez A, Rodriguez-Cuesta J, Miller JF, Cotter PA, Martinez de Tejada G. Evaluation of the role of the Bvg intermediate phase in Bordetella pertussis during experimental respiratory infection. Infect Immun 2005; 73: Stockbauer KE, Fuchslocher B, Miller JF, Cotter PA. Identification and characterization of BipA, a Bordetella Bvg-intermediate phase protein. Molecular microbiology 2001; 39: Relman D, Tuomanen E, Falkow S, Golenbock DT, Saukkonen K, Wright SD. Recognition of a bacterial adhesion by an integrin: macrophage CR3 (alpha M beta 2, CD11b/CD18) binds filamentous hemagglutinin of Bordetella pertussis. Cell 1990; 61: Hazenbos WL, van den Berg BM, Geuijen CW, Mooi FR, van Furth R. Binding of FimD on Bordetella pertussis to very late antigen-5 on monocytes activates complement receptor type 3 via protein tyrosine kinases. Journal of immunology 1995; 155: McGuirk P, Mills KH. Direct anti-inflammatory effect of a bacterial virulence factor: IL-10- dependent suppression of IL-12 production by filamentous hemagglutinin from Bordetella pertussis. European journal of immunology 2000; 30: Hsieh CS, Macatonia SE, Tripp CS, Wolf SF, O'Garra A, Murphy KM. Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages. Science 1993; 260: Abramson T, Kedem H, Relman DA. Proinflammatory and proapoptotic activities associated with Bordetella pertussis filamentous hemagglutinin. Infect Immun 2001; 69: Emsley P, McDermott G, Charles IG, Fairweather NF, Isaacs NW. Crystallographic characterization of pertactin, a membrane-associated protein from Bordetella pertussis. Journal of molecular biology 1994; 235: Parkhill J, Sebaihia M, Preston A, et al. Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nature genetics 2003; 35: Weingart CL, Weiss AA. Bordetella pertussis virulence factors affect phagocytosis by human neutrophils. Infect Immun 2000; 68:

31 Hewlett EL, Donato GM, Gray MC. Macrophage cytotoxicity produced by adenylate cyclase toxin from Bordetella pertussis: more than just making cyclic AMP Molecular microbiology 2006; 59: Pearson RD, Symes P, Conboy M, Weiss AA, Hewlett EL. Inhibition of monocyte oxidative responses by Bordetella pertussis adenylate cyclase toxin. Journal of immunology 1987; 139: Boyd AP, Ross PJ, Conroy H, Mahon N, Lavelle EC, Mills KH. Bordetella pertussis adenylate cyclase toxin modulates innate and adaptive immune responses: distinct roles for acylation and enzymatic activity in immunomodulation and cell death. Journal of immunology 2005; 175: Harvill ET, Cotter PA, Yuk MH, Miller JF. Probing the function of Bordetella bronchiseptica adenylate cyclase toxin by manipulating host immunity. Infect Immun 1999; 67: Weiss AA, Goodwin MS. Lethal infection by Bordetella pertussis mutants in the infant mouse model. Infect Immun 1989; 57: Magyar T, Chanter N, Lax AJ, Rutter JM, Hall GA. The pathogenesis of turbinate atrophy in pigs caused by Bordetella bronchiseptica. Veterinary microbiology 1988; 18: Arico B, Rappuoli R. Bordetella parapertussis and Bordetella bronchiseptica contain transcriptionally silent pertussis toxin genes. Journal of bacteriology 1987; 169: Andreasen C, Carbonetti NH. Pertussis toxin inhibits early chemokine production to delay neutrophil recruitment in response to Bordetella pertussis respiratory tract infection in mice. Infect Immun 2008; 76: Kirimanjeswara GS, Agosto LM, Kennett MJ, Bjornstad ON, Harvill ET. Pertussis toxin inhibits neutrophil recruitment to delay antibody-mediated clearance of Bordetella pertussis. The Journal of clinical investigation 2005; 115: Yuk MH, Harvill ET, Cotter PA, Miller JF. Modulation of host immune responses, induction of apoptosis and inhibition of NF-kappaB activation by the Bordetella type III secretion system. Molecular microbiology 2000; 35: Fennelly NK, Sisti F, Higgins SC, et al. Bordetella pertussis expresses a functional type III secretion system that subverts protective innate and adaptive immune responses. Infect Immun 2008; 76: Russell AB, Hood RD, Bui NK, LeRoux M, Vollmer W, Mougous JD. Type VI secretion delivers bacteriolytic effectors to target cells. Nature 2011; 475: Weyrich LS, Rolin OY, Muse SJ, et al. A Type VI secretion system encoding locus is required for Bordetella bronchiseptica immunomodulation and persistence in vivo. PLoS One 2012; 7:e45892.

32 Belcher CE, Drenkow J, Kehoe B, et al. The transcriptional responses of respiratory epithelial cells to Bordetella pertussis reveal host defensive and pathogen counter-defensive strategies. Proceedings of the National Academy of Sciences of the United States of America 2000; 97: Flak TA, Heiss LN, Engle JT, Goldman WE. Synergistic epithelial responses to endotoxin and a naturally occurring muramyl peptide. Infect Immun 2000; 68: Ishikawa H, Isayama Y. Evidence for sialyl glycoconjugates as receptors for Bordetella bronchiseptica on swine nasal mucosa. Infect Immun 1987; 55: Goebel EM, Wolfe DN, Elder K, Stibitz S, Harvill ET. O antigen protects Bordetella parapertussis from complement. Infect Immun 2008; 76: Barnes MG, Weiss AA. BrkA protein of Bordetella pertussis inhibits the classical pathway of complement after C1 deposition. Infect Immun 2001; 69: Weiss AA, Mobberley PS, Fernandez RC, Mink CM. Characterization of human bactericidal antibodies to Bordetella pertussis. Infect Immun 1999; 67: Barnes MG, Weiss AA. Activation of the complement cascade by Bordetella pertussis. FEMS microbiology letters 2003; 220: Mann PB, Elder KD, Kennett MJ, Harvill ET. Toll-like receptor 4-dependent early elicited tumor necrosis factor alpha expression is critical for innate host defense against Bordetella bronchiseptica. Infect Immun 2004; 72: Mann PB, Wolfe D, Latz E, Golenbock D, Preston A, Harvill ET. Comparative toll-like receptor 4-mediated innate host defense to Bordetella infection. Infect Immun 2005; 73: Wolfe DN, Mann PB, Buboltz AM, Harvill ET. Delayed role of tumor necrosis factor- alpha in overcoming the effects of pertussis toxin. The Journal of infectious diseases 2007; 196: Zhang X, Hester SE, Kennett MJ, et al. Interleukin-1 receptor signaling is required to overcome the effects of pertussis toxin and for efficient infection- or vaccination-induced immunity against Bordetella pertussis. Infect Immun 2011; 79: Carbonetti NH, Artamonova GV, Van Rooijen N, Ayala VI. Pertussis toxin targets airway macrophages to promote Bordetella pertussis infection of the respiratory tract. Infect Immun 2007; 75: Andreasen C, Carbonetti NH. Role of neutrophils in response to Bordetella pertussis infection in mice. Infect Immun 2009; 77: Dunne A, Ross PJ, Pospisilova E, et al. Inflammasome activation by adenylate cyclase toxin directs Th17 responses and protection against Bordetella pertussis. Journal of immunology 2010; 185:

33 Higgins SC, Lavelle EC, McCann C, et al. Toll-like receptor 4-mediated innate IL-10 activates antigen-specific regulatory T cells and confers resistance to Bordetella pertussis by inhibiting inflammatory pathology. Journal of immunology 2003; 171: Mills KH, Barnard A, Watkins J, Redhead K. Cell-mediated immunity to Bordetella pertussis: role of Th1 cells in bacterial clearance in a murine respiratory infection model. Infect Immun 1993; 61: Pilione MR, Harvill ET. The Bordetella bronchiseptica type III secretion system inhibits gamma interferon production that is required for efficient antibody-mediated bacterial clearance. Infect Immun 2006; 74: Wolfe DN, Karanikas AT, Hester SE, Kennett MJ, Harvill ET. IL-10 induction by Bordetella parapertussis limits a protective IFN-gamma response. Journal of immunology 2010; 184: Mahon BP, Sheahan BJ, Griffin F, Murphy G, Mills KH. Atypical disease after Bordetella pertussis respiratory infection of mice with targeted disruptions of interferon-gamma receptor or immunoglobulin mu chain genes. The Journal of experimental medicine 1997; 186: Kirimanjeswara GS, Mann PB, Harvill ET. Role of antibodies in immunity to Bordetella infections. Infect Immun 2003; 71: Kirimanjeswara GS, Mann PB, Pilione M, Kennett MJ, Harvill ET. The complex mechanism of antibody-mediated clearance of Bordetella from the lungs requires TLR4. Journal of immunology 2005; 175: Pishko EJ, Kirimanjeswara GS, Pilione MR, Gopinathan L, Kennett MJ, Harvill ET. Antibody-mediated bacterial clearance from the lower respiratory tract of mice requires complement component C3. European journal of immunology 2004; 34: Mills KH, Ryan M, Ryan E, Mahon BP. A murine model in which protection correlates with pertussis vaccine efficacy in children reveals complementary roles for humoral and cellmediated immunity in protection against Bordetella pertussis. Infect Immun 1998; 66: Bemis DA. Bordetella and Mycoplasma respiratory infections in dogs and cats. The Veterinary clinics of North America Small animal practice 1992; 22: Warfel JM, Beren J, Merkel TJ. Airborne transmission of Bordetella pertussis. The Journal of infectious diseases 2012; 206: Weyrich LS, Feaga HA, Park J, et al. Resident Microbiota Affect Bordetella pertussis Infectious Dose and Host Specificity. The Journal of infectious diseases Warfel JM, Beren J, Kelly VK, Lee G, Merkel TJ. Nonhuman primate model of pertussis. Infect Immun 2012; 80:

34 Nicholson T, Brockmeier S, Loving C, et al. Phenotypic modulation of the virulent Bvg phase is not required for pathogenesis and transmission of Bordetella bronchiseptica in swine. Infection and immunity 2012; 80: Rolin O, Smallridge W, Henry M, Goodfield L, Place D, Harvill ET. Toll-Like Receptor 4 Limits Transmission of Bordetella bronchiseptica. PLoS One 2014; 9:e Rolin O, Muse SJ, Safi C, et al. Enzymatic Modification of Lipid A by ArnT Protects Bordetella bronchiseptica against Cationic Peptides and Is Required for Transmission. Infect Immun 2014; 82:491-9.

35 26 Chapter 2 : The Bordetella bronchiseptica type three secretion system affects transmission via interactions with host microbiota

36 27 Abstract Pathogens are dependent on continuous transmission cycles between hosts for their propagation, and therefore the cost of causing disease during infection must be balanced by virulent characteristics that provide a selective advantage during transmission events. The acquisition and conservation of virulence mechanisms among pathogens suggest that they contribute to the transmission process; however, there is limited experimental evidence to support this conclusion. Animal models of disease often require that pathogen fitness and individual virulence factors be studied in the context of experimental inoculation with doses that overwhelm mucosal barriers and innate immune defenses, and thus fail to consider interactions that contribute to the natural transmission process. The Type III Secretion System (T3SS) has been central to the evolution of virulence in many bacterial pathogens, although its role in transmission of pathogens remains unknown. Using a recently characterized model for transmission of Bordetella bronchiseptica, the ability of the T3SS to contribute to the transmission process in mice was assessed. While a mutant strain of B. bronchiseptica that lacks a functional T3SS was comparable to the wild-type parental strain in growth, persistence and shedding, the mutant failed to transmit from an index case to susceptible individuals in the same cage. Disruption of host microbiota in susceptible mice with antibiotics enabled the mutant to transmit to susceptible hosts. This suggests that host microbiota present an important source of colonization resistance in the respiratory tract and that the T3SS is required to overcome this source upon invasion of new hosts.

37 28 Introduction Continued transmission of classical Bordetella species, B. pertussis, B. parapertussis, and B. bronchiseptica, within vaccinated populations is a significant contributor to the morbidity and mortality related to Bordetella infections in both humans and animals [1, 2]. The number of transmission events originating from an infected individual is dependent on the intensity and duration of bacterial shedding from the infected individual, as well as the susceptibility of new hosts [3]. Identifying factors involved in transmission of the Bordetella is essential for improving current control measures designed to hinder the spread of the infectious agents, not just their clinical manifestations. The recently-developed transmission model for B. bronchiseptica in TLR4-deficient mice has been used to determine host mechanisms that both facilitate and limit transmission of the Bordetella [4-6]. Because the classical Bordetella species express a large number of conserved virulence factors, including distinct secretion systems, we can identify bacterial factors that are involved in the transmission process [7]. In particular, the T3SS is used by many pathogens to translocate proteins across both bacterial membranes and, in some cases, into host cells. The locus encoding the T3SS is highly conserved among the classical Bordetella species, indicating that a strong selective pressure has favored stable maintenance of this locus throughout a long evolutionary history. The T3SS has been associated with a diverse range of virulence phenotypes within the Bordetella species [8-10]. The T3SS of B. pertussis dampens pro-inflammatory cytokine production and downstream Th1 and Th17 responses to delay pathogen clearance from the lungs [11]. In B. bronchiseptica the T3SS is required to kill epithelial cells and macrophages in vitro [12]. During B. bronchiseptica infections, the T3SS causes apoptosis of neutrophils that are

38 29 recruited to the lungs, inhibits dendritic cell maturation, and induces the anti-inflammatory cytokine IL-10 [12, 13]. The T3SS system of B. bronchiseptica also results in delayed generation of protective Th1 responses and prolongs the course of pulmonary infection [14, 15]. In addition to affecting the outcome of adaptive immunity, the T3SS also contributes to virulence during the initial stages of infection, as multiple B. bronchiseptica lineages with increased expression of the T3SS locus have hyper-virulent phenotypes [10]. Although the virulence phenotypes associated with the T3SS are highly variable among the different lineages of classical Bordetella, the conservation of the locus encoding the T3SS is higher than that of any other secretion system [9]. Together these data suggest that some important functions of the T3SS are not evident in the common in vivo experiments using high-dose inoculation of animals. Using our recently developed model for Bordetella transmission a T3SS mutant, RB50 bscn, was found to be defective. The inability of RB50 bscn to transmit between TLR4- deficient mice was not associated with a reduction in shedding, indicating that the T3SS is required for colonization of new hosts. As an important factor contributing to the colonization resistant of hosts, naïve microbiota levels were compared to infected hosts. Hosts infected with RB50 bscn failed to compete with the microbiota in vivo. Finally, the ability of RB50 bscn to invade new hosts was restored by treatment with antibiotics prior to exposure of susceptible mice to index mice, suggesting that the T3SS contributes to transmission by mediating competition with host microbiota.

39 30 Materials and Methods Ethics Statement. This study was carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee at The Pennsylvania State University (#40029). All animals were anesthetized using 5% isoflourane in oxygen and euthanized using carbon dioxide inhalation to minimize animal suffering. Bacterial Strains. B. bronchiseptica strain RB50 has been previously described [16]. WD3, a T3SS mutant strain derived from RB50, has also been described elsewhere and is referred to as RB50 bscn [17]. Bacteria were maintained on Bordet-Gengou agar (Difco) supplemented with 10% defibrinated sheep blood (Hema Resourses) and 20 µg/ml streptomycin (Sigma-Aldrich). Liquid cultures were grown overnight in Stainer-Scholte broth at 37 C to mid-log phase. Animal Studies Four- to six-week C3/HeJ mice were obtained from The Jackson Laboratories and maintained in our specific pathogen-free facility. Inoculations were prepared from liquid cultures grown to mid-log phase and bacteria were diluted to a concentration of 3x10 4 CFU/mL or 1x10 3 CFU/ml. Mice were given 150 CFU in 5 µl of PBS that was deposited on the external nares for transmission and shedding experiments. For comparison of infectious dose 5 CFU was deposited on the external nares in 5µl PBS. To quantify bacteria, organs were excised on the day indicated. Following dissection, organs of interest were homogenized in 1mL PBS and diluted appropriately for culture and quantification on Bordet-Gengou agar or Blood agar. Transmission and Shedding Studies To determine transmission, a single infected mouse (index) was inoculated and placed in a cage containing three susceptible individuals. Transmission to susceptible individuals was determined by dissection and culturing of the nasal cavity tissue on day 21 (limit of detection 5 CFU).

40 31 Shedding of bacteria was determined by swabbing the external nares with a Dacron-polyester tipped swab (VWR International) for 15 seconds. Swabs were placed into 1ml of PBS and vortexed. Statistical Analysis. Shedding and colonization data was fit to a generalized linear model (GLM) to determine the effect of group and day on outputs, such as within-host colonization and shedding using Minitab v16. The effect of group on within-host colonization and shedding at individual time points was determined by ANOVA using a Tukey simultaneous test for significance.

41 32 Results The type III secretion system is required for efficient transmission of B. bronchiseptica. To determine whether the T3SS is involved in transmission of B. bronchiseptica, an individual mouse (index) was inoculated with 150 CFU of either RB50 or RB50 bscn /15 3/15 and placed in cages with three naïve susceptible mice. Susceptible mice were sacrificed on day 21 to determine transmission via culture of nasal cavity tissue. RB50-inoculated index mice readily transmitted to 14 out of 15 susceptible cage mates, while only 3 out of 15 susceptible cage mates became infected when housed with RB50 bscn-inoculated index mice (Figure 2.1). These results suggest that the T3SS is Log10+CFU RB RB50 bscn+++ Figure 2.1. The T3SS is required for efficient transmission of B. bronchiseptica Index mice were inoculated with 150 CFU of B. bronchiseptica strain RB50 or RB50 bscn, and placed with three naive susceptible mice. 21 days after being housed together, susceptible mice were dissected and the quantity of bacteria from the nasal cavity of individuals was enumerated. Dots represent the log 10 CFU of RB50 or RB50 bscn recovered from individual susceptible mice. The proportion of infected mice in each group is listed above. important for facilitating transmission between mice. Efficient shedding, growth, and persistence does not require the T3SS Two important factors contribute to the successful transmission of a pathogen: shedding from the infected host and colonization of a susceptible individual. To test these parameters, groups of mice were inoculated with RB50 or RB50 bscn. Bacterial shedding was monitored from the nares of individual mice for 28 days. Throughout the time-course, the shedding

42 33 Log10+CFU+per+Second+ " 3.5 RB50 RB50 RB50 bscn RB50 bscn # Days+Post+Challenege Log10+CFU RB50 RB50 bscn Figure 2.2. T3SS is not required to enable shedding, growth, or persistence in mice. (A) Shedding was detected by quantitative culture of bacteria obtained from a 15 second swab of the external nares. Symbols represent the mean log10 CFU ± the standard error shed from the nares of mice inoculated with RB50 or RB50 bscn. (B) ID50 of RB50 and RB50 bscn were compared. Dots represent log10 CFU (C and D). Four mice per group were inoculated with 150 CFU of RB50 or RB50 bscn in 5 µl of PBS. Bars represent the log mean CFU ± standard error of bacteria recovered from the nasal cavity, trachea, and lungs of mice at the indicated time-points. intensity of mice inoculated with RB50 bscn was not different from those inoculated with RB50, indicating that the lack of transmission was not due to a discrepancy in the ability to shed (Figure 2.2A). Additionally, respiratory tract colonization was measured at days 8 and 28 after inoculation, and no defects in growth and persistence of RB50 or RB50 bscn were observed in the nasal cavity (Figure 2.2C and D). Furthermore, experimental inoculation with as little as 5 CFU, in 5ul of PBS, of either RB50 or RB50 bscn was sufficient to infect 50 percent of

43 B 34 individuals (ID 50 ) (Figure 2.2B). Together these results suggest that the T3SS is not required for shedding, growth, persistence, and experimental colonization of B. bronchiseptica. The T3SS is required for competition with nasal cavity microbiota Some facets of infection are not easily reproduced during experimental inoculation. Based on previous observations that normal culturable microbiota are no longer recovered following inoculation with 100 CFU B. bronchiseptica in wild-type mice (Weyrich et al), we hypothesized that inter-bacterial Log10+CFU * Naïve RB50 RB50 bscn competition during natural transmission is an important aspect of the successful colonization of a new host. To determine whether the T3SS is required for B. bronchiseptica to compete Figure 2.3. T3SS is required for competition between microbiota in the nasal cavity of mice Microbiota was detected by quantitative culture from the nasal cavity of RB50 or RB50 bscn mice and compared to naïve mice. Bars represent the mean log10 CFU of microbiota ± standard error. Asterisks represent p 0.1. with host microbiota, mice were inoculated with RB50 or RB50 bscn. To minimize the variation in host microbiota, mice in this experiment were derived from the same breeding cage and maintained together in a gang cage for five weeks before inoculation. Three days after inoculation with RB50 or RB50 bscn, culturable host microbiota from infected mice were cultured on Blood Agar and compared to naïve mice (Figure 2.3). A difference in the microbiota levels was observed between naïve mice and the group inoculated with RB50 bscn (p < 0.1). Mice inoculated with RB50 bscn experienced a out-growth of culturable host microbiota, compared to mice infected with RB50. This result indicates that the T3SS plays an important role

44 35 in competing with the host microbiota within the nasal cavity of mice and is likely critical during natural transmission events. Log10CFU /6 6/6 T3SS is not required for transmission to antibiotic treated susceptible mice. If the T3SS is required to invade new hosts by contributing to bacterial competition, then disruption of the microbiota of susceptible mice with antibiotics should facilitate transmission of RB50 bscn to these 0 PBSEnroDloxacin Figure 2.4. Depleting microbiota from susceptible mice with antibiotics facilitates transmission of RB50 bscn. Index mice were inoculated with 150 CFU of RB50 bscn and maintained in isolation for 6 days. 6 days after inoculation of index mice, susceptible mice received either a 10 µl intranasal of PBS or 2 mg/ml of enrofloxacin. 12 hours prior to exposure of index mice for 14 days. 14 days after being housed together, susceptible mice were dissected and the quantity of RB50 bscn recovered from the nasal cavity of individuals was examined. Dots represent the log 10 CFU of RB50 bscn recovered from individual PBS treated or antibiotic treated susceptible mice. The proportion of infected mice in each group is listed above. animals. To determine whether RB50 bscn can transmit to antibiotic-treated mice, susceptible mice were treated with enrofloxacin or PBS and placed in cages with an index mouse that had been inoculated with RB50 bscn seven days earlier. To ensure that host microbiota remained depleted in enrofloxacin treated susceptible mice, streptomycin was added to the drinking water for one week following co-housing with index case. Only two of six susceptible mice treated with PBS became infected with RB50 bscn; however, all six susceptible mice treated with antibiotics were colonized with high levels of RB50 bscn (Figure 2.4). Together, these results suggest that the T3SS-mediated ability to

45 36 compete with the host microbiota contributes to initial host colonization contributing to the transmission success of B. bronchiseptica.

46 37 Discussion Historically, the majority of experimental research has been focused on pathogen interactions within individual, experimentally-inoculated, hosts. However, transmission is the essential and defining aspect of infectious disease, and therefore development of transmission models in the research setting is essential to explore the roles of host immunity and bacterial factors contributing to transmission. To examine the role of the T3SS in B. bronchiseptica transmission between mice, we used a strain containing an in-frame deletion of the bscn gene. The non-functional T3SS mutant resulted in reduced transmission to susceptible mice, as compared to wild-type bacteria. While there were no defects in shedding or growth during experimental inoculation with RB50 bscn, pretreatment of susceptible mice with antibiotics resulted in transmission to 100% of susceptible mice. These findings suggest that the T3SS is required to overcome colonization resistance engendered by the host microbiota for successful colonization of new hosts to occur. Pathogens encounter both host eukaryotic cells and other bacterial species that antagonize their presence throughout the colonization process. The host microbiota provide significant resistance to pathogenic colonization [18, 19]. Successful pathogens have evolved mechanisms that target these adversaries in order to compete and survive. To efficiently cause the relevant disease, experimental infection models often overwhelm the mucosal barrier defenses with large numbers of bacteria, precluding the possibility of observing the natural interactions between microbiota and an invading pathogen. Consequently, few experimental models have successfully observed, documented and manipulated microbiota-mediated colonization resistance in the respiratory tract. Previous studies of the Bordetella T3SS, as well as studies with other bacterial species, have demonstrated numerous interactions with eukaryotic cells, but not directly

47 38 with other bacteria [20]. Our results suggest that the B. bronchiseptica T3SS is also required for competition with the host microbiota to efficiently colonize new hosts. It is unclear, however, whether the T3SS-dependent competition with host microbiota by B. bronchiseptica in vivo occurs by direct interaction with competitor species or whether it is mediated through the host immune system. Recently, it has been suggested that B. bronchiseptica displacement of resident microbiota involves components of host immunity, particularly T cells, (Weyrich et al; unpublished). The displacement of host microbiota helps open physiological space within the respiratory tract, as well as creates increased nutrient availability. Studies with Salmonella typhimurium have determined that stimulation of the inflammation response by Salmonella leads to competition with other host microbiota by creating a new nutrient source that Salmonella can preferentially utilize [21, 22]. B. bronchiseptica is resistant to many bactericidal peptides [5, 23], and its stimulation of inflammation could contribute to the displacement of more sensitive organisms. The T3SS system is known to affect inflammation, and to mediate various interactions with several cell types recruited during an inflammatory response [12]. Few insights into T3SS s role in transmission have been observed. In Citrobacter rodentium transmission studies, a T3SS ATPase mutant resulted in no transmission [24]. In particular, iterative competition experiments with wild-type and T3SS mutant strains of C. rodentium showed a selective advantage for strains containing an intact T3SS during transmission to new hosts [24]. Here we show a role for the T3SS of B. bronchiseptica in the successful transmission between mice. B. bronchiseptica pathogenesis has also been recently studied in the swine model [25-27]. Transmission studies of the T3SS in swine contrast the findings here; however, factor s such as interactions with the host and host microbiota are likely

48 39 to contribute to efficient transmission. Differing sets of resident microbiota may potentially explain the discrepancy. The infectious process likely also differs between mouse and swine models of B. bronchiseptica infection due to basic differences in the structure and organization of the respiratory tract. Mice lack the pharyngeal and palatine tonsils that are important sites during the antibody responses in the respiratory tract of swine [28, 29]. These findings also highlight the broad range of phenotypes associated with different lineages of B. bronchispetica. The Bordetella T3SS contributes to virulence, prolongs the course of infection, and is required for efficient transmission between mice. Despite various phenotypes related to the T3SS in B. bronchiseptica, the structural components of the T3SS are highly conserved between B. pertussis and B. bronchiseptica [8], and because it is involved in pathogenesis at multiple stages, including initial invasion, virulence, persistence, and shedding, the T3SS may be an ideal target for vaccines or therapeutics that could more effectively prevent Bordetella transmission. Immunization against the needle tip complex protein bsp22 reduces the severity and duration of B. bronchiseptica infection following high-dose challenges [30]. Anti-Bsp22 antibodies also protect epithelial cells from the cytotoxic effects of the T3SS [30]. Although prior studies have used traditional approaches to examine the effects of immunity to T3SS components on virulence, the data presented here suggest that this approach could have the important additional effect of protecting against Bordetella transmission. Studies in both mouse and primate studies show that current acellular vaccines do not limit the spread of disease. Therefore, the addition of novel antigens may prove essential for the control of Bordetella transmission.

49 40 Authors and Contributions: William E. Smallridge 1,3, Olivier Y. Rolin 1,3, Laura Weyrich 1,2, Nathan T. Jacobs 1, Eric T. Harvill 1 Authors contributed equally to this work 1 Department of Veterinary and Biomedical Sciences, The Pennsylvania State University, University Park, PA, Department of Biochemistry, Microbiology and Molecular Biology, The Pennsylvania State University. 3 Graduate Program in Immunology and Infectious Disease, The Pennsylvania State University, University Park, PA, Conceived and designed experiments: WES, OYR, ETH Performed experiments: WES, OYR, LSW, NTJ Analyzed Data: WES, OYR, LSW, NTJ, ETH Wrote Paper: WES, OYR, LSW, ETH

50 41 References 1. Crowcroft NS, Stein C, Duclos P, Birmingham M. How best to estimate the global burden of pertussis? Lancet Infect Dis 2003; 3: Long SS, Welkon CJ, Clark JL. Widespread silent transmission of pertussis in families: antibody correlates of infection and symptomatology. The Journal of infectious diseases 1990; 161: Woolhouse ME, Dye C, Etard JF, et al. Heterogeneities in the transmission of infectious agents: implications for the design of control programs. Proceedings of the National Academy of Sciences of the United States of America 1997; 94: Rolin O, Smallridge W, Henry M, Goodfield L, Place D, Harvill ET. Toll-Like Receptor 4 Limits Transmission of Bordetella bronchiseptica. PLoS One 2014; 9:e Rolin O, Muse SJ, Safi C, et al. Enzymatic Modification of Lipid A by ArnT Protects Bordetella bronchiseptica against Cationic Peptides and Is Required for Transmission. Infect Immun 2014; 82: Smallridge WE, Rolin OY, Jacobs NT, Harvill ET. Different effects of whole cell and acellular vaccines on Bordetella transmission. The Journal of infectious diseases Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Parkhill J, Sebaihia M, Preston A, et al. Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nature genetics 2003; 35: Park J, Zhang Y, Buboltz AM, et al. Comparative genomics of the classical Bordetella subspecies: the evolution and exchange of virulence-associated diversity amongst closely related pathogens. BMC genomics 2012; 13: Buboltz AM, Nicholson TL, Weyrich LS, Harvill ET. Role of the type III secretion system in a hypervirulent lineage of Bordetella bronchiseptica. Infect Immun 2009; 77: Fennelly NK, Sisti F, Higgins SC, et al. Bordetella pertussis expresses a functional type III secretion system that subverts protective innate and adaptive immune responses. Infect Immun 2008; 76: Yuk MH, Harvill ET, Cotter PA, Miller JF. Modulation of host immune responses, induction of apoptosis and inhibition of NF-kappaB activation by the Bordetella type III secretion system. Molecular microbiology 2000; 35:

51 Nagamatsu K, Kuwae A, Konaka T, et al. Bordetella evades the host immune system by inducing IL-10 through a type III effector, BopN. The Journal of experimental medicine 2009; 206: Siciliano NA, Skinner JA, Yuk MH. Bordetella bronchiseptica modulates macrophage phenotype leading to the inhibition of CD4+ T cell proliferation and the initiation of a Th17 immune response. Journal of immunology 2006; 177: Pilione MR, Harvill ET. The Bordetella bronchiseptica type III secretion system inhibits gamma interferon production that is required for efficient antibody-mediated bacterial clearance. Infect Immun 2006; 74: Cotter PA, Miller JF. BvgAS-mediated signal transduction: analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in a rabbit model. Infect Immun 1994; 62: Yuk MH, Harvill ET, Miller JF. The BvgAS virulence control system regulates type III secretion in Bordetella bronchiseptica. Molecular microbiology 1998; 28: Dethlefsen L, McFall-Ngai M, Relman DA. An ecological and evolutionary perspective on human-microbe mutualism and disease. Nature 2007; 449: Que JU, Hentges DJ. Effect of streptomycin administration on colonization resistance to Salmonella typhimurium in mice. Infect Immun 1985; 48: Cornelis GR, Van Gijsegem F. Assembly and function of type III secretory systems. Annual review of microbiology 2000; 54: Stecher B, Robbiani R, Walker A, et al. Salmonella enterica serovar typhimurium exploits inflammation to compete with the intestinal microbiota. PLoS biology 2007; 5: Winter SE, Thiennimitr P, Winter MG, et al. Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 2010; 467: Banemann A, Deppisch H, Gross R. The lipopolysaccharide of Bordetella bronchiseptica acts as a protective shield against antimicrobial peptides. Infect Immun 1998; 66: Wickham ME, Brown NF, Boyle EC, Coombes BK, Finlay BB. Virulence is positively selected by transmission success between mammalian hosts. Current biology : CB 2007; 17: Nicholson TL, Brockmeier SL, Loving CL. Contribution of Bordetella bronchiseptica filamentous hemagglutinin and pertactin to respiratory disease in swine. Infect Immun 2009; 77: Nicholson TL, Brockmeier SL, Loving CL, Register KB, Kehrli ME, Jr., Shore SM. The Bordetella bronchiseptica type III secretion system is required for persistence and disease severity but not transmission in swine. Infect Immun 2013.

52 Nicholson T, Brockmeier S, Loving C, et al. Phenotypic modulation of the virulent Bvg phase is not required for pathogenesis and transmission of Bordetella bronchiseptica in swine. Infection and immunity 2012; 80: Kuper CF, Koornstra PJ, Hameleers DM, et al. The role of nasopharyngeal lymphoid tissue. Immunology today 1992; 13: Wu HY, Nguyen HH, Russell MW. Nasal lymphoid tissue (NALT) as a mucosal immune inductive site. Scandinavian journal of immunology 1997; 46: Medhekar B, Shrivastava R, Mattoo S, Gingery M, Miller JF. Bordetella Bsp22 forms a filamentous type III secretion system tip complex and is immunoprotective in vitro and in vivo. Molecular microbiology 2009; 71:

53 44 Chapter 3 : Filamentous hemagglutinin stimulates the recruitment of neutrophils that is required for shedding of Bordetella bronchiseptica

54 45 Abstract Despite transmission being the defining characteristic of infectious disease, little is known about how bacterial factors contribute to the spread of disease. Even with high vaccination rates, Bordetella species still continue to circulate in both agricultural animals and humans, causing substantial economic burden. Previously, the filamentous hemagglutinin of B. bronchiseptica has been shown to be important in adhesion to host tissues and in altering the host immune response. Its role in transmission, however, is not known. Using a recent low-dose laboratory model of infection, we investigated the effects of filamentous hemagglutinin on the natural transmission process of B. bronchiseptica: initial colonization, growth, and shedding. A mutant strain of B. bronchiseptica, lacking expression of filamentous hemagglutinin (RB50 fhab), was compared to the wild-type parental strain in its ability to transmit between TLR-4 deficient mice. RB50 fhab failed to transmit to susceptible individuals. RB50 fhab showed no defect in the number of bacteria necessary to colonize a host (ID 50 ) or with its ability to expand within the nasal cavity of mice; however it was defective in its ability to be expelled from the nasal cavity of mice. Furthermore, the ability of RB50 fhab to stimulate inflammatory responses at the site of infection was compromised, as neutrophil recruitment to the nasal cavity was significantly reduced during RB50 fhab infection. These results indicate that filamentous hemagglutinin stimulation of the inflammatory response and subsequent neutrophil recruitment to the nasal cavity is required for shedding of B. bronchiseptica from the nares of mice.

55 46 Introduction The classical Bordetella, B. pertussis, B. parapertussis, and B. bronchiseptica, are significant respiratory pathogens impacting both the agricultural industry and human health. Transmission of the Bordetella is thought to occur through continuous cycles between infected and naïve hosts, as no known environmental reservoir exists [1]. For successful transmission to occur, pathogens must overcome colonization resistance, expand within the host, and facilitate shedding from the infected host [2]. B. bronchiseptica is highly infectious in mice, providing a model to study pathogenesis of specific Bordetella virulence factors; however, the study of these factors during transmission was limited, as transmission of wild-type mice is rarely seen within a laboratory setting [3, 4]. Recently, defects in TLR-4 were shown to contribute to the lack of transmission, allowing for a model in which B. bronchiseptica virulence can be studied in the context of transmission. Infection by the Bordetella begins with the colonization of the upper respiratory tract. A two-component regulatory system, BvgAS, controls of the expression of key virulence factors that enable B. bronchiseptica to thrive within the host [5]. These virulence factors are expressed during the Bvg + phase. Expression of the adherence factor filamentous hemagglutinin (FHA) is regulated by BvgAS [6]. Mature FHA is a large 220-kDa β-helical protein containing binding domains that include a heparin-binding domain, which promotes attachment to polysaccharides, an Arg-Gly-Asp (RGD) domain that interacts with leukocyte-response integrin, and a cardohydrate-recognition domain facilitating bacterial binding to ciliated respiratory epithelial cells and macrophages [1, 7, 8]. The FHA of B. bronchiseptica is required for colonization of the trachea in a rat model of infection, while the FHA of B. pertussis is important during the colonization of the lungs [9, 10].

56 47 A highly immunogenic protein, FHA is both surface-associated and secreted by a twopartner secretion system suggesting that, in addition to functioning as an adhesion, FHA contributes to other aspects of infection [11]. FHA can modulate the immune response by interacting with receptors on macrophages, resulting in the suppression of IL-12, leading to the persistence of Bordetella within the respiratory tract [12, 13]. In particular, the RGD domain is responsible for the up-regulation of intercellular adhesion molecule 1 by epithelial cells and the up-regulation of CR3 binding activity on monocytes and macrophages by modulating the NF-κB signaling pathway [14-16]. Macrophages that come into contact with FHA inhibit antigendependent CD4+ T cell proliferation [17]. FHA can also lead to the stimulation of proinflammatory and apoptotic responses from human monocyte-like cells and bronchial epithelial cells [18, 19]. The proinflammatory response can lead to pathology of the respiratory tract, thus incorporation of FHA into vaccines elicits a strong antibody response required to limit disease in vaccine recipients [20]. Although studies have shown that FHA contributes to virulence and adherence, previous studies have failed to address its role in the transmission cycle of the Bordetella. In this work, we characterized the function of the B. bronchiseptica virulence factor FHA in transmission by analysis of a fhab mutant in B. bronchiseptica strain RB50. We found that FHA was required for transmission to susceptible mice. No change in the infectious dose was required to colonize mice or in RB50 fhab ability to expand in the nasal cavity. Absence of fhab, however, resulted in an inability to shed bacteria from the external nares of mice. Further analysis revealed that the recruitment of neutrophils to the nasal cavity of mice infected with RB50 fhab was significantly decreased. Together, these results show that in the absence of FHA B. bronchiseptica is unable to stimulate an inflammatory response necessary for the shedding of B. bronchiseptica.

57 48 Materials and Methods Ethics Statement. This study was carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee at The Pennsylvania State University (#40029). All animals were anesthetized using 5% isoflourane in oxygen and euthanized using carbon dioxide inhalation to minimize animal suffering. Bacterial Strains. B. bronchiseptica strain RB50 has been previously described [5]. RBx9, a fhab mutant strain derived from RB50, has also been described elsewhere and is referred to as RB50 fhab [10]. Bacteria were maintained on Bordet-Gengou agar (Difco) supplemented with 10% defibrinated sheep blood (Hema Resourses) and 20 µg/ml streptomycin (Sigma-Aldrich). Liquid cultures were grown overnight in Stainer-Scholte broth at 37 C to mid-log phase. Animal Studies Four- to six-week C3/HeJ and C57/B6 mice were obtained from The Jackson Laboratories and maintained in our specific pathogen-free facility. Inoculations were prepared from liquid cultures grown to mid-log phase and bacteria were diluted to a concentration of 3x10 4 CFU/mL or 1x10 3 CFU/ml. Mice were given 150 CFU in 5 µl of PBS that was deposited on the external nares for transmission and shedding experiments. For comparison of infectious dose, 5 CFU of RB50 or RB50 fhab was deposited on the external nares in 5µl PBS. To quantify bacteria, organs were excised on the day indicated. Following dissection, organs of interest were homogenized in 1mL PBS and diluted appropriately for culture and quantification on Bordet-Gengou agar. Transmission and Shedding Studies To determine transmission, a single mouse (index) was inoculated and placed in a cage containing three susceptible individuals. Transmission to susceptible individuals was determined

58 49 by dissection and culturing of the nasal cavity tissue on day 21 (limit of detection 5 CFU). Shedding of bacteria was determined by swabbing the external nares with a Dacron-polyester tipped swab (VWR International) for 15 seconds. Swabs were placed into 1ml of PBS and vortexed, and excised tissues were harvested and manually homogenized in 1ml PBS before being cultured on BG agar. Flow Cytometry. 10 ml of PBS was used to perfuse systemically. Nasal bones and NALT were excised and placed in 1ml of DMEM (5% FBS + 1mg/ml collagenase D (Roche)). Samples were incubated for 45 minutes at 37 o C and passed through a 70µm mesh screen to form a single cell suspension. Samples were resuspended in 200:1 Fc Block (BD Biosciences), anti-cd45:apc (BD Bioscience), anti-cd11c:pe-cy7 (BD Bioscience), anti-ly6g:fitc(bd Bioscience), and Anti-CD3e:PE (E Bioscience) and incubated at room temperature for 20 minutes. Cells were then washed and fixed with 2% paraformaldehyde. Statistical Analysis. Shedding and colonization data was fit to a generalized linear model (GLM) to determine the effect of group and day on outputs, such as within-host colonization, and shedding using Minitab v16. The effect of group on within-host colonization, and shedding at individual time points was determined by ANOVA using a Tukey simultaneous test for significance.

59 50 Results Filamentous hemagglutinin is required for transmission of B. bronchiseptica To determine whether RB50 fhab can be transmitted, C3/HeJ mice were inoculated with 150 CFU of RB50 or RB50 fhab. Inoculated Log10+CFU individuals were placed with 3 naïve susceptible mice. Upon completion of a 3 week co-housing period, susceptible mice were sacrificed and RB RB50 ShaB+ bacterial numbers in the nasal cavity were determined. No transmission events were recorded in susceptible mice exposed to mice inoculated with RB50 fhab, while 6 out of 6 Figure 3.1. Filamentous hemagglutinin is required for transmission of B. bronchiseptica Transmission of RB50 or RB50 fhab to susceptible individuals after co-housing with index mice. Dots represent the log 10 CFU of bacteria recovered from individual mice susceptible mice became infected when exposed to mice inoculated with RB50 (Figure 3.1). These results suggest that FHA contributes to the transmission process of B. bronchiseptica between mice. Efficient shedding, but not growth of B. bronchiseptica, requires filamentous hemagglutinin Defects in transmission could be due to two types of deficiencies: failure to colonize new hosts, or failure to shed from infected hosts. To determine if RB50 fhab was unable to colonize efficiently, C57/B6 mice were inoculated with 5 CFU in 5ul of PBS of either RB50 or RB50 fhab. As little as 5 CFU of RB50 or RB50 fhab was sufficient to infect 50 percent of individuals (ID 50 ) (Figure 3.2A). Furthermore, inoculation with 150 CFU of RB50 or RB50 fhab in C57/B6 mice resulted in growth within the nasal cavity with levels of RB50 and

60 RB50 RB50 DhaB Log10+CFU Log10+CFU RB50 RB50 fhab 0 Day0(inocula) Day7 Day35 Log10+CFU+per+second RB50 RB50 DhaB Days+Post+Challenege+ Figure 3.2. Filamentous hemagglutinin is required for shedding, but not initial colonization. (A) ID50 of RB50 and RB50 fhab were compared. Mice were inoculated with 5 CFU in 5 µl of PBS. Dots represent the log 10 CFU of bacteria recovered from individual mice on day 7.(B) Groups of 4 mice were challenged and nasal cavity colonization was quantified on days 0, 7, and 35 days post challenge. (C) shedding was detected by swabbing the external nares for 15 seconds throughout the infectious course. Symbols represent the mean CFU +1 (± SEM). Asterisks represent p RB50 fhab reaching approximately 100,000 CFU on day 7 (Figure 3.2B). Differences in colonization were not observed until day 35, confirming previous data, suggesting a role for FHA in persistence [10]. To determine if FHA was required for shedding from the external nares of mice, groups of C57/B6 mice were inoculated with 150 CFU of RB50 or RB50 fhab, and bacterial shedding was monitored from the nares of individual mice for 21 days. RB50 infected mice shed 10 s to 1000 s of bacteria at each time point sampled between days 6 to 14, with numbers peaking on day 7. No shedding was detected from mice infected with RB50 fhab. Together, these results indicate that FHA is not required during initial colonization of mice; however, is required for persistence and shedding from mice.

61 52 Filamentous hemagglutinin is required for recruitment of immune cells to the nasal cavity Previous studies have linked neutrophil infiltrate as a determinate for Bordetella shedding [4, 21]. To determine if recruitment of particular immune cell subsets to the nasal cavity of mice were affected during infection of RB50 fhab, C57/B6 mice Percent CD45+ cells RB50 RB50dfhaB RB50 fhab Neutrophils Dendritic Cells T cells Figure 3.3. Recruitment of neutrophils and dendritic cells to the nasal cavity are altered during RB50 fhab infection. Four mice were assayed 7 days post challenge with RB50 or RB50 fhab. Neutrophil, dendritic, and T cell numbers in the nasal cavity were quantified by flow cytometry Bars represent the mean ± standard error. Asterisks represent p were infected with 150 CFU of RB50 or RB50 fhab and immune cell subsets were analyzed at the peak of shedding, day 7, by flow-cytometry. Of the cell types analyzed, neutrophil and dendritic cell recruitment to the nasal cavity was significantly reduced in mice infected with RB50 fhab (p< 0.05). These results indicated that the failure of RB50 fhab to shed from the nares of mice is likely due to poor stimulation of the immune response during the initial colonization process.

62 53 Discussion Transmission of pathogens requires many steps, including initial colonization, withinhost growth, and shedding of the pathogen to new hosts. Assessment of a previously characterized mutant in fhab revealed that, along with contributing to overall persistence, FHA is required for transmission of B. bronchiseptica. The failure of mice to transmit RB50 fhab corresponded with an inability to shed from infected hosts. Shedding can be attributed to bacterial burden or immune mediated pathology. Colonization of RB50 fhab was comparable to RB50 on day 7, indicating that reduced shedding was not due to deficits in bacterial growth within the nasal cavity during peak shedding. Instead, reduced neutrophil and dendritic cell infiltrate was observed in mice infected with RB50 fhab. Together these results suggest that FHA contributes to immune pathology in the nasal cavity during initial infection, resulting in shedding and ultimately the transmission success of B. bronchiseptica. In vivo phenotypes for FHA have been more difficult to obtain, and multiple groups have reported different findings. Early work showed that mutants in FHA of B. pertussis were indistinguishable from wild type B. pertussis in their ability to both colonize and infect mice [22-25]. Others have reported that mutants in FHA of B. pertussis and B. bronchiseptica were indistinguishable from wild type in its ability to colonize the lungs and nasal cavity of mice; however, these mutants were defective in tracheal colonization [10, 26]. Furthermore, deletion of fhab in a swine isolate, KM22, showed that FHA was absolutely required for colonization of the respiratory tract of swine [27]. The study of adherence factors in the Bordetella is difficult due to the redundancy of multiple factors in adhesion [1]. Diversity within the Bordetella also likely contributes to the varying phenotypes observed. In particular, FHA has been attributed to

63 54 host restriction, as construction of a B. bronchiseptica strain containing FHA derived from B. pertussis failed to colonize the lower respiratory tracts [11]. FHA is often referred to as a Bordetella adherence factor, as mutants lacking FHA have been shown to be less adherent to cultured cells and fail to persist or efficiently colonize the lower respiratory tract of mice [9]. However, FHA is an important immune modulator, likely impacting the successful colonization of mice. FHA and pertussis toxin modulate the immune response to unrelated antigens during B. pertussis infection. FHA also stimulates a strong immune response and is currently used as an antigen in acellular pertussis vaccines [24, 28]. Stimulation of pro-inflammatory factors by FHA may contribute to transmission by causing clinical manifestations of disease within the host. By utilizing a low-dose infection model, we were able to determine that FHA was not required for the colonization and growth within the host nasal cavity. This finding suggests that the adherence of B. bronchiseptica is unaffected in vivo, despite the absence of FHA. Shedding, however, was severely affected in mice inoculated with RB50 fhab. Shedding of B. bronchiseptica and other pathogens is dependent on the immune status of the host [4, 21, 29, 30]. Cell recruitment of neutrophils was altered during RB50 fhab infection, suggesting that FHA is required to induce immune pathology in the nasal cavity of mice. Recruitment of neutrophils has previously correlated to shedding in the mouse model [4]. Furthermore, whole-cell vaccination of mice reduced shedding in this model [31]. The strong antibody response to FHA in whole-cell vaccines has previously been shown [32, 33], and may contribute to the reduced shedding of B. bronchiseptica in this model. An acellular vaccine that includes FHA failed to reduce shedding in mice; however, other factors may have contributed to the failure to control shedding in these mice, such as polarization towards a Th2 response instead of a Th1 response [31].

64 55 Our results indicate that FHA contributes to the infectiousness and transmissibility of B. bronchiseptica. We also show that the low-dose infection system complements previous findings, allowing for a more complete picture by the probing of interactions that are important for the initial colonization of new hosts or infectiousness of colonized hosts. While data for the role of FHA in high-dose infection models is less compelling, more natural doses revealed that FHA is required for shedding and transmission of B. bronchiseptica, likely impacting the burden of this infectious disease.

65 56 Authors and Contributions: William E. Smallridge 1,2, Olivier Y. Rolin 1,2, Eric T. Harvill 1 1 Department of Veterinary and Biomedical Sciences, The Pennsylvania State University, University Park, PA, Graduate Program in Immunology and Infectious Disease, The Pennsylvania State University, University Park, PA, Conceived and designed experiments: WES, OYR, ETH Performed experiments: WES, OYR Analyzed Data: WES, OYR, ETH Wrote Paper: WES, ETH

66 57 References 1. Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Keesing F, Belden LK, Daszak P, et al. Impacts of biodiversity on the emergence and transmission of infectious diseases. Nature 2010; 468: Goodnow RA. Biology of Bordetella bronchiseptica. Microbiological reviews 1980; 44: Rolin O, Smallridge W, Henry M, Goodfield L, Place D, Harvill ET. Toll-Like Receptor 4 Limits Transmission of Bordetella bronchiseptica. PLoS One 2014; 9:e Cotter PA, Miller JF. BvgAS-mediated signal transduction: analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in a rabbit model. Infect Immun 1994; 62: Roy CR, Falkow S. Identification of Bordetella pertussis regulatory sequences required for transcriptional activation of the fhab gene and autoregulation of the bvgas operon. Journal of bacteriology 1991; 173: Prasad SM, Yin Y, Rodzinski E, Tuomanen EI, Masure HR. Identification of a carbohydrate recognition domain in filamentous hemagglutinin from Bordetella pertussis. Infect Immun 1993; 61: Menozzi FD, Gantiez C, Locht C. Interaction of the Bordetella pertussis filamentous hemagglutinin with heparin. FEMS microbiology letters 1991; 62: Saukkonen K, Cabellos C, Burroughs M, Prasad S, Tuomanen E. Integrin-mediated localization of Bordetella pertussis within macrophages: role in pulmonary colonization. The Journal of experimental medicine 1991; 173: Cotter PA, Yuk MH, Mattoo S, et al. Filamentous hemagglutinin of Bordetella bronchiseptica is required for efficient establishment of tracheal colonization. Infect Immun 1998; 66: Inatsuka CS, Julio SM, Cotter PA. Bordetella filamentous hemagglutinin plays a critical role in immunomodulation, suggesting a mechanism for host specificity. Proceedings of the National Academy of Sciences of the United States of America 2005; 102: McGuirk P, Johnson PA, Ryan EJ, Mills KH. Filamentous hemagglutinin and pertussis toxin from Bordetella pertussis modulate immune responses to unrelated antigens. The Journal of infectious diseases 2000; 182:

67 McGuirk P, Mills KH. Direct anti-inflammatory effect of a bacterial virulence factor: IL-10- dependent suppression of IL-12 production by filamentous hemagglutinin from Bordetella pertussis. European journal of immunology 2000; 30: Ishibashi Y, Claus S, Relman DA. Bordetella pertussis filamentous hemagglutinin interacts with a leukocyte signal transduction complex and stimulates bacterial adherence to monocyte CR3 (CD11b/CD18). The Journal of experimental medicine 1994; 180: Ishibashi Y, Relman DA, Nishikawa A. Invasion of human respiratory epithelial cells by Bordetella pertussis: possible role for a filamentous hemagglutinin Arg-Gly-Asp sequence and alpha5beta1 integrin. Microbial pathogenesis 2001; 30: Relman D, Tuomanen E, Falkow S, Golenbock DT, Saukkonen K, Wright SD. Recognition of a bacterial adhesion by an integrin: macrophage CR3 (alpha M beta 2, CD11b/CD18) binds filamentous hemagglutinin of Bordetella pertussis. Cell 1990; 61: Boschwitz JS, Batanghari JW, Kedem H, Relman DA. Bordetella pertussis infection of human monocytes inhibits antigen-dependent CD4 T cell proliferation. The Journal of infectious diseases 1997; 176: Ishibashi Y, Nishikawa A. Role of nuclear factor-kappa B in the regulation of intercellular adhesion molecule 1 after infection of human bronchial epithelial cells by Bordetella pertussis. Microbial pathogenesis 2003; 35: Abramson T, Kedem H, Relman DA. Proinflammatory and proapoptotic activities associated with Bordetella pertussis filamentous hemagglutinin. Infect Immun 2001; 69: Edwards KM, Meade BD, Decker MD, et al. Comparison of 13 acellular pertussis vaccines: overview and serologic response. Pediatrics 1995; 96: Gopinath S, Hotson A, Johns J, Nolan G, Monack D. The systemic immune state of supershedder mice is characterized by a unique neutrophil-dependent blunting of TH1 responses. PLoS pathogens 2013; 9:e Goodwin MS, Weiss AA. Adenylate cyclase toxin is critical for colonization and pertussis toxin is critical for lethal infection by Bordetella pertussis in infant mice. Infect Immun 1990; 58: Khelef N, Zychlinsky A, Guiso N. Bordetella pertussis induces apoptosis in macrophages: role of adenylate cyclase-hemolysin. Infect Immun 1993; 61: Roberts M, Cropley I, Chatfield S, Dougan G. Protection of mice against respiratory Bordetella pertussis infection by intranasal immunization with P.69 and FHA. Vaccine 1993; 11: Weiss AA, Goodwin MS. Lethal infection by Bordetella pertussis mutants in the infant mouse model. Infect Immun 1989; 57:

68 Geuijen CA, Willems RJ, Bongaerts M, Top J, Gielen H, Mooi FR. Role of the Bordetella pertussis minor fimbrial subunit, FimD, in colonization of the mouse respiratory tract. Infect Immun 1997; 65: Nicholson TL, Brockmeier SL, Loving CL. Contribution of Bordetella bronchiseptica filamentous hemagglutinin and pertactin to respiratory disease in swine. Infect Immun 2009; 77: Denoel P, Godfroid F, Guiso N, Hallander H, Poolman J. Comparison of acellular pertussis vaccines-induced immunity against infection due to Bordetella pertussis variant isolates in a mouse model. Vaccine 2005; 23: Pathak A, Creppage K, Werner J, Cattadori I. Immune regulation of a chronic bacteria infection and consequences for pathogen transmission. BMC microbiology 2010; 10: Cattadori IM, Pathak A, Murphy L. COS 56-4: The role of immune mediated co-infections on disease transmission. In: The 95th ESA Annual Meeting. 31. Smallridge WE, Rolin OY, Jacobs NT, Harvill ET. Different effects of whole cell and acellular vaccines on Bordetella transmission. The Journal of infectious diseases Cherry JD, Gornbein J, Heininger U, Stehr K. A search for serologic correlates of immunity to Bordetella pertussis cough illnesses. Vaccine 1998; 16: Cherry JD. Comparative efficacy of acellular pertussis vaccines: an analysis of recent trials. The Pediatric infectious disease journal 1997; 16:S90-6.

69 Chapter 4 : Different effects of whole-cell and acellular vaccines on Bordetella transmission 60

70 61 Abstract Background. Vaccine development has largely focused on the ability of vaccines to reduce disease within individual hosts with less attention to assessing the vaccine s effects on transmission between hosts. Current acellular vaccines against Bordetella pertussis are effective in preventing severe disease, but have little effect on less severe coughing illness that can mediate transmission. Methods. Using mice as a natural host of Bordetella bronchiseptica, we determined the effects of vaccination on shedding and transmission of this pathogen. Results Vaccination with heat-killed whole-cell B. bronchiseptica or B. pertussis inhibited shedding of B. bronchiseptica. Differences in neutrophil and B cell recruitment distinguished sham-vaccine from whole-cell vaccine responses and correlated with shedding output. Both B and T cells were essential for vaccine-induced control of shedding. Adoptive transfer of antibodies was able to limit shedding, while depletion of CD4+ T cells led to increased shedding in vaccinated mice. Finally, whole-cell vaccination was able to prevent transmission; however, an acellular vaccine that effectively controls disease failed to control shedding and transmission. Conclusion. Our results highlight discrepancies between whole-cell and acellular vaccination that could contribute to the increased incidence of B. pertussis, since the transition to the use of acellular vaccination.

71 62 Introduction The incidence of whooping cough, once a common and deadly childhood disease, was greatly reduced following the introduction of whole-cell vaccines in the late 1940s [1]. However, concern about their side effects led to a transition to acellular vaccines in the 1980s [2]. Subsequently, whooping cough incidence has increased to levels 50-fold higher than the alltime low in the US in 1976 [3]. In many countries, even those with high vaccine coverage, the pathogen continues to spread and cause periodic epidemics [4-9], raising questions about the effects of vaccines on the spread of disease. Acellular vaccines provide protection against the most severe forms of whooping cough, but are less effective against B. pertussis infections associated with milder forms of coughing illness [1, 4]. Consequently, there is debate about whether acellular vaccines induce the most effective type of immune response to protect vaccinated individuals and prevent spread of disease [10]. Both whole-cell and acellular vaccines are able to generate high levels of antibodies toward the bacterial components present in each. But only the whole-cell vaccine efficiently activates Th1 cells that generate an effective IFN-γ response that is important in the control and clearance of B. pertussis infection [11, 12]. Acellular vaccination creates a largely Th2 and Th17 response that is less effective in animal models, potentially explaining the increased incidence coinciding with the switch to acellular vaccines[10, 12]. To date, these analyses have been limited to studies within individuals. However, the observed differences in response would also be expected to affect the inflammatory response that could contribute to symptoms, such as coughing and sneezing, the primary mechanisms of transmission of B. pertussis. Therefore, the different effects of the two vaccines could also contribute to the

72 63 observed increase in incidence by affecting the transmission of B. pertussis, although these effects have not been measured experimentally. Detailed molecular studies of Bordetella pathogenesis have been performed in the mouse model because of its simplicity and reproducibility, and have been consistent with the limited work done in humans and other animals [1, 13]. In many cases, Bordetella bronchiseptica, a closely related subspecies of B. pertussis that naturally infects a wide range of mammalian hosts including humans and mice, has been used as a model system to study the infectious process [1, 14]. Since B. bronchiseptica naturally infects mice with as few a 5 CFU, both interactions between bacterial factors and host immune functions can be probed to the molecular level during the infectious process. These infection models have focused on the interactions between bacteria and an experimentally inoculated host, largely avoiding the defining characteristic of infectious disease: transmission. To overcome this limitation, we have recently developed a transmission system in mice, in which we have demonstrated the importance of innate immune activity regulated by toll-like receptor 4 (TLR4) in limiting transmission of B. bronchiseptica [15]. Defects in TLR4 led to increases in both infectiousness of the individual infected (shedding) and host susceptibility (colonization of susceptible mice) in this model. Importantly, this work has provided an experimental system in which transmission can be studied experimentally, allowing direct measurement of the effects of vaccines on various aspects that contribute to transmission. In this study, we examine the effects of vaccination on transmission and examine the immune mechanisms involved in these effects. Consistent with expectations, whole-cell vaccines (B. pertussis or B. bronchiseptica) induce an immune response that limits shedding and blocks transmission from inoculated index cases. Both memory CD4+ T cells, as well as antibodies, are implicated in the control of shedding. Whole-cell vaccination was effective in

73 64 limiting shedding and infectivity of B. bronchiseptica, while acellular vaccination was less effective. Together these results suggest that the resurgence of B. pertussis could be due to two deficiencies of the acellular vaccines: failure to protect the vaccinated individual from infection, only blunting the severity of disease, and failure to prevent the transmission of B. pertussis. The different effects of vaccines on individual and herd immunity should be important considerations for the next generation of B. pertussis vaccines.

74 65 Materials and Methods Ethics Statement. This study was carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee at The Pennsylvania State University (#40029). All animals were anesthetized using isoflourane or euthanized using carbon dioxide inhalation to minimize animal suffering. Mice. Four- to six-week C57/B6, C3/HeJ, MµMT -/-, and TCRβδ -/- mice were obtained from The Jackson Laboratories and maintained in our specific pathogen-free facility. Vaccination. Liquid cultures were grown to 1x10 9 CFU/ml. Bacteria were heat-killed at 65 o C for 30 minutes. Vaccinated mice were intraperitoneally injected with sham (PBS), 2x10 8 heat killed bacteria in 200ul of PBS (referred to as whole-cell), or Adacel (referred to as acellular; Sanofi Pasteur) at 1/5 the human dose on days 0 and 14. Challenge occurred on day 35 after initial vaccination. Bacterial Strains. B. bronchiseptica strain RB50 [16] and B. pertussis strain 536 [17] were maintained on Bordet-Gengou (BG) agar with 10% defibrinated sheep blood (Hema Resources) and 20 µg/ml streptomycin (Sigma-Aldrich). Liquid cultures were grown in Stainer-Scholte broth at 37 C to mid-log phase. Inoculation. Inocula were prepared from mid-log phase liquid cultures. Cultures were diluted to 3x10 4 CFU/ml. Mice were anaesthetized with 5% isoflurane in oxygen, and a 5µl droplet was placed on nares for infection.

75 66 Flow Cytometry. 10 ml of PBS was used to perfuse systemically. Nasal bones were excised and placed in 1ml of DMEM (5% FBS + 1mg/ml collagenase D (Roche)). Samples were incubated for 45 minutes at 37 o C and passed through a 70µm mesh screen to form a single cell suspension. Samples were resuspended in 200:1 Fc Block (BD Biosciences) in PBS + 2% FBS and incubated on ice for 20 minutes. Cells were incubated in the following antibodies in PBS + 2% FBS: anti- CD45:V500 (BD Bioscience), anti-cd11b:horizon V450 (BD Bioscience), anti-ly6g: APC- Cy7 (BD Bioscience), anti-cd11c:fitc (BD Bioscience), Anti-F480:PE-Cy7 (E Bioscience) Anti I-A d :PE (E Bioscience), Anti-CD3: APC (E Bioscience), anti-cd19:percp (BioLegend). Quantification of Bacteria. Shedding of bacteria was determined by swabbing the external nares with a Dacron-polyester tipped swab (VWR) for 15 seconds. Swabs were placed into 1ml of PBS and vortexed, while tissues were harvested and manually homogenized in 1ml PBS before being cultured on BG agar. Adoptive transfer For adoptive transfer, C57/B6 mice were vaccinated and serum was collected on day 28 postmortem from vaccinated or naive mice. 200 ul of vaccine induced serum or naive serum was injected intraperitoneally into MµMT -/- mice directly before challenge of B. bronchiseptica. Depletion of CD4+ cells CD4+ cells were depleted by intraperitoneal injection of 0.5 mg of monoclonal antibody from the GK1.5 hybridoma [18] 24 hours prior to inoculation and every other day thereafter until completion of the time-course (Day 21). Antibody detection. For detection of B. bronchiseptica-specific antibodies 100 ul of PBS was washed through the nasal cavity several times after sacrifice and was analyzed by ELISA, using secondary antibodies for mouse immunoglobulin.

76 67 Transmission Studies Todeterminetransmission,avaccinatedsinglemouse(index)wasinoculatedandplacedin acagecontainingthreesusceptibleindividuals.tomonitortransmission,susceptiblemice wereswabbedeveryotherdayforsheddinguntilcompletionofthetimevcourse.observed shedding was counted as a transmission event on the day detected, and confirmed by dissectionandculturingofthenasalcavitytissueonday21(limitofdetection5cfu). Statistical+Analysis++ Time-course analysis was analyzed by fit to a generalized linear model (GLM) to determine the effect of treatment over the time-course. The effect at individual time points was determined by ANOVA using a Tukey simultaneous test for significance. Minitab v.16 was used for all analyses.

77 68 Results Whole-cell vaccination reduces shedding To determine whether vaccination affects shedding of bacteria from the nares, groups of C57/B6 mice were vaccinated with whole-cell B. bronchiseptica or shamvaccine. 35 days after initial vaccination, mice were challenged with 150 bacteria in 5 ul of PBS. External nares were swabbed to quantify shedding periodically over a threeweek period post-challenge. Sham vaccinated mice shed 10 s to 1000 s of bacteria at each time point sampled between days 6 to 14, Log10+CFU+Per+Second+ Log10+CFU Sham WholeVcell Days+Post+Challenge+ Sham WholeVcell with numbers peaking on day 8. Whole-cell vaccination decreased bacterial shedding by more than 90% (p 0.01) during the peak period of shedding, and throughout the timecourse (GLM; F (10,66) =4.49; p=0.000) (Figure 4.1A). To determine if the decrease in Days+Post+Challenge+ Figure 4.1. Vaccination affects B. bronchiseptica shedding, but not colonization. Groups of 4 whole-cell or sham-vaccinated mice were challenged and (A) shedding was detected by swabbing the external nares for 15 seconds throughout the infectious course. (B) Nasal cavity colonization was quantified on days 3, 7, and 21 days post challenge. Symbols represent the mean CFU +1 (± SEM). Asterisks represent p shedding was due to reduced bacterial colonization, groups of similarly treated mice were sacrificed on days 3, 7, and 21. Surprisingly, the bacterial burden in the nares of whole-cell vaccinated and sham-vaccinated mice were indistinguishable (Figure 4.1B). Together these data

78 69 suggest that vaccination affects shedding by some mechanism other than reducing bacterial numbers. Immune cell recruitment correlates with shedding output Figure 4.2. Vaccination alters the recruitment of B cells and neutrophils to the nasal cavity during B. bronchiseptica infection. Four whole-cell or sham-vaccinated mice were assayed 7 days post challenge with B. bronchiseptica. Neutrophil, macrophage, and B cell numbers in the nasal cavity were quantified by flow cytometry Bars represent the mean ± standard error. Asterisks represent p The immune response can affect symptoms that are known to contribute to shedding and transmission, such as coughing and sneezing. Vaccination, which can lead to rapid activation of immune components at the site of infection, could affect shedding in multiple, different ways. To understand the effects of vaccination on the nasal cavity Figure 4.3. Recruitment of B cells and neutrophils correlates with shedding output during B. bronchiseptica infection. Groups of 4 whole-cell B. bronchiseptica or sham vaccinated mice were swabbed on day 7 post inoculation and bacteria shed was plotted as a function of nasal cavity neutrophil (A) or B cell (B) percents. (C) Shedding is plotted as a function of neutrophils/ml in blood. leukocyte response, nasal cavities of B. bronchiseptica inoculated sham or whole-cell vaccinated

79 70 C57/B6 mice were analyzed at the peak of shedding, day 7, by flow-cytometry. Significant changes in the total leukocyte population were observed in whole-cell vaccinated mice, as compared to sham-vaccinated mice. Significant increases in the percentage of B cells (p 0.01) and decreases in the percentage of neutrophils (p 0.01) were observed in the nasal cavity of whole-cell vaccinated mice (Figure 4.2). To determine if these changes correlate with the shedding output of mice, shedding of each individual mouse was determined prior to sacrifice. Increases in B cells within the nasal cavity correlated with the decrease in shedding observed in vaccinated individuals (p=0.074), while increased neutrophil recruitment correlated to increased shedding output (p=0.05) (Figure 4.3). These results suggest that vaccination alters the recruitment of immune cells, making the environment less conducive to shedding of the pathogen. Vaccines induce adaptive immunity that limits shedding The immune memory generated by vaccination is mediated by B and T cells, which are important in directing an immune response that ultimately controls and clears the infection. The distinct roles of B and T cells on shedding, however, are not defined. Sham-vaccinated mice shed larger numbers of bacteria earlier in infection, but those numbers decreased to near the limit of detection on day 14. Based on their role in regulating inflammation, we hypothesized that memory CD4+ T cells play an important role in this reduction in shedding over time and in the vaccine-induced reduction in shedding. To determine whether the T cells are involved in the effect of whole-cell vaccination on shedding, we compared wild type (C57/B6) and T celldeficient (TCRβδ -/- ) mice. As before, vaccination of C57/B6 mice was effective at reducing shedding throughout the time-course (Figure 4.4A). However, vaccination did not reduce shedding in TCRβδ -/- mice throughout the time-course (GLM; F (10,66) =3.48 p=0.001). Whole-cell

80 71 Figure 4.4. Adaptive immune components limit shedding of B. bronchiseptica during infection. (A) Groups of 4 whole-cell vaccinated C57/B6 or TCRβδ -/- mice were inoculated and shedding was detected throughout the infectious course (B) Depletion of CD4+ cells from groups of 5 whole-cell vaccinated C57/B6 mice. (C) Groups of 4 whole-cell vaccinated C57/B6 or MµMT -/- mice were inoculated and shedding was detected throughout the infectious course. (D) Adoptive transfer of whole-cell induced serum or naïve serum to groups of 8 MµMT -/- mice, shedding was detected throughout the infectious course. Symbols represent the mean CFU +1 of B. bronchiseptica (± SEM). Asterisks represent p vaccinated TCRβδ -/- mice failed to limit shedding even after day 14 (Figure 4.4), indicating that T cells are important in the control of shedding in naïve animals. To look specifically at the function of CD4+ T cells, whole-cell vaccinated wild type mice were administered an isotype control or an anti-cd4 antibody prior to challenge and throughout the time course. Depletion of CD4+ T cells in whole-cell vaccinated mice increased shedding when compared to the isotype control group throughout the time course (GLM; F (7,112) =1.88 p=0.079) to levels similar to that of sham-vaccinated mice (Figure 4.4B). Similarly whole-cell vaccination of mice lacking TNF-α, IFN-γ, and IL-6 failed to decrease shedding, indicating that production of these powerful

81 72 immune cytokines by the adaptive immune response is require for effective control (Appendix Figure C) To evaluate the effects of B cells recruited to the site of infection on the control of shedding, mice lacking B cells (MuMT -/- ) were whole-cell vaccinated and compared to C57/B6 whole-cell vaccinated mice. Vaccinated mice lacking B cells shed higher numbers of bacteria than wild type vaccinated mice from day 9 through at least day 21 (GLM F (10,66) =2.90 p=0.005) (Figure 4.4C). The whole-cell vaccine induced an increase in anti-b. bronchiseptica Ig antibody titers within the nasal cavity of vaccinated mice, as compared to sham-vaccinated mice (Appendix Figure D). To determine whether antibodies alone were sufficient to control shedding, whole-cell vaccine-induced antibodies were adoptively transferred into naïve MuMT -/- mice prior to challenge. Adoptive transfer of vaccine-induced immune serum decreased shedding throughout the time course, as compared to adoptively transferred naïve serum (GLM; F (10,162) =2.16 p=0.022) (Figure 4.4D), indicating that antibodies play an important role in the control of shedding. Together these results indicate that altered host response guided by antibodies and memory CD4+ cells can limit the successful shedding of B. bronchiseptica. Acellular vaccination fails to inhibit shedding Although most of the world still uses them, whole-cell vaccines are no longer used in countries like the US, which now exclusively uses acellular vaccines that are known to induce different immunity than whole-cell vaccines. To Figure 4.5. Acellular vaccination is ineffective at inhibiting shedding during B. bronchiseptica infection Groups of 4 mice vaccinated with whole-cell B. bronchiseptica, whole-cell B. pertussis, acellular (Adacel), or PBS (sham) were challenged and shedding of B. bronchiseptica was detected throughout the infectious course. Asterisk represent p Symbols represent the mean CFU +1 of B. bronchiseptica (± SEM).

82 73 experimentally test the effects of an acellular vaccine on shedding, we used a commercially available B. pertussis acellular vaccine, Adacel. 4 out of the 5 antigens in Adacel are highly conserved and expressed in both B. pertussis and B. bronchiseptica (Pertactin, Fim 2/3, and Filamentous hemagglutinin). Previous studies show that Adacel is effective at controlling disease caused by B. bronchiseptica [19] and produces anti-b. bronchiseptica antibodies (Appendix Figure E). Mice were vaccinated and boosted as before with either sham, whole-cell B. bronchiseptica vaccine, whole-cell B. pertussis vaccine, or the acellular, Adacel, vaccine and challenged 35 days after the initial vaccination. Sham-vaccinated mice shed 10 s to 100 s of bacteria at each time point sampled between days 6 to 14 (Figure 4.5), as observed above (Figure 4.1). Vaccination with whole-cell B. pertussis or whole-cell B. bronchiseptica reduced shedding by >90% on day 8 (p 0.05). In contrast, acellular vaccination did not significantly affect shedding in either C57/B6 (Figure 4.5) or C3/HeJ mice (Appendix Figure F). These findings highlight important differences in vaccines that could dramatically affect transmission rates. Whole-cell vaccination controls transmission Based on the discrepancy between the effects of vaccines on shedding, their effects on transmission were determined using our recently developed transmission model in C3/HeJ (TLR4 deficient) mice [20]. Sham, whole-cell or acellular vaccinated mice were inoculated with B. %+Transmission Sham WholeVcellB.b Acellular Days+Post+Exposure+ bronchiseptica (index mice) and placed in cages with naïve, unvaccinated mice (susceptible mice). Susceptible mice were then monitored via Figure 4.6. Whole-cell vaccination, but not acellular vaccination, is effective at controlling transmission to susceptible individuals. Percent transmission of B. bronchiseptica to susceptible individuals after co-housing is compared for mice previously vaccinated with whole-cell, acellular of PBS (sham). Transmission events were determined by observation of shedding from susceptible individuals.

83 74 swabbing to determine when transmission events occurred. Sham-vaccinated mice readily transmitted to 6 out of 8 exposed cage mates respectively within 21 days, with the majority of transmission occurring prior to day 14 (Figure 4.6). Vaccination of index mice with whole-cell B. bronchiseptica vaccine was effective at inhibiting transmission to secondary mice, as 0 out of 8 exposed susceptible mice became colonized. Vaccination with acellular Adacel, however, was less effective at controlling transmission to susceptible mice, with 3 out of 8 exposed mice becoming colonized relatively early in the time course (Figure 4.6). Together, the results demonstrate that vaccination can substantially affect transmission to exposed animals. Furthermore, these results identify potentially important differences between whole cell and acellular vaccines in an experimental model that could be used to develop vaccines that induce both individual and herd immunity by preventing transmission.

84 75 Discussion The rapid decline in whooping cough clinical cases, after the introduction of the wholecell vaccine, suggested herd immunity played some role in its decline. However, it has been suggested that whole-cell and acellular pertussis vaccines do not provide herd immunity, as cases increased significantly after vaccine scares in countries such as the U.K. [21]. Recent clinical observations conversely point to short-lived immunity created by the acellular vaccine as having drastic effects on the rates of the disease [22, 23]. Clinical studies conducted in cohorts from Australia during the transition to acellular vaccines determined that receiving an initial wholecell vaccine had significant effect on reducing the incidence of pertussis compared to those receiving only the acellular vaccine [24]. Furthermore, recent clinical data has suggested that adolescent boosting affects the rate of whooping cough hospitalizations in infants, though the study did not distinguish between adolescents initially vaccinated with whole-cell or acellular vaccines [25]. These findings suggest that whole-cell vaccines provide improved priming and longer immunity that ultimately affect the incidence of whooping cough cases. Here we suggest an experimental mechanism where current vaccines prevent disease, but are less effective than their whole-cell counterparts at reducing transmission rates. Ultimately, control of whooping cough is dependent on understanding the transmission mechanisms of the Bordetella in naïve and immune populations. Currently, both human and baboon experimental models lack the powerful immunological tools of the mouse model and therefore cannot similarly probe the immunological mechanisms involved in the effects of vaccination of transmission [26]. Understanding these mechanisms is necessary before there can be rational design of improved vaccines and therapeutic interventions to block transmission.

85 76 Although whole-cell vaccination did not reduce bacterial numbers, it prevented transmission of B. bronchiseptica, decreasing both intensity and duration of shedding from infected individuals. The adaptive immune response generated by whole-cell vaccination altered the recruitment of specific immune cell types to the site of infection. Reduced shedding from whole-cell vaccinated mice corresponded with an influx of B cells to the site of infection. Shedding intensity increases correlated with increased neutrophil infiltrate; vaccination reduced neutrophil numbers and shedding of bacteria. B cells were required for vaccine-induced reduction of shedding, and adoptive transfer of whole-cell induced antibodies was sufficient to reduce shedding. Cell depletion of CD4+ T cells demonstrated that these cells that control recruitment and inflammation are also an important component in the control of shedding. We were surprised to determine that an acellular vaccine previously found to affect pathology and colonization of the lungs was ineffective at inhibiting shedding and transmission. This finding has important implication and could partly explain the recent rise in whooping cough cases. Adaptive immunity plays an important role in controlling shedding past day 14 in unvaccinated individuals (Figure 4.1). The type of adaptive immune responses induced by different vaccines affect the outcome of infection. A Th-1 response is characteristic of wholecell vaccination [27]. The IFN-γ secretion during a Th-1 response can lead to the activation of phagocytic cells clearing infection of B. pertussis [28]. Acellular vaccination, however, induces a Th-2 response characterized by IL-4 secretion and a strong antibody response [29]. The largely Th-2 response generated by acellular vaccines has been hypothesized to lower efficiency compared to the largely Th-1 whole-cell vaccines [10]. Based on the important role of this cell type in our mouse model, we hypothesize that the whole-cell vaccination-induced Th-1 response is also responsible for the control of shedding. Results from a recent vaccine study in the baboon

86 77 model similarly show that acellular whooping cough vaccines only blunt the severity of disease within individuals, thus failing to inhibit transmission [30]. Using this mouse model, future studies can determine which immune components whole-cell vaccines are activating to control transmission and why acellular vaccines are unable to activate those components. Similarly, new vaccine targets and adjuvants can be assessed for their effects on pathology, as well as on shedding and transmission. Pathogens are known to manipulate the host response to benefit their fitness. Salmonella typhimurium stimulates an inflammation response to compete with other host microbiota and enhance its transmission [31]. Immune regulation by either the host[32] or pathogen [33] can have important effects on shedding. Memory CD4+ cells have been shown to reduce neutrophil infiltration during persistent Salmonella typhimurium infection, in which CD4 T cell exhaustion can trigger release of large numbers of the pathogen in the feces of mice [34]. Our findings suggest that CD4 T cells play an important role in controlling shedding, likely by their effects on the inflammatory response. Previous studies have shown that immune sera is sufficient for clearing B. bronchiseptica from the lower respiratory tract, but not the upper respiratory tract of mice [35]. Here we observe an increase in B. bronchiseptica specific antibodies within the nasal cavity of mice, but little effect of those antibodies on colonization. Those antibodies, however, do have an effect on the shedding of B. bronchiseptica from the nares of mice, likely by helping to control inflammation [36, 37]. Cases of Bordetella related coughing illnesses are increasing. There are several theories as to why this is occurring, from vaccine-driven antigenic shift to inefficient vaccine response. The data presented here support an alternative explanation; current vaccines do not effectively prevent transmission of Bordetella and thus fail to confer the full benefits of herd immunity in

87 78 reducing clinical cases. Despite acellular vaccination being able to protect the majority of individuals from severe disease, increases in whooping cough cases coincide with the increased use of acellular vaccines beginning in the 1980 s. It is possible that the effects of acellular vaccination could mask symptoms, allowing infected individuals to act as unsuspecting reservoirs for potential spread to more susceptible individuals. Importantly, full implementation of acellular-only vaccination began relatively recently, in the 1990 s. Therefore, the proportion of the population that has only received the acellular vaccine will continue to rise for several years to come, raising the possibility that we may observe further increases in cases. Acknowledgments. Thanks to Dr. Andy Gunderson for providing α-cd4 neutralizing antibodies. This work was supported by National Institutes of Health Grant GM (to E.T.H.).

88 79 Authors and Contributions: William E. Smallridge 1,2, Olivier Y. Rolin 1,2, Nathan T. Jacobs 1, and Eric T. Harvill 1 1 Department of Veterinary and Biomedical Sciences, The Pennsylvania State University, University Park, PA, Graduate Program in Immunology and Infectious Disease, The Pennsylvania State University, University Park, PA, Conceived and designed experiments: WES, OYR, ETH Performed experiments: WES, OYR, NTJ Analyzed the data: WES, OYR, NTJ, ETH Wrote the paper: WES, ETH

89 80 References 1. Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Cherry JD. Pertussis: challenges today and for the future. PLoS pathogens 2013; 9:e Allen A. Public health. The pertussis paradox. Science 2013; 341: Crowcroft NS, Stein C, Duclos P, Birmingham M. How best to estimate the global burden of pertussis? Lancet Infect Dis 2003; 3: Crowcroft NS, Booy R, Harrison T, et al. Severe and unrecognised: pertussis in UK infants. Arch Dis Child 2003; 88: Guris D, Strebel PM, Bardenheier B, et al. Changing epidemiology of pertussis in the United States: increasing reported incidence among adolescents and adults, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 1999; 28: de Greeff SC, de Melker HE, van Gageldonk PG, et al. Seroprevalence of pertussis in The Netherlands: evidence for increased circulation of Bordetella pertussis. PLoS One 2010; 5:e Hellenbrand W, Beier D, Jensen E, et al. The epidemiology of pertussis in Germany: past and present. BMC infectious diseases 2009; 9: Quinn HE, McIntyre PB. Pertussis epidemiology in Australia over the decade trends by region and age group. Communicable diseases intelligence quarterly report 2007; 31: Higgs R, Higgins SC, Ross PJ, Mills KH. Immunity to the respiratory pathogen Bordetella pertussis. Mucosal immunology 2012; 5: Barbic J, Leef MF, Burns DL, Shahin RD. Role of gamma interferon in natural clearance of Bordetella pertussis infection. Infect Immun 1997; 65: Ryan M, Murphy G, Ryan E, et al. Distinct T-cell subtypes induced with whole cell and acellular pertussis vaccines in children. Immunology 1998; 93: Warfel JM, Merkel TJ. Bordetella pertussis infection induces a mucosal IL-17 response and long-lived Th17 and Th1 immune memory cells in nonhuman primates. Mucosal immunology de la Torre MJ, de la Fuente CG, de Alegria CR, Del Molino CP, Aguero J, Martinez- Martinez L. Recurrent respiratory infection caused by Bordetella bronchiseptica in an immunocompetent infant. The Pediatric infectious disease journal 2012; 31:981-3.

90 Olivier Rolin WS, Michael Henry, Laura Goodfield, David Place, Eric T. Harvill. Toll Like Receptor 4 Limits Transmission of Bordetella bronchiseptica PloS One In revision at PLoS one. 16. Cotter PA, Miller JF. BvgAS-mediated signal transduction: analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in a rabbit model. Infect Immun 1994; 62: Stibitz S, Yang MS. Subcellular localization and immunological detection of proteins encoded by the vir locus of Bordetella pertussis. Journal of bacteriology 1991; 173: Dialynas DP, Wilde DB, Marrack P, et al. Characterization of the murine antigenic determinant, designated L3T4a, recognized by monoclonal antibody GK1.5: expression of L3T4a by functional T cell clones appears to correlate primarily with class II MHC antigenreactivity. Immunological reviews 1983; 74: Goebel EM, Zhang X, Harvill ET. Bordetella pertussis infection or vaccination substantially protects mice against B. bronchiseptica infection. PLoS One 2009; 4:e Rolin O, Muse SJ, Safi C, et al. Enzymatic Modification of the Lipid A by an ArnT Protects B. bronchiseptica Against Cationic Peptides and Is Required for Transmission. Infect Immun Gangarosa EJ, Galazka AM, Wolfe CR, et al. Impact of anti-vaccine movements on pertussis control: the untold story. Lancet 1998; 351: Lavine J, Bjørnstad O, de Blasio B, Storsaeter J. Short-lived immunity against pertussis, agespecific routes of transmission, and the utility of a teenage booster vaccine. Vaccine 2012; 30: Lavine J, Broutin H, Harvill E, Bjørnstad O. Imperfect vaccine-induced immunity and whooping cough transmission to infants. Vaccine 2010; 29: Sheridan SL, Ware RS, Grimwood K, Lambert SB. Number and order of whole cell pertussis vaccines in infancy and disease protection. JAMA : the journal of the American Medical Association 2012; 308: Auger KA, Patrick SW, Davis MM. Infant hospitalizations for pertussis before and after Tdap recommendations for adolescents. Pediatrics 2013; 132:e Warfel JM, Beren J, Merkel TJ. Airborne transmission of Bordetella pertussis. The Journal of infectious diseases 2012; 206: Ross PJ, Sutton CE, Higgins S, et al. Relative contribution of Th1 and Th17 cells in adaptive immunity to Bordetella pertussis: towards the rational design of an improved acellular pertussis vaccine. PLoS pathogens 2013; 9:e

91 Torre D, Ferrario G, Bonetta G, Perversi L, Tambini R, Speranza F. Effects of recombinant human gamma interferon on intracellular survival of Bordetella pertussis in human phagocytic cells. FEMS immunology and medical microbiology 1994; 9: Ryan EJ, Nilsson L, Kjellman N, Gothefors L, Mills KH. Booster immunization of children with an acellular pertussis vaccine enhances Th2 cytokine production and serum IgE responses against pertussis toxin but not against common allergens. Clinical and experimental immunology 2000; 121: Warfel JM, Zimmerman LI, Merkel TJ. Acellular pertussis vaccines protect against disease but fail to prevent infection and transmission in a nonhuman primate model. Proceedings of the National Academy of Sciences of the United States of America Winter SE, Thiennimitr P, Winter MG, et al. Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 2010; 467: Pathak A, Creppage K, Werner J, Cattadori I. Immune regulation of a chronic bacteria infection and consequences for pathogen transmission. BMC microbiology 2010; 10: Wickham M, Brown N, Boyle E, Coombes B, Finlay B. Virulence is positively selected by transmission success between mammalian hosts. Current biology : CB 2007; 17: Gopinath S, Hotson A, Johns J, Nolan G, Monack D. The systemic immune state of supershedder mice is characterized by a unique neutrophil-dependent blunting of TH1 responses. PLoS pathogens 2013; 9:e Kirimanjeswara GS, Mann PB, Harvill ET. Role of antibodies in immunity to Bordetella infections. Infect Immun 2003; 71: Russell MW, Reinholdt J, Kilian M. Anti-inflammatory activity of human IgA antibodies and their Fab alpha fragments: inhibition of IgG-mediated complement activation. European journal of immunology 1989; 19: van der Neut Kolfschoten M, Schuurman J, Losen M, et al. Anti-inflammatory activity of human IgG4 antibodies by dynamic Fab arm exchange. Science 2007; 317:

92 83 Chapter 5 : Summary and Significance

93 84 Synopsis Bacterial pathogens have evolved complex interactions with their hosts in order to survive and reproduce. In order to be successful, a pathogen must transmit between hosts. Successful transmission requires components from both the host and bacteria; however, this complex interaction in the Bordetella is still being investigated. This dissertation has analyzed several bacterial and host factors that are required for transmission, and examines the effect of two vaccines on transmission of the Bordetella in a mouse model system. Summary and Implications Bordetella virulence factors contribute to transmission Interactions between infecting pathogens and host immunity have been almost exclusively studied using reproducible high-dose infection models. These studies are ideal for demonstrating components required to establish or repel infections; however, they lack the ability to examine how natural infections begin. Recently, TLR-4 was shown to limit transmission of B. bronchiseptica in mice, as transmission of wild-type mice is rarely seen within a laboratory setting [1, 2]. TLR-4 polymorphisms are known to affect the susceptibility of individuals to E. coli and Meningococcus [3]. Therefore, TLR-4 stimulation may be important to limit the spread of disease. The human pathogens, B. pertussis and B. parapertussis, poorly stimulate TLR-4 during infection, potentially contributing to their circulation within the human population [4, 5]. Utilizing the recently developed low-dose/low-volume infection model, we are able to determine factors that are required for the natural infection and transmission processes. For example, a strain containing a mutation in arnt failed to show any deficiency in the highdose/high-volume mouse model. However, upon examination in the low-dose/low-volume model

94 85 was found to be deficient in transmission, due to an increased mean infectious dose [6]. Previously characterized mutants of B. bronchiseptica were screened for their ability to transmit within our TLR-4 deficient low-dose model. The outcomes of those experiments are shown in Table 1. In chapters 2 and 3, we focused on two factors found to be defective in transmission, the T3SS and FHA, and determined how they contribute to the transmission process of the Bordetella. Strain RB50 RB50 RB50 RB50 RBx9 RB50 RB50 RB50 RB50 RB50 WT ΔbscN ΔcyaA ΔarnT ΔfhaB ΔclpV Δdnt Δprn ΔFim ΔWbm Transmissible Between Tlr-4 deficient mice Growth within Tlr-4 deficient mice Persistence in Tlr-4 deficient mice Table 1. The outcome of transmission growth and persistence in the C3/HeJ mouse model for several B. bronchiseptica mutants. (table adapted from the Thesis of Olivier Rolin entitled How Interactions Between Host Immunity and Bacterial Virulence Genes Affect Transmission of Bordetella Bronchiseptica ) A mutant in the T3SS was found to be defective in transmission, although no defect was observed in components important for transmission: shedding, experiment colonization, and growth. Previous observations in our lab have suggested that displacement of nasal cavity microbiota occurs during B. bronchiseptica infection, and is T3SS dependent in wild-type mice (Weyrich, LS, et al., unpublished data). It was therefore hypothesized that the failure to transmit likely stemmed from an inability to overcome colonization resistance engendered by the microbiota upon entry into a new host. Indeed, mice infected with a B. bronchiseptica mutant in bscn showed an inability to compete with culturable microbiota compared to its wild-type

95 86 counterpart. Finally, we were able to show that depletion of host microbiota in susceptible mice with antibiotics allowed RB50bscN to transmit. These results showed that the T3SS is not only important for host pathology, but also for competition between B. bronchiseptica and microbiota of the nasal cavity by some yet unknown mechanism. Pathogens must overcome multiple barriers to successfully colonize new hosts, one of them being the host microbiota. Recently it was shown that the host microbiota attribute to the host specificity of B. pertussis [7]. Mice treated with antibiotics prior to infection with B. pertussis required 100 fold less bacteria to establish infection in mice. B. pertussis, while unable to complete with mouse microbiota isolates, was capable of competing with microbiota isolated from human individuals [7]. Likewise, antibiotic treatment can cause adverse effects by allowing opportunistic pathogens to grow unhindered [8]. Today, fecal transplants are often used in severe diarrheal diseases cases as a successful treatment [9]. Isolation of host microbiota species that are effective at competing with Bordetella may lead to the development of probiotic strategies for enhancing resistance to infection. Similarly, targeting toxins or bacteriocins that contribute to competition may represent important therapeutic targets for preventing transmission. Finally, a strain defective in the production of FHA was found to be unable to transmit. FHA is an important adhesion factor required for the successful colonization of B. pertussis and B. bronchiseptica in the lower respiratory tract [10, 11]. More importantly, FHA can manipulate the host response and is a potent stimulator of the immune response. FHA causes pathology by inducing proinflammatory pathways and directing the immune response towards other antigens contributing to the persistence of the Bordetella [12, 13]. Here, we determined that the lack of transmission stemmed from an inability to stimulate a robust immune response. Despite being

96 87 an important adherence factor, mice inoculated with RB50 fhab were able to colonize and grow within the nasal cavity of mice, but failed to recruit neutrophils to the site of infection. Controlling the inflammation caused by the particular virulence factors of Bordetella may be an important tool in limiting transmission from infected hosts. The use of anti-inflammatory drugs, in conjunction with antibiotic therapies, may reduce the likelihood of transmission to individuals in the same household than antibiotic treatments alone. Targeting Adaptive Immunity to Control B. bronchiseptica Transmission The discovery of vaccination was a key turning point in the fight against infectious disease. Vaccination has led to the eradication of smallpox by breaking the chain of transmission. B. pertussis and B. parapertussis cause transient disease, and no known environmental or animal reservoir exists [14]. Vaccination, therefore, should be able to break the chain of transmission; however, current vaccines fail to interrupt transmission. Despite high vaccine coverage, whooping cough remains significant and periodic epidemics are becoming an increasing problem [15-21]. Waning immunity and the immune response generated by current vaccines likely contribute to the prolonged asymptomatic colonization of B. pertussis [22-24]. Furthermore, current vaccines only protect against severe disease, allowing older vaccinated individuals to harbor B. pertussis and act as unknowing reservoirs for the transmission of whooping cough [25-27]. One study determined that in 60% of infant cases, the source of transmission was found to be siblings and parents [28, 29]. In chapter 4, we focused on the effects of vaccination on transmission of the Bordetella in a recently-characterized mouse model. While growth of the pathogen within the nasal cavity of mice was unaffected by whole-cell vaccination, shedding from the external nares was significantly reduced in whole-cell vaccinated mice. Furthermore, recruitment of B cells or

97 88 neutrophils to the site was found to influence shedding in unvaccinated and vaccinated individuals. Neutrophil recruitment has previously been shown to contribute to the expulsion of bacteria from infected host [30]. While antibodies were sufficient in limiting shedding, we can only hypothesize a mechanism, as antibodies fail to reduce nasal cavity numbers. B cells and antibodies likely reduce inflammation that contributes to symptomology, such as mucous secretion, sneezing, and coughing [31, 32]. Previous studies show that adaptive immune responses can enhance the innate immune response upon encountering the pathogen for a second time. During secondary influenza, infections memory CD4+ T cells accelerate the kinetics and enhance the intensity of innate immune cytokine and chemokine production, leading to rapid control of the infection [33]. Likewise, we found that the innate cytokines IL-6, TNF-α, and INF-γ to be important during vaccine-induced control of shedding. Furthermore, we showed that the CD4+ cells are an important component in controlling shedding from vaccinated animals. Finally, we determined differences between whole-cell and Adacel, an acellular pertussis vaccine, vaccination that may explain the increased incidence of B. pertussis cases. Acellular vaccination failed to limit shedding and transmission in our model system. Whole-cell and acellular pertussis vaccines elicit two distinct types of immune responses [34], potentially explaining the discrepancies in transmission highlighted here. The Th1 and Th17 immune memory responses, produced during primary infection, are particularly important in combating B. pertussis in multiple models [35, 36]. While whole-cell pertussis vaccines mimic the immune response of primary infection creating Th1 and Th17 immune memory, acellular vaccines create a largely Th2 immune memory response. These results could explain the observed increase in whooping cough cases since the transition to acellular vaccines.

98 89 Conclusions Shedding can be enhanced by the presentation of disease symptoms [37]. Therefore, the evolution of virulence factors can have an important role in the transmission success of pathogens. The evolution of these factors must be balanced, however, as not to decrease host fitness and their contact network. The hospital setting has allowed for evolution to occur in a unique setting. Virulence of pathogens found in the hospital setting is often increased and can be attributed to the increased contact networks found there [38, 39]. Understanding bacterial factors that contribute to the transmission of infectious diseases is important in combating the mortality and morbidity attributed to them. Here we have provided insights to only a few Bordetella virulence factors. Continued studied of Bordetella transmission will help determine conditions, such as population density, host immune status, and host fitness that contribute to the evolution of the Bordetella. Additionally, the use of the model discussed here will allow for further investigation as to the differences between these two types of vaccination. For example, acellular vaccination may limit the overall disease within the host, but fail to limit neutrophil recruitment to the nasal cavity, leading to asymptomatic carriage and shedding of the pathogen. Furthermore, this model will allow for rapid testing of novel routes of delivery, choices in adjuvant, and antigen components that influence the strength, longevity, and anatomical distribution. Currently, particular interest resides in mucosal immunizations for respiratory infections [40]. Mucosal immunization result in long-lived T lymphocytes at the site of immunization [41], and increase production of IgG and IgA isotypes at the site of immunization [42]. To combat the reemergence of whooping cough, novel therapeutic strategies are in development, including novel

99 90 adjuvants, live-attenuated intranasal vaccination, and use of purified non-endotoxic outer membrane vesicles [43-46]. Current vaccines are sufficient at protecting individuals from disease, and therefore re-emergence is the result of incomplete herd immunity. Our transmission model will contribute to the development of novel therapeutics that not only reduce disease, but transmission of Bordetella.

100 91 References 1. Goodnow RA. Biology of Bordetella bronchiseptica. Microbiological reviews 1980; 44: Rolin O, Smallridge W, Henry M, Goodfield L, Place D, Harvill ET. Toll-Like Receptor 4 Limits Transmission of Bordetella bronchiseptica. PLoS One 2014; 9:e Smirnova I, Mann N, Dols A, et al. Assay of locus-specific genetic load implicates rare Tolllike receptor 4 mutations in meningococcal susceptibility. Proceedings of the National Academy of Sciences of the United States of America 2003; 100: Mann PB, Wolfe D, Latz E, Golenbock D, Preston A, Harvill ET. Comparative toll-like receptor 4-mediated innate host defense to Bordetella infection. Infect Immun 2005; 73: Wolfe DN, Buboltz AM, Harvill ET. Inefficient Toll-like receptor-4 stimulation enables Bordetella parapertussis to avoid host immunity. PLoS One 2009; 4:e Rolin O, Muse SJ, Safi C, et al. Enzymatic Modification of Lipid A by ArnT Protects Bordetella bronchiseptica against Cationic Peptides and Is Required for Transmission. Infect Immun 2014; 82: Weyrich LS, Feaga HA, Park J, et al. Resident Microbiota Affect Bordetella pertussis Infectious Dose and Host Specificity. The Journal of infectious diseases Wilcox MH, Cunniffe JG, Trundle C, Redpath C. Financial burden of hospital-acquired Clostridium difficile infection. The Journal of hospital infection 1996; 34: Gough E, Shaikh H, Manges AR. Systematic review of intestinal microbiota transplantation (fecal bacteriotherapy) for recurrent Clostridium difficile infection. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2011; 53: Saukkonen K, Cabellos C, Burroughs M, Prasad S, Tuomanen E. Integrin-mediated localization of Bordetella pertussis within macrophages: role in pulmonary colonization. The Journal of experimental medicine 1991; 173: Cotter PA, Yuk MH, Mattoo S, et al. Filamentous hemagglutinin of Bordetella bronchiseptica is required for efficient establishment of tracheal colonization. Infect Immun 1998; 66: Ishibashi Y, Nishikawa A. Role of nuclear factor-kappa B in the regulation of intercellular adhesion molecule 1 after infection of human bronchial epithelial cells by Bordetella pertussis. Microbial pathogenesis 2003; 35: Abramson T, Kedem H, Relman DA. Proinflammatory and proapoptotic activities associated with Bordetella pertussis filamentous hemagglutinin. Infect Immun 2001; 69:

101 Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Allen A. Public health. The pertussis paradox. Science 2013; 341: Crowcroft NS, Stein C, Duclos P, Birmingham M. How best to estimate the global burden of pertussis? Lancet Infect Dis 2003; 3: Crowcroft NS, Booy R, Harrison T, et al. Severe and unrecognised: pertussis in UK infants. Arch Dis Child 2003; 88: Guris D, Strebel PM, Bardenheier B, et al. Changing epidemiology of pertussis in the United States: increasing reported incidence among adolescents and adults, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 1999; 28: de Greeff SC, de Melker HE, van Gageldonk PG, et al. Seroprevalence of pertussis in The Netherlands: evidence for increased circulation of Bordetella pertussis. PLoS One 2010; 5:e Hellenbrand W, Beier D, Jensen E, et al. The epidemiology of pertussis in Germany: past and present. BMC infectious diseases 2009; 9: Quinn HE, McIntyre PB. Pertussis epidemiology in Australia over the decade trends by region and age group. Communicable diseases intelligence quarterly report 2007; 31: Lavine J, Bjørnstad O, de Blasio B, Storsaeter J. Short-lived immunity against pertussis, agespecific routes of transmission, and the utility of a teenage booster vaccine. Vaccine 2012; 30: Lavine J, Broutin H, Harvill ET, Bjornstad ON. Imperfect vaccine-induced immunity and whooping cough transmission to infants. Vaccine 2010; 29: Higgs R, Higgins SC, Ross PJ, Mills KH. Immunity to the respiratory pathogen Bordetella pertussis. Mucosal immunology 2012; 5: Mooi FR. Bordetella pertussis and vaccination: the persistence of a genetically monomorphic pathogen. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases 2010; 10: Hewlett EL, Edwards KM. Clinical practice. Pertussis--not just for kids. The New England journal of medicine 2005; 352: Strebel P, Nordin J, Edwards K, et al. Population-based incidence of pertussis among adolescents and adults, Minnesota, The Journal of infectious diseases 2001; 183:

102 Raymond J, Armengaud JB, Cosnes-Lambe C, et al. Pertussis in young infants: apnoea and intra-familial infection. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 2007; 13: de Greeff SC, Mooi FR, Westerhof A, et al. Pertussis disease burden in the household: how to protect young infants. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2010; 50: Gopinath S, Hotson A, Johns J, Nolan G, Monack D. The systemic immune state of supershedder mice is characterized by a unique neutrophil-dependent blunting of TH1 responses. PLoS pathogens 2013; 9:e Russell MW, Reinholdt J, Kilian M. Anti-inflammatory activity of human IgA antibodies and their Fab alpha fragments: inhibition of IgG-mediated complement activation. European journal of immunology 1989; 19: van der Neut Kolfschoten M, Schuurman J, Losen M, et al. Anti-inflammatory activity of human IgG4 antibodies by dynamic Fab arm exchange. Science 2007; 317: Strutt TM, McKinstry KK, Dibble JP, et al. Memory CD4+ T cells induce innate responses independently of pathogen. Nature medicine 2010; 16:558-64, 1p following Ryan M, Murphy G, Ryan E, et al. Distinct T-cell subtypes induced with whole cell and acellular pertussis vaccines in children. Immunology 1998; 93: Ross PJ, Sutton CE, Higgins S, et al. Relative contribution of Th1 and Th17 cells in adaptive immunity to Bordetella pertussis: towards the rational design of an improved acellular pertussis vaccine. PLoS pathogens 2013; 9:e Warfel JM, Merkel TJ. Bordetella pertussis infection induces a mucosal IL-17 response and long-lived Th17 and Th1 immune memory cells in nonhuman primates. Mucosal immunology Drexler JF, Baumgarte S, de Souza Luna LK, Eschbach-Bludau M, Lukashev AN, Drosten C. Aichi virus shedding in high concentrations in patients with acute diarrhea. Emerging infectious diseases 2011; 17: Ziebuhr W, Hennig S, Eckart M, Kranzler H, Batzilla C, Kozitskaya S. Nosocomial infections by Staphylococcus epidermidis: how a commensal bacterium turns into a pathogen. International journal of antimicrobial agents 2006; 28 Suppl 1:S Ewald PW. Transmission modes and the evolution of virulence : With special reference to cholera, influenza, and AIDS. Human nature 1991; 2: Holmgren J, Czerkinsky C. Mucosal immunity and vaccines. Nature medicine 2005; 11:S45-53.

103 Stagg AJ, Kamm MA, Knight SC. Intestinal dendritic cells increase T cell expression of alpha4beta7 integrin. European journal of immunology 2002; 32: Rudin A, Johansson EL, Bergquist C, Holmgren J. Differential kinetics and distribution of antibodies in serum and nasal and vaginal secretions after nasal and oral vaccination of humans. Infect Immun 1998; 66: Kindrachuk J, Jenssen H, Elliott M, et al. A novel vaccine adjuvant comprised of a synthetic innate defence regulator peptide and CpG oligonucleotide links innate and adaptive immunity. Vaccine 2009; 27: Asensio CJ, Gaillard ME, Moreno G, et al. Outer membrane vesicles obtained from Bordetella pertussis Tohama expressing the lipid A deacylase PagL as a novel acellular vaccine candidate. Vaccine 2011; 29: Mielcarek N, Debrie AS, Mahieux S, Locht C. Dose response of attenuated Bordetella pertussis BPZE1-induced protection in mice. Clinical and vaccine immunology : CVI 2010; 17: Mielcarek N, Debrie AS, Raze D, et al. Live attenuated B. pertussis as a single-dose nasal vaccine against whooping cough. PLoS pathogens 2006; 2:e65.

104 95 Appendix A: Analysis of transmission within the Classical Bordetella

105 96 Bvg regulation and bacterial factors. Sensing and responding to changes in the environment is important not only for successful survival, but is essential for the transmission of bacterial pathogens. Pathogens have distinct phases during infection that allow for adaptation to changes within the host. During Salmonella enterica serovar Typhimurium infection, mice that have a lapse in the ability to dampened inflammation become super shedders [1]. Sensing changes, such as the host immune response, may help initiate transmission of pathogens, increasing disease burden. The Bvg regulatory system of the Bordetella has been described in chapter 1. In particular the Bvg - phase is important for life outside the host, while the Bvg + phase is Figure A.1 Modulation of the Bvg system in Bordetella is important for efficient shedding and transmission. (A) Groups of 4 mice were challenged as indicated and shedding of B. bronchiseptica was detected throughout the infectious course. Symbols represent the mean CFU +1 of B. bronchiseptica (± SEM). (B) Transmission of RB50, RB53, RB50i to susceptible individuals after co-housing with index mice. Dots represent the log 10 CFU of bacteria recovered from individual mice required for virulence [2]. A preliminary study comparing RB50 and RB53, a Bvg + phase locked mutant, shedding in wild-type mice found that individuals infected with RB53 shed more over the time course than those infected with RB50 (Data not shown). This supports the hypothesis that inflammation is critical for Bordetella shedding. To further study the importance of the BvgAS regulator system during

106 97 Bordetella transmission, the C3/HeJ mouse model of transmission was used. Index mice were challenged with RB50, RB53, or RB50i, a Bvg i phase locked mutant, and placed in cages containing 3 susceptible mice for three weeks. On Day 21, index and susceptible mice were sacrificed and their nasal cavities were cultured to determine transmission. No major difference in shedding from RB50, RB53, or RB50i index mice was observed during this experiment. RB50, RB53, and RB50i index mice were capable of shedding 100 s to 1000 s of bacteria during the peak of shedding (Figure A.1A). Interestingly, transmission to susceptible mice from RB53 and RB50i was less than that observed for RB50. RB50 index mice readily transmitted to 7 out of 9 exposed cage mates within 21 days, while Both RB53 and RB50i index mice transmitted to 4 out of 9 exposed cage mates respectively (Figure A.1B). If shedding is indeed increased as a result of increased virulence, one might speculate that the increase in shedding would result in increased transmission rates. A Bvg + phase locked mutant was unable to transmit as effectively as the wild-type strain. Likewise, a Bvg i phase lock mutant was unable to transmit as effectively as the wild-type strain. These results suggest that modulation is important during infection and may be important to the transmission process of the Bordetella. Shedding from a host is just one step in the transmission process; bacteria must survive in the media between hosts. Bacteria that are not able to transition to conditions outside the host may not survive in sufficient numbers to reach the next host. The Bordetella produce numerous virulence factors. In order to determine their effect on transmission, a panel of already characterized B. bronchispetica mutants were assayed on their ability to transmit to susceptible hosts (Table 1). Mutants in secretion systems toxins, adhesion factors, and O-antigen all fail to transmit in the C3/HeJ mouse model of Bordetella transmission. No defections were detected in growth and persistence within C3/HeJ mice. A select subset was further assayed to determine

107 98 how these factors contributed to the transmission success of B. bronchiseptica. The T3SS is required to compete with the host microbiota during the initial colonization phase and is described in chapter 2 of this dissertation. FHA is required for the recruitment of inflammatory cells to the site of infection, enabling shedding from infected individuals and is discussed in chapter 3 of this dissertation. This model system has also determined a role for arnt in transmission [3]. The shedding and ID 50 of RB50 dnt and RB50 prn were also assayed and found to show no major defects in these transmission parameters (Figure A.2). 1.4 Log10+CFU+per+second RB RB50 dnt Days+Post+Challenge+ Log10+CFU+per+second RB50 RB50 prn Days+Post+Challenge+ Figure A.2. Shedding of RB50 dnt and RB50 prn. Groups of 4 mice were challenged as indicated and shedding was detected throughout the timecourse.

108 99 Genetic diversity of B. bronchiseptica contribute to various transmission phenotypes Similarly, different Bordetella isolates of varying virulence and immunogenicity have been characterized. Sequence analysis of Bordetella will and have provided genetic mechanisms for observed phenotypes. Four sequence strains of B. bronchiseptica were assayed for their ability to shed into the environment from C57/B6 mice (Figure A.3). Strain RB50 was originally isolated from the nares of a New Zealand white rabbit and has been highly studied in the laboratory [4]. Strain 1289 is a hyper-virulent strain that was isolated from the thoracic cavity of a monkey [5] Figure A.3 Shedding of 4 genetically distinct B. bronchiseptica isolates Groups of 4 mice were challenged as indicated and shedding was detected throughout the timecourse. and other B. bronchispetica strains found in the same lineage are hyper-virulent due to an increase in the expression of genes in the

109 100 T3SS [5]. Meanwhile, another lineage of B. bronchiseptica containing 253 is less-virulent, due to replacement of the cya locus with an operon predicted to encode peptide transport proteins [6]. CyaA is a calmodulin-activated, bifunctional adenylate cyclase/hemolysin that interferes with invasivity, phagocytosis, and chemotaxis. CyaA is also important in causing apoptosis of host cells. Absences in cyaa result in decreased pathology, lethality, and the colonization and persistence of B. bronchiseptica. Lastly, a strain isolate from a human patient, MO149, was found to be poorly immuno-gentic, due to a novel O antigen chain [7]. These changes have effects on factors that result in pathology, symptomology, and the outcome of disease within a host. However, the diversity within the B. bronchiseptica lineage has not been correlated with differences in factors contributing to transmission success. To determine whether shedding was affected by the genetic changes observed in each of these lineages, C57/B6 mice were challenged with 150 CFU in 5ul of PBS and shedding was monitored throughout the infectious course (Figure A.3). Using RB50 as a baseline, shedding of 1289, 253, and MO149 was determined. Not surprisingly, 1289 was able to shed at higher levels during the observed time course. This increase in shedding is likely a product of increased inflammation in the nares of infected mice due to the hyper activity of the T3SS. One would expect that there will be an increase in the neutrophil infiltrate of these mice, as their presence has previously been correlated with shedding of B. bronchiseptica; however, further analysis needs to occur to definitively define the mechanism for the increase in shedding seen from 1289 infected mice. Despite the decreased lethality, 253 appeared to have no effect on shedding, as compared to RB50. Support for this finding is revealed in our transmission model, in which mice inoculated with RB50 cyaa are still capable of transmitting to naïve susceptible mice (Table 1). It is possible that CyaA is not required to cause pathology within the nasal cavity of mice. Hester et

110 101 al also show that there is increased cyaa expression in the presences of 5% CO2 [8]. The concentration of CO2 gases likely increase, proceeding down the respiratory tract, and therefore expression of cyaa is likely decreased comparably in the nasal cavity. These results suggest that cyaa is not required for transmission to occur in our mouse model. Finally, MO149 induced decreased shedding when compared to RB50. Immunization with MO149 s poorly immunogenic O-antigen results in little protection against MO149 colonization [9]. The lack of immune stimulation by MO149 s O antigen may be able to explain why it is unable to induce shedding from mice. As a human-isolated strain, MO149 has experienced more acquisitions of transposons as compared to non-human isolates like RB50 [10]. The adaptation of B. bronchiseptica to infect human host may also be playing a role in the lack of shedding. Finally, MO149 is also missing DNT. DNT has dermonecrotic activities, and is thought to be responsible for turbinate atrophy in swine atrophic rhinitis. Mutants in DNT fail to transmit in our mouse model of transmission (Table 1); however, C57/B6 mice inoculated with RB50 dnt are capable of shedding. Taken together, these results show numerous factors that could and can effect transmission of the Bordetella. These preliminary findings will provide a framework to understand Bordetella evolution in the context of transmission. B. pertussis and B. parapertussis shedding The mouse model provides an Log10+CFU+per+Second B.pertussis B.pertussisBaytril B.parapertussis Days+Post+Challenge+ excellent system to study transmission of the Bordetella. While insights can be inferred Figure A.4 Shedding of B. pertussis and B. parapertussis from C3/HeJ mice Groups of 4 mice were challenged as indicated and shedding of was detected throughout the infectious course

111 102 from B. bronchiseptica infection and transmission, the study of transmission parameters for the human adapted pathogens B. pertussis and B. parapertussis in this model would allow for direct evidence for improved strategies in disease caused by B. pertussis and B. parapertussis. Shedding of B. parapertussis was observed in C3/HeJ mice. Mice were inoculated with 500 CFU of B. parapertussis in 5 ul of PBS. These findings indicate that future studies can probe the factors that are important in the shedding of B. parapertussis. Shedding of B. pertussis, however, was unable to be detected when 10,000 CFU in 5 ul was used to inoculated mice. Depletion of host microbiota prior to infection of B. pertussis also failed to induce shedding. Therefore, the study of B. pertussis transmission may only be able to be studied from clinical data and the recent baboon model. Because many of the mechanisms overlap between the classical Bordetella, B. bronchiseptica transmission studies can guide baboon studies to determine factors required for B. pertussis transmission (Figure A.4)

112 103 References 1. Gopinath S, Hotson A, Johns J, Nolan G, Monack D. The systemic immune state of supershedder mice is characterized by a unique neutrophil-dependent blunting of TH1 responses. PLoS pathogens 2013; 9:e Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Rolin O, Muse SJ, Safi C, et al. Enzymatic Modification of the Lipid A by an ArnT Protects B. bronchiseptica Against Cationic Peptides and Is Required for Transmission. Infect Immun Cotter PA, Miller JF. BvgAS-mediated signal transduction: analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in a rabbit model. Infect Immun 1994; 62: Buboltz AM, Nicholson TL, Weyrich LS, Harvill ET. Role of the type III secretion system in a hypervirulent lineage of Bordetella bronchiseptica. Infect Immun 2009; 77: Buboltz AM, Nicholson TL, Parette MR, Hester SE, Parkhill J, Harvill ET. Replacement of adenylate cyclase toxin in a lineage of Bordetella bronchiseptica. Journal of bacteriology 2008; 190: Vinogradov E, King JD, Pathak AK, Harvill ET, Preston A. Antigenic Variation among Bordetella: Bordetella bronchiseptica strain MO149 expresses a novel o chain that is poorly immunogenic. The Journal of biological chemistry 2010; 285: Hester SE, Lui M, Nicholson T, Nowacki D, Harvill ET. Identification of a CO2 responsive regulon in Bordetella. PLoS One 2012; 7:e Hester SE, Park J, Goodfield LL, Feaga HA, Preston A, Harvill ET. Horizontally acquired divergent O-antigen contributes to escape from cross-immunity in the classical bordetellae. BMC evolutionary biology 2013; 13: Park J, Zhang Y, Buboltz AM, et al. Comparative genomics of the classical Bordetella subspecies: the evolution and exchange of virulence-associated diversity amongst closely related pathogens. BMC genomics 2012; 13:545.

113 104 Appendix B: Analysis of B. bronchiseptica strain 980

114 105 Abstract The respiratory mucosa offers a unique environmental challenge for bacteria, as the microenvironments can dramatically change between the upper and lower respiratory tracts. B. bronchiseptica can establish persistent infections within the upper respiratory tract of animals without causing disease symptoms and act as unknowing reservoirs for the spread of disease. Here, we compare two phenotypical distinct B. bronchiseptica isolates for their ability to form biofilms. In vitro analysis revealed complex differences in biofilm formation between strain RB50 and strain 980. The increased biofilm formation may be linked to the secretion of a biofilm-promoting molecule as supernatant from 980 culture s increase biofilm production in RB50. Furthermore, we show that while defective in cytotoxicity of RAW macrophages and colonization in the lower respiratory tract, 980 is successful at colonization of the upper respiratory tract of mice. These results support previous findings that biofilm formation is essential for long-term colonization of the upper respiratory tract. Future genomic and transcriptomic analysis will further the understanding of biofilm formation of the Bordetella during infection.

115 106 Introduction Different lineages of bacteria display a wide range of phenotypes. The most studied are those that cause varied disease states within their host s. Recently, genome-wide analyses, phylogenetics, mutational analysis, and host infection models have identified factors that alter the severity of disease, as well as other phenotypes. Bacteria belonging to the classical Bordetella, B. bronchiseptica, B. pertussis, and B. parapertussis, cause respiratory tract infections in their respective hosts [1]. In particular, B. bronchiseptica results in kennel cough in dogs, and atrophic rhinitis and pneumonia in pigs [1]. Individuals infected with B. bronchiseptica often become permanently colonized, while showing no signs of disease [2-4]. Despite vaccination of animals, B. bronchiseptica still is able to circulate within immune populations [5]. Biofilm formation in the nasal cavity is likely to contribute to B. bronchiseptica s ability to cause long-term, to life-long, asymptomatic carriage and evade vaccine induced immunity. Biofilms are highly structured communities of cells and are the predominant form in which bacteria are found. Formation of biofilms results in increased resistance to components of the immune system, as well as to antibiotic therapies [6, 7]. Biofilm formation within the nasal cavity likely leads to prolonged colonization through production of macromolecules, including polysaccharides, nucleic acids and proteins. Biofilm formation within the Bordetella has been demonstrated with B. bronchiseptica, B. pertussis, and B. parapertussis [8]. Formation of a biofilm in Bordetella is BvgAS dependent. The Bvg regulatory system in Bordetella regulates virulence genes in response to environmental cues. During the virulent (Bvg + ) phase, genes required for pathogenesis are expressed, while genes for motility and life outside the host are active during the non-virulent (Bvg phase)[1].

116 107 Interesting studies show that biofilm formation within the Bordetella occurs under Bvg+ phase condition, indicating that biofilm formation is an important component during the infectious process [8, 9]. In particular, expression of flagellar genes, a Bvg- phase phenotype, is critical for initial adhesion; however, prolonged production of flagellar genes is detrimental for robust biofilm production [10]. Furthermore, the polysaccharide locus, bpsabcd, is essential for mature biofilm formation in the Bordetella, both in vitro and in vivo [11-14]. In B. pertussis this locus aids in the colonization of both the nose and lungs of mice by enhancing adhesion and protecting B. pertussis from complement-mediated killing [15]. While correlation between the expression of virulence factors and disease have been made for particular Bordetella lineages, limited studies have focused on other factors that contribute to successful infection. Here we examine phenotypic differences between two isolates of B. bronchiseptica, strain RB50 and strain 980. RB50 is the prototypical lab strain that was isolated from the nares of a New Zealand white rabbit. 980 is a clinical isolate and is capable of forming a robust biofilm. Confocal microscopy revealed the formation of a thick, complex biofilm containing pillars of bacteria. Analysis in vivo revealed that 980 was defective in colonization of the lower respiratory tract, likely due to a defect in cytotoxicity. The defect in cytotoxicity, however, did not affect colonization of the upper respiratory tract, suggesting that biofilm formation is important for adaptation for life in the nasal cavity. Further analysis at the genomic and transcriptive level will likely reveal novel factors of Bordetella biofilm formation.

117 108 Materials and Methods Ethics Statement. This study was carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee at The Pennsylvania State University (#40029). All animals were anesthetized using 5% isoflourane in oxygen and euthanized using carbon dioxide inhalation to minimize animal suffering. Bacterial Strains. B. bronchiseptica strain RB50 has been previously described [16]. 980 was a clinical isolate of B. bronchiseptica generously given by the CDC. Bacteria were maintained on Bordet-Gengou agar (Difco) supplemented with 10% defibrinated sheep blood (Hema Resourses) and 20 µg/ml streptomycin (Sigma-Aldrich). Liquid cultures were grown overnight in Stainer-Scholte (SS) broth at 37 C to mid-log phase. Animal Studies Four- to six-week C57/B6 mice were obtained from The Jackson Laboratories and maintained in our specific pathogen-free facility. Inoculations were prepared from liquid cultures grown to mid-log phase and bacteria were diluted to a concentration of 1x10 7 CFU/mL. Mice were given 5x10 5 CFU in 50 µl of PBS that was deposited on the external nares. To quantify bacteria, organs were excised on the day indicated. Following dissection, organs of interest were harvested and homogenized in 1mL PBS and diluted appropriately for culture and quantification on Bordet-Gengou agar or Blood agar. Quantification of biofilm formation. Biofilm formation was monitored by adaptation of a microtiter plate assay described previously. 12-well plates were inoculated with 5x10 6 CFU either RB50 or 980 and grown in a total volume of 2mL of SS broth at 37 C for 10 hours. Cells that remained adhered to the wells were stained with crystal violet (CV) and were incubated at room temperature for 15 minutes and washed thoroughly with water. The CV staining the cells

118 109 was solubilized with 1 ml of 95% ethanol. 200ul of ethanol was transferred to a 96-well plate and the absorbance determined at 545 nm. Confocal microscopy was used to determine biofilm architecture. 50 ml conical tubes containing glass slides were inoculated with 5x10 6 CFU either RB50 or 980 in 10 ml of SS and incubated at 37 C for 24 hours. Glass slides were then washed thoroughly and stained with filmtracer green biofilm. Biofilms were visualized using an Olympus fluoview 1000 confocal scanning laser microscope. For displaying biofilm images, the PerkinElmer Volocity 3D Image Analysis Software was used. Cytotoxicity assay. RAW cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Difco) supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, 1% nonessential amino acids, and 1% sodium pyruvate. The cells were grown to 75% confluency in 5% CO 2 at 37 C in 96-well plates. DMEM was then replaced with RPMI medium lacking phenol red, with 2% fetal bovine serum, 1% L-glutamine, 1% nonessential amino acids, and 1% sodium pyruvate at least one hour prior. Bacteria diluted in RPMI at multiplicities of infection (MOI) of 10 and 100 were centrifuged onto the macrophages at 300 g for 5 minutes and incubated in 5% CO 2 at 37 C for the time indicated. The cell culture supernatants were collected, and lactate dehydrogenase (LDH) release was analyzed using a Cytotox96 kit (Promega).

119 110 Results B. bronchiseptica strain 980 form a robust biofilm and potential secrets a biofilm promoting molecule. RB50 Figure B.1 B. bronchiseptica strain 980 form a more robust biofilm than strain RB50 (A) A comparison of 24 h biofilms made by B. bronchiseptica strain RB50 and 980. Bars represent the mean OD 545 of biofilmassociated crystal violet. (B) Confocal microscopy of biofilm formation by RB50 and 980 To quantify observed differences in the biofilm formation between RB50 and 980, a Crystal Violet Binding Assay was used. 980 showed a significant increase in biofilm formation. Culture on Bordet-Gengou agar reveal >99% of cells were incorporated into the biofilm of 980 (data not shown). Furthermore, 10 fold more bacteria were attached to the glass test tube of 980 cultures compared to RB50 when started at similar concentrations and grown for 20 hours. Using confocal microscopy to determine whether the 3-demensional architecture of these two strains differ, we show that biofilm formation is drastically different between RB50 and formed a biofilm that was 5 times thicker than RB50. Most notable, 980 formed complex pillars which are known to contribute to enhance nutrient acquisition of the biofilm community by increasing the surface area exposed to the environment. Biofilm formation is often the result 980

120 111 Figure B.2 Potential secretion a biofilm promoting molecule by B. bronchiseptica strain 980 (A) A comparison of 10 h biofilms made by B. bronchiseptica strain 980 grown in the indicated media. (B) A comparison of 10 h biofilms made by B. bronchiseptica strain RB50 grown in the indicated media. of communication between bacteria, known as quorum sensing. Quorum sensing molecules released by bacteria within a population can coordinate the gene expression leading to biofilm formation, for example. To determine if 980 was secreting a biofilm-promoting molecule, RB50 and 980 were grown in Stainer-Scholte (SS), 1:1 Stainer-Scholte to RB50 supernatant, or 1:1 Stainer-Scholte to 980 supernatant. Filtered supernatant was collected from overnight cultures of RB50 or 980. Growth of RB50 or 980 in Stainer-Scholte was similar to previous observations. Interestingly, RB50 was observed to form a more robust biofilm in the presence of 980 supernatant. Growth in 1:1 Stainer-Scholte to RB50 supernatant had no observed effect on biofilm formation of either RB50 or 980, as compared to growth in Stainer-Scholte, indicating that RB50 is not secreting a biofilm-inhibitory molecule. Interestingly, biofilm formation of RB50 was observed to be increased in the presence of 980 supernatant. These results show

121 112 Figure B.3 In vivo and in vitro analysis of B. bronchiseptica strain 980 Colonization of RB50 verses 980 in C57BL/6 mice at an inoculation dose of CFU in 50 µl in the nasal cavity (A), trachea (B), and lung (C). (D) LDH release assay monitoring cytotoxicity of RAW264.7 macrophages at an MOI of 10 or 100 for 2, 4, and 6 hour incubations with RB50 or 980 phenotypic difference between two strains of B. bronchiseptica. Furthermore, these results are the first to suggest that quorum sensing is an integral part of Bordetella biofilm formation. In vivo and in vitro analysis of 980 To determine whether strain 980 is capable of causing disease in mice, groups of mice were inoculated with 5x10 5 CFU of RB50 or 980 and bacterial numbers were determined in the nasal cavity, trachea, and lungs at days 3, 7, 14, and 28 post-inoculation. 980 colonized the respiratory tract similarly to RB50 on day three. However, by day 7, 980 numbers in the lungs and trachea were significantly lower than RB50. Cytotoxicity of RAW264.7 macrophages by

122 113 strain RB50 or 980 was determined by a lactate dehydrogenase (LDH) release assay. As previously observed, RB50 was capable of causing significant cytotoxicity by 6 hours [17]. Minimal cytotoxicity was observed in RAW cells exposed to a MOI of 10 or 100 of strain 980. The decrease in cytotoxicity observed for strain 980 is likely contributing to defects in lower respiratory tract colonization. Interestingly, we found no defect in the nasal cavity colonization of strain 980. These results indicate that strain 980 is hypo-virulent in the mouse model of infection.

123 114 Discussion The severity of a B. bronchiseptica infection can range from long-term asymptomatic carriage in the upper respiratory tract to severe disease. While virulence is important to the success of pathogens, other aspects also contribute. The ability to form biofilms during the infectious process can offer haven from the immune system, as well as antibiotic treatment. Here we describe an isolate that is characterized by robust biofilm formation, and despite being hypovirulent, able to successfully colonize the upper respiratory tract of mice. Biofilm formation in many species of bacteria, including P. aeruginosa and V. cholerae, are regulated by a phenomenon known as quorum sensing [18, 19]. Quorum sensing allows for the bacteria to act as a multi cellular unit [20]. Many gram-negative bacteria utilize Nacylhomoserine lactones for communication and can be species specific [21]. Another molecule, autoinducer 2, has blurred the lines between intraspecies quorum sensing and interspecies communication, allowing for communities between gram-negative and gram-positive bacteria to come together under one goal [20]. Variation in the concentrations of these molecules can lead to the activation of particular genes necessary for biofilm production. Quorum sensing mechanisms of the Bordetella have yet to be determined; however, data presented here and elsewhere suggest it does occur [8]. Studies of the major virulence factors of the Bordetella show their importance in colonization of the lower respiratory tract. Only mutations resulting in the bvg- phenotype s have been shown to be critical for upper respiratory tract colonization [16]. Because biofilm formation in the Bordetella is Bvg dependent, this suggests that a yet unknown quorum sensing mechanism may be controlled by the BvgAS system. Transmission is defined by the ability of an organism to spread to new hosts. Particular lineages of B. bronchiseptica have adapted to cause severe disease within its host, with one such

124 115 lineage exhibiting increased T3SS mediated virulence [22]. Other lineages have been shown to be hypo-virulent in the laboratory, despite isolation from diseased hosts [23]. The diversity in disease caused by differing lineages of B. bronchiseptica has suggested that some have adapted differing strategies for their success. The ability of virulence factors to cause pathology can maximize transmission by contributing to symptomology, such as coughing and sneezing. However, transmission is often dependent on the survival of the host; therefore, increased virulence could be detrimental to transmission. Likewise, the ability to cause long-term asymptomatic infection may enhance transmission to new hosts, as infected hosts act as unknowing vessels for the spread of disease. The results presented here only describe phenotypic differences between two isolates of B. bronchiseptica. Forward genetics, transcriptional and genomic analysis, will allow for identification of novel mechanisms that are involved in the formation of Bordetella biofilms and their role during the infectious process, as acquisition of novel genes, phage integration, phenotypic variation, gene loss, or mutation can contribute to observed phenotypic changes [24-26].

125 116 References 1. Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clinical microbiology reviews 2005; 18: Akerley BJ, Cotter PA, Miller JF. Ectopic expression of the flagellar regulon alters development of the Bordetella-host interaction. Cell 1995; 80: Kirimanjeswara GS, Mann PB, Harvill ET. Role of antibodies in immunity to Bordetella infections. Infect Immun 2003; 71: Goodnow RA. Biology of Bordetella bronchiseptica. Microbiological reviews 1980; 44: Bemis DA. Bordetella and Mycoplasma respiratory infections in dogs and cats. The Veterinary clinics of North America Small animal practice 1992; 22: Hall-Stoodley L, Stoodley P. Evolving concepts in biofilm infections. Cellular microbiology 2009; 11: Donlan RM, Costerton JW. Biofilms: survival mechanisms of clinically relevant microorganisms. Clinical microbiology reviews 2002; 15: Mishra M, Parise G, Jackson KD, Wozniak DJ, Deora R. The BvgAS signal transduction system regulates biofilm development in Bordetella. Journal of bacteriology 2005; 187: Irie Y, Mattoo S, Yuk MH. The Bvg virulence control system regulates biofilm formation in Bordetella bronchiseptica. Journal of bacteriology 2004; 186: Nicholson TL, Conover MS, Deora R. Transcriptome profiling reveals stage-specific production and requirement of flagella during biofilm development in Bordetella bronchiseptica. PLoS One 2012; 7:e Parise G, Mishra M, Itoh Y, Romeo T, Deora R. Role of a putative polysaccharide locus in Bordetella biofilm development. Journal of bacteriology 2007; 189: Sloan GP, Love CF, Sukumar N, Mishra M, Deora R. The Bordetella Bps polysaccharide is critical for biofilm development in the mouse respiratory tract. Journal of bacteriology 2007; 189: Conover MS, Sloan GP, Love CF, Sukumar N, Deora R. The Bps polysaccharide of Bordetella pertussis promotes colonization and biofilm formation in the nose by functioning as an adhesin. Molecular microbiology 2010; 77: Conover MS, Redfern CJ, Ganguly T, et al. BpsR modulates Bordetella biofilm formation by negatively regulating the expression of the Bps polysaccharide. Journal of bacteriology 2012; 194:

126 Ganguly T, Johnson JB, Kock ND, Parks GD, Deora R. The Bordetella pertussis Bps polysaccharide enhances lung colonization by conferring protection from complement-mediated killing. Cellular microbiology Cotter PA, Miller JF. BvgAS-mediated signal transduction: analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in a rabbit model. Infect Immun 1994; 62: Mann P, Goebel E, Barbarich J, Pilione M, Kennett M, Harvill E. Use of a genetically defined double mutant strain of Bordetella bronchiseptica lacking adenylate cyclase and type III secretion as a live vaccine. Infect Immun 2007; 75: Davies DG, Parsek MR, Pearson JP, Iglewski BH, Costerton JW, Greenberg EP. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 1998; 280: Hammer BK, Bassler BL. Quorum sensing controls biofilm formation in Vibrio cholerae. Molecular microbiology 2003; 50: Straight PD, Kolter R. Interspecies chemical communication in bacterial development. Annual review of microbiology 2009; 63: Bassler BL, Losick R. Bacterially speaking. Cell 2006; 125: Buboltz AM, Nicholson TL, Weyrich LS, Harvill ET. Role of the type III secretion system in a hypervirulent lineage of Bordetella bronchiseptica. Infect Immun 2009; 77: Buboltz AM, Nicholson TL, Parette MR, Hester SE, Parkhill J, Harvill ET. Replacement of adenylate cyclase toxin in a lineage of Bordetella bronchiseptica. Journal of bacteriology 2008; 190: Sumby P, Whitney AR, Graviss EA, DeLeo FR, Musser JM. Genome-wide analysis of group a streptococci reveals a mutation that modulates global phenotype and disease specificity. PLoS pathogens 2006; 2:e Park J, Zhang Y, Buboltz AM, et al. Comparative genomics of the classical Bordetella subspecies: the evolution and exchange of virulence-associated diversity amongst closely related pathogens. BMC genomics 2012; 13: Fitzgerald JR, Musser JM. Evolutionary genomics of pathogenic bacteria. Trends in microbiology 2001; 9:

127 118 Appendix C B. bronchiseptica specific antibody titer in the nasal cavity of whole-cell or shamvaccinated mice during B. bronchiseptica infection Nasal washes collected on days 3 and 7 after challenge from sham (open) or whole-cell (closed) vaccinated mice were assayed for anti-b. bronchiseptica antibodies by ELISA.

128 119 Appendix D Anti-B. bronchiseptica antibodies in serum. Total B. bronchiseptica specific antibodies 28 days after vaccination.

129 120 Appendix E Effect of vaccination on shedding by TLR4 deficient mice. Whole-cell and acellular vaccination showed similar effects on TLR4 deficient and wild type mice.